Learning an Occupation: Practices and Policies 9783112402214, 9783879974108

Table of contents :
Table of Content
Preface
Chapter I. The Art and Science of Training
Chapter II. Training Policies that Make Sense
Chapter III. The Latin American Panorama
Chapter IV. Multilateral Agencies: Understanding the Beasts
Monitoring Training Projects
References
Full Table of Content

Citation preview

Claudio de Moura Castro: Learning an Occupation

In Memoriam Klaus Schaack

A Note from the Publisher

The publication of this book is dedicated to the memory of Klaus Schaack, who passed away suddenly on September 14, 2010. Klaus had initiated, and was planning to edit, a new series of publications in Klaus Schwarz Verlag that are relevant to the realm of Technical and Vocational Education and Training issues. His own book Why Do German Companies Invest in Apprenticeships? (ISBN 978-3-87997-348-4) was the first contribution to this undertaking. Klaus was also the person who proposed the publication of the present volume, which integrates several of Claudio de Moura Castro's most important articles. Professor Castro is a highly experienced Brazilian expert in the field who has worked in leading positions for the ILO, the World Bank, the Interamerican Development Bank and other international and Brazilian institutions. I am very thankful to him for his close cooperation in the preparation of this volume. I would also like to express my special gratitude to Klaus' wife, EunSook Ahn, and to his close friend Arthur Grossmann for their generous financial assistance, which made the publication of this book possible. Gerd Winkelhane Executive Director Klaus Schwarz Verlag

Claudio de Moura Castro

Learning an Occupation - Practices and Policies -

KLAUS SCHWARZ VERLAG • BERLIN

Bibliographische Information der Deutschen Bibliothek Die Deutsche Bibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliographie; detaillierte bibliographische Daten sind im Internet über http://dnb.ddb.de abrufbar.

Alle Rechte vorbehalten. Kein Teil dieses Buches darf in irgendeiner Form (Druck, Fotokopie oder einem anderen Verfahren) ohne schriftliche Genehmigung des Verlages reproduziert oder unter Verwendung elektronischer Systeme verarbeitet werden. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, transmitted or utilized in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without permission in writing from the Publishers.

© 2011 Klaus Schwarz Verlag GmbH ist Edition 2012 Working editor: textintegration.de Cover illustration: Carmen Steiner, fotolia.com Printed in Germany

ISBN

978-3-87997-410-8

Table of Content Preface

7

Chapter I The Art and Science of Training How Trainers Train - How Educators Teach Training Ideas from Oklahoma Schools with Technology or Schools versus Technologies? Skills Training: Where Simulations are at Home Chapter II Training Policies that Make Sense Training in the Developing World: Issues and Policies Have Training Policies Changed? Training Dilemmas and False Dilemmas The Cost of Equity and the Cost of Efficiency Training the Poor When There Are Not Enough Jobs Is Training in Developing Countries Different from Training in Developed Countries?

9 11 17 25 50 55 57 78 84 93 96 102

Chapter III The Latin American Panorama Community Colleges and Latin America: Clone, Reject or Inspire? When Brazilian Employers Control Training, Many Things Can Happen

146

Brazilian Technical Education: The Chronicle of a Turbulent Marriage

154

Chapter IV Multilateral Agencies: Understanding the Beasts Does Skills Development get Short-Changed in Post-Jomtien Days? Do Training Institutions Learn from Experience? Can Multilateral Banks Educate the World? 5

Ill 113

175 177 180 207

Can Development Banks Promote Social Reform?

235

Can Training Bring about Organizational Change?

238

Monitoring Training Projects: A phylogenetic Approach

251

References

280

Full Table of Content

286

6

Preface

This book started as a collection of papers written since 2000. But since they did not sit comfortably side by side, they needed to be adjusted, mostly to eliminate duplications. Therefore, this is a collage of essays, massaged to perform two roles. First, they must tell their stories without interfering or conflicting with the others. Second, there must be a thread, a storyline. The author faced a dilemma when confronted with different papers that discussed the same institutions or cases. To eliminate all duplications would cripple some papers, since their logic required concrete examples (that were also referred to in another). But to leave all duplications would lead to boring repetitions. The choice was to minimize but not to eliminate repetitions. The papers reflect the past experience of the author, working in Brazil, for the ILO, for the World Bank and for the Interamerican De velopment Bank; in addition, as an outside observer, watching SENAI (the Brazilian Industrial Training Service) and other similar institutions in the country. They are neither theoretical exercises nor efforts to count heads. Instead, the papers result from the effort to understand training, as it happens inside schools and training centres. But also, looking at a broader picture and asking what policies are appropriate to deal with them as a group. The last chapter deals with multilateral banks, in their efforts to educate and train. The essays share the perceptions of the author while he was a member of their staff. This is delicate territory and the approach is very personal. In many cases, but not all, hard data simply does not exist. The effort to update statistics was very limited. This is acceptable because most of the papers were not about taking quantitative stock of a situation. Since some papers are older, one may question whether they still describe the real world. There is no certainty that some analysis may describe a reality that no longer exists. But since the papers are more about ideas and insights, the risks of obsolescence are not major. This is not to say that all the papers written by the author have survived unscathed by history. What is meant is simply that when real history collided with a paper it was not a strong candidate for inclusion in the present volume. 7

Chapter I The Art and Science of Training

How Trainers Train - How Educators Teach

Education and training have often been considered as polar extremes, the first being the development of the mind and the latter the mastery of strictly practical endeavours. But the two worlds of practical and conceptual endeavours are less distant than it may seem and these simplistic views of education and training are misguided. Indeed, there are definitional problems concerning education and training, leading to misguided policies. There is a need for a clear understanding of the overlaps and contrasts between the two concepts. There is a long and old controversy in the literature opposing education and training. Quintilian claimed that oratory was more useful than philosophy, thereby stating the superiority of training over education. But for many centuries education was closer to philosophy than to applied endeavours. Some educators use the word training in a derogatory way, as if to suggest that the learning is intellectually shallow, or that it goes with attempts to educate the poor. In contrast, some trainers refer to education as vacuous, fuzzy and rambling learning that is good only for wasting the time of students. However, both views are too narrow and misleading. When dealing with less schooled students, vocational subjects can be used to motivate and to create an environment which is familiar to them. Good training may function as a conduit for the best possible education for students less ready for abstraction. By using practical situations as a starting and end point, abstract concepts can be introduced and mastered by students who otherwise would be very low achievers in academic schools. The environment created by good vocational schools can give students a sense of getting closer to a concrete job. This can in turn generate a degree of motivation and sense of self-efficacy that is conducive to the mastery of abstract concepts that would leave students cold and aloof when taught at academic schools that have difficulties in recreating environments that motivate low achieving students. Good vocational training makes use of the context of the practical subjects to teach mathematics, writing, reading and science. Students are asked to read the instructions of what they are doing and write down the procedures they will execute. Concrete workshop situations are conceived, for instance, to make students convert inches to centi11

metres, Fahrenheit to Celsius etc. In other words, proportion will be learned as a by-product of solving shop problems. Mathematics is smuggled into the practicalities of shop work (Castro, 1988). In fact, good training institutions have different versions of mathematics, one for machinists, another for electricians and so on. As research in the psychology of learning suggests, the mastery of subjects increases when the contexts in which phenomena are examined are fully familiar to the students (Raizen, 1991). Experiments have shown that a physical principle is better understood when the students are given the broad context in which it applies. For instance, it has been shown that students acquired a better grasp of the concept of density when they were shown a clip from the film Indiana Jones in which the hero has to replace the golden skull sitting on a platform by his bag filled with rocks that added to the same weight. Students were asked to estimate the weight of the golden skull by measuring the approximate volume of a human skull and multiplying it by the relative density of gold (Vanderbilt Group, 1991). What good training does is to present inside the workshop such concrete problems based on concrete needs arising in the practical tasks to be performed. However, what training offers is merely the possibility to tap on this potential. There is nothing automatic about it. Training can fail to use these opportunities. Training that is only training is bad training or merely too short to go beyond the transmission of some dexterity. How to put out a fire or how to unclog a pipe is useful knowledge in its own rights and need to be taught. But they are essentially different from longer training programmes that contain more conceptual and theoretical structures (Castro, 1988). The "basic skills" movement consists of improving the knowledge of the fundamental literacy and numeracy skills of workers who are learning a trade or have already mastered the more practical and manual aspects of their occupations. However, the essence of the successful strategies is to use the same workplace operations as a scaffold on which to build the conceptual or cognitive skills that are missing. The worker learns how to read by reading the same manual that he needs to read to perform correctly his job. Vocational contents can be an ideal context in which to plant cognitive development of a higher order. Thinking skills, good reading and writing habits can be developed while doing practical tasks that lead to marketable skills. 12

By the same token, academic education may also resort to practical endeavours in order to carry the more general message. Laboratory classes try to do this and the Indiana Jones example illustrates more deliberate attempts to bring context to learning. Let us not forget that theory is generalization and conceptualization on real world observations. Formulae written on the blackboard merely displays a packaged and sanitized version of the intense intellectual effort that was required to arrive at them. The idea of having the students "rediscover" physical principles goes in the same direction. This reasoning implies that the differences between education and training have always been exaggerated and that most reputable training programmes are education as much as training. However, recent developments in technology and work organization seem to be blurring even further the distinction between education and training (Carnevale et alia, 1988; Elan, 1989; Eliasson, 1987; Hoffman, 1989). In industrialized countries, a very significant share of manufacturing activities have considerably changed and presently incorporate new technologies, particularly those based on microprocessors and the variety of automation techniques that result from them. Some successful industrializing countries are definitely moving in the same direction. New production technologies require more reading, more writing, more applied mathematics and more science (Castro and Oliveira 1991). In the past, all these cognitive skills were, at best, a means to master these trades (e.g. one needs to know how to read to take the machinist course because some of the instruction is written in books or handouts). But presently they are becoming part and parcel of the occupational profile. E.g., reading is directly useful for the performance of the core tasks of the occupation. Could it then be said that reading and mathematics are now vocational subjects? For that reason, most training programmes offered today could benefit from a little more emphasis on language, mathematics and science, as occurs in the best courses and apprenticeships. This is increasingly happening in Germany, the American Tech Prep and the new generation of SENAI courses in Brazil. While learning an occupation, the trainee may have an ideal opportunity to develop the same general skills that are taught in academic schools, i.e. general education. But this will not happen spontaneously. The integration of theory and practice, of shop activities with general principles of science, can only be the result of deliberate 13

and well-informed efforts. Training programmes should not underestimate the potential offered by such integration or the difficulties of achieving it. But there are good examples of these ideas: for instance, the new versions of the traditional Latin American "methodical series," as well as new methods developed in countries such as the United States (Tech Prep, School to Work) and Germany (key qualifications) have good track records. In so many words, vocational subjects can be used to motivate and to create an environment that is familiar to the students. Good training may function as a conduit for the best possible education for students less ready for abstraction. By the same token, academic education may also resort to practical endeavours in order to carry the more general message. There are conceptual differences between the roles of vocational training and education. Yet, as mentioned, the borderline between training and education is quite blurred. In its purest version, education is knowledge removed from practical applications (e.g. learning astronomy is pure education, except for those who plan to become professional astronomers). At the other extreme, pure training is a version of skill preparation that does not explore the theoretical implications of tasks being learned (e.g. learning how to use a saw and a jack plane without learning drafting and the requisite mathematics). However, in most cases the two are combined. Good training and a good education are equally good - and actually very similar in nature - when they promote the broad conceptual and analytical development of the trainee. By the same token, a good education is often linked to applied endeavours that turn the theoretical knowledge into a practical skill. In fact, in order to truly grasp basic concepts, it is necessary to apply them in real world situations be they in the world of work or not. The difference is mostly one of intention. Education uses the practical or occupational content to obtain a deeper mastery of theory, being somewhat unconcerned with the application of the knowledge in the marketplace. Training starts with the clear goal of preparing for an existing occupation, the theory being a necessary component to prepare a better worker for that position. Yet, despite all the merits of training, it is not a cost-efficient substitute for good schools for all. By contrast, a solid basic education is the best preparation for a wide range of jobs. In addition, it shortens the length of training required. In other words, the need to develop a 14

good training system does not replace the (perhaps) stronger imperative to develop a good general education system. Workers with a good mix of practical skills and conceptual understanding of technology can adjust more easily to new and different occupations, grow in their careers, and adjust to technological changes. The real issue is not general versus super-specialized training but the solidity and depth of the basic skills that go together with specialized training. One first element to understand the differences and similarities between training and education is to consider that the presence of training contents that may be applicable at the work place does not vary inversely with the presence of fundamental concepts and abstraction. Both poetry and solid-state physics are rich in abstraction. The first has scarce direct applicability at the work place. The second has ample utilization. Basket weaving has hardly any abstraction or conceptualization and finds little demand in modern societies. Cutting hair offers little in abstract thinking but there are ample economic applications for this skill. It necessary to stress that theory and practice are not the extremes of a same continuum but dependent concepts that admit all possible combinations, as exemplified above. Fortunately, to have the high theoretical and conceptual content that educates and sharpens the mind one does not have to forego learning the practicalities of life and work. What is called vocational training and what is called education of all sorts have both their theory and the practice sides. The main hypothesis of this essay is that the occupational training that fetches a good market is as good or better than any other environment to educate the mind in the fundamental concepts that are usually found in good education. Training should not be understood as something poor in theory and conceptualization. It can be rich or poor. Education should not be understood as something helplessly unpractical. In fact, it may be removed from immediate applications or it may be very close to them. There are no good reasons to be concerned with the differences between education and training instead of offering learning opportunities that have both. To sum up, abstract subjects that are removed from the everyday life of students offer a more arid ground for learning. Latin declensions, French irregular verbs, underground geological layers, the successions of kings of France, the capitals of African states are not the 15

subjects that fascinate the average student. Hence, they are not the ideal subjects to develop the broad basic skills that constitute an education for a modern society. Vocational schools can avoid these motivational difficulties by bringing in the world of the factory, with its practicalities and the inherent motivation of learning skills that have immediate market value. Nevertheless, not everything that happens in the factory is ideal for the process of learning. In particular, the factory routines teach mostly how to deal with repetitive activities. This is a worthy objective of short training courses and for the preparation of workers who lack the educational prerequisites for further development. This may be justified in many cases but it is not what is considered the optimal environment for broad learning. But equally important to understand is the fact that many interesting, motivating or even fascinating practical applications of the concepts and theories taught in academic schools may fail to have immediate demand on the market place - even though, indirectly, all good education ends up being valuable in the world of work. Learning statistics by dealing with Formula 1 examples is as good as any form of motivating students and leading them to master complex concepts. But the newspapers are not advertising jobs to analyse Formula 1 data.

16

Training Ideas from Oklahoma

The State of Oklahoma has a very interesting training system. As a latecomer, it could benefit from the experience of others. But it took plenty of dedication and creativeness to become one of the most comprehensive and innovative systems in the United States. In fact, in this unique system, high schools, vocational schools, community colleges and universities are linked to one another, allowing students to move between them, carrying credits for courses already taken. But just as much as this seamless integration, some of details of implementation of vocational and community college programmes are highly creative and deserve a closer look. Visiting the Francis Tuttle vocational school and the Okmulgee campus of the Oklahoma State University some ideas and innovations caught my attention. These are simple schemes on how to improve training. Most of them are attempts to bring training closer to the demand from enterprises. But some also have to do with instructional technology. A. The academic warranty

The relatively new crop of two-year programmes offered by the community colleges lacks the total and complete acceptance by industry that the regular and traditional four year-courses enjoy. Employers still do not know what they are getting. Partly, this is because these courses have open admissions, receiving just about any student who applies. If any state resident with high diploma can enroll, what kind of graduate do we get at the end? Like industrialists who assuage consumers by giving them a warranty that their products perform as promised, courses could also have a written and ironclad warranty. Why not? This is what the Okmulgee campus of the University of Oklahoma did: an academic warranty. The diploma or certificate lists what the students have been taught. If an employer hires them and finds that they cannot perform the intended tasks, the warranty kicks in. The employer can send the trainee back to school. Upon return to the programme where he graduated, the trainee is entitled to study for a maximum of six months, in order to make up for his shortcomings. 17

Neither the employers nor the students have to pay for the additional training - it's covered by the warranty. This policy reflects a correct attitude towards vocational courses. Courses and institutions are expected to respond to industry needs. They have to satisfy the consumer. If they fail to do so, employers can exert their right to the educational equivalent of "satisfaction guaranteed or your money back." B. Teachers accountable for the employment of their students?

The curse of vocational education around the world is the mismatch between training and market demand. In the typical vocational school, once students graduate, the teachers turn to the next crop and erase from their memories those who have already graduated. I have visited several schools in which teachers explicitly say that what the students do after graduation is none of their business. This Olympic disregard of schools for the occupational careers of their graduates has serious consequences. In the absence of tracking and surveying students after graduation, vocational schools revert back to the most embarrassing of all situations: training misses the mark. Schools start graduating students unable to find jobs and industries and societies alike discount the value of training. This is not the way it is done in the Francis Tuttle School. The entire system of training was created to attract and serve industry. Its leitmotiv was to create a workforce so well trained and educated that it would attract industry to the State. Unless taxpayers' money is being put to good use and employers are satisfied, the system is not fulfilling its explicit mandate. Therefore, what students do afterwards is indeed its business. Like any serious vocational school, it carefully helps students in placement and monitors what they are doing and how well they are doing it as they proceed in their professional careers. But Francis Tuttle goes one step further. The instructor himself is responsible for ensuring that the graduates get the expected jobs and the employers are satisfied with them. He who teaches the trade has to ensure that his former students are performing to expectations. Passing the buck to another civil servant in the system is not allowed. He who teaches and approves the students must know where his students are working and how much they earn. The instructors must 18

know what is the proportion of graduates who can find the expected jobs. And if this proportion falls below 85%, the course will be subjected to a technical audit. What is wrong with the course (or with the teacher)? Is the curriculum obsolete, incomplete, and misdirected? Does it need an overhaul? Is the market reaching saturation? C. Advisory Boards: telling schools what employers want

How do schools know what industries want? The answer is obvious: ask industrialists. The best training schools have advisory boards to give a chance to those who employ to tell the schools what skills they need. The Francis Tuttle School is no exception. It has been universally found that boards dealing with a wide variety of occupations do not perform well. Therefore, the school has several boards, one for each family of occupations - such as automotive, business, hospitality etc. The boards meet twice each year to revise the corresponding training programmes. This is also a chance to discuss internships, donations and other forms of collaboration. The fact of the matter, boards composed of managers, entrepreneurs, engineers and personnel department officers keep the supply of training in line with demand. They have real power because community colleges are partly funded out of municipal budgets and these budgets are very vulnerable to disgruntled business communities. D. Courses mutate: cobblers become pedorthic technicians

In a training system so keen to ensure that courses are responding to the market, the obvious question is what happens to courses that fail to meet the market test. In a system that prides itself as being so demand driven, this is the right place to ask that question. It is instructive to notice that the schools visited have not experienced the need to eliminate courses. Instead, courses mutate to something else, they drift into other related trades. Indeed, this has been the hallmark of the most established and successful training systems across the globe. Take a most unimpressive course offering such as shoe and saddle repair. Training cobblers at the post-secondary level seems to be a strange proposition when high tech is the shibboleth of all modern training systems. Not surprisingly, the demand for this specific programme has been weak. Yet, as it has been adapted to meet changing circumstances and demand, interest in the trade has been rekindled. 19

As it turns out, people who have feet malformation or other related problems are entitled to have their custom-made orthopedic shoes paid for by Medicare. However, insurance companies require that the shoe manufacturer holds a degree in pedorthics (the technology of making orthopedic shoes). This occupation is a natural mutation for cobblers. It gives this new generation of more sophisticated cobblers a captive market that require far more complex skills than making or repairing regular shoes. E. Teachers with diplomas or teachers with experience?

The expected profile of a community college teacher depends on the subject. To teach English, a Masters' degree is required. However, for practical subjects, teachers are expected to have a bachelor's degree and at least eight years of experience in the occupation they intend to teach. This ensures that teachers have a good theoretical background as well as ample practical experience. However, it is not always possible to find this ideal profile. Some teachers have degrees but not enough experience. Others have experience but no degrees. How to choose? The decision of the Oklahoma community colleges was taken long ago: choose experience. Subsequently, the teacher may be encouraged to return to school to complete a higher degree. But this is an afterthought. What matters is experience. These are practical schools and in practical schools, it is practice that matters. F. From roadside mechanics to automotive engineers

Community colleges are forced to receive as students graduate from any Oklahoma high school, in fact, a most heterogeneous bunch. Enrolment requires 8th grade reading level. But, in fact, remedial courses have to be offered to a significant proportion of the students who fail to reach this basic level. In other words, if community colleges ask graduates of a school system lasting 12 years to read at least at 8th grade level, much can be read between the lines, with regard to the standards of American secondary education. That is to say, post-secondary schools catering to clienteles in the lower quartile of the academic aptitude distribution have a difficult task in their hands. Not only do the community colleges have the task of preparing their students for the labour market, they also have the (perhaps more diffi20

cult) job of reinforcing basic skills that should have been learned in high schools. In times past, the students of lesser academic aptitude would be sent to struggle with car suspensions or hammer dented fenders. However, cars have become more complex machines. Perhaps, removing a suspension remains as physical as ever. However, brute force is of little help when struggling with the web of microprocessors that control a modern automobile. The head of the auto-mechanics department of Francis Tuttle School presented the following statistics: 1950

Present

20% of jobs required higher education 20% of jobs required technical training 60% on-the-job training or unskilled

20% require higher education 60% require technical training 20% on-the-job or unskilled

In other words, technical occupations are growing faster than any other category - at least in the United States. Chances are that the entire world is moving in this direction. After all, even in simpler societies the automobiles are not any simpler. Troubleshooting a modern automobile requires sophisticated skills. Reading the manuals and operating the computer that finds the faulty elements can hardly be done with 8th grade reading level. In fact, to graduate as a mechanic, students must be able to read at a level equivalent to 14 years of schooling. This poses a major challenge to schools like these that pride themselves as producers of highly trained graduates and receive students who are quite modest in their academic skills. G. Do we still need lectures in the classrooms?

Francis Tuttle has been an innovating school from the beginning. It always had as many teachers as any other school. But these teachers never lecture to students. Students learn from printed materials, watching videos, in computer simulations, in workshops and in laboratories. But we never find them sitting through lectures. The teachers are there, present and helpful, talking to the students and working 21

with them (the student/teacher ratio is 1:15, not much different than what is observed for regular instruction). Interestingly enough, the school does not prepare videos, computer programmes and other such means. It has no need to. Thousands of individuals, schools and firms are preparing courseware around the world. The inventory of training materials is far greater than the use that is being given to it. There are many more people preparing instructional materials than there are courses using them. This approach makes sense: why reinvent what already exists - especially if what already exists has benefited from the best minds in the business and captured the best from years of trial and error? One of the reasons this mountain of materials remains idle is no different from the reasons why putting a piano in the living room will not create a pianist in the house. The supply of hard or software is just that, and not the same thing as education. To be put to good use, it is necessary to have something in between. This is the role of a set of detailed instructions on how to go about the business of studying and learning (called LAP, short for Learning Activity Packet). The LAP makes the link between the training materials (chosen from the vast existing inventory) and the sequence of tasks the student will have to perform. This packet is what distance education is about. It takes the teaching raw materials (videos, written handouts or books, tasks for the labs or workshops) and tells the student exactly what to do with them. Like the Francis Tuttle courses, correspondence education does not have to be conducted with materials especially written for it. As the example of the Open University shows, the critical step is to construct a guide telling the students what to read, where, in what order and what they have to learn in these readings. For every course offered, the school has a set of LAPs. This is what the school itself has to produce. The original teaching materials are purchased off the shelf. The teachers are there to help when the students get stuck. No more the tedious repetition of lectures, day after day, year after year. Students learn by doing and receive help/input from teachers as needed. Each step in learning has to be mastered at 85% before the student is allowed to move on to the next. When he finishes the last module, he graduates. His grade is inversely proportional to the total time taken to complete the modules. 22

The system has been in operation for quite a number of years. It seems to be working well. The school has been so successful that it has opened another large campus using the same teaching methods. H. What if your teacher is in another town? Schools and universities in the State offer a number of disciplines the student can choose from. But what if the discipline is only offered somewhere else in Oklahoma? The solution is straightforward. In Oklahoma, campuses are linked by fibre optics. Curiously, with the depletion of petroleum reserves, the network of underground pipelines has been given a new use: to conduct fibre optics. A camera is set in the classroom where the course is offered and the prospective student attends the lecture in a distance classroom from his own school. This classroom has TV sets displaying what is happening on the other end. This distance classroom also has video cameras and microphones to allow the student to participate in the class, as if he were present. He can flag the teacher and ask questions - setting in motion a camera that puts him on the monitors of the live-class site. The equipment of each classroom cost about US$ 90,000. But this price is expected to fall in the future. In addition, in typical American exaggeration, there is far more hardware installed than is necessary to obtain similar results. With far fewer monitors, gadgets and cameras, equivalent results could be obtained. This is an idea that can be implemented in many different scenarios, even in Latin America today. It does not make economic sense to offer a discipline remotely to a small group of students. However, it should be far less expensive to link classrooms via TV and microphones than having a live class somewhere else. Indeed, Latin America's success with programmes such as Telecurso and Telesecundaria give reason to think- along these lines. When checked, PictureTell charged around $500/hour of international video-conference. This is far too expensive for the average course of study. If ten students attend, it costs $50/student (good quality vocational training in Latin America may cost about $3/student hour). This is also too expensive for all but the most sophisticated courses. For sites that do not have fibre connections, there are already other modes of transmission that are much less expensive. On the 23

other hand, better off Latin American regions are fast moving to fibre connections. This is certainly a technology to watch for. It is feasible already in many places and offers endless possibilities in the future, as the technology for transmitting video images in conventional TV lines continue to improve. The use of satellites is also offering another set of technical alternatives, which tend to be less expensive. What is worth retaining in this note is not the particular technology used - which is changing at a fast clip - but the idea of linking an on-going classroom to another school, allowing far more students to attend the course. This can be done with far simpler technologies and remain an efficient and attractive possibility.

24

Schools with Technology or Schools versus Technologies? This paper explores the turbulent marriage of new information technologies with education. It tries to show that technology has much to offer to education, but Utopian dreams of a takeover by technology or an easy cohabitation have become hard to believe. Let us go back one century, to the small town of Itabirito, in the Brazilian countryside. My great grand father, a young engineer, moved in to run the country's first blast furnace. But in this remote town, his two daughters could not attend a school, because none existed there. He had to hire preceptors for them in Rio de Janeiro and bring them in for extended periods of time. At that moment, there were two solutions to teach children: tutors and the traditional classroom (invented by LaSalle) and still in use. This individualized mode of education was the only possibility he faced. Classrooms with students sitting in from of a teacher were not widespread in the country and had not reached this remote town. The Lancaster method had been tried and failed. One century later, we are faced with a bewildering array of technological alternatives to educate. No wonder, we get confused. This paper is an attempt to sort out these issues.

A. What do we expect technology to do for us? The first questions that come to mind are: What do we want the technology for? Why would we want to mess up with expensive machines that are hard to use and prone to fail? There are two answers for that. 1. To make serious education even better Without exceptions, all the information and image technologies we are referring to were developed in the richest countries of the world. Not surprisingly, these are the same countries that have the most performing education systems. Therefore, if technology is to make any sense there, it must help make these education systems even better. They must be the topping on the cake. Otherwise, why bother? But in order to improve education, interventions will have to be complex and delicate, after all, the simpler stuff is already in place. 25

After trying to imitate teachers or drill students in repetitive tasks, computers were given much harder goals. Now students must play with turtles that move around, do research with the Internet or use simulation games to develop complex modes of reasoning. Nothing wrong with those uses, on the contrary. However, most of them require highly trained teachers and an intensive interaction between them and the students. Therefore, we are dealing with expensive forms of teaching. It is not so much the cost of the machines - that is dropping fast - but the cost and scarcity of the excellent and dedicated teachers required to handle such teaching styles. For the rich countries, these approaches make all the sense in the world. They have used up all the tricks without technology. Now they must try something else and technology is what is left. However, developing countries tend to imitate these models of utilization of technology. This is an unfortunate approach. The sophisticated teachers required to deploy such teaching strategies are in very short supply and the costs of using technology in such ways are a very high proportion of per student costs. Hence, the few experiments that are conducted here and there, regardless of whether they are successful or not, cannot be replicated to a significant number of students. This is not a promising path to poorer countries. 2. Technology

to reach farther with respectable

quality

Less affluent countries have better options to use technology in education. As a rule, they have some excellent teachers, but they are few. What technology can do is to bring the creativity and charisma of these rare teachers to a wider range of students. Instead of having their benefits restricted to the ridiculously small number of students they can teach, technology can bring them closer to the masses. To do that, there are several paths. Talented and experienced individuals, be they content-writers or super-teachers, should be able to support or supplement the average classroom teachers, making them more effective. Or, replace them altogether, if conditions are particularly adverse. And this can be done at costs that are much lower than the individualized tutoring done in richer countries. The use of computers should depend less on the creativity and sophistication of individual classroom teachers. Chosen technologies must be easier for the teachers to apply and allow more students per 26

teacher. In practice, it means a more structured teaching. And also, it is preferable to choose software that is easier to use. For instance, even though drill & practice software is scorned by some constructivist educators, it is easy to learn and easy to use. At least to get started, it is a convenient choice. Even though computers are incredibly versatile tools, they do not seem to be the ideal way to freeze and replicate the genius, charisma and creativity of the best teachers. Image technologies have shown much more tangible results. CDs, DVDs, videos and Television bring to the most remote areas and to the poorest students, the image of eloquent professors, scholars and scientists. I was one of the lucky few that could have Amartya Sen as a professor of Economics. Nowadays, one can download podcasts of this Nobel laureate speaking on different subjects. In addition, his videos are sold for a little more than cost of the magnetic medium that stores them. As will be mentioned subsequently, education TV has been a great success in countries like Brazil and Mexico. The magic it does is to freeze and package the very best teaching, done by the best soap opera actors. Students in the remotest villages in the Amazon region can watch a class that is far better that any class by any superb teacher in the most elitist school. This is because the best minds were summoned to prepare the content, the best actors present it and world-class television standards are secured to make the programmes attractive. To sum up, rich countries are well justified in their attempts to top off their education with technology, even if it is expensive and delicate to manage. By contrast, less affluent societies have fewer reasons to pursue the same tracks. They are better off using technology to make their best minds reach farther into poorer and more remote schools. B. Test of replicability: the pilot project works ok, b u t -

One of the most recurrent mistakes in school innovation is to propose solutions that are difficult or too expensive to replicate. Indeed, if they cannot be replicated to the target population, why bother? Pilot projects are useful, in order to test whether an idea works at all. They are not ends in themselves. They make sense only as a testing ground for something that can be applied to a large fraction of the target clientele. Otherwise, they will not make a difference. When a 27

pilot project is conducted, we should be always considering its future replication to a wider clientele. Therefore, the following criteria apply to all pilot projects. They test whether they are replicable: 1. The cash flow test Suppose we are considering a project to bring computers and other technologies to all schools. If the project were to double the costs per student, it follows that education budgets would have to double. From a political point of view, this is a naive proposition. It is not possible to double appropriations for education, even in the span of one generation. Therefore, in this example, replication is impossible. If the resources to replicate the experiment to the target population cannot be made available or if it is politically impossible to appropriate the additional funds, we are dealing with an idea that does not merit much consideration. In other words, if one cannot realistically procure the resources to replicate the technology at the scale commensurate with the target population, trash it! 2. Doctor Frankenstein's test Dr. Frankenstein was a brilliant German Jew who developed a method to increase the intelligence of his students (The name and the story are real! I met him personally). After World War Two, he migrated to Israel and continued his work at a local university. The fact of the matter is that his students gained points in their IQs, after taking his course. Not only that but the gains seemed to be irreversible. It was a great success. There was only one problem with his method: he was never able to prepare someone to replace him and do likewise with the intelligence of his students. His method was not replicable with other teachers. We are using here Dr. Frankenstein's anecdote as a metaphor for methods that might be creative and show splendid results. However, they are hard to replicate at a scale. Any teacher can conduct their students in the initial steps of LOGO. However, we do know that a little bit of LOGO has an irrelevant impact on the students. To bring concrete benefits, S. Pappert postulated that students have to spend a long time with it. Considering that, once it becomes more complicated, it is impossible to find the 28

number of teachers required to reach a broad coverage of students. Therefore, LOGO turns out not to be a very practical use of computers. If a method fails to be replicable with average teachers, trash it!

3. Institutional sabotage test Some techniques may look very promising. Their results on pilot projects may be solid and beyond dispute. However, their success may depend heavily on some critical actors in the system. If, for whatever reasons, these actors are unhappy with them, are they able to sabotage the entire process? This is not just a theoretical possibility, but a common situation. For instance, the introduction of computers in classrooms depends on the cooperation of teachers. If they do not perceive a gain in their utilization or they don't want to learn how to use them, the means to force compliance tend to be very weak in regular academic schools (in contrast to the armed forces or business enterprises where the means to enforce the requisite behaviour are much stronger). Very often, using computers is far more demanding on the time of the teachers than traditional methods. Perhaps, preparation time is longer. Or, the grading of open-ended homework takes more time. If that is the case, the probability of teachers sabotaging the use of computers is quite high. The fact of the matter, this seems to be one of the main reasons for the frequent failures of computer programmes in schools. Therefore, if there are people who, for whatever reason, are able sabotage technology if they are disgruntled by it, chances are that they will. If that is the case, trash it!

4. Integrity test A traditional classroom is an unbelievably robust technology. It can survive under a tree, without books and without electricity. In contrast, digital and image technologies involve several relatively complex and vulnerable links. To operate properly, all links in the chain must work, at the same time. In any of them fails, the system comes to a halt. E-learning requires that computers work properly, broadband connections are reliable and many other technical details are fulfilled. If any of these links break down, no e-learning takes place.

29

E-learning can be a splendid mode of learning. Under the proper circumstances, it may be the best choice. But it is quite demanding, in terms of having all links in the chain working properly. It suffices for one of them to fail to bring to destroy the entire process. Surely, according to Murphy's Law, if anything can fail, it certainly will. The issue here is probability. At present, in some places, systems are quite robust and breakdowns are rare. However, this is not always the case and failures in some link in the system can be frequent indeed. Therefore, if a method is too dependent on a complex chain of vulnerable procedures, trash it! To sum up, we need to ensure that we are proposing technologies that have a good chance of going beyond the testing stage. It is not enough that it presents good results when all odds are in its favour, as is the case with pilot projects. It must be replicable and robust. C. The technological tools Since time immemorial, people have been learning, one way or another. In fact, young monkeys learn with their parents and peers, mostly by imitation. Probably, the greatest breakthrough in learning has been the book. The next was the classroom technology we use, invented more than two centuries ago. Blackboards and other appurtenances also appeared along the way. The postage stamp, permitting the creation of practical mail systems was essential for the creation of correspondence schools, more than one century ago. This was the birth of all subsequent forms of distance education. The radio offered an extraordinary means to take education far away at very modest costs. It is still a very efficient mode of delivery. Unfortunately, it is less used than deserved by its merits and low costs. We could perhaps consider the podcast as its high-tech successor. Image technologies started with television. Like the radio, it was expected to bring a major revolution in education. To some extent, it did. But as shown later in this paper, it did not appear where one could expect. The last major breakthrough was the computer. Like all other such innovations, while not created to convey education, it was also announced as the definite revolution in learning. 30

The Internet allows the exploits and features of the computer to break the geography barrier. This, of course, opens the way for Elearning in all its varieties. Soon, virtual reality will allow us ways to have a vicarious presence and almost first hand experience. Simulations of the real world are getting better and prices are falling. Graph I allows us to see in a simplified way how all these elements fit together. The two lines separate face-to-face from distance forms of education. Graph I Exeperiential Face to face

Text-based

On the job training

Integrated

Image

Classroom Powerpoint Correspondence

Distance

Virtual reality

Computer

E-leaming

On the first column we have experiential education, i.e., doing the real thing, as in "learn-by-doing", apprenticeships and internship. The second column shows the varieties of text-based of education. At the most traditional mode, we have textbooks and other paperbased materials. Correspondence education belongs to this category of written technology. Computers also use text. The difference is that it is on a screen. Later on, they became more capable of handling images, but to a large extent, they remain a mode of delivery based on written words. E-learning is nothing else but written words sent by electronic means. Hence, it is a later-day correspondence education, done by electronic means. We could have added a third column for sound or voice. Face-toface classrooms have the voice of the teacher. Radio takes it anywhere in the world, making radio-schools a very respectable means of dis31

tance education. The Podcast can be used to bring audio classes anywhere at very low costs. However, for the sake of simplifying the graph, we have left out the voice column. The fourth column (we return to the third later on) contains the image-based teaching technologies. To a significant degree, face-toface classrooms are also image-based. After all, the teacher is there, alive, acting and talking. Replacing the blackboard and flipchart, come triumphant the PowerPoint. Strictly speaking, it is an image technology. Television takes the image elsewhere. It can beam the image of a teacher, an actor, or whatever fits the occasion, like a laboratory to illustrate a science class. At its essence, television is a means of transporting images (and sounds). The next iteration of image technology is the video, where the television image is frozen in a magnetic media. Subsequently, this frozen life migrates to CDs and then to DVDs. Surely, flash memory alternatives are just as convenient to record images. Tele-conferencing allows some degree of interactivity between the teaching and the learning end of the process. Talking about interactivity, let us understand that, from the moment we have more than a dozen or so students, this will always remain a limited possibility. Some traditionalists complain that distance education lacks interaction and student's participation. Technophiles point to the existing - and expensive - solutions to get interactivity. However, both are partly wrong. We have to consider that the moment a teacher is dealing with a large classroom, not too many students can ask questions, otherwise the class comes to a halt. This is true in distance education but it is also true in regular classrooms. Therefore, lack of interaction is an inherent constraint of education delivered simultaneously to many students, whether it is face to face or distance. Looking along the lines of the graph, when we move from live teachers to television, we are surely moving from face to face to distance education. The same happens when we go from books to text sent by mail and then, sent by computers. The teacher is multi-media by nature. It integrates image, voice and text. However, once we move from face to face to distance, there is a gap between text-based and image-based technologies. At least in the beginnings, computers were not good with images. And television had to forego the written word (except as courses had to supply books 32

at the side). Therefore, these two branches of education technologies imposed a serious option: image or text? More recent developments allow the combination of all these technologies in almost seamless packages. E-learning can go together with image technologies in many different solutions. When dealing with hundreds or thousands of students, high quality images can be delivered through the air or by satellites, at reasonable costs. Otherwise, small images, with You-Tube quality can be used with broadband Internet. A preferred solution in Brazil is to group in classrooms students from the same geographical areas. A local teaching assistant provides support to the students and helps create a learning environment. This allows for the development of good study habits, interaction between students and the local tutors, as well as a boost to motivation - always the number one problem in distance education. Looking at the Graph I we can see the alternative modes of teaching and learning. One century ago, my great grand father only had one solution to educate his daughters. If we consider all the alternatives available today and the different modes of combining the pieces, we have dozens of different ways to educate someone. We are all the better for this wealth of alternatives. The downside is that intelligent choice becomes more difficult. How do we match each individual circumstance to a given combination of teaching technologies and strategies? In what follows, the paper goes into greater detail on the available technologies and the criteria for choosing them. At the end, it will discuss two issues that are far more important than hardware and software. The first is the science and art of teaching, with whatever technology. The second deals with the sociological clashes between computers and the academic schools.

D. Choice of technology is constrained by the numbers of students Old teaching technologies consisted of small variations on having one teacher in front of his students. A major feature of this solution is that just about all costs are variable. That is to say, the number of teachers increases proportionately to the number of students. Once the classroom is full, another teacher will have to be hired. Therefore, a school with two teachers - or one hundred - will have the same costs 33

per student as a single-classroom. Plotting the figures in a graph, costs per student are represented by a line parallel to the horizontal axis, because more students require more teachers, in the same proportion. In contrast, when we deploy educational technologies, fixed costs up front tend to be much higher. To start with, hardware is often expensive. Computers, projectors, cameras and other such equipment have to be paid for, even before the courses start. By the same token, preparation of courseware becomes an expensive proposition. A wellprepared teacher will walk into the classroom and deliver his class. Printed handouts may take five to ten hours of preparation for every hour of classroom. A video takes at least 30 hour to prepare, for each hour of class. In other words, more technology implies higher fixed costs. Very often, technological solutions save on teacher costs. For all these reasons, the cost equation is quite different. Fixed costs are piled up at the beginning of the course. When we calculate costs per student, the total costs will have to be borne by the students. If there is only one student, he will have to pay all the fixed costs incurred. With two students, the fixed cost will be divided into two. As the number increases, the denominator of the cost equation expands, while the total fixed costs are the same. Therefore, costs per student falls, because more and more students share them. Graph II (next page) shows per student costs and illustrates their variation. Being divided by an increasing number of students, fixed costs per student decrease. This is represented by a rectangular hyperbole. As mentioned, a line parallel to the horizontal axis, showing that it is constant, represents variable costs, usually teacher pay. Adding the variable and the fixed, we obtain a decreasing line, but higher up in the graph (not depicted). This is all we need to know. With few students, a traditional faceto-face education tends to be the most appropriate alternative. The contrary is true for large enrolments. More students justify investing more on preparation or equipment. This choice may save on labour costs, because it becomes possible to have equivalent results with less excellent teachers.

34

Graph II More technology Cost per student V \

\

Conventional classroom

Less technology

M

\

Variable costs

\

Video conference

"*——

FrH

E-learning Correspondence, TV # students

For instance, Telecurso 2000 is a television programme to prepare younger adults to get the equivalent of a primary or secondary degree (more on that later). The video production cost over 30 million dollars. However, to this day, over five million students have graduated from the programme. If it were to be interrupted today, TV costs per student would be six dollars. If it continues for more years, they would be further reduced. The lesson is very clear and forceful. Scale of operation is a major determinant in the choice of education technology. Its importance should never be underestimated. E. Computers are devious fellows, they don't always teach Since the moment computers became widespread, their educational potential was perceived. The first experiments in using them for teaching began in the sixties. After almost forty years, much has happened. 35

First of all, their speed and memory have grown spectacularly. Second, they became far more reliable. Third, interfaces became more user friendly. In fact, the appearance of graphic interfaces was a turning point. Fourth, they became less expensive. And fifth, quality, variety and sophistication of software have grown continuously, since the early days of mainframes. If we were to extrapolate the initial faith in computers in education with the spectacular advances in all these five areas, it would be inevitable to predict an extraordinary success story. Yet, this is not at all the case. Much research has been conducted to evaluate the performance of computers in education. There is a large inventory of position papers, white papers, controlled experiments, official reports and informal reports. Putting all these sources together, we find a clear pattern. However, it is a double pattern. Results point to opposite directions. Ample research, with control groups and all other methodological orthodoxies has been conducted to examine specific cases of computer applications. By and large, these case studies show positive results. In other words, students learn more with computers, compared to their performance without them. Not always, but some technologies tend to show solid superiority for students utilizing them. In sharp and dramatic contrast, studies trying to evaluate the impact of computers on education systems (be they in countries, provinces or towns), usually fail to show significant improvements. In other words, promises are made, millions are spent and nothing happens. In other words, bringing computers to schools turns out to be ineffective. This is so, despite the above-mentioned results of research with specific applications showing good results. Just as perplexing, PISA analysis suggest that computers in the homes of students improve their academic standing - provided there is no overuse. Could we say that students themselves know how to use computers to improve their education better than schools? Having access to chat, pornography, games and whatnot, students choose a productive path. Schools, with total control of usage, cannot get better results. What an irony! What can we infer about teachers and their ideas of education? The programme One Laptop per Child tries to blend the home with the school approach. Countries like Uruguay have delivered 380 thousand computers, one to each student and teacher. However, it is too early to expect an evaluation of its impact. 36

Let us take a look at the different possibilities of using computers in education. Mind you, under favourable conditions, almost all of them have solid evidence in favour of their capabilities to boost learning. 1. Tutorials and Computer-Aided Instruction (CAI)

This is the grandfather of computer applications in education. Theoretically, these uses derive from Skinner's ideas about learning which were popular in the sixties. The icons of this strategy were the teaching machines. Patrick Suppes spelled out the main ideas in wellknown papers. In a nutshell, a computer could replace a teacher. It could eventually teach better than its human counterpart, having more patience for repetition and more inspiration from the best minds summoned to programme it. All these statements remain true, to a significant extent. However, tutorial programmes turned out to be very expensive to develop. And they are even more expensive if higher order learning is to be achieved. The massive quantity of content that is to be taught in schools is way beyond the practicalities of developing tutorials for all or most of them. And the fact is, the usual content to be taught is relatively simple. Good books also teach them. In other words, expensive software competes with inexpensive books and displays meager superiority. Intellectual fashions also played a role. Skinner's theories about stimulus and response were criticized by a motley collection of scholars, having loosely defined constructivism as a common faith. For all these reasons, tutorial programmes have fallen out of fashion. Some still exist, but this line of usage seems to be a dead end. Notwithstanding the difficulties with computer tutorial programmes, there is a consequence of these ideas that deserves more than a second look. This is what has been called Programmed Instruction. In its essential lines, it is not truly different from the computer tutorial programmes. However, it is geared to more mature students who can use existing materials or materials that require only minor adaptations. More advanced students can be assigned chunks of regular books or textbooks. Once they read one major idea, they are given a quiz. If they get it right, they move on to the next idea. If they fail, they go back to easier quizzes and to the original materials. Since it is

37

text-based, no elaborate programming is necessary. It is just linking content and verification, both in small doses. Programmed Instruction became fashionable for a while. But it fell into disrepute later on, when educators revolted against the mechanical teaching that was being conveyed by the available software of this genre. Indeed, Programmed Instruction was used to teach computer programming and other very shallow disciplines in which memorization was the main challenge. As a result, it became an ugly duckling, that no serious educator dared to propose. But the truth of the matter is that the criticism is unfair and unfortunate. The idea of having short chunks of knowledge put in a book (or computer), the student asked to read and then be tested for his mastery is perfectly sound. It is in line with the finding that feedback should come as soon as possible. This is not too different from Mastery Learning, a cousin with a better reputation. There is no reason in the world to have the teaching shallow and the same applies to drills checking whether the students learned what was explained. Higher order knowledge can be put into programmed instruction format and this makes a lot of sense. Students can be assigned to read the Categorical Imperative of Kant and asked how does it fit with alternative situations presented by the computer. Smart content writers can cook up interesting questions that require mastering higher order reasoning to answer. In a way, this is not different from what is done at institutions such as the Open University. Therefore, readers are well advised that Programmed Instruction remains a respectable and promising means of education. There was another major consequence of these early experiments. The same proponents of the half-failed tutorial programmes came up with something that is as state of the arts as anything coming later: individualized instruction. As P. Suppes showed, computers allow students to move at their own pace, in contrast to face-to-face education where this constitutes an intractable problem. Perhaps no other development in the deployment of computers offers so many promises as individualized instruction. Students have different learning speeds. Or, they don't have the same time available to spend on their education. Allowing each to move at his own natural pace is a wonderful breakthrough.

38

2. Drill and Practice The tutorial programmes described above can be decomposed into two phases. The first is the tutorial proper. The second is the drill to ensure that the materials have been mastered. As it turns out, the exercises are easier to prepare than the tutorials and are more in line with what computers can do better than humans. They have endless patience to repeat and repeat allowing students to progress or have them review what they failed to master. Therefore, drill and practice divorced from the tutorial stage. The former took on a life of its own. Thousands of pieces of software were prepared to drill students in many subjects. Of course, some were crude, while others became quite sophisticated. But like Programmed Instruction, this line of utilization of computers fell into disrepute, by the same reason and by the same people. However, to a large extent, criticism is not warranted. Drill & Practice tends to work best with materials that need to be memorized. To the extent that we want education to focus on higher order learning, this is not the number one priority. However, we still need to learn lots of materials that are nothing by memorization. This is the case with multiplication tables, spelling, irregular verbs and much more. This dull stuff taxes the endurance of teachers and students. Drill & Practice software makes life easier for both. Students enjoy the rewards of playing little games when they get the right answers. The lesson is clear, let us not dismiss this form of computer learning. It is not the most intellectually exciting and teachers have been told to hate it. By it still has its place. 3. Simulation and Games Piaget said it with more authority than previous authors: playing is part of the development of intelligence. Therefore, computer games have a respectable theoretical rationale. Early games, like playing ping-pong or tennis, had hardly any educational content. But the development of computers with more speed, memory and ability to process moving images has brought the technical end of games to a very sophisticated level. At the same time, games became more intelligent and introduced practice and experimentation with the concepts that are supposed to be taught at school. For instance, Oregon Trail was a very well known exemplar of such 39

games. Sim City came much later and brought in much greater sophistication. From a pedagogical point of view, educational games are a very welcome idea. They have, however, two problems. The first problem is that youth used to games are not interested in understanding the theories that they are supposed to teach. They often adopt an iterative strategy, pushing keys at a rapid pace, without trying to understand what they are doing. They eventually get to the right answer, without knowing what the theoretical problem is. S. Turkle comments that in order for such games to become educational, teachers have to tutor the students to understand the underlying concepts, rather than rushing through the game. The second problem is the ecology of the school with its watertight disciplines. Games cross disciplines; hence they do not fit any course. Teachers are quite reluctant to use them in class, for the above reasons and for many others. Therefore, games remain a vastly underutilized resource for education. 4. Development of critical skills

This is a vague concept. It is being used here because the term is fashionable and teachers are expected to teach such skills. They have to do with problem solving, with complex modes of communication and with research done by the students using computers. There is nothing wrong with such goals. In fact, it would be hard to find more worthy objectives for a school. The problem is that they are difficult to accomplish with the average teacher. They require very individualized tutoring, being incompatible with the number of students that need a good education. To the extent that these goals lack a clear definition and well-codified sequence of procedures, they are not good recipes for the less advanced countries. Unfortunately, teachers and researchers in poorer countries mimic what they saw in the First World and try to recreate the delicate and vulnerable strategies that are not replicable at the requisite scale. 5. Basic skills and the preparation for the labour market

Having in view the systematically mediocre results of putting computers in schools, one can ask a simple but pertinent question: is there 40

some set of uses that will improve the odds of having some impact on students? Actually, there are two strategies that can be combined into one single set of activities. First, basic skills are the first order of priorities for schools. Fluent reading and writing, good understanding of the written word and using numbers to deal with everyday problems are some of the most central contents that students should master. Secondly, poorer students do not have computers at home and will need to use them in the labour market. Teaching how to use computers is a more than justifiable goal for such students. At least, they must become fluent with operational systems and productivity software. Many programmes to introduce computers to schools do exactly that. These introductory programmes, however, tend to be quite mechanical and lack depth. One learns how to use the computer to do what? One acquires fluency doing what? The good idea here is to combine these to goals. Productivity software should be used to write, to read and to develop understanding of increasingly complex texts. There is ample evidence that word processors can improve writing skills. Much can be accomplished along these lines, without overburdening the teachers. The same is true with the use of spreadsheets to solve real life problems in projects that motivate the students. This is mathematics at its best. By the same token, non-relational data basis can teach many concepts (for instance, principles of Boolean Algebra). In other words, students can learn how to use computers and, at the same time, they practice some of the most critical skills in real life. For some reason, despite its appeal, this strategy remains vastly underutilized. 6. E-Learning and virtual classes

The Internet is progressively replacing the mail system as a means to take the written word to distance students. Considering that correspondence education has a strong record of success, it should be no surprise that the Internet should also turn out to be an efficient mode of distance education. Considering all the hype that surrounds e-learning, we need not glorify it even more. On the contrary, it is worth mentioning that neither correspondence nor e-learning are always a preferred or even viable solutions. They should not be oversold. 41

Some students have the profile to take courses at a distance. Others do not. Maturity and self-discipline are perhaps the critically decisive traits. But considering the low levels of reading abilities of many students in developing countries, this can be a serious constraint to elearning. In fact, this is learning with books, the only difference being that they are digital books. A poor reader needs the strong support of a live teacher. Perhaps as important is the combination of content and pedagogies. Tired or dead class notes transferred to the Internet remain exactly what they were before: tired or dead education. A teacher may compensate for a dead book, by bringing the spark of his own charisma and performance. A dead book migrated to the Internet does not have this second chance. But mind you, there is no guarantee that the teacher will compensate for inert content. In the e-learning mode, everything has to become a written text. The words must promote the learning. One can make it more entertaining by adding graphs, movements, chirping birds, photos and whatnot. But here, what is valid for Powerpoint presentations is also true in e-learning: Bad content cannot be compensated by bells and whistles. On the contrary, such distractions may take the focus away from the learning. 7. The digital textbook Textbooks are expensive. Higher education students can hardly afford them. However, the production technology of books is stable and does not suggest that dramatic reductions in price are possible. On the other hand, once the books are written, at whatever costs, reproducing them electronically has marginal costs that approach zero. Devices such as Kindle or just reading on the screen of regular computers could bring a revolution in the costs of books to students. We observe an increasing number of experiments to replace books. Research suggests that younger students have no major problems reading in the screen. The problem lies with the book editors who fear the paperless book. Whether they are justified or not, it is early to tell. However, the worst scenario is mass pirating of books. Digitalization is becoming less and less expensive, especially when done in China. How are the book editors going to prevent mass copying of textbooks, similarly to what happened with music? In fact, books are much easier to trans42

mit through the web, since files are smaller and encryption is absent. This is beginning to happen. Allegedly, 5000 Brazilian websites are displaying pirate copies of books. In Brazil, more than 90% of higher education students have access to computers. More than half of secondary students also have access. In addition, the production and sales of computers is increasing at a fast clip. The digital natives are reaching higher education and the new generations have more and more computers. Therefore, whether one thinks of digital books, e-learning or plain computers using educational software, it is difficult to imagine that education is going to remain as it is presently. The stalemate of computers not breaking into the core of the education process will probably be overcome. Perhaps classrooms will never really use computers. But surely, students will be using them for a multitude of reasons, including books, legally or illegally obtained. 8. Computers to manage schools This is a no-brainer. Doing by computers what was traditionally done with ledgers, tables and pieces of paper saves an incredible amount of time and trouble. This includes enrolments, budgets, grades, classes attendance and many other repetitive tasks. Therefore, there is a sharp contrast between the lukewarm or absent impact of computers in learning and its outstanding success in handling the paperwork of schools. For that reason, it is not uncommon to find that the only working computers in some schools are busy helping in the management. Ultimately, a computer is a solution looking for a problem. We need to find the right problem, expect no miracles and be aware of the sociological collisions that are happening all the time. F. Image Technologies: Some poorer countries do much better To a very significant extent, educational television has become a dead end in the advanced countries. There are very few examples of television being used to teach in the academic streams of education. What exists is rudimentary and restricted, as is the case with some American programmes. The uncontroversial success of Sesame Street is not an exception, as it caters to students before they go to school. This absence has created the conventional wisdom that, in the rich countries, television is not the way to go. However, this generaliza43

tion is particularly inappropriate for countries in which a universal and high quality basic education system has not yet been achieved. Two very impressive exemplars show that TV is not dead. These are the cases of Brazil and Mexico that share some common features. They are both large countries with many millions of students. Therefore, they have the scale required for heavy investments in the fixed costs of preparing the TV classes. They also have inadequate systems of education, both in coverage and quality. And finally, both have export-quality commercial television, providing the know-how to prepare highly sophisticated programmes. In particular, the soap opera tradition can provide the dramaturgy that makes classes interesting and motivating. Telecurso 2000 in Brazil and Telesecundaria in Mexico have millions of students, deploy professional actors, use soap opera styles and are highly successful. The solution combines refined TV images with classrooms where teachers or teaching assistants help students and manage the class. The model works very well. With some adaptations, this model has migrated to distance higher education in Brazil. Satellite links create significant networks of local classrooms equipped with antennas and decoders. Students meet a few times a week to watch the transmission and do the assignments, under the supervision of a local professor (not necessarily someone knowledgeable in the subject being taught). Interactivity is usually by email, allowing the professor to answer questions on video, after the lecture. The major difference is on the televised lecture. In contrast to the high cost and sophistication of Telecurso and Telesecundaria, these are plain lectures by teachers. No actors are involved and the technique is barren, being no more than what is known in the TV milieu as "talking heads". Be that as it may, the results are respectable. In Brazil, graduating students of higher education must present themselves for a test covering what they are supposed to learn in the course. The scores of this test (ENADE) are tabulated for each institution, allowing the Ministry of Education to assign a grade to the course. Recent evaluations have shown that distance courses perform just as well as face-to-face education. The above shows two major styles of utilization of TV. But image technology has a much broader range of possibilities, either comple44

menting regular education or replacing it. Let us review the most obvious uses of images. /. Talking heads: a teacher, a camera and a tripod This is the simplest and most direct way of shipping images elsewhere. It is also one of the alternatives with the lowest fixed costs. Good teachers with the mastery of the subject to be taught can usually be found. Even if they are expensive, the large numbers of students provide the denominator that lowers the per capita costs. The whole package is as good as the quality of the teacher in front of the camera. In TV circles "talking heads" have a horrible reputation. Spectators are said to lose interest after a few minutes. While this can be true, two considerations apply. First, this solution is meant for mature students who have a stake in finishing the course. They are not TV spectators who can surf quickly to the next channel. And compared to the local teacher who either does not exist or is less competent or even more boring, this turns out not to be a bad solution at all. Secondly, some teachers are outstanding. The series of courses from the American-based The Teaching Company (www.theteachingcompany.com.br) illustrates how captivating a "talking head" can be. A variant of this mode replaces the satellite transmission by broadband Internet. The drop in cost is precipitous. But the drop in the quality of the image is also devastating. How important to learning is a big, high definition image? As far as I know, there are no good answers to this query. 2. Recycling live courses In a famous experiment, live classes offered to master students of electrical engineering from Stanford University were recorded. Subsequently, the tapes were reproduced to HP engineers. Both groups took the same tests. Surprisingly, the results for the HP students were slightly superior. Simple minded as this solution might appear, it should always be considered. It is inexpensive and can produce good results. An alternative that is being increasingly used is to have live transmission of real lectures to classrooms elsewhere. This has the added bonus of allowing the students who are away to ask questions and get immediate answers. 45

3. Teacher styles Some teachers are outstanding in front of students. For them, being a talking head on television imposes no significant loss of interest. Finding those teachers is one of the tricks to generate effective and inexpensive distance education. It has been known in the television milieu that, in front of a camera, actors posing as teachers perform better than real teachers. Telecurso and Telesecundaria have opted for actors. Some other educational TV programmes try to locate teachers with the requisite showmanship. 4. Demonstrations and live examples I was struck by a video prepared to show future waiters how to bone a trout. Let us think for a moment of a written handout explaining the intricacies of removing all the bones of the fish and reassembling it without breaking the flesh. When we watch the video, everything is so clear and easy, in the hands of an experienced maître d'hôtel! In that sense, if a picture is better than a thousand words, a moving picture is worth how many more words? Think now of a teacher trying to explain to Brazilian students about volcanoes and their eruptions. Now think of a video clip from Discovery Channel showing volcanoes, both calm and angry. Business courses often use short videos demonstrating this or that principle of administration. When well prepared, they convey the ideas far more forcefully than a teacher or a book can do. I remember a video of an executive who does everything wrong and ends up dying from a heart attack. Arriving in heaven, an angel shows him all his mistakes and gives him a chance to go back to earth and do things right. A silly plot, to be sure. But it delivers the lessons better than the dry textbooks that circulate around. 5. Dramaturgy: soap opera migrates to education As mentioned at the beginning of this chapter, soap-opera styles migrate very nicely to education. If we agree with the belief that story telling is education at its best, soap operas are story telling at its best. There is drama, there is humour and wit. Images are compelling. All this allows the message to be built into the story, in a seamless way. Telecurso has bits and pieces of plots. There are recurring sketches with the same actors, there are animations and other tricks with im46

ages. However, it is not a regular soap opera with a single long-drawn plot. In contrast, a later production of Telecurso, aimed at students not covered by the original programme opted for a full-blown soap opera style, with a single plot that unravels along the series of 15 minutes classes. An early experiment along these lines was Bank Street's The Voyage of the Mimi. It had a single plot, in which an adapted fishing ship, out of the coast of New England, does a study of whales. Since this author studied carefully these videos and took part in the original conception of Telecurso, one can safely say that the Bank Street programme is its father or grandfather. Compared to Mimi, the major difference is cost of production. While the American programme was a competent but low-budget production, Telecurso deployed the full weaponry of expensive commercial television, with soap opera actors and sets. These exemplars show the very high end of this mode of utilization of images. But it would be wrong to imagine this is the only way. Far more modest production levels can be fully justified. G. Do machines know the science and art of teaching? Distance education and computer-based education are just forms of education. The difference is in the means to bring it to the students. Above all, we should not make the mistake of believing that technology has much to do with the science and art of teaching. It makes some difference, but only in the sense that each technology is better or worse suited to the deployment of this or that teaching strategy. To a very large extent, what makes for good face-to-face teaching is the same that makes for good education using technology. And vice-versa, for bad conventional teaching. In other words, lack of sound strategies to lead the students to master the content taught is a fatal flaw of both of face-to-face and distance education. As a very broad generalization, we can say that there ought to be two main concerns in any teaching: 7. Contextualization Very few people can learn abstract concepts by facing them directly. For the majority of the mortals, we learn when the new ideas can be seen as a continuation of something that is familiar to us. This re47

quires good examples, metaphors, story telling, Project-Based Learning and so on. Abstraction delivered with new technologies is just as indigestible as abstraction in the blackboard. Computers can contextualize and bring to life cold mechanical formulae and show evocative images. But the author must understand that this has to be done for the classes to be effective. Graph III depict the well-known taxonomy of education objectives, originally formulated by Benjamin Bloom. This is as good as any other means to show a hierarchy of objectives. Remembering facts is just the beginning. It has to be done, but this is not real education, as we want it to be. Understanding concepts is the next step in the ladder. Application is already a respectable level. Above that there is analysis, evaluation and creation. The goal of a good education is to climb the stairs of the taxonomy. This is as true for face to face as for instruction that incorporates technology. The legitimate criticism against programmed instruction was that it remained very low in the taxonomy (mostly memory). However, the mistake of the critics was to assume that it could not move upwards, just as much as any other forms of education. I make the claim that, in mass education, programmed instruction could push students farther up, compared to modes of instruction totally based on the initiative of the teacher. H. The sociological wars: schools versus computers

If we list all the institutions that use teaching technologies, we can see different patterns of adoption. The highest levels of technology adoption are in the armed forces, particularly in the richer countries. Next come enterprises in their training programmes. Still showing good disposition to adopt them are the technical and vocational schools. Quite reluctant to introduce them are regular primary and secondary schools. The most impervious to changing their old modes of teaching are the higher education institutions, especially the research universities. Why is this so? A superficial look at the sociology of power inside these institutions shows patterns that just as clear. How much power and freedom have the teachers and instructors in these institutions? The armed forces tell their instructors to use this or that technology and it would not occur to either side that orders are not be obeyed. 48

Business firms also tell their instructors what they are expected to use. Perhaps they do not risk disciplinary action if they fail to comply, but in general, it is not a good idea. Vocational schools work with norms that are influenced by those of business firms. There is the expectation that the lines of command are there to be followed. In addition, the familiarity with technology makes it more natural to deploy new learning devices. In contrast, in the academic world, the sacrosanct principle of teacher freedom is for real. Teachers do not feel any obligation to follow rules that tell them how to run their classes. Therefore, in practical terms, teachers can choose not to use computers, videos or whatever is available. And indeed, this is what happens too often. It is the belief of this author that the failure of computers to improve learning has more to do with such factors than with their intrinsic lack of efficacy. In practical terms, teachers are free not to use them. They can challenge the policies to deploy them, but more often that not, they merely say yes and plod along with their old practices. Therefore, perhaps we can say that it is not true that the use of computers in schools is ineffective. It may be more accurate to say that the non-use of computers yields no improvements in learning. Ultimately, when dealing with actors that, in practice, are free to do as they please, we need to take that into account and design styles of utilization that are attractive to them. The army captain does not have to do that. But in academic education, teachers have to be prepared, persuaded and given forms of utilization that make their lives better and easier. Increasing their workload or imposing other constraints is a sure recipe for failure. To sum up, the ultimate battle in the use of technology in classrooms has less to do with the technologies themselves, than with the art and science of teaching. But also, it has much to do with the sociological clashes between the values and routines of schools and the behaviours that favour the uses of computers.

49

Skills Training: Where Simulations are at Home

The cohabitation of education and instruction technology has been turbulent if not outright antagonistic. By contrast, skills formation has had a far more benign interaction with all forms of instructional technology. In particular, trainers have used simulations for a long while. In simulations, a concrete situation is reconstructed in what we now call a "virtual reality". Why would anybody want to replace the real world by a fake version? Essentially, there are at least three reasons. The first reason is safety: the real world is dangerous. Students should not risk their lives (or that of others) or get scared in the process of learning their trades. The second reason is simplicity, ease and time savings achieved with the "virtual" version of training. Some experiments take a long time to set up and conduct. In others, the results are not so clear-cut, due to too many unmanageable variables. Simulations can be a convenient and convincing way to synthesize the real world. The third reason is economy. Simulations can be less expensive than learning in the real world, particularly now that computers are increasingly inexpensive. Simulations may save on expensive labs or on consumables such as metal, electronic components or welding electrodes. The first significant use of simulations was to train airplane pilots. The flight simulator was invented by Mr. Edwin A. Link in 1929 and used by the Army from 1934 on. Working in a family business that manufactured organs, he adapted its bellows system to make the model lean along the three axes. In this case, the first and most forceful reason to use simulation was safety. Pilots needed to know how to react to life-threatening situations. Yet, turning off a turbine or disabling a rudder control in a jet liner in order to test a pilot's reaction is not a good idea. A modern flight simulator is a multi-million dollar machine, often not much cheaper than real airplanes. But, of course, nobody thinks of how much cost a flight simulator when deciding to use it. The reason to resort to simulation is that it permits to reproduce flight under very dangerous conditions. It provides the pilots with a chance to acquire the proper reactions, in cases where such radical conditions were to be reproduced in real flights. 50

Another family of simulations that are becoming increasingly important is in health training. For instance, Intensive Care Simulators have become very sophisticated and realistic. A rubber patient lies at a hospital bed, wired to the conventional monitoring devices used in Intensive Care rooms. The instructors may simulate a cardiac arrest. This condition has to be identified by the student who needs to learn how to quickly follow the standard protocols, for instance, with defibrillators or electric shocks. Her behaviour and speed is monitored by the system and by the instructor. In this case, the justification is the same as that of the flight simulator. By the same token that one cannot set a turbine in fire to see how the pilot reacts, it is not possible to stop the heart of a patient to check how fast the nurse can take the proper steps to revive him. Convenience and economy are the justifications for other types of simulators, like those that reproduce the operation of numerically controlled machine tools (known as CNC machines). An apprentice becomes familiar with a conventional late by handling it under controlled conditions. At first, he will turn simple parts, always being careful to keep the tool far from the faceplate. But accidents happen. An extra turn of the lever and the tool may hit the fast-turning plate. Be that as it may, a broken bit and a scratched faceplate in a learning lathe is not much of a loss. Yet, CNC lathes - that are programmed like a computer - cost several times more and are more prone to serious accidents. A wrong line of code may zoom the turret towards the faceplate, provoking a horrendous collision and causing serious losses. The school has to write off the losses and instructors report that students become traumatized by the crash. Therefore, the obvious first idea was to couple the CNC to a simulator that would trace on paper the trajectory of the cutting tool. The paper drawing would immediately reveal an eventual mistake. Such a contraption was created and manufactured by a Swiss technical school at Sainte Croix. Only after the simulation shows the programme to be devoid of gross mistakes will the real movements be attempted. With computers becoming more common, a monitor replaces the paper plotter. The obvious follow up development is software that simulates the entire process, dispensing with the real lathe altogether. Clearly, this applies to milling machines and the whole gamut of CNC-controlled machine tools. Today, computer simulations of CNC machines are very common, being quite sophisticated and inexpensive. If properly used, they can 51

speed up the training and lower the costs significantly because trainees can learn much from them and they require a lot less supervision. To what extent they dispense with first hand contact with real life CNC machines is a controversial subject. Research on learning suggests that virtual practice can be surprisingly effective. Past studies suggest that soldiers who aim and pull the trigger of handguns without ammunition improve their shooting accuracy to a surprising degree. The same is true with other similar activities, including in sports. That is not to say that real practice will become obsolete. Manipulation of objects and tools is a key component of just about any learning. The key issue is how to blend less expensive simulations with the real thing, is such a way as to minimize costs, while retaining solid results. Perhaps the case of CNC machines is less controversial. There is little disagreement on the overwhelming importance of real life learning on basic lathes. The challenges of moving from a manual lathe to a CNC version reside at the programming end, not in handling the machine, since the CNC version requires little human input once programmed. That being the case, it makes little difference on whether the programming is for a machine simulated in the monitor or a reallife machine. Another very common family of simulations is for electric and electronic circuits. Vocational schools frequently use panels where components are plugged, reproducing the typical electric wiring, for example, of an automobile. After students understand the circuitry, the teacher may introduce faults in the circuit, either by disconnecting wires or by inserting malfunctioning components. Students have to troubleshoot the defective circuit and find the faults. Obviously, this is much more convenient and fast than working in real automobiles, where access to components and wiring consumes far more time. In more recent versions, the defects can be introduced electronically, by means of central controls managed by the instructors. There are also simulations of defects in real life automobiles or tractors that have been wired to a computer that simulates the faults. In such cases, simulations are a convenience to recreate in a training environment the kind of situation that is likely to happen in real life. Like in the case of the flight simulator, it packs together, in a short time interval, events that would take much longer to occur 52

spontaneously. If one were to learn in real automobiles how to troubleshoot faults, infrequent defects could fail to appear during the training cycle of the apprentice. In addition it often takes a long time just to remove a component, before it can be tested. A less usual manner of simulation tries to teach manual dexterity, without incurring into the costs of consumables. For instance, arc welding requires a steady hand to keep the electrode at a constant distance from the parts being welded. At the same time that the hand has to move at constant speed, it has to adjust for the distance, as the electrode shortens. This operation requires hundreds of hours of practice, burning expensive electrodes. Machines that simulate a welding machine and permit significant savings in consumables have been offered in the market. But perhaps the most impressive developments are coming from the use of computers to simulate electrical and electronic circuitry. One can use a mouse to pick up electronic components in a virtual storeroom and connect them in any way desired. A virtual battery or power supply is then connected and the circuit energized. Gauges will display the properties of a real system, from turning on a light bulb to far more complex roles. Then, using a virtual multimeter or oscilloscope, the student can make any measurement in this circuit, as if it were a real circuit. The Electronic Bench is the best-known software of this type. With it or with other similar programmes one can quickly assemble an infinite variety of virtual circuits and watch them work. Not only this avoids damages to real world components but the speed of assembly is much greater, even compared to panels where no soldering is required. At the limit, in digital electronic simulations, the student can build a computer that works just like in real life. The parts are picked up with the mouse and connected, creating digital circuits, starting from flip-flop gates, and/or switches and moving up to more complex microprocessors. In so many words, on the screen of a computer one can assemble and operate a computer. The software simulates the hardware. Ultimately, this is no different from a major thrust in real computer design, i.e. the use of software to emulate hardware. As an increasing fraction of the tasks requiring training involve electronic circuits and components, the frontier between the real thing and simulations are becoming blurred. Take imaging, for instance. In the realm of silver halide photography, when shooting a still or movie picture, by pressing the shutter button, one unleashes an irreversible 53

process, consuming film, chemicals and photographic paper. Studio photographers used Polaroid film to check results, before committing themselves to more expensive process with real film. This was already a sort of simulation of results by means of a quick and less expensive alternative. However, in digital photography, the very idea of simulation loses its sharp edges. One can record and re-record indefinitely the images, without consuming anything. Retouching an image or negative with fine brushes required a steady hand and errors could be irreversible. By contrast, working with Photoshop one can "undo" anything and everything. The border between real and simulation becomes blurred and immaterial, by contrast to the tangible difference between the simulation of a CNC machine and the real thing. Overall, by contrast to academic education, always at odds with new instruction technology, vocational training has had a long history of easy-going coexistence with such teaching contraptions. Rejection is less frequent and there is a long trajectory in the use of simulations. Low and behold, not all simulations are equally popular with trainers. Flight simulation has been part and parcel of pilot training. But the highly intriguing and realistic Electronic Bench (or its equivalent) is not so widely used. The same is true with welding simulators which offers significant potential for saving consumable electrodes. One possible explanation is that schools in wealthier countries are under little pressure to cut costs while poorer schools have access neither to simulations nor to information about them. But it may very well be that conservative trainers shun those varieties of simulations that merely reduce costs but subtract from the realism obtained in real life situations. To sum up, simulations in skills training have a long history. Trainers have welcomed them, by contrast to academic education where rejection is so common. Overall, it is a conspicuous case of successful use of technology in learning. Resistance from conservative groups seems to concentrate in areas where the benefits are purely economical rather than in the quality of learning.

54

Chapter II Training Policies that Make Sense

Training in the Developing World: Issues and Policies

This essay is a review of several major issues in training and an evaluation of policies from the point of view of their effectiveness. While examples come mostly from Latin America, the broader focus of the paper is the developing world, under the assumption that issues and problems are relatively similar. A. Economic development without training? There is ample consensus among researchers that education and training have been critical factors of production and that their importance has been amplified by the latest wave of technological transformations that have swept the world. In the second half of the century, Latin America invested in the creation of training systems that have been an invaluable component of development policies. During most decades past, high growth ensured employment for all the graduates. In fact, quantity was more critical than quality. But after a succession of economic crises that began in the 1970s, quality became more important and the fine-tuning of training to existing demand became even more important. This turning point took too long to be understood by many countries where old training solutions persisted and institutional inertia was the norm. The ensuing sequence of economic disasters provoked stalemate and crisis in training institutions. They took a very defensive stand and prepared to fight any encroachment on their territory. Conversely, planning and financial administrators became increasingly critical. But the world did not stand still. The new wave of economic transformations has made training at once more important and more difficult to calibrate to the new and more stringent requirements of the world of work. Enterprises working at the leading edge of technological change have become avid producers and buyers of training. Traditional industries face the threat of open borders, internationalization and fierce competition. Without significant improvements in the quality of their work force, which is required for modernization, these industries risk being wiped out of the market, as has happened in many cases. The new forms of organization that would allow them to 57

survive require workers with much higher levels of education and training. It is instructive to notice that from the 1990s on firms displayed their interest in better qualifications of their workforce by hiring workers with an ever-increasing level of formal schooling. A higher demand for specific technical skills should go hand in hand with this tendency on the formal education level. B. Training policies: general principles Careful observation of the world of training allows us to distil some general principles to understand the world of training and steer it in the right directions. These are not necessarily new ideas, but they have not always been obvious to those in charge of training institutions. 1. Training requires an enabling

environment

The success of any training programme depends on a broad range of circumstances in the country. In fact, most of the policies and actions required to create a favourable environment for the success of training programmes are beyond the legitimate range of its interventions. Usually, they have to do with the political and economic environment. In fact, training by itself will not be effective unless the conditions for the deployment of learned skills are favourable. If such preconditions are not present, it may be impossible to implement a successful training programme. A case in point was Russia, during the last years of communism. The country boasted huge and serious system of vocational education. However, productivity was low and growth stagnated, due to an economic and political system that generated perverse incentives for workers and managers. The first and foremost precondition for training to produce good results is the existence of jobs. Unless jobs are being created, training runs a high risk of being ineffective. Yet, the opposite argument can be made: an overall environment favouring productivity growth is pitifully incomplete if workers lack requisite skills. Thus, even if training does not immediately lead to employment or the labour insertion rates of graduates are less than spectacular, some training may still be justified. This argument has strong implications for the content of training and calls attention to 58

the need to take extra precautions when designing training courses in times of economic decline. If the benefits of training take some time to materialize, improvisation and stopgap policies are not justified. What matters are the skills that last and not all training is equally durable or effective in the long run. If trainees are unable to find employment immediately upon completion of their programmes, then the real value of training may be in its provision of a more durable core of basic skills. In some cases, then, the necessary precautions may include a greater emphasis on basic skills; in others, depending on the clientele, a greater emphasis on education or self-employment may be warranted. It is worth stressing the main message of this section. Training alone is pitifully ineffective. It needs the right economic and political environment to bring results. Therefore, in lean times, it becomes far more delicate to decide what training to offer. 2. Training does not replace a good education

Before setting the groundwork of vocational training policies, it is important to sort out the respective roles of vocational training and education. Although there are conceptual differences between the two, the borderline between training and education is quite blurred. In its purest version, education is knowledge removed from practical applications (e.g. learning astronomy is pure education, except for those who plan to become professional astronomers). At the other extreme, pure training is a version of skill preparation that does not explore the theoretical or technical implications of the tasks being learned (e.g. learning how to use a saw and a jack plane without learning drafting and the requisite mathematics). However, in most cases, theory and practice are combined. In other words, a good electricity programme includes manual skills as well as the study of circuits, electrical theory and applied mathematics. Good training and good education are equally useful - and actually very similar in nature - when they promote the broad conceptual and analytical development of the trainee. Good education is often linked to applied pursuits that give the theoretical knowledge a practical meaning. It may use applied or occupational contents to obtain a deeper mastery of theory, being somewhat unconcerned with the application of the knowledge in the marketplace. Training starts with the clear goal of preparing for an existing occupation, the theory be59

ing a necessary component in the preparation of a better worker for that position. The difference is mostly one of intention. Having said that, it must be stressed that training does not make economic sense, as a substitute for good schools for all. Although training can serve this purpose in limited cases (and can be a convenient way to give context to theory, as explained in the next section), this is not a general solution, given its higher cost and the fact that not all occupations need vocational or technical training. By contrast, a solid basic education is the best preparation for a wide range of jobs. In addition, it shortens the length of the training required. In other words, the need to develop a good training system does not replace the (perhaps) stronger imperative to develop a good general education system. Some orthodox economists have been preaching the notion that all that is needed is a good education. For them, training is something one learns on the job. This is not true since, in many instances, onthe-job learning is not feasible and the complexity of some subjects requires a long learning period and a heavy load of theory. To circumvent a long discussion, let us just note that, not one single developed country has failed to invest heavily and for a long time in training. In fact, all industrialized societies have massive and expensive systems of training. 3. Good training is also good education

Modern economies require strong cognitive development as a foundation for vocational skills. Learning an occupation increasingly requires higher levels of understanding of scientific theories and the technological component of occupations. Part of this education should precede training, thus facilitating and shortening it. But, in addition, most training programmes offered today could benefit from a little more emphasis on language, mathematics and science, as occurs in the best courses and apprenticeships. This is increasingly happening in Germany, the American Tech Prep programmes and the new generation of SENAI courses in Brazil. Recent developments have shown that the integration of skills training with conceptual development is possible. Moreover, this integration creates a learning environment that is particularly favourable to embedding theory. The learning context should resemble the context of application. Learning can be triggered by practical problems that are interesting by themselves. These are the principles be60

hind the "basic skills" movement, as well as "applied academics" and the "contextualization of learning." While learning an occupation, the trainee may have an ideal opportunity to develop the same general skills that are taught in academic schools, i. e. general education. But this will not happen spontaneously. The integration of theory and practice, of shop activities with general principles of science, can only be the result of deliberate and well-informed efforts. Training programmes should not underestimate the potential offered by such integration or the practical difficulties of achieving it. But there are good examples of these ideas: for instance, the new versions of the traditional Latin American "methodical series," as well as new methods developed in countries such as the United States (Tech Prep, School to Work) and Germany (key qualifications) have good track records. Workers with a good mix of practical skills and conceptual understanding of technology can adjust more easily to new and different occupations, grow in their careers, and adjust to technological changes. The real issue is not general versus super-specialized training but the solidity and depth of the basic skills that go together with specialized training. 4. Training pays

Careful studies have shown that good training provided at the right moment to the right group pays back its costs. Actually, those studies merely confirm what Adam Smith said, more than two centuries ago. Training increases productivity and, hence, the income of workers. It also tends to improve their employability and adaptability to different occupations. By the same token, from a social point of view, it is as good an investment as any other, if not better. Well-trained workers can be more productive, as long as they work in an environment that allows them to deploy their higher skills. And an environment with well-trained workers breeds good habits that benefit everybody. An economist would call it external economies. Training consists of imparting not only cognitive skills and dexterity but also in developing the requisite values, attitudes and behaviors, in other words, the ethos which are typical of the occupations taught and essential for superior performance in them. Acquiring the values and the skills takes place simultaneously and with interactions between cognitive and non-cognitive sides. One learns to value an occupation, at the same time that one becomes proficient in it. Being 61

close to those instructors who have the proper values and attitudes of that occupation is essential for that. Beyond preparing competent workers, a very significant role of training is tantamount to a transfer of technology. Technology may be embedded in machinery but it can also be brought to the workplace through training. The best training not only reproduces the skills endowment of a country (or industry or firm) but also upgrades it. This essay takes the position that the chief justification for training is its long-run impact on technology, know-how and productivity. Increases in wages of graduates is just one of the consequences of training. Rates-of-return analysis, while a useful tool for understanding training, is only one of the criteria for making intelligent decisions, given the presence of externalities and long-run effects which cannot be detected by looking at costs and earning profiles. Given the heterogeneity of training modes and skills, and the sector of the economy to which it is targeted, aggregation risks masking the truly important findings. Circumstances may be so fundamentally different and the modalities of training may have so little in common that adding disparate results can lead to meaningless averages. By the same token, the fact that training is successful (or ineffective) in one place or at one level, tells little about its success (or failure) elsewhere. For instance, in a survey I conducted in two Russian cities, 90% of those trained as machinist found suitable jobs in one city and less than 10% in another. To state that, on average, half of the graduates found jobs is worse than useless information, it is misleading. Training targeted to the manufacturing sectors tend to be expensive, long and caters predominantly to men. However, expected growth in enrolment is very modest. Training for the services industries (the full gamut of office technologies included) is more flexible, less expensive and easier to organize. This training grows faster and has a greater participation of women - a clear predominance in some areas. Training to improve the so-called soft technologies in management constitutes an increasingly important area (ISO 9000, Total Quality, Just-in-time, etc.). Training for entrepreneurs and for self-employment is a more recent development and its optimal profiles and success rates are still not well known. Training for the rural sector has features that may approach one of the above modalities but also has its own peculiarities. In dis-

62

cussing training, one is well advised not to mix all these different modalities. Notice that industrial enterprises include service sector activities (accounting, cooking in the cafeteria, etc.), and that the strong movement to outsource services pushes typical manufacturing activities (such as plant maintenance) to the tertiary sector. It is important not to confuse sectors with the intrinsic nature of the training to be imparted and not to extrapolate characteristics of one sector onto another. 5. No demand, no training The number one problem of training is the mismatch between training and jobs. When those who receive training cannot find a job where their skills can be used (directly or indirectly), training is a bad investment. It is always more expensive than education and if the skills learned are not utilized, the benefits are doubtful. Programmes lacking good targeting abound. Therefore, obtaining a good match is a sine qua non condition. The rule should be simple and forceful: no demand, no training. To get a good fit, the incentives have to be right and the mechanisms that drive training to respond to market demand must be in place. The efficient use of resources should be rewarded at all levels by means of hard and soft incentives. Socially useful behaviours within the training system should be rewarded and, conversely, those that are not good should be penalized. Some recent practices deserve attention. For instance, requiring that a certain proportion of the graduates find jobs where they use the skills acquired, establishing the employability of graduates as a condition of financing and providing additional funds to schools that achieve better links to labour markets. Training in independent centres, as is common in Latin America and the Caribbean, tends to pay at least some attention to demand. By contrast, those forms of training linked to the formal systems of education (vocational and technical schools) tend to be greater sinners in ignoring the markets for its graduates. I asked a vocational school principal in a Maghreb country whether he knew where his graduates were. He was surprised with the question and answered that this is something for the Ministry to be concerned, not him. As it turns out, a survey commissioned by the World Bank showed an employment rate below 10%! 63

This may be one of the reasons explaining a strong tendency to move the training end of these programmes outside of academic schools. One of the most widespread solutions is to move training to the post-secondary level, having it become a short higher education career. Another solution is to dilute the occupational component and merely offer a secondary education that introduces students to some broad occupational lines. Still another alternative is to separate the training from the education end of technical/vocational programmes, allowing academic schools to focus on academics and move the training somewhere else.1 This latter solution was adopted by Brazil in the mid 1990s. Demand-driven training does not mean that the training institutions sit and wait for the demand to appear or that they accept the usual reticence of traditional business. Like aggressive firms promoting a new product, training institutions need to market their training and convince employers that training pays. Good training meets the present needs of firms but also takes some steps ahead of present needs in order to make training a channel for change. Like many other economic activities, training needs an aggressive marketing effort. Contrary to current belief, most firms know neither their training needs nor the profile of the skills that would best help them. By the same token, trainers, isolated in their schools, have an imperfect notion of firm's requirements. But working together, schools and firms can put their comparative advantages to good use and develop the right training. Many private technical schools offer firms programmes tuned to their needs and work together to define a suitable curriculum. Worldwide experience shows that quality is paramount to meet market needs. Improvisation and sloppy practices do not pay in the long run. Successful training imparts a sense of perfection and responsibility that cannot be learned unless the training environment itself displays these traits. Last, but not least, graduates of high quality training programmes tend to be chosen for the existing jobs, even if the economy is not creating new ones, hence, from an individual point of view, it is always a good investment.

1

For more details, see Claudio de Moura Castro and Martin Carnoy, "Secondary Education and the Transition to Work."

64

Demand-driven training can lead to a process of selection of courses and people that may secure economic results but at the cost of avoiding those areas and people who, for equity reasons, are more deserving. Administrators have to be aware of this tendency and set up programmes that strike a balance between efficiency and equity. C. Those who benefit should pay

There are good reasons to encourage those who benefit from training to pay for at least a small part of the costs. This makes the budget load lighter and can boost equity. Moreover, those who pay will expect more, be it in terms of quality or targeting, putting additional pressure on providers. But it is quite clear that the ability and the willingness to pay depend much on the clientele and the type of training. One of the most important reasons to charge users is to introduce a self-regulating mechanism in the provision of training. Unless paying users perceive a worthwhile market at the end of the course, they are less likely to enrol. In fact, paid training (both in private and public institutions) is far less likely to generate the classic pattern of mismatches with demand. Employers should pay for that training which generates results they can immediately appropriate and where the risk of losing the investment to free riders is small. This usually means short and highly specialized training or training in which, the employers can capture most of the results. The specialized literature calls this "firm-specific training1." Many firms are conservative, reticent about training or overwhelmed by more urgent demands, failing to benefit from increases in productivity that can result from higher levels of skills. In addition, firms tend to be short-sighted and do not usually foresee long-run economic trends and adjust their training to them. Even so, limited initiatives, such as accepting interns or apprentices, may have positive consequences: some service sector courses should be offered free of charge to vulnerable clienteles, and training for some blue-collar occupations may require some payment.

1

The concept of firm-specific training has been proposed by Gary Becker, Human Capital: A Theoretical and Empirical Analysis, With Special Reference to Education, Chicago: University of Chicago Press, 1993.

65

The presence or the possibility of free riders and poaching severely limits the amount of training individual firms are willing to undertake. This is the classical case of the "Prisoner's Dilemma" where uncoordinated action leads to inaction. No entrepreneur wants to pay for training if competitors can poach their trainees. This leads everywhere to little "generic" training, even though entrepreneurs would be willing to pay if they knew others would also pay. The levy on the payroll is one such "social contract" to get all parties to pay their part. In fact, it was originally proposed by entrepreneurs in Brazil, the first Latin American country to adopt such an approach. Controversies1 aside, the payroll levy also ensures some degree of equity in training systems. Training providers follow funding sources and adapt their offerings accordingly. Barring payroll levies or public resources, training is likely to drift upwards, both in terms of the clienteles and markets addressed. Payroll levies provide a means to keep training focused on the needs of those industries and employees paying the bill Those who take the training should pay as much as they can afford. Middle-class students taking short courses geared to the service sector (secretarial, computing, etc.) typically can pay a significant part, if not all, of the costs. In fact, there are numerous proprietary trade schools offering such courses. Blue-collar workers taking long courses in industrial arts typically cannot afford to pay much, requiring considerable subsidies from the government. In such a case, it will often be pointless to require payment, considering the foregone earnings already incurred and the low level of income of the clientele. But the rules should be flexible, the above are just illustrations of typical differences in clienteles. There are reasons to believe that good quality training generates at least as many external economies as education. There is no more rigorous empirical research to back this statement than there is to support the opposite view. But one should consider that well-trained workers bring sound working habits and techniques to the workplace, and that the best craftsmen are given the task of helping younger staff. In fact, the finest training systems in Europe justify their existence and expenditures on the grounds that they are needed to develop a technological culture. Hence, training should be treated in the same 1

Literature on tax burdens questions the belief of entrepreneurs that the levy impinges on their profits, but this is immaterial for the argument presented above.

66

way as education and a proper balance between expenditures for general education and training should be sought. There are no industrial countries in which the public budgets for training are modest. (For the purpose of the present paper, we are assuming payroll levies to be public funds). This is one of the reasons why training systems should not aim at breaking even through cost recovery. The principle should be to raise revenues as much as permitted by the financial means of the clientele but not beyond the point where the reduction in the amount of training will hurt the economy. Additionally, various schemes should be considered to diversify sources of revenue. These may include student loans for higher-level courses, selling of services to enterprises and perhaps even alumni fund-raising. Public funding of training is not the same as public operation. The first is unavoidable the second is often avoidable. Significant efficiency gains are possible when delivery is in the hands of institutions that can be held accountable to outside agencies or actors. Private training firms, NGOs or the same firms that employ the trainees are likely candidates to deliver training. Also, it may be a public institution, as long as it is subjected to the same incentive and accountability rules that are usual in the private sector. In fact, leading enterprises in the United States are increasingly divesting their training activities to public institutions, such as community colleges, which have comparative advantages in their lines of business and have to comply with performance requirements set out in contracts with firms such as Boeing, Caterpillar and the major carmakers. If such strong accountability can be implemented, there are no compelling reasons to avoid public delivery. But in most countries across the region, efficiency gains through private delivery are still a more feasible alternative. The government always will have a key role to play, both in funding and in policy-making. Training cannot be funded solely by the private sector or from the payments made by trainees, even though there should be an effort to increase their financial participation. The payroll tax has been a common and successful means of financing training. Yet, new styles in public finance resist funding mechanisms that tie revenues to uses. Governments' record in the coordination of training has been mixed. Training boards have been effective in some countries but not in others. The tools to give effective power to boards are not obvious. At the same time, we can only rejoice that some boards have been in67

effective, given their prejudices and poor judgement. I have been a member of a high level board in charge of skills preparation and was delighted with the fact that it was totally ineffective. Otherwise, it would have done more harm than good. As a rule, one ministry is powerless to influence whatever activities take place under the auspices of another ministry. The bottom line for coordination is not very optimistic: there are no winning formulas. D. Finetuning the delivery of training This section presents a number of proposals for improving training and reforming institutions. The examples and the advice provided derive from recent observation, predominantly from Latin America. However, these proposals could be valid as guidelines for other regions. 1. Improving the performance of existing providers of training A few decades ago Latin America sported a dynamic and effective set of training institutions. It was a just cause for pride, and experimental attempts for exporting this system to Africa were made. Unfortunately, some of the training systems have lost their shine, dulled their ability to respond to market needs and become politicized, heavy and unmanageable. With some exceptions, the large training institutions in Latin America became slow, inefficient and expensive, even though they still delivered good quality training. They were in need of structural transformations to make them more acutely aware of the need to better target their output, become quicker to respond to market changes and obtain more results from given inputs. In addition, the reform needed to emphasize the introduction of self-regulating mechanisms. But there was no single formula for reaching these objectives. Reform does not bring about an abrupt or overnight change. Rather, it is likely to happen in stages. The goal of reform should be to move along the lines described below, at whatever speed is possible, understanding that these are not mutually exclusive trajectories. All training institutions, bar none, should monitor closely the market for their graduates. Tracer studies of former trainees — formalized as in SENAI/Sao Paulo or ad hoc — are one of the easiest and most effective means to keep track of market evolution. But other methods 68

are also available and are equally useful. Large, centralized and expensive tracer studies are not necessary. Informal contacts with former students and close interaction with the enterprises that typically hire the graduates can work quite well. What matters is a sense of proximity and intimacy with the market, and the belief that it is incumbent upon the schools to adjust the supply of training to existing demand. However, it must become clear: information is a necessary but not a sufficient condition for change. If penalties for generating unemployed graduates do not exist, schools operators have no interest in getting information or in using it when provided. Thus, the implementation of a policy mandating the monitoring of markets depends more on changing incentive structures (creating penalties for poor targeting) than on the nuts and bolts of generating information. Breaking the monopoly of the large public training institutions is always a good policy. The creation of competitive training funds is a very attractive complement that will increase the efficiency of the system. Respectively in Chile and Argentina, Chile Joven and Proyecto Joven, (described later in greater detail) are prime examples of this strategy. As a rule, there is ample justification for the survival of the "S and I" institutions, but equally strong reasons to make them a smaller player in the training market. They should compete with many other providers but also retain a major role in training trainers, preparing high quality training materials and playing a leadership role. In order to reduce the mismatches between supply and demand for training, employers should be given a stronger voice in the decisions of training institutions. The token presence of business representatives on boards or working groups is not enough. Real entrepreneurs with real power should be involved in making these decisions at school level. For the most part, the participation of unions has been modest in all but a handful of countries. However, this participation is more than welcome and there are good reasons to encourage a greater role for workers' unions. Conditioning some components of the budgets of training institutions to their success in adjusting training to demand can increase efficiency. There are several mechanisms to grant prizes or to reduce budgets according to performance, in particular, by linking funding to the success of the graduates in finding suitable jobs, as the Chile Joven project has done. 69

Under some conditions, training institutions could shift their emphasis to a greater normative role instead of delivery. However, the agency that creates and enforces norms should not also provide training (unless its norms affect sectors where it does not operate). Given the reluctance of the traditional training institutions to withdraw from the conventional delivery of training, standards should be the purview of the government, enterprise associations or of a new agency or board created for that specific purpose. 2. New modes of apprenticeships On-the-job learning acquired a more deliberate and purposeful format, beginning with the establishment of apprenticeships in the Middle Ages. The idea of using the job environment as a learning place and the established craftsmen as trainers remains as valuable today as it was in the past. In Germany, a classroom component was added to the learning-by-doing system, giving rise to the expression "dual system" to denominated the practice of alternating between classroom and work environments. Much effort has gone into reproducing the "dual system" in Latin America and elsewhere. However, the demanding requirements on employers' and workers' associations, as well as on the government, have led to modest results. As an exaggeration, we could say that it is either small and good or big and bad, due to the difficulties of scaling up the system and the complexity of the required articulation between different social players. There are good reasons to pursue apprenticeships, closely modelled after the German format, in sophisticated or up-market areas where the small numbers more or less correspond with the demand, and the complications and costs are justified by the strategic importance of the skills. Yet, the idea of adding some structure, technology and basic skills to existing on-the-job learning and apprenticeships remains quite promising. (Tunisia has adopted a very well conceived programme along these lines but its implementation has been inadequate.) One of the driving forces behind this approach is to add to the on-the-job experience cognitive skills. Subjects such as mathematics, drafting, and reading are poorly, if at all taught in traditional apprenticeships. The public sector provides a significant amount of training to its employees. The problem with training in public institutions (sometimes including parastatals) is that it tends to be seen as an entitle70

ment. Workers are trained because there is a mandate to train, not because there is a concrete need for additional skills. This leads to waste and ill feelings, since nothing happens after training takes place. The rule to fix these mistakes is simple: training in public institutions should follow exactly the same principles adopted by the best private enterprises, avoiding the waste of aimless training and the myopia of conservative enterprises. Public policy should stimulate all forms and modes of learning. This includes regulation, such as the certification of skills or policies to increase transparency, to protect consumers, and to create a market reserve for those properly prepared. In dealing with regulation, we should not forget that the legal framework may kill or boost interesting initiatives. A common situation is the legislation passed in order to protect apprentices. With the best intentions of protecting them, in many developing countries, the burden created for the enterprises discourages the hiring of apprentices. At the very least, the government should refrain from doing more harm than good, by means of ill-conceived legislation. 3. Promoting lifelong learning

Modern societies offer a myriad of education and training opportunities outside the official and public institutions. Proprietary schools offer courses in computers, secretarial skills, accounting, languages and other subjects. Distance education has a long and successful record throughout in Latin America, having been present in the form of correspondence schools and now migrating towards television and the Internet. Radio, television, cable television and satellites are being used for specialized courses and mass education, including extension courses on farming and small business operations. Universities, higher education institutions and technical schools already offer extension courses, both to the public in general and under contract with companies. Last but not least, many serious NGOs cater to the lower end of the society and have played a priceless role. Some estimates, methodologically flawed as they may be, suggest that the amount of informal education and training is several times larger than that offered through academic and formal channels. Regardless of the seriousness of the public sector's efforts to provide training, in most societies, this will constitute a small fraction of the learning opportunities available. In fact, proprietary institutions, NGOs and employers offer various training programmes. Cor71

respondence programmes, television and videotapes offer unlimited opportunities for learning. While there are no statistics for these training alternatives, there are good reasons to believe that in a modern economy the sum of these private and scattered initiatives may enroll several times more students than official public training. Workers learn by doing. The fact of the mater is that the workplace is also a learning place. All policies that encourage workplace learning are welcome. Once workers reach a certain threshold of education or training, self-learning becomes a major source of knowledge and skills, be it by buying books and magazines, listening to the radio, going to the library or through the Internet. Given the proper incentives at work, the availability of self-learning opportunities and the mastery of reading skills, the potential for additional learning is great. Despite the significant variance in quality, the overall impact of this motley set of initiatives is not to be underestimated. There are very good reasons to encourage further expansion of such activities (in some Scandinavian countries they reach as much as half of the population, while in the United States they reach one third). Therefore, countries are well advised to enact policies and create an overall environment in which self-learning can flourish. Protecting consumers from fraud, amateurish efforts and deceptive advertising should be part of the training policies adopted by all countries. But outright regulation and licensing of such initiatives, in most cases, will do more harm than good. 4. New forms of delivery for forgotten

clienteles

Latin America has had considerable success in developing institutions capable of offering high quality training for the classical manual trades. Yet, their record in providing for the lower end of the labour force (both formal and informal) has been mediocre. In particular, the challenge to train the workers for many segments of the informal sector has not been met. The fact is, given their resources and costs, the ability to train a significant proportion of these workers is limited, even if they were to use their entire budgets of existing training institutions. The traditional training institutions may only hope to reach this huge clientele (which is about half of the labour force in many countries) by establishing training technologies which would allow for a dramatic reduction in costs per trainee. Distance education is a pos72

sibility that has been tried at relatively small scales in different countries. The use of mass media, such as radio, television and videos, is another variant that has worked well in the past and shows continued promise (Tecnologico de Monterrey offers training in this manner). The possibility of franchising training to small operators is another path. These initiatives were pioneered by the Brazilian Iazigi Language Institute and informally tried by SENAI in some of its simpler trade courses. In recent years, Microlins of Sao Paulo has put together a training franchise with 730 local facilities for teaching computer skills. Given the reticence of traditional training institutions, ministries of labour are moving in and already sponsoring a significant share of the training for low-end clients. An increasingly popular approach is to have ministries of labour (or local governments) purchase training by competitive bidding in the private and public market, instead of trying to create their own training system or rely on the regular training institutions. Another interesting line of action is given by initiatives launched by some ministries of labour to contract NGOs and other small institutions to offer training in simple occupations. Financing from the ministries of labour could greatly increase the scale of operation of these activities. Other areas of concern include training of older workers and the unemployed. This is, by far, the most failure-prone area. Indeed, the record of retraining these clienteles is mixed. It is relatively expensive and what works in one setting may fail miserably in another. By and large, the experience of industrialized countries along these lines tends to be ineffective in leading to job creation or higher income (see OECD, 1996). Thus, in training older workers and the unemployed, countries are well-advised take a more cautious approach. To minimize risks, the most reasonable policy will be to focus on well-targeted and selected trainees and, in close collaboration with employers, train only for well-specified purposes. 5. Upgrading training for the modern economy

Economic modernization requires increasingly complex forms of training. Training institutions must upgrade some of their courses in order to cater to new needs such as increasingly complex technologies. Technician training, undergraduate programmes and, eventually, post-graduate courses need to be offered in areas such as CAD/CAM, robotics, welding technology and industrial automation. As pro73

grammes with more sophisticated technical components begin to be offered, there is a tendency to also offer services that go beyond regular training. These include quality control, technical assistance and, in the best schools, applied R&D (there are some good examples of this, particularly in Brazil: SENAI and a few federal technical schools). The models of the U.S. community colleges and French IUTs deserve particular attention. Modern training requires many bridges between school and enterprise. Traditionally, there was an abrupt transition between training and the job market. Even when internships were offered, they tended to be formal afterthoughts, not true links between school and work. With the increasing complexity of technologies, schools cannot provide the environment required for learning and enterprises are unable to offer the full range of theoretical preparation which new technologies demand. Therefore, various bridges between training and work have to be established, particularly in the case of more complex forms of occupational training and technical schools. Interesting approaches involve teachers spending time in factories, as well as company engineers spending time in schools. 6. Developing materials and training trainers

The provision of training requires previous and expensive investments in trainers, in methods and in training materials. The most outstanding training programmes are backed by heavy outlays in course development, long periods of fine-tuning the teaching materials and ambitious programmes for preparing trainers. There must be institutions capable of preparing teaching materials, and they must have adequate funding. Training through competitive contracts is desirable to improve the efficiency of delivery and targeting. Yet, such short run contracts do not justify serious investments in training materials or training of trainers. In order to do that, countries need institutions - not necessarily public - capable of operating with longer time horizons and with mandates to provide the pool of materials and instructors required for maintaining high quality training. From the point of view of training markets, such expertise and R&D investments are essentially public goods. This role does not have to be performed by public institutions or even by national institutions. In fact, there are many institutions, private and public, that prepare materials for other countries. They tend to be in countries such 74

as Germany, United States and France that have a long tradition of training and of international technical assistance. 7. Training as social policy but not to create jobs In the last decades there has been a tendency to resort to training as a social policy tool or to justify it on moral grounds. Indeed, training should not be an exception to the imperative that equity must be a permanent target for public spending. But the intention of equity is not the same as effective improvements in equity. It makes much sense to target resources to the disadvantaged. But what matters is not the intentions but the consequences. Unless the trainee can benefit directly from the training received, little is accomplished. Mainstream Latin American training institutions have, to a large extent, failed to reach the underprivileged classes. Their training has been mostly targeted to what one could call a "blue-collar elite," even though they have often done an excellent job in preparing world-class skilled workers. This is no minor achievement. Yet, in countries where the informal sector encompasses close to half of the labour force, there is a moral imperative not to forget this side of the economy. Since the seventies, with the oil crisis, training has often been considered as a tool to fight unemployment. It shows initiative on the part of public administrators and it creates the impression that the problem is being solved. Unfortunately, there is no tangible evidence that training alone creates jobs. Jobs are created when all requisite factors come together, not merely by offering training.. The key idea to keep in mind is that training is essential to improve productivity and competitiveness and, hence, contributes to the health of the economy. To the extent that economic growth creates jobs, training may, in fact, make a strong contribution to job creation. But this indirect and powerful potential impact on growth should not be confused with the immediate impact of training programmes on employment. In all training programmes, however strong the goals of equity may be, there must be a significant probability of finding jobs at the end of the training or within a reasonable period of time. Unless that is the case, general education or other forms of support for the target clientele may be a better idea. While some programmes may fail to reach both equity and efficiency goals, there is usually a point at which the trade off between 75

equity and efficiency will surface, whether we like it or not. To make training more equitable it is usually necessary to make it less efficient (or more wasteful). Whatever the clientele, training institutions have to strive for efficiency. Significant improvements can be obtained, particularly in trying to respond to demand. But there is a limit to efficiency gains without equity losses. There comes a point in which difficult decisions will have to be faced: more efficiency or more equity? As economic crisis hits countries and as unemployment soars, these tradeoffs become steeper. To what extent should training institutions sacrifice efficiency in the attempt to cater to the lowest end of the social scale, i.e. those who need it the most but are the least employable? As public budgets become scarcer in times of crisis, should they be used to train groups of people in which very few will get jobs, instead of responding to the needs of enterprises which require candidates with different profiles? There are no easy answers, and this strategy is not trying to induce any alternative but merely to point out a serious issue in training aggravated in times of economic crisis. In periods of unemployment or where chances of finding employment for all graduates are slim, training should emphasize basic skills. The longer the expected waiting period before a job is found, the greater the chance of a mismatch between the training and the job and more is forgotten of what was learned. Hence, training should be less specialized and, instead, focused on skills that are of a more generic use. Even though the overall record is poor, there have been some positive experiences in Latin America and the Caribbean with training targeted to underprivileged groups. Perhaps the most significant examples are Chile Joven and Projecto Joven in Argentina, both supported by the Interamerican Development Bank (IDB). [see in the last chapter of this book "Monitoring Training Projects: A phylogenetic approach"]. These are classical initiatives in which the potential providers compete for the contracts but in order to bid they have to convince an employer to hire the graduates or offer them an internship lasting as much as the training programme. Both programmes have shown rates of subsequent employment significantly higher than those found in the control group. The challenge facing the managers of such programmes is to strike a balance between the motivation to reach down 76

to deprived clienteles and still obtain a satisfactory level of subsequent employment of the graduates. Self-employment training is the only clear exception to the proposition that training does not create jobs, and therefore deserves considerable attention. Nevertheless, the record is mixed. A careful selection of the candidates is critical for achieving an acceptable success rate. Youth just out of school have few chances of being successful in creating their own businesses. Nevertheless, adding courses introducing all or most students to business practices and promoting self-employment initiatives remains a good idea. Even if few ultimately benefit from the exposure to these ideas, this is an inexpensive policy that makes much sense. Past experience suggest that successful programmes provide far more than training to the chosen candidates (e.g., they also provide financing and technical cooperation). It has also been found that programmes that try to improve the performance of already established small enterprises fare better than those trying to enable trainees to create their own firms. The down side is the fact that training institutions have little comparative advantage in offering more (i.e. integrated training packages) than training. Despite these difficulties, there are strong reasons to pursue policies to promote self-employment.

77

Have Training Policies Changed?

New economic and technological changes suggest that it is not a good idea for training institutions to repeat the same old formulae, no matter how successful they might have been in the past. On the other hand, this paper does not suggest doing everything differently, promoting a revolution in training. What is called for is updating and revising the customary procedures. We always have to ask: Does it still make sense? Perhaps it does, perhaps not. A. General principles are the same

Well-focused training is investment in human capital at its best and is indispensable for economic development. This statement is the rationale for trying to formulate and implement forceful training policies. If there is a second rationale it would prescribe that in training, what matters are the results, not the intentions. Regardless of clientele, training is justified when it leads to jobs; it should not be considered a form of social assistance. But of course, disadvantaged clienteles should deserve careful attention, considering, at the same time, the equity imperatives and the greater difficulties of matching their profiles to jobs. Evaluation and monitoring are critical to identify, to correct errors and to learn the lessons needed to avoid repeating the mistakes of the past. In this respect, it is more convenient, easier and more rigorous to design evaluations beginning at the early stages of the project. The usual enthusiasm and hurry on the part of administrators tends to relegate evaluation design to a stage when it is too late to include the "before" appraisals that are so important in measuring the impact of a project. In principle, policies should favour systematic reforms rather than individual projects. This requires a stronger commitment from critical stakeholders. Policy dialogue with all concerned stakeholders will be the entry door to any project. As discussed elsewhere in this book, reforms have political costs. Unless some groups are willing to pay them - or are not in a position to sabotage the project - not much will be achieved. For a long time, the association of its own employees sabotaged reforms proposed by the Colombian SENA administra78

tion. Ambitious reforms proposed by the IDB in the training system of Paraguay never got off the ground, since political objections were overwhelming. There are no single solutions that are good for all countries or even for the same country at different moments in time. For instances, in the case of the IDB loan to Paraguay, what worked were independent schemes to fund training offered by private centres. This is because they bypassed the bureaucracy of the public training system. Policies ought to respond to specific needs and avoid one-size-fits-all solutions. Needs and priorities follow geography. Nevertheless, given the existence of large, expensive and somewhat dysfunctional training organizations in several countries, reforming them is often one of the more central priorities. B. New priorities are emerging The following areas deserve greater attention in policy design: 1. Reforming training institutions is the number one priority. Training is a privileged form of technology transfer. It is a tool to increase efficiency and competitiveness. Creating new institutions is always easier and politically less painful. However, in many cases, institutions already in existence are sorely in need of reform. Dealing with them can be more arduous but may have a greater impact. Many traditional training institutions throughout the region have an outstanding record but have failed to keep up with changes in the economy and the labour market. While there are fundamental differences in their performance, many of them are embattled and deeply involved in divisive controversies. Legitimacy has been corroded and their efficacy questioned. Some outside observers believe that they should disappear altogether, reasoning that a good academic education is all that is required. Notwithstanding, for better or for worse, these institutions have enough power to repel lethal attacks and, at the same time, avoid the necessary changes. The first aspect to consider is the high costs of the traditional training institutions, regardless of whether they are productive or not. Indeed, in Latin America they typically consume from one to two percent of the payroll, a sizeable sum by any yardstick. If these institutions fail to deliver good and well-targeted training, the waste of re79

sources is considerable. Therefore, money and effort invested in reforming them could have a huge pay off, as it leverages the productivity of the entire manufacturing sector. Past experience shows several cases of successful institutional change. Some Latin American institutions have undergone major reforms or improvements. Peru is a case in point, but there are several others. In other words, reforms are possible and are taking place. On the other hand, attempts to reform by brute force or to close training institutions have failed. Under any circumstance, understanding these entities is a first step in the right direction. This will require careful institutional analysis and the examination of the political feasibility of different scenarios for change. It is no less important to analyse the market for training (tracer studies are the most compelling tools). It is difficult to imagine that major projects in this area can be undertaken without this previous effort. Experience shows that institutions are not changed from the outside. Multilateral agencies can support change but not promote or initiate it. Strong commitment to change by the major stakeholders is a precondition for any initiative in this direction. Activities by international agencies to support change need a constant and intimate dialogue with all those who yield power. Trust between the agency and their partners is essential. It is better to obtain incremental improvements than to risk a stalemate in head-on confrontations with embattled institutions. Let us not forget, changing institutions has less to do with their specific mission than with inherent features of organizations. Hospitals, social security or education, it does not matter. What counts are the common features, such as inside politics and bureaucratic inertia. 2. Programmes to reach the poor

Given the discrepancy between the large number of workers who need training, particularly in the informal sector, and the relatively modest means of training providers, only the deployment of different delivery technologies can offer hope for a solution. Latin America has considerable experience in distance education, particularly radio and television. Hence, at least in this region, it makes sense to create training programmes using distance education, radio, television and satellites that may allow reaching farther, lowering costs and improving 80

the impact of training. Computers, the Internet and other information technologies have also shown their efficacy. For the same reasons, training policies should favour those alternative organizational patterns that lower costs and extend access. This is the case with the selective use of NGOs, for-profit organizations and any other arrangement that may extend the reach of training. Franchising (formally or metaphorically) is a distinct possibility to combine the technical expertise of the traditional institutions with a radical decentralization of training and lower costs. A few informal experiments along these lines warrant further support. The Sao Paulo SENAI is expanding its programmes in this area. By the same token, simplified forms of apprenticeship offer considerable promise. The French AFPA has designed some very well thought-out schemes along these lines, for the Maghreb region. Yet, the rule "no demand, no training" applies here as much as anywhere else. In that respect, outsourcing schemes remains an intriguing idea. Instead of operating training institutions, buying training by competitive bids is a very pertinent approach, even though it is not a panacea or a replacement for conventional training. 3. Improvements in classical vocational training

Despite the emergence of occupational profiles that require higherlevel training, it still makes sense to dedicate much attention to the traditional trades. Contrary to flamboyant articles in the press, we will continue to need mechanics, electricians, welder and pipefitters. Competency-based training, modularization of training and several other technical innovations in delivery deserve attention. Also worth considering are projects to introduce basic skills in conventional training, even though this is a new area and there is lack of inhouse expertise. As occupations became more complex, the ability of workers to write, calculate, read instructions and blueprints, draw and work in teams increased considerably. Even if formal training programmes offer significant preparation in these skills, the majority of the workers learned on the job. Therefore, they may have learned the "hand" side of the trades, but not the "brain" side. "Contextualization of training," "applied academics," "work-keys" and "key qualifications" are part of what has to be developed and implemented, in order to complement the manual skills of existing workers who already have the manual skills and experience. Such 81

programmes only need to deliver the basic cognitive and affective skills. Once the monopoly of the large training institutions is eroded, certification becomes more important since neither trainees nor employers can easily evaluate the degree of preparation of people trained in a wide range of institutions. In fact, in order to promote decentralization and multiplication of independent training initiatives, quality control via certification is necessary. Yet, due caution is required, to avoid the rigidities and undue "credentialism" that might be introduced by certification. The fix can be worse than the problem when it creates a market reserve for those who have the right diplomas but may not be the only nor the best qualified persons for the job. Accreditation of training institutions also has a place, despite the bad reputation of most attempts in the area. But in general, given the high specificity of content and the fragmentation of suppliers, certification of individual trainees is a better idea. Certification is more reliable because it tests outcomes rather than processes. Moreover, since certification is centralized and requires a simple bureaucracy to enforce and it is easier to operate. On hindsight, it seems that the success of the Latin American training institutions has had a restraining effect on the movement towards certification. Since they produced most of the well-trained skilled labour and kept serious standards of quality, this circumstance eroded the efforts to certify. However, informal training has mushroomed and other training institutions have appeared on the scene. These developments militate in favour of a renewed effort to create a certification system. Of course, in critical areas such as pipeline and naval construction welding, ISO standards have become common. 4. Funding the private sector

It certainly makes sense to design policies to encourage diversification in the provision of training, both by enterprises and selected NGOs. It will also encourage self-learning at large and apprenticeships. These stimuli may be tax incentives or subsidies, but above all, a sound regulatory framework, involving issues such as lower salaries for trainees, indenture contracts and so on. The record of tax rebates is mixed. It should be pursued with due caution. The French experience with payroll levies to contract training with outside providers is eminently positive case. It is not too different from the Latin American schemes to divert part of the levies to 82

competitive training contracts (which have found much opposition in several countries). Direct subsidies to firms that train their workers, as done in Chile, are also a possibility. Notwithstanding, attempts to redirect the revenues of the payroll levy to private providers have met fierce opposition from the government bureaucracies. The World Bank has attempted to do it in the Maghreb region, with hardly any results. For the bureaucrats who control the training budgets, it goes against the grain of their ethos to reduce their own budgets in order to fund private operators who will compete with them.

83

Training Dilemmas and False Dilemmas

The training world is rife with dilemmas and false dilemmas. Dilemmas mean that one has to give up something to achieve something else, when both are desirable. False dilemmas give the same impression but contain an error of reasoning or a misunderstanding. This note explores a few of these, encountered by the author, along several years of work in the area. Some of the issues are new. Others are old. Nevertheless, they still mesmerize trainers and observers. A. Training the few or training the masses?

When most training systems were created, the menu was relatively narrow. The task at hand was to train in the classic trades. Turners, machinists, woodworkers and many other occupations were needed, with higher skills or in numbers that could not be provided by traditional apprenticeships. In those countries that were successful in industrializing, progressively, training requirements became more demanding. A lot more theory and abstraction was needed. More basic skills became necessary to deal with technology. It went from workers capable of handling CNC machines to high-level technicians. However, the simple-minded idea that the informal sector was a temporary stage turned out to be false. Even in the most successful developing countries, close to half of the labour force remains in the informal sector. There, levels of complexity and remuneration are modest - but not homogeneous, by any means. Therefore, given limited resources, training agencies faced new and thornier choices. Training gaps remained in all three sectors. Welders and mechanics are still needed in significant numbers (in Brazil, initial wages for pipeline welders are higher than for engineers). Nevertheless, without highly sophisticated technicians, the countries lose their competitiveness. And what to do with the other half of the workers who labour in the informal sector? Latin America experimented profusely with creative schemes to train in the informal, perhaps, more frequently than in any other region. Nevertheless, while some formulae were quite successful, they were not replicated in numbers commensurate with the size of the informal sector. One reason was the high fixed cost of most training 84

agencies, created to cater to modern-industry workers. At the same time, Ministry of Labour ad hoc schemes were not sufficiently robust and did not succeed either. Sorting out the training priorities remains a real dilemma for most developing countries. Traditional programmes drift towards higherlevel training. Those in favor of more forceful social policies want more training for everybody but do not know how to do it. B. Cohibitation of the old crafts with high tech Most European training institutions are quite segregated in terms of the complexity or sophistication of the training offered. Institutions preparing skilled workers are not the same as those training the unemployed or preparing technicians and higher-level workers. A case in point is France, where the youth are prepared in the traditional vocational and technical schools. In contrast, the unemployed and older workers have their own completely independent agency, AFPA. In contrast, in less affluent countries, institutions tend to cater to all levels of workers. In some, quick training for simpler trades are offered, together with technician training. Some even offer postgraduate programmes for engineers. This is a good idea, as it permits economies of scale and a more intense deployment of instructors and equipment. Yet, it often creates problems of status conflicts. Recently, I visited a very good school, preparing workers for natural gas trades. It is a joint venture of SENAI and the Brazilian oil company, Petrobras. In addition, it deals with norms and does interesting research and development in gas applications. A pair of top engineers showed me the school. They took me to their laboratories and research projects. When the tour came to an end, I realized that we had not visited the workshops where workers are prepared to adapt gasoline automobiles to burn natural gas. This was a symptom of the dilemma faced by this school. Obviously, the top echelons of the school do not think training workers is so important. Therefore, we can easily imagine that workers being trained in these simpler trades perceive that they are the lowly end of the school and are treated as second-class citizens.

85

C. Structured training of teachers "creating" their own materials?

Piaget never thought much about education, therefore, never wrote about it. Nevertheless, his followers have taken leads from his theories and gone deep into education. One of the lines partly derived from his ideas took shape in the "constructivist" movement. It would be beyond the scope of this note to delve into the intricacies of this theory. From the point of view of the arguments presented here, we can agree or disagree with constructivism, it does not make any difference. However, what matters for the present argument is the transmutation of the original ideas on the epistemology of learning into an almost religious movement. In contrast to American constructivists, Latin American apostles have become zealots and taken the original ideas much beyond what seems reasonable and deductible from mainstream theorizing along such lines. To them, since students must "construct" their own understanding, each one learns differently. So far, so good. However, the implications derived by this "sect" are quite unreasonable. Educators espousing these ideas - and they are many - are adamantly against textbooks, particularly when they are detailed and stepby-step. They preach that teaching must follow the individual learning styles of students, not the "authoritarian" line prescribed by textbooks. In moments of greater fervour, burning all textbooks has been suggested. This strategy plays havoc with learning, in a Continent where teachers are poorly selected and poorly trained. At the same time, as more and more research accumulates, there is mounting evidence that the more structured and step-by-step the programme, the more students learn. In the last decade or so, as training had to incorporate broader skills in reading, writing, mathematics and science, several training agencies started bringing in educators. And the newly hired teachers brought in their devotion to the idea that teachers have to "create" their teaching materials. Formerly, engineers and experienced technicians organized training. They believed in detailed blueprints and learning sequences. With the influx of the new crop of educators, a lot less attention was devoted to learning materials. In some cases, entire departments that produced handbooks and manuals were disbanded. 86

The new gurus admit training modules. In other words, the teachers preparing their classes can go to the Internet and have access to hundreds of files with models and suggestions. This is the standard answer to those who regret the disintegration of the neatly woven curricula, handbooks and practical activities. The theory might be fine - although this author disagrees - but the practice is not. The ultimate result is a lot more improvisation in classes and in the handouts to students. For some mysterious reason, there is little reaction to this dominance by ideological educators. As far as I know, there is no dilemma here. Instead, there is a plain error. The superiority of training that benefits from structured and detailed materials have been fully demonstrated by research. In contrast, the "new school" styles suffer from repeated improvisation by a vast majority of teachers incapable of preparing quality materials (why should we expect otherwise, in a world with increasing division of labour in all spheres of life?). D. Structured learning curbs creativity?

The standard argument against the structured style of training is that creativity suffers. According to critics, students are given fixed questions and they must answer them precisely. No flights of imagination, no invitation to create, to invent, to speculate, to bring their own answers or to disagree. To the constructivist educators, this is the death sentence to training that offers detailed plans for all activities. This argument contains a logical error. Let me illustrate it with an ill-famed technique: programmed instruction. Some time in the fifties, the idea of sequencing exposition and drills became very fashionable. It was a spin-off from the ideas of B. F. Skinner with his "learning machines". The initial rationale for programmed learning makes much sense. Teach a titbit and test, to ensure that it was learned and the student gets the proper feedback. Only then move forward. This is the ultimate manifestation of structured learning. As it turns out, this method was taken up on earnest to teach computer programming. The task at hand requires knowing the exact answer to questions that are not imaginative or inherently complex. Understand and remember: that is all that was needed. Coming from that milieu, programmed instruction acquired a bad reputation among educators who wanted their students to fly high. 87

We all agree that learning programming languages does not involve high-flying intellectual pursuits. Yet, was anybody asking whether programmed instruction could be also used to teach Comparative Literature? Literature taught by programmed instruction would be better or worse than following the meanders of each teacher's imagination? There is more to this question than can be dealt with here. But we can forge ahead, with due precautions. Taking Bloom's taxonomy, we agree that programmed instruction used to learn computer programming was good to learn facts and rules, and to remember them. But can it be used to apply knowledge, to analyse, to use one's imagination? It is my belief that not only this is possible, but also it is far superior to what can be attained with the run-of-the-mill teachers and instructors that exist today in schools and vocational training institutions. In other words, I want to argue that it is the other way out. Programmed instruction or other structured versions of teaching can better protect the fragile nature of higher order learning. Suppose students must learn the concept of acceleration of gravity. The average teacher, improperly trained, as he tends to be, will go to the blackboard and show the formula. Next, students will get drills in applying it to different sets of numbers. However, they are not learning physics. They are merely training in the resolution of simple mathematical equations. In structured learning, it is possible to bring many real life examples of acceleration. For instance, the students can receive instruction on how to construct real life inclined planes, release balls and time their descent and acceleration. In other words, student can be given real applications. And we know full well that one learns the principles only when one applies them to problems. It is no longer applying numbers to formulae, but looking at problems and situations and being prompted to apply to them what was learned. Surely, individual teachers can do exactly that in their classes. But are they prepared to do as well in a variety of situations, considering the wealth of detailed examples and well-conceived projects that can be built-in the written materials (not to speak of audiovisuals and computer simulations)? We should also consider another advantage of structured learning. Students must be tested. And if they are to learn high order skills, they must be tested on them. It defeats the purpose to try to teach 88

high order skills and ask them to remember formulae on the tests. And this often happens, since it is extremely difficult to create good tests dealing with higher order learning. Teachers have neither the time nor the preparation to formulate such questions. The result is well known, tests ask for recall of what was learned, with no analysis or, even less, creativity. In structured learning, tests can be prepared by experts who have seen hundreds of ways to teach whatever is to be taught and how to test for higher order learning. It is not the superiority of a pile of paper over a real life teacher. Instead, it is the inevitable superiority of a significant planning staff, in which each member is specialized in a segment of the task at hand and having access to the best information anywhere in the world. Last but not least, teams have budgets and time to do a good job. Let me illustrate what a good test can be, moving into a field that is far less concrete than the usual course in vocational training. Consider a real example of a course on Ethics. The teacher guide, after indicating that students should read Plato's defence of Socrates, asks the students to answer the following questions: Was it fair to condemn Socrates? Considering the stalemate faced by Socrates, did he have any other alternative? If the judgement were today, would Socrates be considered guilty? Could the average teacher prepare a test that is as thought provoking, considering how much time he can devote to such tasks? Could he prepare equivalent sets of questions for every chapter of the course? Therefore, for the average teacher or instructor, detailed instruction on how to conduct the class is helpful and welcome. There is a minor controversy here. Educators defending "constructivist" approaches claim that teachers are not challenged when they receive detailed materials. However, when asked, teachers always say that they love whatever makes their classes easier. Suppose we agree that structured materials improve learning, in the case of the average teacher. But what about the highly creative teacher? The answer is very simple. There is no reason to constrain the creative teacher to use this or that material when he can to do better with his own inventions. No dilemmas here, just an erroneous notion on the part of some educators. Structured leaning can help teachers deal with higher order 89

skills. The average teacher can do better with such materials and the gifted teachers can go their own way, if they so desire. E. What if workers lack basic skills? Increasingly, skilled workers need sound reading abilities, as well as the competencies to use numbers to solve practical problems. And they also need to deal with abstraction, with theories, and with higherorder concepts. The problem is that vocational schools recruit students in the lower half of the academic achievement distribution. Therefore, they are chronically weak in these basic skills. Complaining against the public schools is a patriotic duty of good citizens in the developing world. But not much is going to happen in the short run. Vocational schools have to deal with the students they receive. And even in countries like the United States, the proportion of functionally illiterate students entering higher education is alarming (about one third have to take remedial courses). Therefore, training agencies in developing countries face a thorny predicament. In order to offer the training that is required by modern industry, they have to do what primary or secondary schools should have done, but did not. In other words, they have to become also regular schools. Do they have to add on to their programmes the conventional curricula of regular schools? Do they need to become two schools, one vocational and another academic? This is seen as a serious dilemma for training agencies. That they have to teach basic skills is unavoidable. It is not an either or dilemma. Whether they like it or not, it is something that has to be done. Low and behold, the idea of running an academic and a vocational programme, side by side, is ill advised. One of main reason they did not learn basic skills in elementary or secondary schools is the way they were taught. In fact, in the typical bad school, concepts are presented in isolation. Situated or contextual learning is absent - or present only in the official rhetoric of programmes. And as we know full well today, unless teaching is contextualized, real understanding is not achieved for most students, particularly those coming from modest backgrounds. At best, students memorize words, formulae, 90

facts and principles. But since they were not fully understood, they are useless fragments of memory, not tools to work with in the real world. As it turns out, the privilege of vocational training is to offer a natural, motivating and intuitive context to learn concepts and abstractions. The study of dilation of metals remains a vague concept, when learned in books and the blackboard. But it comes to life, when students have to adjust the gaps in the rings of a piston, according to the temperature they will reach inside the engine. Therefore, the idea of two schools is a dead end. It implies that the strongest means to transmit basic skills will not be used in the classrooms. By merging the two, theory and applied situations, students of vocational schools achieve a much better mastery of the concepts, compared to what happens in regular academic schools. The same happens with reading and writing. In a good vocational school, students must read manuals and instructions on how to perform their everyday tasks. And they must write about what they are doing in the workshops. Those learning opportunities lead to a better mastery of language skills than the vague and pointless readings of regular schoolbooks. In other words, the shop manual teaches language skills better than the poetry book of the academic classroom. This is true under any condition, however, language learning can be improved if there is a deliberate effort to maximize those reading opportunities, as well as to test for an exact understanding of the written text. Of course, the best training systems in the world intuitively knew it, all along. If one enters a Mathematics class in a serious electricity course, what the students think they are learning is "electricity maths". This stuff is immediately useful and they perceive it as such. Low and behold, it is regular mathematics dressed up in electrical applications. I visited a course in metal work in the American State of Oklahoma. Students were building a helical stair. They had learned all the calculations required to get it right. What was not told them is that they were learning trigonometry. In fact, one of the teachers told us that if they were to be told that this was trigonometry, their minds would freeze in panic, after having unsuccessfully struggled with that subject in high school maths classes. The best Latin American training systems have done contextual learning for a long time and this is part of their strategy. But when 91

traditional educators organize training, the maths tends to be an addon programme, achieving the same mediocre results that were observed in their prior schooling experience. The same happens with language skills. Therefore, to the extent that learning basic skills in the context of trade training is practised, there is a dilemma only in the sense that resources have to be allocated to this, adding to the costs. But it is not a dilemma when we consider that this is the best way of leaning basic skills.

92

The Cost of Equity and the Cost of Efficiency

Governments and multilateral agencies have a clear mandate to fight poverty. When it translates into training projects, equity becomes a clear priority. One expects that training becomes more equitable. And we also expect it to be efficient and demand driven, after all, resources are scarce and should not be wasted. But, really, can we be super equitable and super efficient at the same time? Can we have the cake and eat it? This note suggests that there is often a trade off between equity and efficiency. To make training more equitable it is usually necessary to make it less efficient (or more wasteful). If we want to be ultra efficient, the recipe is well known: Find the area where there is greatest scarcity of skills. Find the most appropriate candidates to learn the requisite skills. Train them. And in all cases, no demand, no training. This formula will maximize the impact on productivity and obtain the best ratio between costs and benefits. However, equity may suffer. Let us review how this may happen in a few concrete examples. One the surest recipes to make training more relevant and more demand driven are to bring it closer to enterprises and to its constituencies. The Brazilian SENAI, belonging to employers' association is eminently demand driven, since employers force it to respond to their needs. However, the unemployed, the informal sector and the very uneducated youth do not have easy access to its courses. In other words, the price of placing in good jobs an overwhelming majority of its graduates is to cream off the candidates. The programme selects those who have the highest chance of getting jobs. The Chilean INACAP, considered by the government as an inefficient performer, was privatized, its budget being given to enterprises to purchase training, wherever they so desired. INACAP survived and even prospered after privatization. Yet, it went up-market. It started to offer post-secondary courses and it increasingly catered to the needs of wealthier and more sophisticated enterprises. Pre-employment training for future skilled workers dwindled. In other words, efficiency won, equity lost. SENATI of Peru was part of the spoils politics of that country and was considered particularly ineffective. Industrialists put pressure on 93

the government and succeeded in changing the rules for the appointment of its CEO. It became a prerogative of the association of industries to choose the manager of SENATI. Efficiency dramatically increased. However, it also acquired some of the up-market features of INACAP. Equity suffered. Industrialists created CADERH, in Honduras, with the support from USAID. It was to be a counterpoint to the INFOP (Instituto Nacional de Formación Profesional), considered by many as wasteful, politicized and inefficient. CADERH performed well, acquired the reputation of being a serious and efficient institution. However, when the USAID grant came to an end, the institution had to sell its services in an open market and the obvious purchasers of training did not include those people who would need training the most. Chile Joven and Projecto Joven (Argentina), already mentioned, made a serious effort to target its courses to the unemployed, to the poor and to the less-educated youth. It largely succeeded in this endeavour. However, as internships and jobs materialize, employers to some extent "untarget" the candidates, screening them for health and other disguised means to cream off. The Brazilian PROEP - similar in many ways to the Argentinean and Chilean projects - contracted training providers like SENAI and others to deliver training. It targeted the candidates by criteria similar to those of the above-mentioned projects, i.e., the most destitute. Even though some of the courses were delivered by SENAI, it is interesting to realize that this clientele is different from the regular SENAI clientele. The latter is screened, so as to better fit the needs of enterprises. By contrast, PROEP students rank lower in the social scale. While reliable statistics are not immediately available, it seems that SENAI obtains more efficiency and less equity handling the courses paid out of its own budgets and more equity and less efficiency on the PROEP courses. It is interesting to notice that when crisis hit Latin American training institutions from the late seventies on, their usual reaction was to increase the share of upgrading training and reduce pre-employment training, since they were not able to place their recently trained youth. This is an efficiency-driven strategy, but obviously, equity suffered. Surely some programmes can be incompetent both in reaching equity and efficiency goals. But in serious programmes, there is usually a point in which the trade offs surface, whether we like it or not. 94

Therefore, summarizing some of what was said above: 1. Whatever the clientele, training institutions have to strive for efficiency. Significant improvements can be obtained, particularly in trying to fine-tune the programmes respond to existing demand. 2. There is a limit to efficiency gains without equity losses. There comes a point in which difficult decisions will have to be faced: More efficiency or more equity? We have to be clear and explicit about these perplexing decisions. As economic crisis hits countries and as unemployment soars, these trade offs become steeper. To what extent should training institutions sacrifice efficiency in the attempt to cater to the lowest end of the social scale, i.e. those who need it the most but are the least employable? As public budgets become scarcer in times of crisis, should the government train groups of people in which very few will get jobs, instead of responding to the needs of enterprises which require candidates with different profiles? There are no easy answers and this note is not trying to suggest any alternative, but merely pointing out to a serious issue in training, aggravated in times of economic crisis.

95

Training the Poor When There Are Not Enough Jobs

Societies that are growing and creating jobs tend to be relatively happy with their training systems. However, when the economy slows down and less jobs are being created, training systems are heavily criticized, even though nobody has a clear idea on how to fix them or what to replace them with. In fact, this is a serious predicament of training systems. Amongst respected researchers, there is ample consensus that training does not create jobs. Therefore, if in a given environment, jobs are not being created, there are perplexing issues concerning what to do with existing training systems and what to offer to those who are not getting jobs immediately. In particular, there is a gap between the moment a youth typically leaves school - around 14 to 18 years of age - and the age in which he typically start working in regular jobs - after 18. The best general advice one can give about training is the harsh rule: "no demand, no training." However, does it mean that during bouts of unemployment we should close down training programmes and ignore the potential candidates for training? Even more difficult is deciding what to do with the poor who need jobs and cannot find them. This essay discusses these problems and attempts to present some suggestions. There are promising paths and some that, at least, have not been proven wrong. However, it stands to reason that there are no easy solutions for this problem. This is perhaps the most vexing problem in training policies. A. Education with a bit of training or training with a bit of education Before we move to the core topic of this paper - how to train the poor under conditions of economic stagnation - it is worth reviewing some fundamental ideas about training, education and its overlaps. This is because the fuzzy border between the two relates to the essence of our main topic. Therefore, this section is a digression from our main theme, but a useful one. Industrialized countries have traditionally provided the models for secondary technical education. To present matters in the simplest possible way, either we add vocational content to academic training or 96

we add academic content to job preparation. The first alternative is exemplified by the French system of technical and vocational schools where vocational content is added to secondary schools. The alternative of enriching on-the-job learning is the essence of the German Dual system, based on combining real work and school. Of course, there are many other intermediate and variant formulae. Yet, few countries are entirely happy with the way they mix academics, technical and vocational contents. In fact, there are many reforms taking place, including in the most educationally advanced countries. Another instructive finding is that there are no clear convergences in the new solutions. Countries are moving each in their own directions. The idea of adding practical skills to regular schooling is an old one and is found in almost every country in the world. But the problems resulting from this combination are equally old and universal. Concomitant with what level of education should training be offered? If too early, the accusation of premature specialization may apply. If too late, those interested in the trades taught will have already left school, whereas those still enrolled hope to achieve higher status occupations. In the case of secondary technical education that offers good quality education, the overwhelming proportion of its students aspire to university education, thus devaluing the technical trades taught. The expansion of secondary education exacerbated the already persistent ambiguities in preparing youth to become skilled workers. The occupational profiles of the graduates prepared by industrial vocational education were never very clear. As industrial training moved up to the secondary cycle, the focus of such education became even less clear: Should it produce skilled workers? Technicians? Supervisors? Ambiguities in focus have resulted in graduates who are neither competent skilled workers, nor technicians, nor mid-level supervisors, nor competitive at the entry tests for higher education. If nothing else, there is not enough time to handle properly all those tracks in the same programme. As vocational/technical education and training gradually shifted up into the upper secondary cycle, the practical track has become less "vocational" and more "technical." In essence, this has meant less shop training and more academic curriculum, with some of the academics related to technical subjects. Practical training has tended to shift to more "theoretical" preparation in the context of technical education 97

and, within occupational training, from more specific towards much broader definitions of occupations. One liability of this tendency is that graduates are definitely unable to be employed as highly skilled labour. There is no reason to reject outright this model. But unfortunately, most technical schools provide low quality preparation in academic subjects (even lower quality than most academic/general secondary schools). In addition, being more "theoretical" implies that it may become an inadequate preparation for the market and, often, merely offer rote learning of technical disciplines. Instead of using machines, students read handouts about them and memorize the names of their parts and functions. This has less to do with pedagogical theories or something inherent in technical careers than with the fact that these are low status schools, staffed by unprepared teachers who are, most of the time, school teachers who received second rate training in schools which are themselves removed from the realities of the labour market. This happens in all continents - Latin America, Africa and Asia. The changing role of secondary schools, from a relatively elite system to mass education, is prompting reformers to rethink its structure and the kind of teaching and learning that takes place there. Both university-preparatory secondary school and secondary technical education, whether their students ultimately enter university or not, tend to be of a quality that is insufficient to prepare students for a 21st century economy and society. To sum up a complex and controversial subject, even under conditions of relatively prosperous economic systems, the formulae to deal with training and education of the poor remain elusive. B. Wrong education and wrong training If we prepare students for an occupation and the jobs are not there, something is wrong. Offering training under those conditions only leads to waste and frustration. What is the point of spending significant amounts of money to prepare people for jobs that do not exist? It is an unwise expenditure of public money and disappoints those who, after building up expectations, end up just as unemployed - or underemployed - as before.

98

In support of this scepticism, we must repeat that training does not create jobs. Politicians often ignore this and mandate training to those who are unemployed. Substantial resources are utilized and, predictably, nothing happens. This is essentially what happened in Europe and the United States after the oil crisis of the seventies, when numerous and expensive programmes were offered. Ample research suggests that they were largely unsuccessful. When we look at the reality of poorer countries, it seems clear that training is too expensive as a gimmick, merely to remove youth from the streets, while they attend classes. If it fails to help them to get jobs, surely, there are better uses for the resources. Keeping youth in academic schools for a longer time can be a better answer. Education is less perishable than training. Training that is not put to good use right away risks being abandoned and the skills completely wasted. By contrast, education being far more "generic" can be expected to have a longer shelf-life. Yet, in developing countries academic education remains extraordinarily removed from the real world and from practical applications. We are not referring here to the fact that the schools do not teach the skills that can be immediately applied in real factories and offices but to the fact that whatever is taught, it fails to be translated into usable skills in the world of work. In fact, what passes as education is little more than rote learning. C. Promising solutions

Realistically, what can relatively poor countries offer to their poorer youth when the job market is tight? Repeating what was said before, this remains the toughest of the questions facing those who deal with education and training policies. Surely, it would be folly to imagine that this paper is going to bring solutions to this problem. However, there are some paths that are more promising than others. What is presented below is an overview of alternatives that are worth considering. Some programmes contain mechanisms to target the poor for a few existing jobs. Others focus on the idea that education is less perishable than job-specific training, therefore, more suited to those who have no jobs when they graduate from the programme. Still another category prepares for self-employment. 99

1. Scorch the job market and target the poor: the "Joven Projects"

Starting with Chile and followed by Argentina and other countries, the IDB funded a number of programmes in which training was outsourced to any operator who could offer a decent training project and a firm commitment of a job or an internship to graduates. Of course, since the internships were subsidized by the IDB loan, under condition of economic difficulties, it was much easier to find an internship than a job. But the internship had to last at least as long as the training programme. Evaluations showed that thousands of small (sometimes improvised) training firms roamed the country in search of enterprises that would take interns. And this effort paid off because they found many opportunities. Three outcomes make this model a promising alternative: (i) The programmes were targeted to the poor, who have usually greater difficulties in finding jobs and suffer a more dramatic life situation when they do not. Hence, even if they were poaching jobs from other groups, from an equity point of view, this was a good deal. (ii) The internship is at least as useful as the training per se, as job preparation. (iii) Compared to control groups, after six months of graduation, the employment levels of the trained youth was substantially higher. 2. "Practical" education in academic

schools?

To our understanding this is the least explored and one of the most promising paths. School is inexpensive, compared to serious training. Schools are everywhere. School attendance is growing, both in terms of the proportion of the cohort enrolled and the length of stay. Why do schools have be so boring and removed from the everyday life of students? Why the schools that serve the poor cannot take seriously contextual learning? Why can't the curricula of such schools be shorter, simpler and more focused on those academic skills that can be translated into practical skills? Notice that we are not at all proposing that the school teaches "practical" skills or occupations. What is being proposed is a far cry from teaching students how to grow cabbage in the school backyard because this is useful knowledge for the poor. As previously emphasized, we are stressing the practical side of the same academic skills (E.G. learning to use mathematics to deal 100

with everyday problems, rather than mathematics cast in purely abstract terms). Unfortunately, not much is done along these lines in developing countries. It is easier to find good attempts in this direction in the United States (E.G. CORD curricula, Tech Prep or School to Work programmes) and also in German schools (Realschulen). An interesting example of this approach is the Aula Galileu (originated from Spain and adapted to junior high schools in Uruguay). This programme requires an integrated workshop/laboratory that puts students through a series of stations, each of them dealing with one family of skills: wood, metal, electricity, electronics, computers, pneumatic, hydraulics etc. The purpose is not to learn trades but prepare small practical projects in each of these stations (or in more than one combined). The projects are meant to be an opportunity to link the hands-on activities with academic content, such as building a continuity testing apparatus and having students write instruction manuals, both in Spanish and English. At present, the Aula Galileu has been part of a junior high school programme that remains heavily academic in the other courses taken by students. The programme, as a whole, is too difficult and frustrating to a significant part of the age cohort that has dropped out or never enrolled in this level of education. The Methodical Series of SENAI is a heavyweight programme, lasting hundreds of hours and strictly geared to job preparation. However, it takes the idea of project-centred training to the ultimate extreme. The entire set of basic skills, including reading, maths, technology, drafting and scientific contents are built into the projects. There are no stand-alone "academic" courses in these subjects.

101

Is Training in Developing Countries Different from Training in Developed Countries?

Academic schools in the industrialized world are not particularly different from their counterparts in poorer countries. Except in very destitute societies, school buildings look similar and the same is true for classrooms. Curricula and the overall organization of studies tend to follow similar lines. The fundamental difference is in what we cannot see, namely the students learn a lot more in the richer countries. What could we say about TVET (Technical and Vocational Education and Training)? Do the same similarities and differences prevail? The present essay is an attempt to answer this question. By no means is it the result of a careful and systematic inquiry. The basic approach is deceptively simple-minded: What did I see that could help me sort out the issue, in my numerous trips, as an officer of the ILO, the World Bank and the Interamerican Development Bank? What has struck my mind as being particularly different, for better or for worse? A. Many hits, many misses: heterogeneity is the norm My overall impression is that differences in TVET are far bigger than in academic education. This is the case both in the overall structure and organization of the systems and in results. In Developing Countries (henceforth, DCs), academic education is a relatively homogeneous and non-descript mass of public schools, offering uninspired education (the exception are the few schools catering to the elites). However, TVET offers ranges from sublime to totally useless or decrepit. Take Brazil and Argentine. Secondary public schools in Argentine have a long tradition of offering serious education and, despite the ups and downs of the country, they survive with not too many bruises. In contrast, too many Argentinean technical schools are in shambles. Brazilian public education is mediocre at best. Yet, a few hundred kilometres from the Argentinean border, some of the SENAI vocational schools reach European standards. 102

In other words, in TVET variance is much greater. Some courses offer excellent preparation for existing jobs. Others miss the target in many different ways, such as no jobs, wrong clienteles, wrong preparation or bad quality. B. Weak targeting, worse correction of fire

Why would TVET schools in DCs miss the target so often, compared to their counterparts in the industrialized world? The first, off-thecuff, answer would be their poor targeting of markets and students. They miss because their aiming devices are poor. Perhaps so, however, it seems more plausible to postulate that they lack mechanisms to correct their targeting, after a miss has been recorded. Naval artillery rarely hits the first time. What makes it such a formidable weapon is that the miss is used as the information that helps in the correction of fire. In DCs, there is a considerable lack of concern with the fate of the graduates. Schools do not think it is their responsibility to prepare students for real jobs. But can we blame schools individually or are we dealing with a broader problem where the schools are part of a system with logic of their own? Rarely do schools have the mechanisms to trace former students and they make little effort to undertake them. Doing this is not considered to be their job. In fact, they tend to do nothing when they find that the training was worthless or mistaken. When students do not get jobs, this is not considered to be their problem. In two Maghreb countries I am familiar with, tracer studies indicated that less than 10% of the graduates found the jobs they where were expected to occupy. When confronted with such unsavoury results, principals claimed that this was neither their business nor their fault. But whatever the culprit, they were not at all worried about such misdirected targeting. Another chronic infirmity of technical school, particularly in Latin America is the mismatch between what students expect and what the schools offer. Many students enroll in secondary-level technical schools, just because is tuition-free public education. They have no interest in the occupations taught or on the corresponding jobs. Unfortunately, this happens too often. Principals are not penalized when graduates fail to get the jobs they expected. In contrast, some 103

American vocational programmes make teachers accountable for the jobs obtained by graduates. It is a fact that TVET schools in DCs often fail to train the right people for the right jobs. Yet, the courses offered tend to be of reasonable quality. There is considerable expertise and tradition in the operation of TVET. The main problem is the lack of mechanisms for correction of mismatches. In other words, there is a the lack of feedback and accountability. To a large extent, individual schools cannot be blamed for this. The fault lies with weak governments that fail to create mechanisms for accountability and correction of course, along the way. After all, accountability is not a self-imposed agenda. Governments must make schools accountable for their performance. And this is what they fail to do. C. The mind and the hand do not learn together

TVET systems in DCs tend to follow two patterns. There are those that are associated with the Education Ministries and tend to be part of academic systems. In other words, they are often packages offering academic and vocational components of courses that issue secondary degrees of different sorts and levels - more often than not, the graduates are called 'technicians' ("técnicos" or other terms). These schools have the ethos of academic schools. Typically, their vocational or shop teachers can have two different profiles. They may be 'instructors', who are experienced workers. In this case, they earn less, have lower status and are looked down by the white-collar teachers. Their blue coveralls are the visual dimension of their stigma. Inevitably, their low status lowers the aspirations of students to pursue the trades they teach. It is a self-defeating formula. Dozens if not hundreds of these schools are the brainchildren of the World Bank. Building them was the official gospel of the Bank in the seventies. When a large survey of TVET was commissioned in the late eighties, they were shown to be the least functional modality of TVET. To the Bank's credit, there was a 'mea culpa' and an abrupt change in priorities. Alternatively, and far more common, the shop teachers are just regular teachers who received an improvised and second-rate training in the vocational subjects. Not being experienced professionals themselves, they cannot give the students a true professional training in whatever trade is being taught. 104

Lo and behold, the academic education may be correct or even good. But the students are poorly trained and unprepared for the labour marked. A usual consequence is that they become obvious candidates for higher education. To sum up, they are neither competent in the trades taught nor well equipped to compete with other students who could dispose of all their time for academic pursuits. Many technical schools in Latin America and Africa belong to this category. The other extreme case comprises the trade schools, usually sponsored by Labour Ministries or Vocational Training Institutes. They are descendants of apprenticeships or craft schools, originally created to cater to the poor and the orphans. They tend to have decent professionals as instructors, recruiting them in the world of work. However, the complementary education in language, maths, drafting and science tends to be inadequate. Quite often, the students have weak previous schooling background and come from a much lower social extraction. Overall, their status is quite low. Yet, they tend to land in jobs that are closer to the trades they learned. Many Asian countries have such schools, including those in the Indian sub-continent. They also exist in Africa and in NGO-sponsored programmes in Latin America (they are not the mainstream training solution). When we apply this dichotomy to developed countries, we may see that the same broad features apply. There are technical courses for the better off students. They may be weaker in shop activities, as is the case of the United States. And there are the trade schools that recruit academically weaker students - many European and Russian schools are in this category. However, the differences are not so extreme. The shop activities of the academically strong schools are not so bad. And the theory-end of the craft programmes are not so watered down. The 'mind-versus-hand' gap is less profound and dysfunctional. It is as if the schools of DCs are a caricature of those of developed countries. D. Poor students, under-funded schools

Just about everywhere, vocational training is the underdog of education, even in industrial economies. But in developing countries, this is much more so. Training that is attached to the Ministry of Education plays second fiddle to "serious" academic education. It gets the less capable 105

managers and its budgets are squeezed more often. It has less access to the Minister and to the top brass in the bureaucracies. To wit, 'technical' schools are poorer and worse staffed than the academic secondary programmes in the same bureaucracies. This can be immediately seen in the school buildings that are poorer and more derelict. Such differences can be seen, for instance, in Argentina, Brazil (in the case of state schools) and Chile. In addition, there are thousands of trade schools sponsored by NGOs. In many cases, they offer training of respectable quality. However, they tend to be Spartan and operate with very narrow budgets. Paraguay is a good example of a country where the virtual bankruptcy of the official training system lead to the creation of a significant number of well-functioning schools operated by NGOs. In the former French colonies, the local versions of the Office de la Formation Professionelle tend to be substantial institutions, with respectable budgets and competent staff. However, the same internal power politics that created under-funded vocational schools in the ministries of education may be repeated inside the Offices. The more sophisticated technical courses tend to get the lion's share of the budgets. Often, these are the World-Bank-funded schools. At the same time, the lower-status trade training is short-changed. An interesting illustration of that kind of discrimination was observed in two Maghreb countries, where apprenticeship was offered, under programmes carefully designed by the French AFPA. These programmes were far more effective in helping students to find internships and jobs, compared to the regular trade schools. Yet, they were vastly under funded and the implementation of the French project was never completed. Nobody in the Offices paid much attention to them. A striking exception to this 'poor TVET versus rich academic' disjunction is the network of training systems of Latin America, pioneered by the Brazilian SENAI. Being relatively independent from the government and being funded from a levy on the payroll, most of these institutions are quite robust and affluent. In fact, they can offer very high quality training and boast wellequipped schools. Yet, some of them, such as the Paraguayan SNPP, have undergone serious crisis. Most of them have had abrupt ups and downs and some remain lukewarm. But the five sisters of the Brazilian system operating under this model, the Colombian SENA, the 106

Costa Rica INA and the Chilean INACAP are model institutions, with ample resources to allow them to offer the very best training. E. The logistics are more vulnerable

An academic school can function with very little in terms of material goods. Variety can be chalk of different colours but the classes can proceed with white chalk alone. And, at the limit, they go without chalk. However, TVETs cannot function without a wide range of inputs. Therefore, their vulnerability is much greater. Suppose the V-belt of the lathe bursts. Suppose the supply of electrodes comes to an end. Such trivial events halt the process of training. One cannot become a turner with a lathe that does not turn or learn welding without electrodes. Typical vocational schools may have over one thousand different parts and supplies that need to be periodically replenished. Machines need maintenance. Otherwise they break down. The longer they go without maintenance, the more serious and expensive the repair. Last but not least, machines become obsolete. If real world factories are deploying CNC lathes, the school that prepares turners needs some kind of numerically controlled machines, even if they are not the latest or more expensive models. And it is well known that schools in developing countries are weak at logistics and maintenance. The problem is compounded by the extra rigidities of the civil service. In poorer countries, bureaucratic red tape to buy replacement parts in TVET institutions can be formidable. Maintenance habits are hit and miss and appropriations to maintain and repair are chronically deficient. However, tractors last much longer in Switzerland than in Nigeria, even though Nigerians are much poorer and less capable of affording the waste. In other words, there is a wide gap in habits of maintenance, repair and procurement practices between advanced and less advanced countries. And these same differences reappear in TVET and hit them very hard.

107

F. The ever-elusive informal sector

The informal sector in most DCs is quite vast - in fact, often bigger than the formal sector. When most training systems were conceived and created, the informal was considered to be a dwindling sector, doomed to disappear with progress and modernity. Whether progress took hold is something to be argued. But the fact of the matter is that the informal sector kept growing explosively in practically all DCs. In some African countries, it comprises over three fourths of the labour force and no serious observer believes that this situation will change in the foreseeable future. Most training systems were modelled after institutions of the industrialized world, where the informal sector is residual or non-existent. Yet, in DCs it took long to realize that the informal sector was not going to disappear. Under those mindsets, the training institutions took shape in ways and modes that were appropriate to train for the formal sector. The training is geared to modern industry, with relatively sophisticated machinery and extended division of labour. Inevitably, it is costly and rigid. For over three decades, Latin American systems struggled with attempts to retool their training systems, in order to cater to the informal sector. Some observers believe that nowhere in the world were there more experiments and new schemes to cater to the informal. Unfortunately, not much has been achieved in this direction. As a result, countries with large informal sectors are in a bleak predicament. The TVET systems have difficulties in converting their traditional courses to the informal sector. G. One-Stop training for all

Training institutions in the industrialized world were created to offer high standards of preparation for the youth coming out of academic schools - at whatever level this happened. They were pre-service-training institutions and they held tight to their original mandates. This is true in just about every European country. But, on occasion, unemployment went up and workers had to be retrained. Even more important, the pace of technological change increased and workers needed additional training. The European tendency was to create new institutions to deal with adults, in-service 108

training and with the unemployed. The French AFPA and the network of training institutions associated with the German Labour Ministry are clear examples of that. By contrast, in DCs, another path was more frequent. The same institutions that trained youth were asked to retrain the unemployed and to supplement the preparation of adult workers. In that sense, there is a significant gain in efficiency, as expensive staff and equipment are better deployed. For instance, one can walk into a welding school and find semi-literate workers learning the basics of arc welding in a workstation. A job in a back-road welding shop may wait for them. Technicians may be learning about new methods of MIG or gas welding in another workstation. Engineers may be found in another classroom, studying quality control methods in welding. Post-graduate courses for engineers may be taking place, where they learn about x-ray techniques to detect invisible cracks in steel casts. Finally, the school may have contracts with insurance companies to determine the causes for the failure of a critical part that provoked an accident. This is clearly an area in which the DCs are ahead of the industrialized countries. There is much more efficiency and cross-fertilization in this one-stop formula. H. Conclusions? TVET systems in DCs are transplants and adaptations from a few industrialized countries. There is nothing wrong with it. In fact, industrialized countries also copy freely from one another. As we know, all institutional solutions are compromises. You gain here, but lose there. One formula yields more efficiency, but equity loses. Another may facilitate entry to higher levels of education, but they shortchange students in the preparation for the labour market. Ideally, countries could freely choose the models they want to import. In practice, the model comes with the country that offers technical cooperation. In theory, international agencies, such as UNESCO, ILO and the bilateral banks are neutral. In practice, they are influenced by national practices. Germans spent several decades trying to export their famous 'dual system'. They also preferred to support individual institutions, by sending highly experienced instructors to work in them. UNESCO liked the technical schools, very much along the lines of the French 109

system. The World Bank, with a large contingent of American officers, had a special preference for comprehensive high schools, even though, where tried, they turned out to be unmitigated failures. Brazilian technical cooperation supported several projects to support training in other countries. In all of them, they reproduced the SENAI model. There is nothing intrinsically wrong with most of these models. In fact, some of them have been highly successful. What has to be kept in mind is the fact that the degrees of freedom of the recipient country fall precipitously when they choose the organization that will provide technical cooperation. Such organizations tend to be stuck with their own models. And all models are not appropriate to all circumstances. It seems as if the copies and the copies of the copies tend to exaggerate the weaknesses of the original and lose some of their strengths. TVET in poorer countries suffers from weaker governance. The vulnerabilities in the countries where the solutions were created reappear, but enhanced by local institutional fragilities. For instance, the original French technical schools were good in the academics but had relatively weak links with the labour market. Where this solution was adopted, the academics may survive, but the complete lack of connections with the world of work is striking. The purpose of the present essay was to underline differences, not pass overall judgements on the soundness of TVET in developing countries. Predictably, training in these countries has more problems and weaknesses. However, there is also much that is sound and socially beneficial. What is in the future? TVET has had ups and downs along the last several decades. International and bilateral agencies love it and then hate it. In the nineties, it went down, in terms of international loans. The reverberations of these tides hit countries that, in addition, have their own fashions and oscillations. In the recent past, it has gone up again. For those reasons, it seems pointless to try to guess what is going to be the next move.

110

Chapter III The Latin American Panorama

C o m m u n i t y Colleges and Latin America: Clone, Reject or Inspire?

1

This chapter discusses the lessons that Latin Americans can derive from the American Community Colleges. It argues that cloning is impossible and pointless. Rejection is a waste of good ideas and fruitful experiences. Like in other such cases, they should be conceived as a source of inspiration for the region. Community colleges are educational institutions that provide postsecondary training through shorter programmes than are offered at four-year universities. More formally, a community college is an institution of higher education in which the associate degree is the highest degree conferred.2 This paper describes the U.S. experience with community colleges and then discusses them from the standpoint of Latin America, which is preparing to institute similar types of education. Argentina, Chile, and Venezuela, already offer it.3 The trend towards such programmes is already a global one. In Europe4 and the United States,5 community colleges account for at least one half of all graduates of post-secondary education. Should Latin America be any different? Are there any unique local circumstances that would keep us from following along the path being pursued everywhere else? 1 2

3

4

5

Article authored with Andres Bernasconi, Researcher of the Universidad Andres Bello (Chile). In the US the associate degree is granted upon completion of a two-year programme. Canadian community colleges offer also three-year certificate programmes, an option not available in US community colleges. Curiously, these schools are referred to simply as colegios (colleges) in some Latin American countries. While on the surface this might appear an innocent enough simplification, it could lead to misunderstanding on two counts. First, to the extent that college in the United States refers to traditional four-year schools rather than the two-year curriculum that is typical of community colleges. Second, insofar as colegio is elsewhere in Latin America the name of private schools offering both primary and secondary education. Jean-Pierre Jallade, L'enseignement supérieur en Europe: Vers une évaluation comparée des premiers cycles, Notes et études documentaires, N° 4929, (Paris, 1991). The Carnegie Foundation for the Advancement of Teaching, A Classification of Institutions of Higher Learning (Princeton, 1994). 113

Probably not. The risk we run, however, is that of mixing and matching models that do not reflect our specific situation and needs. The prudent strategy would be to examine the best experiences and learn from the lessons they hold out to us. In this regard, a recent article by Daniel Levy on isomorphism in education is especially enlightening.1 The author observes that there is a tendency among institutions of higher learning to imitate each other (both in the same country and elsewhere), either because they are required to do so by law, because the seek vicariously the legitimacy enjoyed by older institutions, or because they lack imagination and initiative. Although the article's comments are not directed specifically at Latin America, they apply perfectly to our situation. Schools in the less affluent suburbs imitate upper class schools; and private colleges slavishly and mindlessly copy public ones, with few exceptions. They might imitate because the government has instructed them to, but at the same time, they do not protest nor do they seek new formulas that might be better suited to their structure and means or to the labour market. When complaints are lodged, they are rarely accompanied by proposals for alternatives and, even then, there is no commitment to fight for them. So, if the winds of change are indeed blowing in the direction of shorter curricula, we would do well to devise our own system correctly from the beginning. Otherwise, we run the risk of copying what is not appropriate for our end of the world. Clark Kerr—one of the mavens of U.S. higher education—views community colleges as the century's most innovative experience in the educational field. It would therefore behove us to draw the right lessons from this rich and valuable experience. However, it is not possible or desirable to copy all the details of the American model. The private sector will not be able to replicate the system outright, given the likely absence of public subsidies for many of the activities involved. At the same time, it appears it will not be possible for the public sector to do so either, given the lack of flexibility and entrepreneurial spirit among most public institutions stifled by ill-advised, Byzantine legislation. Still, we have much to learn from community 1

Daniel Levy, "When private education does not bring organizational diversity: case material from Argentina, China and Hungary", in Private Prometheus: Private higher education and development in the 21" Century, P. Altbach, editor (Boston: Center for International Higher Education, Boston College, 1999). 114

colleges even though we would not want to imitate many of their features. In order to fully understand the community college system, we need to bear in mind that, even in the United States, it represents a category of institutions that has been the subject of very limited study.1 The elite universities have focused their research on elite universities—not on community colleges, which remain almost invisible. In the heated debate on four-year programmes, community colleges are neither criticized nor praised. The magazine U.S. News and World Report, which ranks institutions of higher education, doesn't even include community colleges in its rankings. This situation is reflected in students' views as well: "I couldn't go to a 'real' college, so I went to [a community college]," were the words of a student from Northern Virginia Community College.2 Community colleges are in fact snubbed by the elite universities and suffer from a chronic status problem. The left views them as performing a cooling out function, as if they were some sort of consolation prize or bone thrown to the poor.3 They are accused of being envious of four-year universities and of trying to imitate them, which might be true in some cases. There is no denying that some of them suffer from the so-called "academic drift", that is to say, a desire to copy and act like prestigious universities. While this may all be true, it does not distort the general picture, bearing in mind that there are also trends in the opposite direction, i.e., community colleges that deliver much appreciated practical education to a clientele that best benefits from it. In Latin America, the picture is even bleaker. Shorter programmes are viewed with outright disdain, when they are not simply ignored by legislation focusing solely in university degrees, as the case is in some Central American countries. In Chile, Latin America's most advanced country in terms of structure of higher education, the Presidents or Rectors of these programmes are not allowed to sit on the 1

2 3

Thomas Kane and Cecilia Elena Rouse, "The Community College: Educating Students at the Margin between College and Work," Journal of Economic Perspectives, col. 13, no. 1, Winter 1999, pp. 63-84. Many of the statistics and other data used herein were drawn from this article. Jim Naughton, "Super NOVA" (Washington Post Magazine, November 14,1999). See Michael W. Apple, "Ideology, Reproduction, and Educational Reform," Comparative Education Review 22 (1978): 367-87. Also, Jerome Karabel, "Community Colleges and Social Stratification," Harvard Educational Review 42 (1972): 52162.

115

board of Presidents, thus creating a situation in which higher education is discussed and decided upon without them even being present. A similar situation exists in Venezuela: they do not sit on the board of Presidents and are not invited to meetings with the Minister at which higher education is discussed. A. An overview of Community Colleges 1. Origins: Junior Colleges At the beginning of the twentieth century the Joliet Township school board authorized the local high school to offer the first two years of post-secondary education. The example of Joliet was soon followed by the state of California, which first authorized high schools to offer the first two years of college, and then in 1917, recognized the right of school districts to establish separate public "junior" colleges. Several other states followed suit during the twenties, and thus the junior college was born—later called "community" college to avoid the negative connotations of "junior" and to stress its community-based mission.1 One of the reasons for the emergence of junior colleges was the effort to facilitate access to higher learning without overburdening the four-year programmes. This, of course, was possible in the U.S. system, where most four-year college graduates are not offered a programme preparing them to the specifics of the workplace. The first two years of post-secondary education are virtually identical for all four-year programmes, regardless of the speciality (or major) selected for the last two years. It should be noted that - contrary to what happens in Latin America - a major refers solely to an area of concentration and not to a true preparation for the labour market, which is actually what U.S. master's programmes do. For the most part, junior colleges were small schools with a viable scale of operations, precisely because the first two years of almost all majors were the same. Small cities, for instance, might have a junior college but no four-year programme because they lacked the infrastructure or the number of students to justify offering the various disciplines required for the third and fourth years.

1

Vaughan, George, The Community College Story, the Community College Press, Washington, DC, Second Edition, 2000, p. 23.

116

With the return of soldiers and the creation of the G.I. bill,1 after the World War II, demand for such schools mushroomed, since most of the demobilized soldiers did not display the typical profile of college students. Millions of vouchers were distributed and enrolment in these schools doubled between 1944 and 1947. In the period since then, enrolment at community colleges has been increasing at a much faster rate than at four-year schools. At the same time, community colleges have undergone a fundamental change in their structure and functions, as will be discussed below. At the turn of the Millennium, the US 1,132 community colleges had a combined enrolment of 5.4 million (10.4 million if not-for-credit students are included), representing 44% of all undergraduates, 38% of all post-secondary enrolment and 45% of enrolment at state-affiliated education. With their shorter programmes, these schools account for one half of all post-secondary diplomas. Worth noting is the fact that 95% of post-secondary enrolments of short duration are in public institutions.2 Only 12% of community colleges in the US are private, which in large part is due to the various roles played by these colleges, which have to rely heavily on government funding. The profile of students and teachers at community colleges is quite different from the four-year schools. There is a high percentage of part-time students (64%) and, more importantly, 84% of the students have jobs. The average age of students is 29, and 33% are 30 years of age or older. Over one half the students attend courses for periods of one year or less (there are many degrees that can be earned in just one year). The percentage of teachers working part-time is also high (two thirds). Many courses are offered in the evening or on weekends. Great effort is made to offer courses close to where students live or work, by means of branch campuses. For instance, Strayer University, which offers mainly one- and two-year programmes, has 13 campuses in the Washington, D.C., area. Lastly, community colleges offer more distance-learning courses than do four-year programmes, which con1

2

A system of public subsidies whereby the federal government made monthly payments to demobilized soldiers who wanted to pursue post-secondary studies. Even so, there are an extraordinary number of similar courses that are offered by private educational institutions and companies, although they do not offer equivalent certificates. 117

firms that they cater to older students who are already part of the labour market. The inner workings of community colleges will be discussed further on. Here, it is enough to appreciate that the system is very extensive (over 1,132 institutions) and the target clientele is definitely of more modest means that in the "normal" educational system, to the extent that—for these students—education is an activity that is concurrent with working. Students are older and study part-time. Teachers are less academically oriented and also work part-time. Interesting, too, is the fact that some of the less prestigious universities now offer short programmes conferring associate degrees; for all intents and purposes, they are running the equivalent of a community college within their walls. Text from an advertisement November 14, 1999:

in The Washington

Post,

Feeling trapped by your current j o b ? Well, shift gears - enrol in our Information T e c h n o l o g y or C o m p u t e r Science Programs. Pursue your associate's, bachelor's or master's degree in information s y s t e m s or computer science. Get certified in Oracle, Unix, C, C + + , Year 2 0 0 0 C o b o l , Visual Basic, w e b development and more. [...] Southeastern's experienced faculty of industry professionals embrace a hands-on approach to teaching. Southeastern is a private, accredited university that is affordable a n d conveniently located close to metro stops. Day, evening and w e e k e n d classes are available. Call today to register. S O U T H E A S T E R N UNIVERSITY

2. Financing Unlike the situation with conventional public higher education, community colleges have a differentiated financing model. With some variations, the classic formula is 40% funding from the state, 10% fed-

118

eral funds, 20% local funds, and 30% from students' tuition and contract services.1 Payment of tuition is a universal principle in U.S. higher education (in both the public and private sectors). In this respect, community colleges do not add anything new. State-government support also is a standard feature of public universities. The component that is most different is the local government's participation, and this is where the word "community" comes from in the expression "community colleges". Given the strong role of community participation in the preparation of local government budgets, this is one of the sources of legitimacy and responsiveness of these colleges to local needs. Community colleges are answerable to the community. And the community is a tough customer, one that demands results. At a dinner attended by one of the authors, offered by Amarillo Community College (Texas, U.S.A.) in honour of local community leaders, the school administrators were visibly worried about making a good impression. At the dinner, the school's results and plans would be presented and there was concern that the community would be sufficiently impressed which was required to approve the school's new budget. Indeed, local government officials have to be convinced that the college is doing a good job. The overall cost of education at a community college is roughly US$4,000 per student per year (dollars of year 2000), which is equivalent to one half the average annual tuition at public four-year institutions. Community-college students pay on average US$1,500 per year. As with four-year programmes, there is an abundant supply of financial aid, visibly in the form of student loans, provided mainly by the federal government. There are also ample opportunities for scholarships and tuition discounts. The system is admittedly geared towards a lower-income clientele, and its accessibility in terms of price is one of its most appealing features. One factor that helps keep costs down is the use of part-time teaching staff. Many teachers have other jobs as well, often better paid and more prestigious. Community colleges take a clear and deliberate position not to promote research systematically. In the words of the president of 1

In Canada the typical funding breakdown is 60% provincial sources, 20% local support and 20% tuition and services.

119

Northern Virginia Community College, "I think there is a place for researchers and thinkers ... and a place for the worker bees and the folks who want to get down and dirty with it." Only 4% of community-college teachers are engaged in research. Community-college instructors are also less academically qualified than their counterparts at four-year institutions: most hold a master's degree but not a doctorate. In fact, community colleges, as a rule, openly avoid hiring Ph.D.'s. Even when such candidates are available (which, actually, is often the case), they are felt to lack the experience and motivation required for a course load as heavy as 5 courses per term.1 According to some administrators, after spending so much time preparing a dissertation, Ph.D.'s would become frustrated by the student profile and the non-academic atmosphere. Notwithstanding, the eschewal of Ph.D.'s is not universal. In fact; many community colleges tap them for manning their faculties of arts and sciences, where students who want to transfer to four-year colleges enroll for a liberal arts education. The "academics" of a community college are, therefore, catering to those students who plan to transfer to a four-year institution. Remedial education is done by adult educators, technical programmes are taught by professionals in the field who, in addition, may have a master's degree in education administration or similar. The vocational trades are the provinces of practitioners who have undergone practical instructional training. 3. Economic benefits for students

Sufficiently reliable data are available that allow for a comparison to be drawn between the performance of community-college graduates and other students who either have not completed post-secondary education or have graduated from four-year programmes. Without a doubt, more education makes a difference. Students who have completed part of the higher education coursework but did not finish their degree programme earn 10% more on average than workers with only a high-school education. Each additional year of coursework completed brings an additional 5%-10% in wages. Compared with four-year programmes, the results are almost proportionate, to the extent that they are equivalent to

1

Union rules limit teaching in Canada to a maximum of 15 hours out of a total of 37 per week.

120

the sum of the increases accruing from each year of study at a fouryear-college. Looking at the particular situation of courses offered at community colleges, they also correlated with a 15% increase in performance by employed adults. More importantly, the initial gains do not dissipate over time. In sum, the economic results are very respectable and equivalent per unit of time to those of four-year-colleges. In addition, they begin to be felt after only a few months of attendance. It is precisely such rapid returns that make these courses attractive to students who are not able to devote much time and money to higher education. 4. The controversial role in social

mobility

Community colleges may be a consolation prize but they are, just as well, a powerful vehicle for social mobility, serving a clientele that is unable to attend four-year programmes. Typically they attract students who are not able to support themselves over a four-year period, especially for a degree that does not actually prepare them for the workplace. They also respond to a clientele that is not interested in or is uncomfortable with theoretical, abstract studies, which are often a component of the longer programmes. Community colleges, furthermore, have a higher percentage of students who are the first ones in their family to attend an institution of higher learning. Community colleges very deliberately tailor their mission and modus operandi to this clientele. Studies have shown that they offer more personalized attention and that their teachers are more dedicated to their students. Survey courses are avoided and the number of students per class is generally kept low. They understand that there are serious problems of motivation among a clientele that has an uncertain and ambivalent attitude towards education; after all, these are emerging groups in the area of higher education. Moreover, these students are academically less gifted and only graduated from high school because most U.S. states do not have minimum performance requirements for graduation. This is an astonishing and not very complimentary feature of the U.S. secondary education system. Students who have problems in a given subject area are channelled towards other, easier subjects that are more practical and less demanding, where it is almost impossible for them to fail. The statistics speak for themselves: 41% of community-college students have to take remedial courses, some of which are actually at the 121

level of functional literacy and basic arithmetic. Many students have not achieved an eighth-grade reading level and thus have to take additional remedial courses. English grammar, for instance, can be taught at a level as low as the seventh-grade.1 Be that as it may, this is the clientele served by community colleges. There is a clear policy of open enrolment for all: A high school diploma is sufficient to ensure acceptance and, for some programmes, not even that is necessary. In fact, high school equivalency programmes for school dropouts constitute a main line of activity for many community colleges. The strong point of these schools is that they work realistically with their clientele, rather than design their courses on the basis of some dreamed standard. Given the current state of the U.S. secondary education system, the role of community colleges is to offer education to all comers in an attempt to make up for that system's shortcomings. And therein lies their worth; it is also what makes this experience instructive for Latin America, which shares these low performance levels in a significant share of its secondary education students. In order to cater to this clientele, there is considerably more concern with teaching methods and innovations to properly serve this clientele. As Clark Kerr pointed out, these are virtually the only institutions of higher learning that produce solid advances in teaching methods, e.g., in the relevance of subjects taught, use of computers, applied academics2, etc.). There is, however, a price attached to this. Students who begin their higher education at a four-year programme are more likely to complete their programme than are students having the same profile who transfer from a community college after their second year. Community colleges have a different ethos—students are less driven towards completing a programme—and this gradually erodes their motivation to continue. On the other hand, there is evidence suggesting that students that do obtain their associate's degree in a community college and then transfer to a four-year college (some 35% of all community college graduates) do better in their junior year than their peers who initiated their studies in the four-year college.

1 2

Dante Chinni, "City College", Washington Post Magazine, ibid. See Dan Hull, Abre tu mente e abrirás las puertas del futuro (Waco, Texas: Cord, 1999). 122

Community colleges are institutions of contrast. If, on one hand, they admit high school drop-outs looking for a second chance and high school graduates unable to make the cut for a university, on the other hand, they also recruit fast-track high school students wishing to enroll while they are still in their junior or senior year of school, in what is called "dual enrolment." Certainly, the customers using this option, as well as the growing numbers of graduates of four year programmes and people with graduate degrees who seek applied technological training not available to them in traditional academic programmes are not the "academically challenged" clientele commonly associated with community colleges. Many researchers and members of universities that are more academically oriented have levelled serious attacks against community colleges, accusing them of undermining students' ambitions and funnelling them directly into the job market (i.e., cooling out poor students by channelling them away from conventional forms of higher education). Many of these same people forget, however, that most of these students wouldn't even be in higher education if it weren't for community colleges. So, while community colleges may indeed stunt the academic careers of a few students who might otherwise have completed a full programme at a four-year institution, the truth is that most community college students would not be studying at all if it weren't for the practical and immediate nature of the education offered by community colleges. B. The changing functions of Community Colleges The discussion thus far has focused on describing community colleges from an outside vantage point, without delving into their course offerings. Actually, the appeal of community colleges lies in their evolution and new roles that they have taken on over time. It can safely be said that their original function of offering the first two years of postsecondary education is but one of the benefits they offer nowadays, especially from a Latin American viewpoint. 1. From Junior Colleges to Multi-Purpose

Institutions

As indicated before, the original purpose of these schools was to allow students from smaller cities (or those with less preparation) to complete the first two years of post-secondary education in their hometown. Since course content at this level was basically the same for 123

all majors, there was sufficient economic justification to operate with the smaller enrolments than would be required for the multiplicity of careers of a four-year college. Gradually, though, the notion and practice of the "comprehensive college" has emerged, a college in which the transfer function coexists, often not as the primary function, with technical and vocational training, remedial education, contract training for firms, continuing education, and recreational learning. The associate degree in liberal or general studies and humanities still constitutes by far the most awarded associate degree with 167,000 degrees conferred in 1996-97, but is a modest share of total degrees awarded. Consider the 77,000 degrees in health professions; 72,000 in business management and administration services; 20,000 in engineering-related technologies and 17,000 in protective services1. Actually, community colleges train today 2/3 of all registered nurses in the US. Community colleges, it is true, offer a diluted level of academic preparation; they cannot compare with the rigour and demands of four-year programmes. But for many students, a more demanding programme is beyond their time availability and academic capabilities. However, the more personalized attention and interest that community colleges can offer such students should not be underestimated. Many go to these colleges precisely because the setting is unpretentious and better tuned to their needs. But this is not what is most impressive about community colleges. Kerr would not consider them the greatest innovation in higher education if all they did was offer two years of watered-down, general studies to economically underprivileged students. In fact, academically speaking, they offer nothing special at all. For someone who wouldn't be able to study in some other kind of school, it could be a major step forward. Overall, however, it is nothing extraordinary. 2. The new path to vocational education Nowadays, the core function of community colleges is vocational education. It is the sphere in which they have grown the most in recent years and which has given them their strongest identity. At the beginning of the twentieth century, the United States took a drastic step, opting for the so-called comprehensive high schools. The voices 1

American Association of Community Colleges, Pocket Profile of Community Colleges: Trends and Statistics, AACC, Washington, DC, third edition, n/d, p. 8. 124

of those who wanted to see vocational schools established as separate institutions from the academically-oriented schools fell on deaf ears. The European model of separate technical and vocational schools was flatly rejected. The decision was taken to create high schools offering various options under a single roof, depending on the skills and interests of the students. Students who are more academically inclined take more science and maths courses and usually choose more difficult courses. For less ambitious students, less demanding courses are available. More significant, however, was the country's decision to entrust vocational training to its high schools, in effect, having a single institution look after all the various aspects of education and vocational training that are offered at secondary level. Such a system is not without its attractions, especially with regard to vocational education. For a country in which manual labour has always enjoyed high social esteem, the commingling of manual and intellectual activities is less problematic than in most other societies, where differences of class and social category are much more distinct (including in Europe). Under the United States' sphere of influence, this "comprehensive high school" model was exported—but turned out to be a major disappointment almost without exception, despite its relative success in its country of origin. (The World Bank tried to promote the model in Latin America, but was unsuccessful). The success at home was not unqualified, though. The fact that some academic subjects enjoyed greater status, coupled with the tendency to push students who did poorly in the sciences and humanities towards vocational training, undermined this role of the high schools. With the increasing complexity and theoretical development of many technical professions, the amateur-level training provided at high schools has lost its relevance and ability to respond to the economy's needs. Gradually, community colleges have been moving into this niche. Vocational training through two-year programmes at the post-secondary level is steadily taking over from the vocational courses at secondary schools. Vocational training has migrated to higher education and community colleges have been at the receiving end of that migration. This is the valuable niche that community colleges occupy today, and this is where Latin America needs to look for useful lessons. "What in the world could we be doing that is more important than getting people ready to make a living? It's got to be greater than 125

studying culture and the arts. It has got to be linked to how the person makes a living and feeds their family1." Also noteworthy is the fact that most major pedagogical innovations are hatched at community colleges. Granted, not all of them are foci of innovation or creative initiative, but if there is one place where such innovations do exist, it is the community college. Conventional four-year programmes tend to repeat conventional teaching methodologies. And the higher the status of the institution, the more conventional the methodology; there's no use in looking for pedagogical innovation at elite universities. Harvard and Yale, for instance, work almost exclusively on the basis of blackboard and chalk. The Media Center at MIT is the world's most advanced laboratory in terms of state-of-the-art teaching technologies. In the classrooms next door, however, the professor holds forth on the latest theories by consulting notes scribbled on pieces of paper and copying words or formulas onto the blackboard. Community colleges, on the other hand, are a paradise of new teaching technologies—and of innovations as well. Some, in fact, do not have a single traditional classroom. On the first day of classes, students are assigned to a workstation equipped with a computer and a course outline showing the order in which their class work is to be done. The teacher walks around among the students and discusses individual problems with each one. Those who advance more quickly finish the course sooner. Around 2000, some 400,000 students were taking courses via distance learning, making ever-greater use of the Internet. Business incubators are being set up, schools are organizing practicums or internships at local companies, there are even businesses that are run by schools. The schools track the market closely, following developments and adjusting their course offerings accordingly. Decisions on what courses to offer hinge on input from business committees (each committee representing a specific branch of industry), and businessmen are often at the helm of the schools. Courses can be quite sophisticated, such as a pilot plant for manufacturing semiconductors; the same school, however, could offer courses in shoemaking or leatherworking. If there is a market, there is a course for it. The corollary is even truer: if there is no market, then there are no courses. Since students always have to make some sort of 1

G. Baker in Jim Naughton, op.cit. 126

payment (usually one third the total cost), enrolment is a good market indicator. It is no secret that students are more interested in the market when they have to pay, even if only part of the cost of their education. In summary, anybody who wants to see innovation in pedagogical methodologies should not waste time going to the prestigious universities: community colleges are the place to go. 3. Contract training In conjunction with the vocational training courses offered directly to students, community colleges increasingly are entering into agreements with companies to train their future (and present) employees. Ninety-five percent of community colleges have such arrangements with companies or the government. A full one fifth of all communitycollege students are studying under contract with an external organization. "We are affordable. We can design programmes quickly. We can do it faster for less money and we still have quality."1 In a way, they've become huge capitalist enterprises that sell courses (in this sense, they are not much different from the best Latin American technical schools and universities). Obviously these contracts help to bring them closer to the companies and place them in a privileged position for understanding their needs. On a recent visit to a community college, it was possible to see, from the doorway of the auto mechanics class, a building that had been put up by Nissan, another by Ford, another by Toyota, and another by General Motors. Each building houses a school that trains technical and management staff for that company's dealerships. In the Ford building, students take apart Ford cars, study Ford manuals, use tools selected by Ford and follow syllabuses prepared by Ford. The same happens in each of the other buildings. The company decides on the course content, class times, and materials. Some even have paper shredders in the classroom, for instructors to destroy the manuals after class and thus ensure confidentiality. The trend among major U.S. companies today is to close their training centres and outsource everything to community colleges. Joining the automobile manufacturers, Caterpillar and Boeing have already moved in this direction. 1

David Pierce, President of the American Association of Community Colleges, in Jim Naughton, op.cit. 127

4. Adult Education On a recent visit to an avionics maintenance shop, visitors saw that elderly men and women upholstering chairs occupied the room next door. Others were weaving baskets, and still others were learning embroidery. Given the community-based nature of financing and decisionmaking, community colleges play a very important role in offering courses to students of all ages. This is clearly a social function and is of inestimable value. It is not vocational training for a profession, but rather a hobby, occupying the time of retirees through veiled or open methods of occupational therapy. Considerable time and resources are devoted to this activity. It is not uncommon for a community college to have 20% of the local population registered for courses in any given year. Course catalogues are prepared each semester, covering a wide variety of interests and duration and are mailed out to the entire community. C. Good for the United States, good for Latin America? A famous American businessman once said that what was good for his firm was also good for America. Can we say the same in the case of community colleges for Latin America? The safest answer to this oft-asked question is that, in principle, no. There is no single model that applies everywhere. The best short statement on the subject is that they seem to be an excellent source of inspiration but not a blueprint. That is to be understood rather than regretted. There are no good reasons to imagine that one can or should transplant an institution from one culture to another, even though we can learn much by studying them, as is the case with community colleges. The first tenet of this paper is that community colleges offer interesting ideas for education in Latin America and should thus be the subject of careful study on our part. The second point is that we are only able to tap into some of the features offered by community colleges, basically because it would be impossible to replicate them exactly in our countries. Our public sectors lack the necessary flexibility and dynamism to operate such a comprehensive and flexible institution. And without public subsidies, our private sectors would be unable to do everything that community colleges normally do (nearly all of which are run by the public sector). 128

In what follows we explore the different dimensions of community colleges, inquiring about its applicability in Latin America. 1. Secondary education for the poor: similarities with Latin America

In Europe, most countries have rigorous examinations at the end of the secondary-education cycle. Students who do not pass or are discouraged by the difficulties of academic secondary programmes are channelled towards vocational training, technical courses, or apprenticeships, none of which allow for subsequent easy access to university-level education. The United States—being the only wealthy country in the world whose high-school graduates are in some proportion functional illiterates—was forced to create a subsystem of higher education to compensate for the poor quality of education offered at the secondary level. Considering that many secondary-school graduates in Latin America as well have not truly mastered the official curriculum (and the reasons for this are varied), it would seem that some programme around the lines of community colleges are a better suited option for Latin America than Germany's Fachhochschulen or France's Instituís Universitaires de Technologie (IUT), especially since these schools are much more elitist than community colleges and have rigorous entry conditions. As mentioned, the jump in secondary-school enrolment in several Latin American countries has created a growing supply of graduates at this level, with the resulting pressure on higher-education enrolment. The only possible outcome in such a situation would be a drop in the quality of education received by this emerging clientele. Even if the quality of the schools remained constant, the schools themselves will be receiving an ever larger number of students from lower socioeconomic strata who are less well equipped to perform at four-year colleges. To think that we can continue to offer the same type of course to students who are progressively more differentiated is to show extreme insensitivity to the world that surrounds us. If we are to meet the needs of this emerging clientele, it will not be through the same traditional courses on Economics, Law, and Management. Moreover, we would be swimming against the tide by doing so, considering what is happening in countries that are educationally more advanced than ours. We must provide an education that is more practical, more concrete, and closer to the labour market. We do not need to reinvent 129

new solutions; the models are here for all to see, and the model of the community college seems to contain some promising solutions. 2. Most people take small steps in social mobility

Natura non facit saltum. Newton claimed that changes in nature are progressive, not abrupt. The statement seems apt for social mobility. Even though spectacular jumps in social mobility strike the public imagination ("rags to riches"), they are the rare exception, not the rule. Most social mobility is by small jumps between one generation and the next. In that sense, the community colleges offer a steppingstone for mobility, a smaller but more feasible advance. Community colleges are the ideal vehicle of social mobility for an emerging clientele of students who are able to finish high schools. Many of the students are the first ones in their family to be attending an institution of higher learning. Courses can be offered at times that are convenient for students and allow them to work full-time while studying. Monthly tuition payments can be made more affordable. And lastly, the courses can be offered close to the market and provide rapid preparation for immediate access to that market. Frustration with the lack of responsiveness of universities to the development needs of societies is mounting in Latin America. Despite the surge in university enrolments in the past 20 years, with the due exceptions, higher education continues to be accessible only to the privileged segments of society. Moreover, Latin American universities have rarely seen their missions encompassing anything other than academic education, and when they have ventured in the field of short technical programmes they have generally "academicized" them to a point that has turned them irrelevant for the immediate requirements of the workplace. By and large, universities have not done a good job of fostering partnerships with local communities, industry and businesses either. On the other hand, as secondary education coverage increases in Latin America, the quality of schooling remains high for the few and rather poor for the many. The diversity in talent and preparation for post-secondary studies among high school graduates will only increase over time, as wider shares of the corresponding cohorts reach this level. A few will be willing and able to undertake traditional university programmes. But the majority will be unable to do so without further support or will prefer to enter the labour force a soon as possible with something more than a secondary school diploma. 130

The alternative of technical schools at the secondary level has been particularly disappointing in Latin America. And the attempts to imitate the American comprehensive high schools has been outright disastrous - as recognized by the World Bank that had advocated them before. It is interesting to notice that the job preparation traditionally offered by American high schools has been steadily migrating to the post-secondary level, via community colleges. The same trend to get training away from secondary technical schools can be observed in Latin America, even though this remains a controversial issue. Be that as it may, the trend to post-pone job preparation is undeniable, creating the need to design the institutions that will receive the students at the post-secondary level. This is where the model of the community college comes to the forefront. This, then, is higher education's new clientele. It is the profile of the new social groups who are nearing the end of their secondary education and are looking towards higher education. The least that Latin American societies can do is to offer these students a kind of post-secondary education that is truly useful to them. To continue offering more of the same would not only be extremely shortsighted, it would be wrong on equity grounds. Latin America's conventional post-secondary institutions have only partially dealt with these needs. Links with industry and business for curriculum development and technical assistance provision are weak or absent, there is little flexibility in programme design and delivery, no effort is made in providing a general education alongside technical skills. Latin American higher education is woefully ill equipped to deal with students of diverse backgrounds and interests. What it offers, at best, is traditional good quality university education for those who can afford it—both economically and intellectually—and bad quality university education for those that can't. Community colleges bring some intriguing innovations to the task of providing post-secondary education for a diverse population. Not only do the programmes contribute to a better match between study opportunities and student needs and capacities, but also the flexible approach to pedagogy, its modest tuition levels, and its proximity to labour markets and partnerships with industry contribute to ensure pertinence.

131

To be sure, postulating the need for community colleges in Latin America is a very delicate issue. Consider a typical accusation: The wealthy offspring of the privileged classes will be able to pursue a university education in traditional majors that will prepare them to take over the reins of economic, political, and intellectual power. Since the poor are only now coming to the doors of higher education, let's give them something simple and unpretentious to keep them content so they won't bring down the level of our elite, public universities. This is the same cooling-out argument, which can be heard in the United States. While the argument above is cast in a very negative tone, it essentially sums up the situation of all the countries that are out ahead of us in terms of education—and there are quite a few of them. Without belabouring the point, there would appear to be only two options from a practiced standpoint: either provide differentiated education, with nuances that allow for varying student profiles (with the poorest students being directed towards education targeted specifically at students from poor backgrounds), or provide equal schooling for all, which would lead ultimately to a more hostile, dysfunctional system for students having less-than-optimal educational backgrounds. No other options have been found in the real world, even though the Utopias abound. As stated by the dean of students at a U.S. community college, "most of us are cognizant that people are looking for a way to make a living, not necessarily becoming cultured. They want to get through their business and computer science courses and get out there and get a job. At least some of them think that studying liberal arts is a waste of time. There is some impatience with things like freshman composition. They don't think they are going to have to write anything."1 Many if not all industrialized countries take very seriously the equity issues, and their educational policies reflect these options. Even so, they do not offer identical higher education for all. They have realized that a family's socio-cultural setting determines characteristics that even the most expensive of schools cannot counterbalance. By the time students reach post-secondary education, they already have different interests, priorities, and—most importantlyscholastic aptitudes. The best that can be done in an imperfect world is to design schools that are better able to further develop each stu1

CyrillaVessey, ibid. 132

dent's potential (here the meaning is not genetic potential, since the impact of the environment is already enormous by the time children reach school age and schools do little to offset these initial handicaps). In Latin America, debate on this topic has been monopolized by those who prefer the comfort of Utopias to the discomfort of the real world. It is sheer hypocrisy or wishful thinking to imagine that Latin America could create a fairer system than the ones already created by nations with considerably more resources allocated to education and with a commitment to equity that is infinitely stronger than ours. If those countries found it necessary to create a divided, differentiated system of higher education, how can we expect to do otherwise in our hemisphere? For this dual-level solution to be acceptable ethically, every effort needs to be made to ensure that all screening is based on scholastic aptitude rather than socio-economic background. The rich should not have an entitlement to the more elite programmes; the poor need to have the same rights of access, and must be admitted, when their academic profile so justifies. Elitism should be intellectual, not social. But above all, the quality of primary and secondary education needs drastic improvements. In order to stress the dilemma, the authors have chosen not to express it in diplomatic or politically correct, terms. The bottom line, however, is that it is better to have short courses of studies for poor students and longer courses for rich students, than to have long course of studies for the rich and nothing at all that will suit the poor. Those who disagree should present real life examples of alternative solutions. One more recycled Utopia is not enough. 3. The financial equation must accommodate the less affluent As will become clearer in the arguments presented in subsequent sections, the coffers of most governments cannot bear the costs of expanding public education at the post-secondary level. Whether we like or not the ethical implication of this financial and political reality, there seem to be no escape from it in the time horizon that we are dealing with. Therefore, we have to seriously consider the practicalities of opening post-secondary education for the new generation of poorer students that increasingly are finishing their high school degrees. The recent Brazilian scene illustrates the situation. It was hoped that the explosive increase in high school graduations and the equally 133

dramatic expansion in vacancies in private colleges would open the doors to a significant change in the social composition of higher education. Until a few years ago, qualified observers of the educational scene were betting on this fresh influx of working class students. Yet, the numbers that are being collated in the last few years have cooled down such expectations. In Brazil, private colleges cannot afford to charge less than the equivalent of a minimum wage per month. Mean tuition values lie around two times the minimum wage. This is more than the new generation of less affluent students can afford. To everybody's disappointment, the new wave of expansion has hardly changed the social composition of students and the expansion in enrolment seems to be coming to a halt. Hundreds of thousands of poorer students who enroll exhaust their family savings and have to quit in the middle. Two-year programmes, if nothing else, last half the time, and therefore, they cost, at most, half the costs. In fact, they can cost less than conventional colleges. In an imperfect world, they are the best available alternative to those students. 4. Pedagogical innovation: different students, different classrooms

Technological advances hold out a vast potential for the education field in such areas as recent developments in cognitive psychology as well as computers, television, and other tools. In wealthy countries, these are just one more of the available luxuries, something that is adopted if, for no other reason, because the technology exists. We, however, can and need to make more rational use of available technology. We need to harness technology, precisely because we lack the high-quality human resources that would allow us to provide a good education for all. Technology broadens the scope of influence of excellent instructors and materials, much more so than would be possible through conventional formats. A class taught by an excellent teacher and recorded on video can be reproduced for thousands of students at negligible cost. And the ultimate bottleneck to growth with quality is excellent teachers. So, we must think of ways to use these new technologies in education, not as a luxury or as the culmination of a process of pedagogical enhancement, but rather as an expedient manner to quickly replicate the best experience around. This is where the model of the community colleges can teach us some lessons. While four-year schools tend to be very conservative in 134

their teaching methods, even in the United States, some community colleges can be a major source of inspiration since the education they provide is low cost, innovative, and tuned to a broad-base of students. They are not a pedagogical laboratory for the rich, with constructivist professors who are meticulously reinventing all the world's knowledge, albeit for a few. They provide education for the masses. As a side comment, eminent Latin American educators often visit the United States. Their hosts are almost always at the elite universities that can afford to be casual - of not outright careless - with pedagogy and classroom management. This is because they receive the very best students who are able to proceed on their own and can overcome poor classroom practices. Therefore, such visitors miss the most exciting developments in the art of teaching, because it takes place in the community colleges that they hardly ever visit. Equally unfortunate, administrators and teachers of poorer institutions and short post-secondary education hardly ever travel. Therefore, our educators miss one of the most powerful lessons of community colleges: their religious devotion to adjusting classrooms to the mass of the poorly prepared students that they receive. Perhaps the most admirable trait of American community colleges is their devotion to the clienteles they have. While few would admit or choose to express it this way, most schools would like to have the best possible students, from the highest possible social extraction. And many succumb to the temptation of trying to move up the academic ladder, what has been called the "academic drift". Latin American schools catering to modest clienteles cannot wait for the time when they upgrade their clienteles and programmes. A most dramatic example is the Brazilian federal technical and technological programmes that effectively became schools for the elites, completely distorting their original roles. While academic drift can be found here and there in the community colleges, what predominates is clarity of purposes and a deliberate devotion to offering what is perceived as the best for the modest clienteles that approach them. Their pride is to be able to work with students who are among the weakest to reach the post-secondary level. Their goal is not to become Harvard but to be better than Harvard in dealing with the students they have.

135

5. The market imperative: no demand, no training

Four-year colleges educate people to adjust to change quickly. However, most programmes are ill focused on specific jobs. At their best, they impart broad and generic skills and prepare their graduates to catch up with the learning required to perform the jobs that correspond to the diplomas. However, with a growth in enrolment that is several times faster than traditional occupation expand, such colleges throw into the market a very large proportion of graduates who will never find a job that corresponds to what is written on the diploma. Ultimately, what was meant to be professional education in Latin America, turns out to be just four more years of education, preparing the graduate to go to a diffuse market and use their "learning-to-learn" skills to adjust to whatever jobs they can find. There is nothing intrinsically wrong with this solution, directed to a significant fraction of the college-bound population. In a modern economy, there are more positions requiring four years of education than those requiring the same length of schooling teaching something closely related to them. In other words, there are few positions where graduates of schools of Economics can deploy the theories learned. Yet, those graduates that cannot find those few jobs are employed in jobs where their ability to read, write, articulate ideas and solve problems is required. Be that as it may, this solution is not appropriate to a large share of the cohort that today attends higher education and, in particular, to those that are only now being able to complete secondary education. And is so for several reasons. • Its ability to deal with abstraction - in the form of theories that do not exactly match the real world - is much more limited. To develop their minds they need far more concrete subjects. It is not that they cannot learn, but their learning needs more contextualization and more concrete subject matters. • Its ability and willingness to wait four years in order to graduate is much less. They need results quicker. They are poorer. • The initial wages of those taking four years of humanities or social sciences are rather low and their initial progress is slow. These clienteles want good wages right away. Therefore, what suits them best are the technical or vocational courses offering a thorough preparation for concrete jobs. As a matter of fact, in modern economies, the growth in middle-level technical 136

jobs is much greater than the growth in demand for the skills displayed by graduates of four-year colleges. In the United States, two-year programmes are leading to wellpaid jobs in their respective fields. The top five programmes by starting salary in 1997 in the US were dental hygiene ($31,750), manufacturing processes technology ($30,675), telecommunications/interactive information specialist ($29,268). Physical therapy assistant and registered nursing follow closely behind.1 Such wages can be significantly higher than those earned by four-year-college graduates in their initial working years. This is because a much higher degree of specialization is called for in these jobs. The graduates need to learn a heavy load of job-specific skills in order to be productive from day one. If the programme does not prepare students for their respective jobs, it is of little use. Since graduates of such technical programmes come to the jobs with a clear-cut and almost complete preparation, they produce more from the start and, therefore, earn higher wages. The flipside of the coin is that this specialized preparation leaves less time for the acquisition of general knowledge - that is the hallmark of a four-year college. The immediate payoffs are greater because of the quick transfer of skills to the performance of the jobs. However, there is less flexibility and less margin for error. If the two-year course of studies does not land them in a job where the skills can be used, they are hurt by their inferior ability to learn quickly and adjust to whatever job they get. In other words, when they hit the target, the prizes are higher. If they miss it, the losses are also larger. Therefore, in order to be a good investment in human capital, courses lasting less time have to be much more concerned with the market for their graduates. An overarching concern with the job market is one of the core features of these shorter vocational programmes. In this particular, community colleges offer many lessons and experiences in the fine-tuning of course offerings to local job markets. Latin Americans are well advised to watch carefully what graduates do, regardless of the particular model of short-duration programmes they choose to adopt.

1

American Association of Community Colleges, Pocket Profile of Community Colleges cit., p. 8.

137

6. Titles and status do matter

When talk turns to human capital and investments, there is a tendency to focus on the aspects of education that can have an impact on productivity and overlook what the sociologists tell us. Actually, the individual decision to continue one's studies is influenced very deeply by the symbolic value of a degree. A degree brings status, which is a perception of self-worth or a feeling of belonging to a higher stratum. It is not a question of agreeing or disagreeing, but rather acknowledging that higher education is much more than just rates of return on investment. That said, the name given to the degree is an important issue, maybe not for the right reasons but it is important and it affects individuals' behaviour. Courses that are viewed as leading to less prestigious degrees fail to attract potential candidates. The British got it right when, with one swoop of the pen, they changed all their polytechnics into university colleges. Cost-free, simple, and painless. Schools everywhere soon followed suit, adding that magic word to their name: university. So, the name to be given to these courses is an important matter and not one without consequences. There is no intrinsic value in any specific name: the distinction between post-secondary, higher education and university is all in people's minds. There is nothing inherent in any of these terms. No amount of tweaking will create meaning where there is none. As K. Popper would say, use alone will define meaning. What name should they be given? Technical schools? Technological institutes? Associate degrees (to use the term adopted in the United States)? Unfortunately, this is an issue of not negligible importance. But it is not, as many people in Latin America believe, the definitive factor upon which hinges the failure or success of the community college model. In countries, as Chile and Argentina, where there is a large sector of institutions devoted solely to two-year programmes, enrolments on such institutions are a significant proportion (up to 40%) of the overall higher education enrolment. These students must have had to overcome the prestige issue, at least in the measure necessary to tolerate a lesser diploma, for a greater overall gain. Secondly, nothing attracts students as good quality. It is quite possible that this reticence to enroll in short programmes is due to the entirely accurate perception that existing programmes have little to offer 138

by way of marketable skills. In this scenario, students are not evading vocational and technical schools for considerations of status, but because they are rational consumers. What one has learned in the process of getting a diploma matters much. But the doors opened by a university degree have also to do with having the degree. Yet, this is not the case with technical diplomas that traditionally do not carry much in terms of status or privilege. Therefore an investment on a degree without skills, as justifiable it might be in the case of a university degree, is an entirely useless outcome of a technical programme. Community colleges in Latin America seem to be more constrained by supply, not by demand. There are not enough places being offered. Persuasion needs to be levelled towards potential suppliers, rather than potential customers. Having said that, let us repeat, there are good reasons for being concerned with the symbolic meanings of the diplomas. It nothing else, it is cheap. Argentina presents an interesting example of using the law to boost the prestige of diplomas. A relatively recent development in Argentina worth following is the Colegio Universitario. Argentina had 1,800 non-university post-secondary institutions, called Terciarios, the majority of which were public. They were allowed, in 1995, to partner with a university to offer transfer programmes to their students. A Terciario with a transfer agreement with a university—which entails a supervisory relationship, called accreditation—can call itself University College. Immediately after these legal provisions, some 300 Terciarios have transformed into university colleges.1 7. Certification and accreditation in Latin America

In the United States, most professions lack anything comparable to the European tradition of certification, covering virtually all occupations. In fact, very few higher education occupations in the United States are governed by legislation or even less by federal norms. The exceptions are those fields that pose risks or security issues for service users. These are usually governed by federal systems for individual certification, e.g„ the health professions, airplane pilots and mechanics and truck drivers. Some voluntary certification systems also exist, 1

See Alberto C. Taquini (hijo), La Transformación de la Educación Superior Argentina: de las nuevas universidades

a los colegios universitarios,

Nacional de Educación, Buenos Aires, 2000.

139

Academia

e.g., for automobile mechanics. Fields in which labour unions have a strong presence (e.g., the construction industry) also have such systems, which are often linked up to apprenticeship schemes. But overall, there is little government accreditation and certification. In contrast, Latin America has profuse legislation on accreditation of institutions and licensing for professional practice. This is not the place to discuss this feature of Latin American education and labour markets but merely to notice that it introduces great rigidities and make responses to the market slower and more imperfect. However, the situation is somewhat different for the occupations in which two-year programmes provide training. The regulated professions are mainly those with four or five years of university preparation. The so-called technical professions have not yet been taken over by entry restrictions based on diploma. The construction industry is perhaps the one most fraught with professional certifications and regulations. The newer fields are still relatively unregulated. In fact, for such programmes, there is less control in the form of certification or accreditation of the institutions. The main constraints are the legal imposition of official curricula. This is an obvious problem, especially in new fields where technology is changing so quickly. Furthermore, it complicates experimentation and incremental changes in curricula, programmes, and content. Anyone who is interested in offering such courses essentially has one important decision to make: follow the official curriculum and accreditations, or leave all that aside and risk the test of the market? The good choice depends on how rigid are existing curricula and norms and how removed from the market realities. In different countries, the situation might be different. On the side of officialdom is the comfort, peace of mind and market benefit of being able to say that the course follows official guidelines. The other side of the coin is that these curricula are always outdated and, when all is said and done, one still has to pass the market test. 8. Bridges to Four-Year Programmes: are they possible?

In the United States, two-year programmes were designed as a bridge that provided a link to traditional four-year programmes. Although this preparatory function has gradually fallen by the wayside, the bridges continues to exist, i.e., the possibilities of transferring to a 140

four-year programme upon completion of the two years of study at a community college. Furthermore, some public universities have established branch campuses within the campus of a community college, making it possible for a transfer student to continue with his college junior and senior year in the same location where he obtained his associates' degree. The first operational engineering courses offered in Brazil were created according to the philosophy that they should provide some sort of bridge to regular engineering. To be sure, the reason that they failed is due precisely to the distorted logic that was applied, whereby the curriculum was tailored to the requirements of the bridge or transfer function.1 In order for these technicians' curriculum to be equivalent to the first two years of engineering education, the programme was bogged down with theory courses, leaving no time for practical training. Graduates would lack any meaningful practical training. They would not have the necessary practical background that would enable them to find a job. Furthermore, they would also be lacking in maths and physics compared to regular engineering schools. It ended up being a mini-course in engineering, one in which both theory and practical training were sacrificed. If applied and technical training is shortchanged in an attempt to make the student "university-ready," a serious mistake is made, because what gets lost is the courses' practical-training component. Such courses need to remain geared towards the labour market. The possibility of transfer did not materialize overnight in the US or Canada. It had to be, and to some extent continues to be fought for every programme and with every university, which often entails negotiation of curricula. Removing obstacles to transfer is the job description of transfer offices in community colleges throughout the US and Canada. There is, however, significant progress in this matter in some US states, where four and two-year colleges share the same code for transferable courses. Transfer in North America is not restricted to those who obtained an associate degree in general education or humanities. Technical alumni can transfer as well, but because of the applied nature of their instruction, they cannot accomplish it right away and need to take extra classes, called "bridges" in Canada, through which they acquire the academics of the next level of educa1

See Cláudio M. Castro and Fernando Spagnolo, "Carreiras superiores curtas na área tecnológica: erros e acertos da experiencia brasileira" (mimeograph). 141

tion. This is why vocational and technical programmes are not "terminal." The traditional lack of general education in university programmes in Latin America and the early professionalization of curricula, pose a major obstacle to transfer. The first two years of an engineering school are crowded with theory courses (physics, calculus etc.) which are required in order to take the courses offered in the last two years. Very rarely are there optional courses. Just about all of them will have to be taken by all students. That being the case, a twoyear course that aspires to have its student transfer to a bachelors programme will be forced to offer just about the same courses. In other words, it will be a clone of the first two years of the bachelors programme. The obvious implication is that it can hardly offer anything applied or different. In such rigid careers as engineering, the chances of creating bridges are close to zero. Two years of liberal arts or vocational courses will simply not coincide with the first two years of law, psychology, engineering or veterinary medicine. The Canadian "bridges" notion could be deployed in Latin America to allow for the possibility of transfer from a technical programme to a university. But the fact remains that transfer is the most elusive aspect of the model when we think of its chances of taking root in Latin America. Ultimately the problem lies with the highly structured and vocational nature of four-year programmes. And given the asymmetry of power, four year programmes are unlikely to change in order to accommodate the needs of the students for short programmes. Any progress in this area is not within the power of those who operate or regulate short programmes. The exceptions that can be observed in Latin America are all in institutions that operate both the short and the bachelors' courses. They have the interest in attracting their short-programme students into two additional years of study. Therefore, they adjust the curricula of the bachelors' courses in order to accommodate the transfers. 9. What should we expect from the public sectors?

Public higher education in Latin America is in the midst of a serious crisis brought on by increased demand, tight budgets and a legislation that has created both distortions and limitations. One of the lessons from American community colleges is that rules need to be flexible and able to adapt in a way that responds to the needs of students. 142

Costs may be too high for what is actually achieved in many courses. In other cases, while per student costs are modest, results are very disappointing. The public sector lacks funds and hesitates to spur innovation, change paradigms and upset the status quo. If that were not enough, public universities in some countries are voicing strong opposition to shorter courses on ideological grounds. Given the default of the public sector, the full-fledged communitycollege model cannot be implemented by the private sector. The comprehensive nature of this model would require public funding in order to offer the broad menu of programmes that serve its core clientele that comes from disadvantaged socio-economic backgrounds. Consequently, in most countries, the private sector has taken the lead in reshaping higher education. Yet, the private sector will only be able to implement those parts of this model that can be funded by the students themselves. In particular, what is thriving in recent years are the two or three year course in business, health professions, computers and electronics. In conclusion, this is an area in which the public sector has an undeniable calling, but under current conditions it is highly unlikely that the public institutions will take a significantly more proactive approach. This leaves a vast market open for the private sector, which, conceivably, will be the fastest growing segment in the coming years. It would be unrealistic, however, to expect the private sector to tackle this task equipped only with its own financing, especially since short-duration courses are normally offered by small-scale operators. As is the case with small businesses, public subsidies will be necessary in order to defray the start-up costs involved in creating quality courses with good materials and properly prepared instructors. Indeed, even if the public sector cannot or is not willing to become a major player in the operation of such programmes, it still has a major role to play. A critical point for public-sector responsibility in this area is that, in order to set up serious two-year programmes, a sizeable investment needs to be made in developing curricula, preparing teaching materials, writing textbooks, and training instructors. Without such investments, which were made by the public sector in the case of Europe and the United States, the effort becomes an exercise in improvisation, as occurred in Argentina and Chile. In these two countries, the bulk of responsibility for post-secondary education was left to the private sector. The schools that took an interest in the area were mostly small and relatively poor. They had 143

neither the resources not the intellectual capabilities to invest in developing high quality programmes. The outcome was that most courses are improvised and are lacking in any kind of creativity or quality. This has hurt them and risks marring the reputation of an area that has not yet had a chance to consolidate and has yet to affirm itself and establish its status.1 D. Community Colleges should inspire Latin America

As should have become clear from what was presented above, arguments to clone community colleges in Latin America are pointless. In principle, institutions are not transferable to other cultures. They always need significant adaptations and transformations to be able to thrive elsewhere. Many of the functions of community colleges need public funding, due to the modest economic status of its typical clienteles. And at present, except in a few countries, it does not seem that such subsidies are forthcoming. That leaves the private sector with the task of developing two to three year programmes at the post-secondary level. And funding its activities out of tuition from students alone, what can be done is much more limited than the broad menu of services offered in the community colleges. One would hope that governments would at least support the development of materials, teacher training, curricula etc. Convincing the government that such tasks are beyond the means of small colleges and schools is one key element in the relevant policies in this area. But even within this narrower menu, the long experience in the United States with associate degrees and post-secondary technical programmes yields many important lessons that can greatly benefit Latin America: • It makes much sense to offer post secondary courses that are much shorter than a bachelor's and much more focused on the specific occupations. This statement should be understood as a plea for massive growth in such programmes. North America and Europe graduate more students in such shorter courses than in the regular four-year programmes. The economies of Latin America are not that different. 1

See Claudio de Moura Castro and Juan Carlos Navarro, "Will the invisible hand fix Latin American private education?" in Private Prometheus, op. cit.

144

•

•

•

•

•

•

•

Short programmes require much more attention to markets. The rule is simple, no demand, no training. When well targeted, they are splendid investments. When they miss the market, the loss is much greater than in the case of longer courses where the "learning to learn" component is often more important than the specific content of the programme. Short courses tend to be a privileged path to social mobility. They are far more appropriate to less affluent students, who in their majority, cannot afford long courses, do not want to wait much longer for results and have difficulties in coping with abstract content. But the ideological barriers are still serious. Like in other levels of higher education, the symbolic content of the diplomas and certificates is powerful. Rates of return are only a part of the equation. Therefore, programmes can benefit from titles that convey status and importance. Contrary to the American scene, official curricula, rigidities in accreditation and excess bureaucracy can be serious hurdles that cannot be underestimated. Given the rigidities in the curricula of bachelor's programmes, the much touted transfer function of community colleges is not at all applicable in Latin America. Courses, in principle, do not permit such transfers, unless they are disfigured beyond the point where they retain some market functionality. Attention to teaching and classroom technology is one of the distinguishing features of community colleges that Latin Americans are well advised to consider. Community colleges are devoted to their students and take as pride in their ability to bring them to their fullest potential. This is in stark contrast to other institutions that disdain such students. This may be one of the most powerful messages that community colleges can convey.

145

When Brazilian Employers Control Training, Many Things Can Happen

The most stubborn problem in training is the chronic mismatch between supply and demand. This is the "supply-driven disease," infecting most training systems. Therefore, examining real-world vaccinations for this disease is a most useful exercise. One obvious but rarely used solution to this problem is to put employers in charge of training systems. If employers pay the bill and consume the output of vocational schools, it stands to reason that it is in their interest to ensure that training is properly attuned to their needs. The gulf between supply and demand is bridged by the radical solution of giving full control of training to its users. Indeed, the fact that Brazil is the only major country to adopt this solution demonstrates just how radical it is. When SENAI1 was created in the early 1940s, it was the result of an initiative by Sao Paulo industrialists. They persuaded the President Getúlio Vargas to create an industrial training system, fund it through a 1% levy on the pay roll and hand its development and management over to the Federations of Industries. In so many words, industrialists taxed themselves to fund their training system - it is not a voluntary contribution. Each Brazilian state has its own independent SENAI and a National SENAI coordinates their actions. In theory, the Ministry of Labor, either directly or through a Manpower Council, oversees SENAI. In practice, SENAI pays only lip service to these bodies of higher governance. This basic equation remains very much same. SENAI is more than 50 years old and the Federations of Industry continue to call the shots. Each Brazilian state has its own SENAI and, in most states, industrialists have kept their SENAIs efficient and responsive to demand. Interestingly, almost all other Latin American countries have created similar institutions with just about the same rationale and styles. Almost 50 years later, Chile handed over some of its technical schools to employers' associations and in Peru there was a decision to allow 1

Servigo Nacional da Aprendizagem Industrial (National Service for Industrial Apprenticeship).

146

employers to appoint the management of SENATI. These are welcome but partial and belated moves when compared to the once-and-for-all approach taken by Brazil. Yet, none of them has taken the bolder step of turning over their training systems to those who employ its graduates. The subject of this paper is the replication of the model within Brazil. In fact, SENAI only covered the manufacturing sector and similar institutions for other areas of the economy were subsequently created. In the late 1940s, SENAC was created. With the "C" standing for "commerce", the idea was to cater to the service sector of the economy (from waiting tables to using computers). SENAC was put under the responsibility of the Federations of Commerce. Later on, SENAR was created, the"R" relating to the rural sector. Much later, SEBRAE was created to deal with small enterprises and SENAT to cater to the transportation sector. Having all essentially the same structure, these institutions have some key points in common. Perhaps what distinguishes them from most other training institutions is the greater weight given to the role of training as a means to increase productivity. Being run by employers, the training-productivity link is explicit; social and equity issues enter only tangentially into the equation. Training is clearly biased towards what these employers need. And Brazilian employers need workers with relatively sophisticated skills. Since its inception, SENAI has been training a blue-collar elite. These are the high-skill manual workers needed by the modern firms. Training those who need the most has never been high on the agendas of these institutions. And since experience shows that training the more dispossessed workers tends to be fraught with problems and mismatches, they have kept a high level of efficiency throughout their histories. The trade off is clear: less effort to reach the bottom, higher impact on productivity and low wastage due to training that is not utilized. The formula was practically the same for all of them. Hence, one could have predicted the same trajectory and the same styles of operation. But this would have been a wrong prediction. The real world has led these institutions in different directions and each has acquired significantly different styles of operation. There are five institutions, all of them owned by employers' associations. And there are five distinct forms of operation, each closely reflecting the historical moment of its creation. In what follows, all five are examined in greater detail. 147

A. SENAI: industrial training

When SENAI was created, there was hardly any other alternative in vocational training. It had to start from scratch and create, little by little, a large network of schools, adding today to over 500. Teaching materials had to be developed and trainers trained. The style was definitely patterned after the German and Swiss tradition, since the first director of the pioneer SENAI of Sao Paulo was a Swiss engineer with strong roots in those countries. Today, SENAI trains over two million workers each year, from back street welders to PhDs in welding who want to perfect their skills in quality control. In many ways, SENAI remains a successful institution, even though it has changed little its organization styles. Industrialists try to keep it under a close watch, are very protective and defend SENAI from its many adversaries (mostly from the education left) who dislike the idea of public money being given to private enterprise (the 1% levy is indeed public money). But SENAI is not without problems and dilemmas. Micro-management by Federations of Industry is a frequent complaint. Worse, weak or politicized Federations of Industries inflict serious damage to the corresponding SENAI, by definition, vulnerable to any weaknesses of their sponsors. To some extent, this has happened in some states in the less developed Northeast. Weak employers lead to weak SENAIs. Closer links between well-organized chambers of specific industries (e.g. machine building, textiles etc) and schools catering to them tend to lead to better training. This often produces a win-win situation for all sides: schools produce graduates that are able to find jobs in those trade for which they were trained and employers are generally satisfied with the calibre of student they receive. A major trend in the last decade has been a broadening of the range of services provided. Instead of offering training to industries, the menu has become much longer. First came custom-designed training. Then the manufacture of components or repairs requested by firms that required state-of-the-art equipment, not owned by them. Consulting in industrial layout and troubleshooting followed, as did information services for new technologies or general trends in industries (e.g., sending observers to international textile and fashion fairs to report on upcoming trends). Then came more specialized services. Several school units sprang up to support industry in implementing ISO 9000 and Total Quality 148

programmes. Finally, about 40 school units have launched serious R&D programmes. In so many words, SENAI has moved from a training institution to an agency catering to multiple needs of firms. For decades, the threat of elimination of the payroll levy has been hanging in the air. Financial authorities dislike tied expenditures and the noisy left dislikes the idea of subsidies for private industry. Thus far, industrialists have fended off all attempts along these lines and are poised to do likewise in the future. But if the levy were to disappear, the most likely scenario would be a clear specialization of SENAI along the lines described in the previous paragraphs and the creation of voluntary or semi-voluntary arrangements industry-by-industry. This would obviously create a vacuum of initial training for the simpler occupations. B. SENAC: The trainee-driven solution for services

Since its inception, SENAC had the same structure as SENAI. The Federations of Commerce owned and managed it. However, as time went on, SENAC administrators realized that service-sector firms were less committed to training its workers than their peers in the industrial sectors. For some time, then, SENAC has taken a slightly different approach. Rather than directly responding to member firms, it has catered to individual students who want to invest in their future. In addition, Federations of Commerce are less possessive towards SENAC. They are not perceived as being ready or able to fight on its behalf, as is the case of the very protective and belligerent industrialists. With the ups and downs of politics and ideology, SENAC managers have begun to fret about their future. Threats of cancelling the levy revenues have led them to adopt fail-safe strategies. Some years ago, SENAC again deviated from its SENAI-like roots. It decided to charge student fees (most SENAI courses were then offered free of charge). This basic change was motivated by the very need for survival, especially if the left had its way and the levy was eliminated or transferred to the government. Student fees, would create a level of revenues that would allow SENAC to survive without its captive source of revenues - if this worst scenario were to materialize. Depending on state and moment, the fees charged to students add one third to half of SENAC's regular budget. These increased revenues have allowed SENAC to expand its course offerings and increase 149

the number of students enrolled. However, perhaps the most important consequence has been to give SENAC a powerful mechanism to bring supply in line with demand. Paying students are much more concerned with the market for their newly acquired skills. Even though SENAC charges less than full cost, it is not free training. Therefore, the market has to be carefully monitored to ensure that each and every course has enough takers. This, in turn, provides an all but automatic check that the training provided is in line with its demand. C. SENAR: Why run schools when you can buy training?

For reasons that are unclear to this author, after creating SENAI and SENAC as private institutions, SENAR was created as a fully public service, housed in the Ministry of Labour (Perhaps the association of rural entrepreneurs was not considered as a legitimate or competent institution). It was born as a full-fledged public bureaucracy, with all the vices and shortcomings of civil service. Needless to say, it was far less competent than its two older siblings. Nobody was happy with the way it operated and complaints were very frequent. Along with the sweeping changes in civil service that took place in the early 1990s, SENAR was closed down and, from its ashes, a new privatized version was created. It was attached to the association of rural employers who, by that time, seemed to have acquired a better reputation. Therefore, SENAR finally came in line with the other two. Yet, when SENAR was recreated, the training scene was not the same in Brazil and its new style reflects these changes. At the time of its rebirth, the training environment had considerably evolved, compared to the period when SENAI was born. Finding a very active training market, the new SENAR never saw the need to create and operate a large network of schools. Instead, it found it more expedite and efficient just to contract training from the market. Focusing mostly on mechanization of farming, it could contract courses from SENAI, from SENAI spin-offs and from other training institutions that abound in Brazil. Thus, despite the nature of its rebirth, SENAR has not become a full-blooded member of the SENAI/SENAC family. Rather, it has become an active buyer of training from the market. These now philosophy seems to have been appropriate to the current realities of the 150

Brazilian market. Indeed, there are few complaints about SENAR today. D. SEBRAE: Employers' support to micro-enterprise The history of SEBRAE runs parallel to that of SENAR. Created as CEBRAE with its own public bureaucracy, it provided support to small and micro-enterprises. In contrast to SENAR, it was never a SENAI sibling, but rather a more distant relative. Much like SENAR, however, its association with and to the public sector was a serious liability that hampered flexibility and left the door open to constant harassing by spoil politics. The same administrative earthquake that privatized SENAR, privatized CEBRAE as well. Full management - budgets and all was transferred to the Federations of Industries. SEBRAE, too, finally came in line with the other "S" institutions. Yet, like SENAR, SEBRAE remains more of a distant relative than a sibling. Again, reflecting the moment of its creation, SEBRAE works essentially as a contracting institution. It has not build a physical network of trade schools like SENAI and SENAC. But, in contrast to SENAR, it gets more involved in studies, development of teaching materials and experimentation. E. SENAT: Distance education to truck drivers SENAT is a branch of SENAI that became independent. SENAI's mandate has always been to train for all industrial sectors, including the transportation industry. Transporting firms always paid the levy and expected the corresponding services. However, employers in the transportation industry were unhappy with SENAI and thought they were being short-changed in the distribution of the training effort. Indeed, training truck drivers and shipping clerks was not quite in line with SENAI traditions and, in this respect, they got less than they paid for. As it turned out, SENAI was not sufficiently alert to respond to this situation and avert the secession of the transportation sector. SENAI was unable to mend its fences properly and, as a result, became a bystander in the successful revolt of the transportation entrepreneurs. With this secession, the levy revenues from transportation 151

were redirected to a new institution entirely dedicated to the needs of this particular sector. SENAT thus was created. Being little more than a dissident wing of SENAI, SENAT was created under its rules and legal framework. Yet the timing of its creation led SENAT down a somewhat different path. Instead of erecting training centers or contracting training in the market, SENAT has done something completely different. It rents transponder time and has created a set of training programmes almost entirely based on distance education. At the time this paper was written, for eight hours each day, a satellite transmits training programmes to truckers and transportation firms. Driver safety courses, materials handling, handling of hazardous materials are some of the usual offerings. More than 1,000 transportation firms have been provided with parabolic antennas and decoders. Even some large service stations provide means for truck drivers to watch the programmes. F. Lessons: Ample variety, the same legal formula Training in Brazil started with a single formula that was applied to all. If not at first, eventually all five "S" institutions came in line with the same model. A levy on the pay roll funds the operation and the chambers of employers run the institutions with full independence and under private sector statutes. By contrast to most other countries, this has been and remains a vastly successful formula. Employers turn out to be better than governments and independent institutions in running a training system catering to their own needs. However, the original recipe did not lead to equal forms of operation. Each of the five members of the "S" system has developed its own style of operation. And each of these styles, more than anything else, reflects the moment in which the institution was created. The older SENAI created a wide network of vocational schools. SENAC, running under the auspices of the less committed employers of the service sector, charges tuition for all courses offered - in an effort to decrease its dependency on the payroll levy. SENAR and SEBRAE, have checkbooks instead of training centres. They simply buy training in the market. And, finally, SENAT completely bypasses conventional classes. All courses are in the form of distance education, beamed by satellites. With the fast transformation of the Brazilian economy, all "S" institutions require constant adaptation. While we may quibble with 152

their speed of their reactions, they remain largely efficient and functional. The future is unclear. There are as many threats of eliminating the levy, as there are alert and powerful defences, particularly from the all-powerful associations of industry. All of these institutions are fairly-well managed, attract competent staff and have very high esprit de corps. Thus far, they have battled and resisted all assaults. They seem to be no less prepared to fight new battles. If, by some political development, they were to lose their levy revenues, most of these institutions could find alternatives sources of funding. Chances are that they would survive. But this would inflict a severe blow in equity: only the most affluent, aggressive and modern firms would be willing to voluntarily devote significant outlays to training. The majority of workers from the lower classes would be left with few alternatives for high quality training, particularly pre-employment training.

153

Brazilian Technical Education: The Chronicle of a Turbulent Marriage Technical schools combine a high school degree, technical information and hands-on vocational experience. This chapter narrates the turbulent cohabitation of the academic and the vocational curricula of technical schools in Brazil. In the final section, the balance between different levels and modalities of technical education is also discussed. In Brazil, as in many other parts of the world - to wit, elsewhere in Latin America, the Arab World and Sub-Saharan Africa - technical schools are an unstable solution. They are supposed to be a combination of academic and practical training and, in most cases, the balance between the parts is lost, becoming something else that is less than satisfactory. While some European countries have managed to have moderate success with this formula, it tends not to work in most developing countries. The formula survived in Brazil for many decades but was abandoned in the mid 1990s, creating some confusion. After the initial perplexities, a new system is beginning to take shape and show promising results. However, the new incumbents of the Ministry of Education, following the change of government in 2002, tampered with the delicate balance obtained. A. The Catch 22: The more you improve, the more dysfunctional it becomes Around 19711 was sitting in my IPEA1 office when a visitor from the World Bank arrived. The man was very proud of the new Bank policies to invest in technical and comprehensive education in Brazil - as well as elsewhere in the world. As he explained, by combining an academic and a vocational curriculum, students would get, at the same time, a solid academic education and whatever it takes to prepare them for jobs. As it happened, I had on my desk a number of tables from a survey of Ginasios Orientados para o Trabalho (GOTs), a network of ju1

Instituto de Planejamento Económico e Social, the policy and research branch of the Planning Ministry. 154

nior high schools combining academic and vocational preparation.1 The data showed the percentage of GOT students who wanted to take up the occupations offered by their schools. On average, only 2 per cent of the students were interested in getting jobs related to the skills they were learning. Instead, almost all of them wanted to go to higher education. I showed him the tables and insisted that the same would happen with the World Bank schools. Needless to say, the World Bank officer was not persuaded by my tables. The Bank went on to invest heavily in technical and comprehensive high schools around the world. In the late 1990s, John Middleton and Arvil Van Adams commissioned a vast number of surveys to find out what was happening with technical and vocational schools all over the globe. The results, used in the policy paper of the World Bank,2 showed that comprehensive high schools were an unmitigated failure and technical schools were almost as disappointing. The Bank shifted gears and stopped investing in these models. In the early 1970s, when my research - mentioned above showed that students were not interested in skills training, I asked a few of my graduate students to write term papers looking at what was happening in the federal technical schools of their home states. The numbers found anticipated what the World Bank found a decade later. Indeed, the schools were not producing the expected results, i.e. technicians who would work as technicians. In a country with a very small enrolment in secondary education and a dismal quality in public schools, technical schools were offering a winning formula for students: free secondary education of high quality. Considering that the only other alternatives to quality education were the expensive private schools, it did not take long for students to realize that practically the only tuition-free schools offering a high quality academic education were the technical schools (most were federal, but a State such as Sao Paulo also had an extensive network of public technical schools). With the increasing demand for enrolment in the relatively few such schools (around one hundred in the

1

2

Castro, Claudio de Moura, Milton Pereira de Assis, and Sandra Furtado de Oliveira. (1978). Enseñanza técnica: Rendimientos y costos. Montevideo, Uruguay: CINTERFOR. The World Bank, (1991). Vocational and technical education and training. Washington, DC: The World Bank.

155

federal system and another hundred in the Sao Paulo network), requests for admission expanded enormously and schools began to impose admission tests (known as vestibulinhos). Candidate to vacancy ratios soared to 10:1 and more. Cramming programmes to pass these tests also appeared. Affluent students, particularly those from private primary schools, began to corner the market for technical schools. These institutions became, de facto, elite schools, preparing students for the most competitive entrance examinations for higher education. A survey conducted in the late 1980s indicated that among the ten schools that produced the largest number of approved students to the most competitive Sao Paulo programmes (e.g. medicine, law and engineering at the University of Sao Paulo) one of them was an industrial arts federal technical school. Lavish resources were dispensed to make them better, including a substantial loan from the Interamerican Development Bank. However, the better these schools became, the less technicians they produced, because they were co-opted by the elites who were able to pass the vestibulinho. No matter how sophisticated the laboratories and workshops, the students were clearly higher education material and did not take very seriously the skills instruction. At best, they would take a technical job for a few years, to ease their personal finances, while attending elite universities. This was the Catch 22. The better they became in everything, including their laboratory and vocational programmes, the less their students took interest in the technical dimension. This was because they also became equally good at academics. These were expensive schools, costing five to ten times more than the local public secondary schools. To sum up, a set of close to one hundred federal schools were preparing technicians who hardly ever became technicians. They tended to come from upper class backgrounds and had nothing in mind but the examination for the best universities, be they in law or medicine. Hence, the schools lost the equity battle. The few students from a more modest origin who managed to enroll were quickly convinced that going to higher education was a better deal, having come so far up the education ladder in a country in which so few reached that level. Hence, industry was also on the losing end, since they were not getting the technical staff they needed. 156

Given the due differences, the situation was not too different from what was described by Philip Foster in his well known 1971 paper on the "The vocational school fallacy in development planning."1 Technical schools in Africa did not prepare for technical jobs. In fact, in a field visit to a technical school in the Ivory Coast, I was told that when the students began to face a tougher job market, the solution was to prepare them better to enter university. The situation persisted for decades in Brazil. In the mid 1980s, the Minister of Education appointed a task force to discuss technical schools and I became a member of this short-lived group. Most members were, one way or another, connected to the Ministry of Education. It is interesting to notice that this group was totally unconcerned with such a distortion vis a vis the original role of technical schools, namely, to prepare technical cadres for industry, services or agriculture. Instead, they reflected the general mood of the faculty of technical schools. They were proud of the academic excellence of their student body. Not becoming technicians was almost a mark of success. This was in line with what Ronald Dore (1997) called the "academic drift".2 At that time, I suggested that the schools be split into two separate tracks. One would offer the academic programme, attracting the upper classes. The other would only offer the skills training. Not being interested in practical skills, the upper class students would not apply for the technical track, allowing those truly interested in skills to find vacancies. Perhaps the more modest students could not compete for the academic tracks in these schools, and would have to take their secondary education elsewhere, but at least, they would find vacancies in the technical tracks. It is instructive to notice that the members of the task force did not even take the time to understand my proposal. Their mindset was elsewhere. There is nothing wrong with preparing engineers or medical doctors who have had to file, solder or hammer. Using one's hands is not a bad idea. However, in a country with over one hundred million inhabitants at that time, two hundred or so technical schools was a very scarce resource. These schools are much too expensive and scarce to 1 2

Philip Foster, in Readings in Economics of Education (Paris: UNESCO, 1971) pp. 614-630 Dore, Ronald (1997) The Diploma Disease. London: Institute of Education University of London.

157

impart a hobby or familiarize students with hand work. Industry needed skilled labour and had no other alternative solutions. While Latin American education has a lot in common, both good and bad, this is an area where Brazil and her neighbours part company. Almost everywhere in the world, there is a close correlation between student's socio-economic status and the excellence of the schools they attend. It should be no surprise to state that the poor attend less expensive and worse schools. Technical schools in Latin America tend to be the poor cousins of the academic secondary. They offer lower quality education and cater to lower status students. As a result, there is much less interest in higher education. However, they have other problems. Their curricula are neither here nor there. They are not good enough at the academic level. They tend to be perfunctory and outdated at the technical level and the hands-on activities are not sufficient to prepare skilled workers: Jack-of-all-trades, master of none. By trying to offer too much, they end up offering a watereddown preparation on all three counts. By contrast, Brazil invested very heavily in its relatively few technical schools. They are very good and very expensive. However by boosting the quality of the academic side, they became a magnet to the elites. Hence, the Catch 22: the better they became, the worse they performed their original task of preparing technical personnel. B. World trends in technical education

As mentioned, technical schools were having the same troubles in many other countries. Granted, Europe has had technical schools for many decades and they tend to work relatively well close up. But the conditions are quite different. Firstly, Europe has a much higher proportion of the corresponding age cohorts in secondary education or some other alternative at the same level, making the diploma less scarce. Secondly, Europe has very high quality academic public schools. There are no serious incentives for someone to attend a technical school in order to have access to the best higher education programmes. Thirdly, Europe has a clear segmentation of secondarylevel technical schools. According to academic qualifications, students are tracked to programmes that are stronger or weaker in academics and more or less geared to job preparation. Even so, countries such as France, that have something similar to technical schools, still feel the weight of the academic bias in their 158

technical programmes. Contacts with enterprises are not always easy. Inertia in changing occupational profiles is endemic. Therefore, a trend away from the conventional academic-technical schools can be discerned even in Europe. The French baccalauréat is divided into broad areas, such as commerce, biological or natural sciences, mathematics, and humanities. But they should not be confused with technical schools, despite the 'technical baccalauréat', because they give only the 'flavour' of occupations, along the corresponding lines, but are a far cry from job preparation. Argentina and Mexico seem to be going along this "soft" career education. The comprehensive high school of the United States is an interesting case. In theory, it offers the best of all possible worlds. Academics, technology and shop, all under one roof. What could be better? But in practice, social cleavages are all under the roof of the same high school - while hidden in Europe by school segmentation. Tracking is rampant inside the schools and each track is typically attended by each social stratum. Working class students, as well as low-performing students go to shop classes. And both groups attend watered down mathematics and science courses. Being dumped into one of them can be seen as a punishment. As a result, they have low status and can be demeaning. To sum up, comprehensive schools, on the whole, perform poorly at job preparation. The more or less spontaneous response to the poor job preparation achieved by high schools has been the highly successful community colleges. In other words, job preparation is moved to the post-secondary level. But this trend to push occupational training up to the postsecondary is not only American. It has become universal, with Lycées Techniques and Institutes Universitaires Technologiques in France, the Fachhochschulen in Germany and the Technical Colleges in the United Kingdom. This trend has reached Latin America where in Chile, Argentine and Venezuela, around one third of higher education enrolment is in post-secondary short courses. Another major trend is to offer job preparation in specialized training centers, unrelated to academic schools and not offering an academic degree. The Brazilian SENAI was the pioneer Latin American institution to create vocational schools outside the academic system and was cloned by practically all countries of the region. The rationale of this system is quite robust. It offers job preparation after one leaves regular academic school, whenever it might happen. Typically, it enrolled youth with four years of schooling, in the 1940s, when 159

SENAI was created. As Brazilian mean school achievement increases, SENAI now receives students with as much as complete secondary, in the most affluent states and in the more complex occupations (e.g. electronics). For the record, as much as SENAI (and SENAC, the corresponding institution for the service sector) perform quite well in their traditional skills training programmes, they also operated technical schools where they met the same predicament of the federal network. In most cases, students became infatuated by the academic track of the programmes and often went straight to higher education, largely frustrating the original intention of the course. All in all, the standard formula of the technical school has not performed too well and is being progressively replaced by other initiatives. This is less the case in Europe where the highly segmented technical schools still resist. Of course, Germany never adopted this model, except for a few such schools in Badem Württemberg. C. Finally, divorce!

In the mid 1990s, after Paulo Renato de Souza became Minister of Education, the idea of a loan to beef up technical education was in the air. The Federal Technical Schools had lost their leading edge in technology and equipment and there was a chance to prepare a substantial loan, using FAT1 money as counterpart funding. But the IDB was reticent about putting more money in an elitist institution that had failed to deliver a credible product. At this time, I was an employee of the IDB and also advising the Minister on matters of education policies. To counter the reluctance of the IDB, the idea of splitting the academic and the technical segments of a technical programme was exhumed. Indeed, this was an obvious way to avoid the problem. If students could take the academic track by itself or the technical track by itself, or both, the reasons for university-bound students to enroll in the technical track would disappear. Why would a high status student want to spend time in shops or labs, learning an occupation that was not desired? For such students, it would be a much better idea to spend more time preparing for the academic degree, increasing the chance of passing a competitive vestibular. If that were the case, they would leave the technical track free 1

Fundo de Auxilio ao Trabalhador, a fund that accumulates a tax imposed on the payroll of workers.

160

for more modest students who could be interested in the occupations taught. These were the students unable to compete with the elites in the vestibulinho. This was more than a hunch. In fact, the Sao Paulo SENAI had conducted a survey among students of three of its technical schools1. Two were regular technical schools, offering the standard fare of integrated technical and academic curricula. The third was called a Curso Tecnico Especial, meaning that it was targeted to clienteles that already had a complete secondary degree. Therefore, they only took the technical courses, making the programme much shorter (one year, instead of three). As it turns out, this natural experiment created the evidence that was needed to increase the confidence in the plan. In fact, the Curso Tecnico Especial, is a spontaneous version that anticipated the split technical programme. Comparing the socio-economic status of the three schools, it became obvious that the two conventional programmes had a clientele that was substantially different from the third, the Especial. The latter had students from a much more modest origin. In other words, by removing the academic track, the door was opened for more modest students to enroll, as they did not have to compete with the others. The IDB agreed with the policy but wanted to ensure that enrolment at the academic tracks inside federal schools would be contained, to avoid distorting even more their original role of preparing technicians. This created a considerable tension between MEC and the IDB. Nancy Birdsall was the Executive Vice-President and had been previously involved in a research piece on Brazilian technical and secondary education. She was adamant to ensure that the loan would not do more of the same that already existed. At one point, the whole project almost fell through. But finally, a working solution was found. D. Implementation and confrontation

For about a quarter of a century, technical education was stuck in the Catch 22 predicament described above. The decisive factor to bring about change was the possibility of a 250 million dollar loan from the IDB. A loan can be a powerful weapon to break a deadlock. 1

SENAI SP 161

Multilateral bank loans have what is called a conditionality: that is to say the banks tell the countries that they are free not to take a loan. By the same token, the banks are entitled not to offer a loan, unless the country is willing to accept certain conditions. Conditionalities are powerful weapons. They can be used for positive or negative ends. They can be the catalyst of change or create horrendous crises and confrontations in the borrowing countries. And very often, the consequences of cancelling a loan because a conditionality was not satisfied are so dire that the banks pretend that they do not see its lack of enforcement. In other words, conditions imposed with a loan can be a good thing if well conceived and all goes according to plan. They can even be a blessing for a Minister who has to deal with recalcitrant actors and cannot afford to pay the political price of confronting them squarely. A positive conditionality reflects the wishes of the Minister - and hopefully the needs of the country - but enables the 'blame' to be diverted on the banks. PROEP (Programa de Expansao da Educa?ao Profissional) was a very benign case of conditionalities being used to boost reform. Before becoming Minister of Education, Paulo Renato de Souza was the Director of Operations of the IDB, being fully familiar with all the trappings of development banking. Breaking the Catch 22 that preserved the elitism of technical schools was his goal, as much as that of the IDB staff involved in preparing the loan. The writer was, at the time, an officer of the IDB and was as much involved in this loan as anybody else in the Bank. Therefore, the present paper has all the advantages and disadvantages of being written by an insider in the reform that took place. Once the split between the technical and the academic tracks of the technical schools had been decided, the rest was a matter of sorting out the details and finding a legal form to do it. The recently approved LDB (Lei de Diretrizes e Bases da Educagao), the broad legal framework for education policy, made matters much simpler than initially thought. It was agreed, at the beginning, that students would take the technical track after completing secondary. But it was subsequently decided that they could also do so while taking the last two years of secondary. It was feared by the IDB that this latter alternative could create some loopholes to preserve the old system. But on hindsight, the fears were exaggerated. 162

A more contentious matter was to decide what to do with the academic tracks to be offered inside the federal technical schools. Technical schools could respond to the new regulations by expanding them and shrinking the technical track, making them as removed from skills training as before. After much debate, it was agreed that enrolment in the academic track had to be reduced to half its present level. This was not a politically easy decision. A major source of reaction when the new plans were first spelled out was an increase in the total work load of students who took both the academic and the technical tracks, compared with the previous load in the integrated programmes. Critics chastened the Minister for imposing on the poorer students - who wanted to get a technicians degree - a heavier work load than before. It is true that the work load increased. But the reason for the increase had absolutely nothing to do with the split imposed by the reform. It was the LDB that increased the work load of all secondary education, technical or otherwise. Had the LDB not been enforced, the work load of the split technical added to the work load of the academic would be exactly the same as before. Once the broad outlines of the reform were agreed with the IDB, the Minister invited a number of leading principals of Federal Technical Schools to present the new plans. The present author was at this meeting and was able to notice that the reactions from the principals were mild and more sympathetic than otherwise. The only complaint they expressed was the loss in contextualization of subjects in the academic track. Some of the best technical schools had managed to bring the technical and the academic subjects closer together, using examples and illustrating the drier and abstract theories of the academic world with concrete examples from the technical end. Soon, however, the tide turned against the Ministry. The same principals, who were so docile at the beginning, probably got much flak from their teachers and teacher's unions, particularly from those leaning to the left. As a result, they began to oppose the reform. Their arguments were less of substance than process. They claimed that they had not been consulted beforehand. This is only partly correct. It is a moot point, whether asking a representative group to participate in an open discussion on the proposed reform is a consultation or not. However, we all know that the alternative of calling in a broad con163

sultation, with students, teachers and unions leads to stalemate, due to the long tradition of obstruction from the organized left. The leitmotif of the reform, namely, to make technical schools less elitist and more able to fulfill their mandate to supply technicians to industry, was never directly challenged. Principals claimed, unconvincingly, that the schools were not so elitist and that not all graduates went to higher education, though the data to support this never materialized. In other words, decade after decade, technical schools never went to the trouble of finding out how elitist their students actually were and how few graduates became technicians. Deep inside, they ultimately liked being an elitist institution, catering for the ablest students who had the most success in the most exclusive universities and who mostly came from upper class families. Ultimately, what convinced them not to take their objections to the reform too far was the 250 million dollars they could tap into. A well-prepared request to the Ministry could bring to the school a handsome amount of money, allowing reforms, expansions, even new laboratories and workshops. It is interesting to ponder on the fact that the real losers in the battle to implement the reform never participated in the confrontations that ensued. If schools were to split the two tracks and the academic track had to cut to half its enrolment, the losers were clearly the elites who were getting a free ride on the best academic education offered by the public sector. Why did they not organize and obstruct the reform? Curiously, the opposition came from the left, which subscribes to an ideology where privilege is the enemy. The hard core opposition to the reform came from a group of leftwing education researchers and professors. They ignored the elitist nature of the technical education being offered by almost all the technical schools. Why would the left choose to ignore the fact that a school designed to cater to modest youth trying to become technicians had become so elitist? Their arguments focused on two issues. The first was the alleged imposition of a reform agenda by a multilateral bank. Acacia Kuenzer, a well-known author in such matters, dedicated several chapters of her book on technical education to the so-called 'evil' influence of the World Bank on Brazilian technical education affairs. She discusses at length how the Bank dumped its cookie-cutter formulae on Brazil. Unfortunately for the credibility of her arguments, she got the name of the Bank wrong. The World Bank never considered lending money 164

to technical education in Brazil. It was the Interamerican Development Bank that made all the negotiations and the loan.1 As previously mentioned, it is true that the IDB wanted to see the reform approved, as a pre-condition to make the loan. But the reform was designed by Brazilians, inside and outside the IDB. And it had roots in past attempts to move along equivalent lines. But, ultimately, the argument hinges on the moral and political justifications for the conditionalities that are part of multilateral bank loans. This involves value judgements. It cannot be proven right or wrong. The second argument is far more abstract and goes back to this group's conception of education. At least a decade before, much attention was paid to something called politecnia. Maybe the roots are in Proudhon, but the real father of politecnia was Gramsci, in the late 1920s, while in an Italian jail. Politecnia is a school conception in which the academic and the vocational, the study and the work are all combined in a seamless process. To those espousing the principles of this school of thought, splitting the academic and the technical tracks was seen as straight on rejection of this principle, and, therefore, it was flawed. There are many problems with the defence of politecnia. First of all, it was conceived in the 1920s, before the significant modern technological revolutions took place. In addition, practically all the vocational and technical education systems of the world took shape after that and all made honest attempts to bring the world of work closer to the world of schools. But more importantly, politecnia is a utopia, since there are no schools anywhere in the world that adopted this model in its pure form. Comparing any existing system with a utopia does not yield a fruitful discussion, because we are comparing something that bears the imperfections of the real world with the purity of something that only exists in the imagination. If we want to look at the closest real life materialization of politecnia, the American comprehensive high school comes to mind, or the polytechnical schools of the former USSR, with vacations spent working in factories. But on both counts, the existing research does not show very flattering results. American high schools disappoint in

1

A. Kuenzer, Ensino Médio e Profissional (Sao Paulo: Cortez Editora 2001) pp 4647.

165

their attempts to impart an occupation to students and the experience of Russian students in factories is even worse.1 To the present author - who is far from neutral in this discussion - the opposition from the politecnia group makes no sense. The quest for intellectual integration of academics and technical subjects is real and important. However, it was neither achieved automatically by the previous Brazilian system nor is it impeded by the reform that split the two. In fact, in some of the technical schools where the students take the academic and the technical tracks at the same time, administrators claim that they have achieved a good degree of integration. Furthermore, good academic education is a combination of theory and its applications. Students should practice the application of theory as part of the learning process. Good education does not need a parallel goal of job preparation to be credible and useful. Using one's hands is a good way of learning about the world - which includes all sorts of theories. In fact, it has been said that theory comes through the hands. But this only requires simple laboratories and workshops to do practical projects. It is very different from job preparation. E. New confrontations and growth The reform process commenced in 1995, and after several years this is a good time to take stock of what happened after that date. There are three main issues worth commenting. One has to do with the numbers. What happened with enrolment figures? The second has to do with the clientele of the technical schools. Have they become less elitist? The third has to do with subsequent changes in legislation and the changes it brought. The reform played havoc with enrolment, immediately after it was implemented. Of course, it caused the integrated programs to be pared down, as prescribed in the law. It also provoked an immediate crisis in private technical education right after it was implemented. This was confirmed by the administration of one of the largest private technical schools in the country. In fact, this school had to open very quickly the four-year undergraduate programmes, in order to survive the crisis. In the public technical schools, the results are more mixed. 1

Claudio M. Castro, co-authored with Marina Feonova and Anna Litman. "Education and Production in the Russian Federation: What are the Lessons?" International Institute for Educational Planning, UNESCO. Paris. 1997.

166

State secretaries of education that tended to dislike technical education used the opportunity to get rid of their technical schools. Despite serious difficulties with statistics, due to changes in classification, it seems safe to assert that a sharp fall in enrolment took place, immediately after the legislation was implemented. However, as the dust settled, growth resumed. Present numbers seem to show a very steep increase in enrolment. In 2002, 560 thousand students were enrolled. From 2007 to 2010, enrolment went from 693 to 924 thousand students. In the same period, academic secondary remained almost constant, with enrolments around 8.3 million students. In other words, from a purely quantitative point of view, the reform has succeeded, even if belatedly. It is instructive to note that a very large share of the enrolled students take the course after having already finished their secondary cycle. Therefore, for all practical purposes, the technical course has become a one-year post-secondary programme. Some institutions like SENAI, offer the academic side by side with the technical track but try to tune their academic programme to the specific needs of the technical curriculum. On the elitism issue, we know little about the clientele of the federal technical schools, since they were never keen on collecting such data. In contrast to federal technical schools, the public system of technical schools from the state of Sao Paulo has always been much more forthcoming with such studies. It has released some very instructive statistics, showing major changes in the social status of its students. Table I Income levels of students (Sào Paulo): 1995-2002

Family Income

1995

2002

0 - 5 times the Minimum Wage

32%

57%

11 - 20 MW

21%

8%

21 - 30 MW

5%

1%

167

The table above shows a very precipitous decrease in the participation of students from the higher income brackets and an equally sharp increase in the share of low-income students. If making the clientele of these schools more in line with the occupations taught was the goal of the reform, the table is an unmistakable proof of success. That is to say, at least in the State Schools (Sao Paulo), the appropriation of technical schools by the social elite was terminated. A new government was inaugurated in 2002, bringing to the Ministry of Education a large group of people who were at odds with everything that was happening in technical education - in other areas as well. Many of them opposed the split of technical schools in 1995. A few names associated with the politecnia movement acquired much prestige inside the Ministry. Their goal was to reverse the main thrust of the reform, i.e. to integrate the technical and the academic tracks once again. This resulted in much discussion and disagreement, as state, SENAI and private systems did not see good reasons to return to the old system - despite the initial traumas. Like King Salomon, the final decision was to split the solution. The Federal technical schools went back to the old integrated system and the remaining networks of schools kept the new separate tracks. The thrust of the Federal network is to drift upwards in the academic cycle, with more short higher education programmes and less emphasis on secondary-level technical education. Considering that its share in the total enrolment is modest, not that much has been changed. F. The elusive balance: the staircase theory

In very traditional systems of education, it is as if God had determined that occupations either require a complete secondary education or a higher-education programme lasting at least four years. Nothing in between exists. In all mature societies, education systems respond to the obvious facts that the intrinsic difficulties of occupations widely differ and the time it takes to prepare a serious professional to discharge them can be less than four years. Therefore, the systems adjust by creating shorter courses. Countries usually come up with more than one alternative that lasts less the than the classical four years. Indeed, there are no good reasons to suppose that all in-between occupations take exactly two or three years. 168

In France, there are the Lycées Techniques but there are also the Institutes Universitaires de Technologie, with different durations. In the United States, there are one-year technical courses and also the two-year associate degrees. These are interesting variants on the response to complex situations. However, the presence of several postsecondary alternatives in any one country brings obvious difficulties of establishing for each of them the rules, requirements and privileges for the diplomas and certificates. For want of a better word, I am calling the balancing act necessary to address such complexities the staircase theory. The crux of this notion is that we need stairs composed of different programmes, each step taking a little longer than the previous, with no huge gaps in the middle. More complex occupations are taught in higher steps in the staircase, requiring longer courses of study. But since education markets are seldom unregulated - even in the most unregulated country, the United States - there are rules and regulations, pertaining to each of the steps of the staircase. In most cases, these regulations set minimum number of credit hours, sometimes curricula and syllabi and the rights and privileges that go with each modality. The balancing act has to do with how the regulations of each step compare with those of the others. If a step taking fewer years grants too many privileges, compared to that immediately higher, the incentives to enroll in the higher step are reduced. For instance, if the law were to allow the same privileges to two-year courses that are given to four-year degrees, the reasons to spend four years in school would be sharply reduced. If the lower step has too few benefits, in terms of what one can do with the diploma, it will languish and enrolments will dwindle, reducing the range of effective alternatives. The reform that created the new generation technical schools disentangled from secondary education - was implemented at the same time that two additional categories of post-secondary education took shape and began to grow. Brazil has presently three categories of post-secondary courses, lasting less than the conventional four-year programmes: Technical Education This is the category focused in this paper. It can last from one to two years (or even more), depending on the complexity of the occupation. Business degrees can be completed in one year, automation and robot169

ics in two. The certificates are not officially considered as higher education. In addition to the loss in status, graduates cannot use the credits, if they move on to higher education. Sequential education This is a possibility opened by the LDB (Education Law) approved in 1995. Institutions accredited to offer four-year degrees can offer twoyear courses in any subject for which they are accredited. They do not need to request permission to open such courses and there are no official curricula to be followed. It is a very flexible alternative. However, the law was too cryptic in defining such programmes, leading to much discussion and disagreements. In a controversial decision, the National Council of Education decided that this is not true higher education. Or perhaps we could say that it is higher education but not quite on par with other alternatives, since the graduates are not allowed to go on to post-graduate schools (be it the standard master or doctorates or what is in Brazil called especializagao, i.e. an abbreviated version of post-graduate programmes). Technological Education This post-secondary alternative has existed for many years. But it has had a bumpy ride, from its early start in the 1960s. The first course on Engenharia de Operagoes (operations engineering) was created emulating the American community colleges. In recent years, the effort by the Ministry of Education to regulate, develop curricula and streamline this modality has resulted in their explosive growth, from the late 1990s. One of the legal provisions that boosted their growth was the higher education status, allowing graduates to move on to post-secondary programmes of any variety. On the downside are the strict requirements to open such programmes, such as the case with four-year programmes sponsored by institutions that do not have the status of universities or Centros Universitarios. To get permission, institutions have to submit complex projects to the Ministry of Education and the delays in processing them are one of the most critical determinants of expansion of different levels of education and different status of institutions. Therefore, the balancing act that the educational authorities have to manage involves these three modalities. Students can leave secondary school and go to the market, they can take the one-year technical 170

courses, they can take the two-year Sequenciais, they can take the two to three year Tecnológicos or they can enroll in regular four-year schools. The privileges and limitations of each level have to be managed, so that the staircase is preserved and one step does not endanger or distabilise the others that lies above or below. From the perspective of the students there are elements of status, transferability to higher levels, duration and labour markets. A diploma after one year is in itself attractive. But its status cannot be the same. The possibility of transfer is always desirable, but at what price? For older students, is investing four years to get a degree worthwhile? Sequenciais are more focused and specialized, a boost to those who have a clear market niche in mind. But they do not allow for enrolling in graduate schools and may end up having less status. For those who operate the schools, the difficulties of each alternative also have to grow stepwise, lest mutually negative effects arise. This is true not only for the modalities of education considered above but even more so for the types of institutions. The privileges of universities - that can open courses freely - and the unbelievable amount of red tape to shepherd through a request for a new programme for those that do not have this status has been a source of profound distortions in the higher education scene. Private universities grew 356 per cent between 1985 and 1998. In the same period, stand-alone programmes hardly grew at all. Such rules may distort the stepwise pattern in any of the three technical modalities. In the present Brazilian scene, expansion of public schools of any of these steps or varieties is severely constrained. There is the financial stalemate of Education budgets and the high costs per student of public education, bound by inflexible personnel rules, privileges and many other political restrictions. Therefore, the growth is practically dominated by private schools (be they for profit or of the not-forprofit variety, the latter including many disguised for-profit operations). To a very large extent, the growth of each modality will be determined by legislation that constrains or makes it easy to open schools or operate them. Therefore, the National Council of Education has been the stage of fierce - if somewhat disguised - battles to shape legislation pertaining to each. Universities are free to create four-year programmes and Sequenciais, without asking permission to the government. The decision of the National Council of Education to consider that Sequenciais are 171

not pure-blood higher education (no access to post-graduation courses) was a boost to the Tecnólogos, that are higher education. But the law requires a permission to open up a Tecnòlogo from all institutions, including Universities. This has hurt the universities, because it lowered the market value of Sequenciais, that can be opened by universities without previous permission. University lobbies tried to force the National Council of Education to increase the minimum workload of Tecnólogos. This would increase the duration of the courses to three years, making a new and still unknown modality of education too close to the traditional four-year degrees. It would be a strong disincentive to its growth. There is also a new proposal to lower the minimal duration of regular higher education programmes to three years. Combined with the increased duration of the Tecnòlogo, it would be fatal to the latter. Another twist is that the Council wanted to keep the Tecnòlogo shorter and also reduce to three years the regular bachelor's degree. With that scenario, the Tecnòlogo could still prosper. While the present paper is essentially about technical schools, now that technical schools are, de facto, post secondary institutions, their situation cannot be properly understood unless we consider the entire set of post-secondary institutions with which they compete. Any misbalance in the restrictions or privileges granted to each technological degree may endanger or even kill them. This is more so due to the relatively fluid and unpredictable nature of legislation concerning such matters. The staircase theory suggests that the manipulation of incentives and legal restrictions to each modality of technical education is one of the most critical factors in determining their future. Unless there is a stepwise progression in difficulties, prizes and privileges, the balance between each modality may be destabilised, resulting in the crippling of one step or another. And in a highly regulated school environment, labour markets are not necessarily the decisive factors in determining the balance between them. G. The rough ride of technical education This chapter has reviewed the trajectory of technical schools in Brazil. It is somewhat different from that of other countries but, in the end, equally plagued with chronic problems, of which the main ones are summarised below.

172

•

Planners, parents and equalitarians are all seduced by the all-inone formula proposed by technical schools. A secondary diploma, together with technical culture and job preparation seemed like a winning proposition. However, this formula is increasingly being perceived as a hopeless solution - except in a few advanced countries. The balance between the three ingredients is difficult if not impossible to achieve. Often, it is overambitious, trying to do too much and achieving too little. • In most developing countries, technical schools tend to be poorer than their regular academic counterparts and end up being weak in academics, obsolete in technology and insufficient as a preparation for skilled jobs. • Brazil is an exception, having expensive and tuition-free technical schools, offering high quality education. This has generated a Catch 22 type of situation in which the better the school becomes, the more it attracts elite students and, therefore, the less interest the graduates have in the technical jobs for which they are prepared. Instead, they go to the best university programmes. Both equity and efficiency suffer. • With the political thrust offered by an IDB loan, a reform was undertaken to split technical education into two tracks. In other words, technical schools were to offer a stand-alone technical programme for students who already had a secondary diploma or were attending another school where they could earn it. The purely academic track could continue to exist and could continue to attract the elites. However, the elites would not want to enroll in the purely technical track, because they are not interested in jobs. Therefore, less affluent students could find vacancies. • The reform was initially resisted but was firmly implemented. At first, there was much confusion and enrolments went down. Progressively, as the dust settled, growth has resumed. Just as important, the social class level of the technical track has gone down substantially, meaning that the new students have a social profile that is more in line with the status of the occupations offered. In other words, the reform has succeeded in increasing equity and in graduating youth interested in the jobs for which training is offered. • At the same time that these changes were taking place, two other post-secondary modalities were officially created and their legal status defined: Technology programmes and sequenciais. Since 173

these three courses of studies and their diplomas are heavily regulated, the pacific co-existence of the three modalities, plus the traditional bachelor depends on the balance between the time it takes and the privileges granted to each course. In other words, the future of the new technical education is contingent on all the factors that regulate their operation, vis a vis the other modes. Now that this difficult and sensitive reform has been undertaken there is the opportunity for Brazil, as the leading economy in Latin America, to operate a technical education sector that is beneficial to both the demands of industry for increasing skilled labour, and the national population in terms of access, opportunity for skills development and regular income.

174

Chapter IV Multilateral Agencies: Understanding the Beasts

Does Skills Development get Short-Changed in Post-Jomtien Days?

This note is post script to the Jomtien meeting1 and to subsequent paper "Skills for growth, work and poverty reduction" produced by the EFA (Education for All) movement. For many decades, training was under the umbrella of the ILO and kept away from UNESCO - that only dealt with education. However, in the nineties, training (now called "skills development") flirted with education. EFA (Education for All), an eminently educational initiative, opened a little bit the door to training, the fiercely defended feud of the ILO. It might have seemed a gesture of peace and generosity, on the part of UNESCO. However, those behind the scenes knew that it was the heavy hand of the Word Bank - and several others - pushing this idea down the throat of a politically weak UNESCO. While the WB was not, at the time, a vehement supporter of vocational training policies, it was not about to embark in the in fights of two UN agencies (UNESCO and the ILO). It considered both education and training with a neutral attitude and changed the wording of EFA official documents to put education and training closer together. The battles to get the wording right and to find a compromise in the direction different constituencies wanted took much time and was difficult to understand from the outside. I was at Jomtien, but that is not to say that I knew what was going on. In subsequent years, the World Bank had other axes to grind, on numerous issues, paying little attention to EFA. Yet, for UNESCO it became a central theme, perhaps even more than during the Jomtien event, when it was dragged into this forced marriage by the WB and UNICEF, both stronger intellectually and economically. From what I remember, the Jomtien would have taken place, even if UNESCO dropped out.

1

In 1990, delegates from 155 countries, as well as representatives from some 150 organizations agreed at the World Conference on Education for All in Jomtien, Thailand (5-9 March 1990) to universalize primary education and massively reduce illiteracy before the end of the decade. 177

It does not seem that the ILO has paid much attention to EFA before or after the conference. Even the decision to participate in the conference was a last minute decision, with little planning or sense of mission towards getting the declaration right. I represented the ILO and received no briefing whatsoever on what to do or to say. Therefore, considering the natural drift of such matters, training has become a watered down goal in EFA, merging with life skills - a vague concept. Therefore, UNESCO has won the battle, by attrition. With the creation of UNEVOC (UNESCO - International Centre for Technical and Vocational Education and Training), with support from the German Government, the borders have become even more blurred. ILO's claim to have some kind of monopoly on training matters eroded further. Since such "market reserves" are quite pointless, the world is not worse off for that. But the world of training is a bit more convoluted. Training has always been a low-key activity. The lofty and grandiloquent discourse of educationalists draws a lot more followers. The construction of Utopias, the grand plans, the glossing over the real life details capture the imagination and loyalty of millions of followers. Indeed, Education has far more appeal than training. In contrast, training by whatever name is a weaker flag than education. The people and the institutions behind it are less colourful. The messages carry less charm and charisma - actually, they tend to be dull. It is too down to earth, compared to the exalted rhetoric of education. It tends to lean to the right, dangerously close to business, markets and the evils of capitalism. Therefore, outside the narrow paths where trainers move, it carries less weight. Perhaps for those reasons, the training establishment tends to be too weak to contain the push of politicians and administrators to create training fireworks when employment falls and they are expected to do something. "No demand, no training": This was the WB mantra. And it remains a good policy. But in times of crisis, this is not what happens. As a result, the credibility of training is even more eroded. Collin Adamson quotes a letter written by a certain Diouf to his son in which the wise man advises him on career choices. He urges the son to avoid manual occupations and prefer those involving writing. This letter was written in hieroglyphs during the heyday of the Pharaohs. It shows how old is the prejudice against manual occupations. Not a surprise that training has a lower status and gets shortchanged. 178

The above arguments hint at a decadence of training or skills development, as a proud and forthcoming activity. But nothing could be farther from the truth. To illustrate the point I will mention the case of Brazil. An associate and I tried to take stock of how much training is being offered. We did a survey of training in Brazil, including only what was not in the official statistics. As it turns out, the amount of nonformal, non-official training adds up to 4,5 percentage points of GNP. This is more than what is spent on public education. Adding the training expenditures included in official statistics, the volume of training is even larger than what is spent on education. In other words, there is training all over the place. But it remains mostly invisible in the statistics and in the minds of policy makers.

179

Do Training Institutions Learn from Experience?

This essay is concerned with organizational learning and, in particular, policy learning. Sometimes, organizations learn; sometimes they do not. Sometimes they use their acquired knowledge, but this is not always the case. Inside organizations, particular values, rules and incentive systems are developed that may collide with the new learning. In addition, organizations need to structure their responses to the outside world. The main challenge of this essay is to make sense out of a web of interests, prizes, penalties, habits and inertia that may drive the impulse to learn. Individual creativity and the types of organizational climates that stifle or stimulate, learning are also be considered. The paper will present a number of concrete cases studies of both policy learning and non-learning, situations. In some cases, the focus is on the learning that takes place within multinational agencies. In, others, learning inside national training institutions is examined. The following lessons stand out: (i) Organizational culture matters. (ii) The critical issue is not learning but using what was learned. (iii) The chances of learning are greater if the payoffs are positive. (iv) Power inside institutions influence learning and non-learning. To sum up, there is logic to the system but no single and simple pattern emerges. We take for granted the fact that individuals learn from experience. Sometimes, however, we can observe that they do not glean any learning from experience; or they fail to use their acquired learning to change their attitudes and behaviour. What about organizations? What about skills-development organizations? This essay is concerned with organizational learning and, in particular, policy learning. At risk of seeming unduly naive, I would postulate that this form of learning is not very different from what has been observed among individuals. Sometimes, organizations learn; sometimes they do not. Sometimes they use their acquired knowledge, but this is not always the case. Organizations are made up of people. But inside organizations, particular values, rules and incentive systems are developed. In addition, organizations structure their responses to the outside world. The main challenge of this essay is to make sense out of a web of interests, prizes, penalties, habits and 180

inertia that may drive the impulse to learn. Individual creativity and the types of organizational climates that stifle or stimulate learning will also be considered. Indeed, if those involved in an organization do not reflect critically on a particular process, the chances of learning are minimal. The paper will present a number of concrete cases studies of both policy learning and non-learning situations. They were not selected randomly or by any other "scientific method", but instead reflect the past experience of the author, working for the International Labour Office (ILO), the World Bank (WB) and the Interamerican Development Bank (IDB). In some cases, I focus on the learning that takes place within multinational agencies. In others, learning inside national training institutions is examined. As the concluding chapter will show, there is logic to the system but no single and simple pattern emerges. A. Do institutions learn? A selection presented below is a collection of case studies on national training centers, bilateral agencies and multilateral institutions. There has been no systematic attempt to fully document the cases with references or footnotes, as they are meant to merely illustrate typical situations within these types of institutions. 1. The holy German Dual System Several years ago, the following joke circulated in the German training circles. Three training experts were caught by African cannibals and were ready to be put into a huge cauldron, when the chief decided to grant them one last request. The British expert wanted to have a scotch before being cooked. The French wanted to drink a grand cru and sing the Marseillaise. The German training expert, however, wanted to give a last talk, defending the Dual System. The joke is all the more revealing when we learn that it was given in a speech to a large public by the former president of the BIBB, Germany's Federal Institute for Vocational Education and Training. Respected observers believe that the Germanic system of apprenticeship or Dual System represents ultimate perfection in terms of skills development through apprenticeship (Greinert, 2002; Federal Ministry of Education and Research, 2007). As it functions very well in the three German-speaking countries, it is not surprising to learn 181

that German Cooperation has relentlessly tried to export this system to developing countries. For several decades, attempts have been made to implement Dual Systems in all continents. Latin America and Africa have received many German programmes to replicate the model. As the joke suggests, when German trainers meet, there will always be at least one vehement advocate of this system. However, the historical records shows, the results of such a system have become predictable: after a German project is put into place, one incredibly competent and experienced trainer comes along and starts to run the programme, and the end results tend to be adequate or even very good. Yet, the programme remains minuscule. Even after decades of operation, it fails to grow or to be replicated. Or, it is altogether abandoned. The seeds of the Dual System do not bloom into broad and successful training programmes (Castro et al., 1993). And the reasons for this are well known. This system is far too complicated and too difficult to manage for most countries, including certain European nations. It requires the harmonious cooperation of too many institutions. Labour Unions must work side by side with trade associations (which hardly exist in many countries) or with employer's associations, in contexts where some of these institutions are not even on speaking terms. It also requires Labour and Education Ministries to cooperate, another difficult proposition. Klaus Schaack (2008) explains the utter complexity of these institutional arrangements in terms of post-World War II policies: fear of a resurgence of authoritarian governments in Germany lead to policies that attempted to dilute the power of any individual branch of the government, by means of many checks and balances. In other words, the institutional arrangements of the Dual System were indeed meant to be intricate and there may be a good reason for that, the rationale being that no one institution should be given excessive power. That being said, the system works well in Germany - and also in Austria and Switzerland. However, when exported, the institutional and bureaucratic burden becomes overwhelming. Local clones wilt under its weight. In addition, the system heavily relies on a"Meister", who is a regular employee in the firm where apprentices work. The "Meister" acts as a teacher and guide for the apprentices, requiring a high level of skills that most firms in developing countries may not be able to count on among their existing workers. 182

It is surprising to realize how long German cooperation has tried to recreate Dual Systems in developing countries, despite its mediocre results. The strong-willed advocates are still present and continue to make eloquent speeches in its favour - hence the above-mentioned joke. Granted, in the last decade, policy guidelines from German Cooperation agencies have tuned down their support of such programmes or altogether abandoned this path. 1 But perhaps even more surprising is the realization that few if any attempts were made to prune the excessive complications of the model and develop a much simpler version of apprenticeship, for export purposes. Also surprising is to note that the "Berufsakademien" were never exported, an equally complicated programme that caters to higher education students. However, for this type of apprenticeship, small numbers do not hurt, as enrolments in higher education systems are always modest. The inflexibility of the Dual System stands as a solid example of non-learning by high-level and competent institutions in a country known for its scholarship and technical prowess. I do not have the inside knowledge of German institutions to venture a good hypothesis that explains why it took so long to realize that the model was not exportable. Perhaps, the German institutions in charge of skills development abroad were oversupplied with former "Meister" and technical people with a long experience managing such systems. In other words, if the bureaucrats in charge of deciding which programmes to offer were former Dual System participants, they are likely to have a built-in bias in its favour. If that is the case, a purely technical reasoning prevailed over public-policy logic - which must consider institutional constraints and results. 2. The French AFPA exports "alternance training"

French training is typically institution-based. Technical Lycées and CAP (Certificat d'Aptitude Professionelle) programmes, among many others, take place in schools. The few apprenticeship programmes that prepare young workers for the workforce tend to be housed in the German firms that operate in French territory, or concern only the

1

"BIBB aims to be proactive in promoting the advantages of the German Dual System, without necessarily advocating its crude 'export' elsewhere" (Dybowsky et al., 2007). 183

construction trades, following the medieval traditions of the Compagnons du Devoir (Nubler, 1997). However, in addition to institutions that train they youth, France has another agency whose mandate is to develop skills among unemployed populations: the AFPA (National Association for Adult Professional Training). In addition to its domestic activities, the AFPA engages in cooperation efforts, particularly to support former French colonies. I had a chance to visit two AFPA apprenticeship programmes implemented in Algeria and Tunisia, visits that were complemented by discussions at AFPA headquarters in Paris (Castro, 1993a, 1994). In contrast to the French tradition of school-based programmes, these two programmes follow the classic formula of signing up young apprentices in participating firms and complementing their work experience with classroom activities. They are therefore much closer in style to the German apprenticeships, but much less ambitious. Although I was not able to review the Algerian materials, the AFPA training materials in Tunisia were uncomplicated and of high quality. They were also in tune with the low levels of schooling among the apprentices. For both programmes, the subsequent employability of the apprentices was excellent, as reported by the local authorities. In Algeria, around two-thirds of the graduates remained employed after the end of the apprenticeship contract. This is in sharp contrast to the unemployment level of the graduates from school-based training institutions that predominate in the country. According to the official information, more than 90% of them were unable to attain a job in the formal sector. By this standard alone, the programme was a great success. Yet, in both countries, the programmes were poorly implemented and looked down upon by the authorities of the local Office de la Formation Professionelle. In the Tunisian programme, the classroom activities were never organized in a serious manner and the programme run short of funds before it was fully implemented. The authorities never expressed interest and made no attempt to hide their prejudice against this low-end programme. The Algerian programme was equally treated with little respect and gained insufficient funds. As with Tunisia, the high employability of the graduates failed to impress authorities.

184

While the AFPA had done its homework, adapting European apprenticeships to the realities of Maghreb countries, the otherwise competent local training offices could not disentangle themselves from their past experience with regular French training models, which they accepted without blinking an eye.1 In France, the AFPA was intended to move the country away from the purely classroom formulae for skills development. Being targeted to older workers and unemployed youth, it adopted the French version of the "modèle d'alternance". However this very different model landed in Maghreb institutions completely sold on the French classroom model. The two cultures did not blend. Another possible explanation for the failure to recognize the excellence of these programmes has to do with status distribution within institutions. It has been observed again and again that programmes that cater to low status students have themselves a low status within their host institutions. We all know that, within Ministries of Education, primary education seems always to be short-changed, vis a vis higher education. The same seemed to occur with the apprenticeship programmes created inside the French-styled Offices de la Formation (in Tunisia and Algeria). These apprenticeships cater to the poorest students and officers often complained about their low levels of education, without recognizing that they ultimately gained employment where graduates of classroom-based courses did not. Another part of the problem may have to do with external constituencies. The Office de la Formation (in Tunisia and Algeria) tends to offer technical schooling and other programmes that cater to higher status students. From the visits, it was clear that these courses garnered much more attention from management - and certainly, far more resources2. Be that as it may, it is easy to conclude that France learned its sociological lesson. It created an independent agency (AFPA) to operate apprenticeships. In other words, it did not mix two different philosophies of training within a single institution. AFPA also learned its Maghreb lesson, by adapting the programmes to local conditions.

1 2

Nubler's (1997) discussion of the cultural aspects of French and German training perfectly fit the two Maghreb institutions. For a discussion on power struggles within education and training institutions, see Castro and Alfthan (1994).

185

However, a mistake was made. By landing inside institutions that were used to classroom-based instruction and catered mostly to somewhat higher status students, the apprenticeship programmes suffered, despite being more successful than the others in terms of employability of its apprentices. This is a clear case of partial learning from experience. The programme was properly adjusted to the expected clientele but was created in the wrong agency. 3. How multilateral banks learn and teach (preach?)

The following case studies focus on two multilateral banks, the WB and the IDB, which merit a brief introduction on how they operate. Let us begin by explaining how negotiations are conducted with countries for projects or programme development loans, an area that is relevant to the issue at hand as such negotiations involve a good deal of exchanges, in terms of both experience and knowledge.1 First of all, it is always pertinent to remind the readers that banks lend money. They do not grant money, other than on a very small scale. Therefore, countries that decide to take loans must determine what sector of the economy will be chosen. In the last several years, there has been more money available in the banks than the demand for loans. Bank officers therefore try to peddle loans for projects that would be conducted by their own departments, and are also in competition with other banks. Mind you, this is formally against explicit rules. In fact, bank officers are not supposed to proactively offer loans to countries, but this is exactly what they do, in a discrete manner, through their conversations with the higher ups in the relevant Ministry. They attempt to scan the perceived needs of the Ministry, suggesting opportunities for developing loans to address these needs. There is a significant element of learning involved in these multiple exchanges. The bank officers offer a menu of suggestions that are inspired by what other countries may be doing. These conversations can have a high educational content, since the bank officers are exposed to best practices from all around the world and have learned about mistakes in previous loans. In addition, if there is sufficient interest in the propositions, seminars and study visits can be promoted. These are valuable opportunities for local staff to educate themselves in state-of-the-art approaches to different problems. Even if the loans do not materialize, such exchanges open a 1

For a brief description of how multilateral banks operate, see Castro (2002).

186

window of opportunity for learning by officers who otherwise have relatively restricted access to international experience. When banks promote these study visits, lectures and workshops for the local staff, they expect the lessons to be quickly assimilated, so that all those involved can swiftly begin project preparation. Bank officials become impatient and frustrated when all their preaching fails to convince their local counterparts. But learning and being convinced of new ideas is a process that can take a long time to brew. The same French WB officer who blames the "natives" for inertia forgets that the French University Reform, proposed in 1968, has yet to take place (Schwartz, 1983). Low and behold, sometimes up to 10 years down the road, the ideas become ripe for implementation. This is what happened with the health reforms proposed by the WB for Latin America. Ten years after the first discussions, Colombia reformed its health system and other countries eventually followed. In other words, perhaps the most profound lessons taught by multilateral banks take a long time to percolate into society and government bureaucracy, to the great frustration of bank professionals who have tight agendas and targets. Such exchanges can also provide precious insights for the bank officers. They are offered a chance to understand the local context, the constraints, the political environment, and many other informational elements. Just as important for the success of the loan, is getting to know the people dealing with the project or in the power structure of the Ministry. As a broad generalization, facts about education systems tend to be understood and involve lessons that tend to be well learned. But the political and institutional environment is often overlooked. This was more noticeable in the past, but remains a last lesson not fully learned. If negotiations for a loan obtain a green light, project preparation begins. On average, it takes about 400,000 dollars and at least 1 year to get a project ready for approval by the bank and the borrower. Part of the expenditures goes to the detailed planning that is required. But a very sizeable chunk goes to study the sector that will receive the loan. Banks engage their own staff and hire high-level consultants to study specific aspects. We can say with confidence that the sector studies that precede the loans tend to be the best and the most accurate appraisals of the sector. Authors are quite careful with the num187

bers and the results present a quite balanced view of the situation. This is learning at its best. In contrast, the sector loan process tends to be weaker on appraising the probability of project or programme implementation. Compared to technical studies of the sector, evaluation of political and institutional constraints is lacking in strength, which ultimately hinders project implementation. Morocco signed a loan for its Office de la Formation Professionelle, in order to create a network of technical schools. I visited some of them towards the end of the loan and they were faultless, both in terms of their facilities and in the way that they operated - with close links to the market. However, the loan also included a provision by which the Office was expected to set aside a fraction of its budget for funding private training institutions, a provision imposed by the WB. The minimal chances of implementing this provision could have been easily predicted by anyone familiar with the local environment. Authorities of the Office would never cut back their own activities to spare resources for funding private institutions. This is a good example of non-learning by the bank project officers. Development banks often fail to appraise realistically the chances of implementing loans that require changes in institutions, that ignore political constraints or mandate reallocation of resources that go against the grain of power structures. As demonstrated throughout parts of this paper, learning and non-learning occur in tandem in the context of bank activities. 4. The World Bank and its comprehensive high schools

In the 1960s and 1970s, the World Bank was infatuated with the American-inspired Comprehensive High Schools. In the United States, this model was implemented in the early years of the 20th century and predicated that high school and vocational training would be offered under the same roof and within the same schools. While we will not elaborate here on the success or failure of Comprehensive schools in the United States, suffice to say that no European country adopted this model of a single school for all youth of secondary-school age. It also seems to be a consensus in the United States that, in most schools, the vocational training component has been imparted in an amateurish manner and did not prepare students for the workforce (National Commission on Secondary Vocational Education, 1985). In 188

fact, the rapid development of Community Colleges can be interpreted as a failure of the Comprehensive High School model. None of this deterred the World Bank from trying to lend money for its replication abroad: the American model was thus accepted and promoted, without any consideration for its shortcomings at home. Indeed, this represents an episode of little learning by the WB. Wherever they were built, these schools failed to prepare graduates for taking up the jobs for which they were being trained. The reasons were and are very simple: students who are able to attain a respectable-level of secondary education prefer to try their luck at higher education, rather than take up the jobs of a much lower status, for which they were being prepared. In other words, the economic and symbolic value of a high school diploma was and is greater than the value of the technical skills. But the situation was a failure in a much graver sense. The WB schools were built and equipped with the highest possible level of sophistication, and provided with well-paid teachers. As public schools, they do not charge tuition and therefore their admissions procedures become very competitive: who would not want to enter an excellent school that is also free? The end result is inevitable. They become schools for the elite, because the academic entrance tests favour students from high economic and social status who had access to the best primary schools and the privileged cultural capital of their home environment. Brazil had its network of junior secondary schools, inspired and funded by USAID. It also has a significant network of federal technical schools, all of them very elitist. Colombia had its INEMs, patterned after the same model. Students put up with the nuisance of the shop classes, as long as these schools help them in passing the tests needed for the most desirable higher education courses (Castro, 2004). When the WB was preparing its Policy Paper on Vocational Education (World Bank, 1991), in the late 1980s, it commissioned a large number of papers to attempt to uncover what worked and what did not work. The results were quite striking: vocational training in secondary education and comprehensive schools were the least successful, because of the reasons mentioned above and well known to many practitioners. This was the end of this model for the WB, as far as new projects were concerned. The WB operational departments never learned their lesson, even though they were supposed to be following up on their projects. 189

However, the Policy Paper was merciless against such schools and signed their death sentence. Checks and balances eventually worked. 5. IDB training: new project ideas and administrative stalemate

The IDB funded a significant number of skills-development projects for many years. Some of them were high-school level technical schools, but it also funded many training institutions offering training detached from formal education. Indeed, this kind of training is the hallmark of the Latin American systems of skills development. When I was working for the IBD, with Aimee Verdisco, we prepared a paper that provided an overview of all training projects (Castro and Verdisco, 2000a, presented later in this same chapter). We wanted to see to what extent a training project learned from the previous one (not necessarily in the same country, but involving project officers that are often the same and communicate with one another). Our goal was to analyse projects from the point of view of lessons learned for the following years. After all, loans are complex operations, with hundreds of details that can go wrong and that do go wrong, more often than Banks are willing to admit. Given all the reviews and field visits, perceptive professionals should be able to find certain project aspects that are less than adequate or altogether wrong. We looked at all the education and training projects ever supported by the IDB. From this universe, we then selected the training projects from which we were able to identify three categories: technical education, training centres, and the projects targeted to unemployed youth. What came out of our review was the clear delineation of a pattern, and in fact a double pattern. This is what makes this case very curious. Project design, to a significant degree, incorporated learning from previous experience. It eliminated elements that were problematic and introduced new ideas that were gathered along the way. In contrast, little was learned to improve the loans at the implementation stage. Let us first consider design. After the creation of the Brazilian SENAI, a long string of loans to fund similar institutions was signed with Brazil and other countries. These were highly successful institutions, for quite a while. Subsequently, they became slow and fat, losing their ability to target courses to real job opportunities. Showing good judgement, the IDB 190

abandoned this line. Much later, many institutions underwent reforms and new methods were devised for bringing training closer to markets. The IDB, once again, funded projects of this nature. Technical schools had a similar cycle of initial successes and failure to perform. As problems were clearly identified and corrected, the IDB again funded new initiatives. The first programme targeting unemployed youth was Chile Joven. In view of its success in developing a decentralized and self-correcting mechanism for targeting training programmes to job opportunities, it was replicated in Argentina, Peru and Uruguay.1 However, as fresh loans reproduced the model, corrections were applied. For instance, in the Argentine replication of Chile Joven, as a pre-condition for the funding of training providers, internship offers were required, instead of finding real jobs. Due to the high levels of unemployment, finding internship offers was more realistic than job offers. All in all, these are good examples of policy learning. Each subsequent project was corrected based on the perceived faults of the previous project. However, the same cannot be said of loan implementation. The implementation of multilateral bank projects is riddled with traps, dead ends, inefficiencies, delays and all sorts of unsavoury accidents. In contrast to the vociferous critics of multilateral banks who scream against imperialism and ideology, to my understanding and that of many others, this is not where the problems arise. The real shortcomings of the banks are in implementation. At this stage, everything can go wrong. And everything does go wrong. Projects planned for full disbursement in four years may still be running a dozen years later. Very few disburse within the prescribed time. Problems abound among the receiving countries as well: congress may take forever to approve the loan; counterpart funds may not materialize, bringing the project to a halt; and reforms may be impeded by ideological or conservative forces - inside and outside the Ministries. Bidding for contracts may get stuck in legal controversies. Once these initial barriers are overcome, the everyday hurdles continue to hurt and delay the project. Weak institutions are unable to handle the complexity of the bureaucratic procedures. Yet, we observed a chronic failure to appraise the ability of the local staff to implement the loan.

1

For a more complété description of these programmes, see Castro and Verdisco (2000b).

191

Time and again, very demanding reforms were included as part of the projected activities, without a serious effort to gauge the political will to implement them or the administrative competence to handle the complexities prescribed in the loan document. Repeatedly, training was seen as an antidote to weak administration with significant funds allocated to it. Enough evidence is clearly available to show that training does not fix institutional fragility. Training helps when everything else is moving in the right direction. The unavoidable result is that the training component of loans is not taken seriously and often it is not properly implemented. This was shown to be the case in the World Bank and less reliable evidence suggests that the same is true in the IDB. Tendering and contracting procedures are unbelievably complicated, as well as control procedures that ascertain that funds are being properly used. Countries that are in dire need of the loan waste an enormous amount of time with a frightening amount of paperwork. Given the weakness of the ministries in administering the loan, there is a standard practice of creating "executing agencies" to obtain better levels of efficiency and expedite implementation, but there is mounting evidence that these units create more problems than they solve. There is no agreement on whether they are a good idea at all and few promising alternatives exist. Low and behold, every new loan repeats the procedures that all experienced officers know, for sure, will bring about troubles or delays. Very little changes in the administrative routines and, if changes occur, it is not due not to any form of accumulated experience that might suggest the need for repair, but rather is the result of new - and often inexperienced - people taking over the management of some area of the bank and imposing their own operating procedures. In other words, there is no policy learning at the implementation stage. This might seem too radical a simplification, but it is my contention that this is the case: banks do not learn at the implementation stage. Two hypotheses can explain what happens with administrative routines. Either the banks do not learn from past experience or they learn but takes no action. Which one is closer to the truth? It would be pretentious on my part to offer an answer to a chronic deficiency among multilateral banks. However, I can speculate on some organizational issues inside the banks.

192

Project officers have, most of the time, a substantial background in research. They have been educated to ask questions and to probe for results. They have the right mindset to be part of learning organizations. And indeed, perhaps a bit less than they ought, they do improve the new loan designs based on their previous experiences. They can also produce a piercing analysis of everything that is wrong in the borrowing countries, with an array of medicines on hand to cure the ills denounced. However, when it comes to their own bureaucracy, the story is quite different: procedures are created and perpetuated by bureaucrats who are not speculative spirits. They want to play it safe as well as protect their turf and peace of mind. If it takes longer, three times as long, it does not matter, provided that the little bureaucrat is protected from the accusations of lack of oversight, petty corruption or arbitrary decisions, either on his part or on the territory he is expected to manage. They do not maximize economic development, efficiency or whatever. They maximize their own safety and comfort inside the bureaucracy. Technically minded project officers feel helpless in front of a perverse machine they barely understand - at least, that was my case. They do not dare challenge the old-time bureaucrats who might trip them up on any day based on some arcane internal rule. Worse, they are aware of the intricate network of petty officers with the ability to send to bureaucratic hell the naive professionals who dare to challenge the set routines. In my own trajectory inside the IDB, twice I fell in such traps and had the most unpleasant experiences, facing the ugly side of a punitive bureaucracy. Both times, I wanted to move quicker and did not have the patience to wait for the proper procedures. In favour of multilateral banks, it must be stressed that, periodically, they perform self-evaluations, which can be severe and scathing (Castro, 2002). In the case of the research I authored - that is being summarized in this section - the staff did not like what they read. However, the piece was not censored or suppressed. Of course, nobody was very eager to circulate it widely. This entire issue brings to the forefront internal motivations and incentive systems. Why would the banks want to evaluate their own performance in this or that aspect, considering all the pain and embarrassment that it brings? One powerful reason for maintaining such as system of evaluation is that the banks feel pressured or obliged to 193

have such mechanisms in place, if nothing else to give satisfaction to the external and vociferous critics. The results of periodical evaluations are often quite fierce, but what does the bureaucracy do with them? Designing new projects, with new ideas hurts nobody inside the bank. In fact, it enhances the reputation of the bank as innovative and at the leading edge of new trends. New projects demonstrate that considerable learning has occurred and that the acquired knowledge has been put to good use, without challenging old systems. However, who gains by trying to change administrative procedures? Negative reactions tend to be powerful, albeit discrete. Managers can collide with smaller bureaucrats who fear change, particularly changes that place their small empires at risk, and who know how to make life unpleasant for those who dare encroach on their territory. The possible personal gains from improving bureaucracy are overshadowed by the losses resulting from clashes inside the administration. Hence, organizations are not good at learning about problems concerning their own internal administrative procedures. This is true everywhere, not only where it concerns skills development. More often than not, an uncomfortable or demolishing evaluation is quietly ignored. Or, any initial moves towards change eventually wear out, to everybody's relief. Learning, yes, but not much is done to correct course. 6. Is SENAI a learning organization? Founded in the early 1940s, the Brazilian SENAI is one of the oldest Latin American training institutions. Its original model was remarkably innovative, showing a great instinct to learn from other countries.1 From its origin in Sao Paulo, SENAI spread to all Brazilian states and, in addition to the original initiative that focused on manufacturing trades, other parallel institutions were created, to cater to the service sector, rural areas, transportation and small enterprises. The model was soon replicated in almost all Latin American countries. SENAI, therefore, offers ample ground for inquiry into issues of learning or not learning from experience. As will be shown, it does indeed offer good examples of learning, but also cases where no learning took place or where errors persisted. 1

For a more detailed description of SENAI and its origins, see Lopes (1982) or Castro (1993b).

194

During World War II, Brazilian manufacturing was booming as a result of restrictions on imported goods, but the education system was in terrible shape and no significant vocational training existed. The Ministry of Education proposed the creation of its own network of vocational schools. However, Brazilian industrialists were not comfortable with the idea and offered to create a system of their own, funded with public monies. This was a revolutionary idea. Indeed, Brazil is the only country adopting this solution of private training with public funds (only much later, Chile also adopted this model with limited scope). Despite the opposition, the strong-willed President of the nation accepted the scheme proposed by industrialists. SENAI was born, initially in the State of Sao Paulo, where industry was strongest. A Swiss mechanical engineer was appointed as its first executive. Given his previous position as a training supervisor for a local railroad, this person came up with an old idea that was developed in the 19th century for the Russian railroads: the "methodical series". This method consists of presenting to the student a sequence of practical projects, going from those requiring the simplest operations to more complex and difficult tasks. In each module, a new skill is introduced and the previous ones are reinforced. SENAI developed this idea much beyond its original Russian conception. Skills such as reading and writing, blueprint reading, applied mathematics, physics, and technological information were built seamlessly into each module, following the rule of increasing complexity. The programme has no curricular grid, but is designed as a sequence of projects. Every skill to be learned is built in smoothly within the routine of working on the sequence of projects. When an apprentice finishes all the practical projects, he is entitled to a certificate and is ready to work in the occupation where learning was acquired. No further tests are required, since each project has its own established minimum performance levels. To move to the next level, the apprentice has to reach that minimum level. In light of current learning theories, this model remains state-of-the-art. It is still better than anything else available, as it allows for the seamless integration of different spheres of knowledge as it progresses from the concrete to the more abstract. A third innovative element was the use of a 1% levy on industrial payroll, in order to finance SENAI. In simple terms, public money is used to fund training institutions operated by industry. This puts the training in the hands of those who hire the graduates. In addition, it 195

creates a simple rule, yielding a predictable budget that avoids the endless battles in Congress. Instead, the budget is simply defined at 1% of payroll. Altogether, SENAI brought forward substantial innovations, setting Brazil ahead of just about any other country of equivalent economic development at the time. It was born as a solid learning organization, which had tapped many ideas from different sources and has sought to improve on them. Subsequently, every Brazilian State created its own SENAI, as an exact copy of the Sao Paulo original. From the beginning, this classroom formula, based on a sequence of practical tasks, shared the floor with apprenticeships schemes. The latter was a soft or light version of the Dual System where twice a week, in the mornings, young apprentices with a work contract in factories would attend classes in Portuguese, mathematics, drafting, etc. While no rigorous evaluations of this system of apprenticeship are available, it seemed to have worked well. However, it progressively lost steam and little of this model remains today. It seems that the main culprit is the inability of those in charge of labour laws to understand both the logic of markets and the issues of age. An inflated pro-apprentice bias progressively eliminated the incentives for firms to hire apprentices, and a 14-18 years-of-age window for apprenticeship lost touch with reality. SENAI is not guilty here, as the problem rather lies with "bleeding heart" attitudes on the part of Labour Ministry legislators and staff. This illustrates a wrong form of learning. In the 1970s, the SENAI from Rio de Janeiro came up with a significant innovation: it maintained the methodical series concept, but instead of having students progressing lockstep, it allowed for individualized trajectories. Each apprentice could move at his own pace, going from classroom, to drafting room, and to shop, according to his learning speed. This allowed faster students to finish more quickly, liberating their slots to new students. In addition, due to dropping out along the course, idle seats accumulate. With students working independently, it becomes possible to fill immediately the slots of dropouts (Agudelo, 1980). All evaluations indicated that this was a significant innovation. The system became more efficient, turning out more students and costing less (Oliveira and Castro, 1991). Everything was ready for replication elsewhere, as was the case with the original Sao Paulo model. However, this did not occur and I was never able to gain the full 196

story. One possible reason is that the innovation took place at a moment when the State of Rio de Janeiro had already lost much of its industrial leadership. It never managed to sell its solution to Sao Paulo, with its powerful SENAI that welcomed close to half of total enrolment in the country. As the manufacturing industry of Rio lost more and more weight, the individualized-instruction model wilted. In so many words, little guys do not impose solutions to big guys, no matter how good these solutions might be. This is a case of foregone learning. In the 1990s, a very bizarre development took place. With the growth of the Brazilian graduate schools of education, SENAI acquired a rambunctious group of critics. Its private ownership and public funding through tax money was anathema to these people who espoused leftist ideals. To protect itself against such boisterous critics, SENAI began hiring graduates and postgraduates from education schools. It also believed that these individuals could bring significant improvements to the quality of instruction. To my understanding, this was a gross mistake. These educators brought in the Brazilian (bastardized) version of constructivism. They are adamantly against textbooks, handouts, or step-by-step and structured materials. Since they claim that students "construct their own knowledge" and each constructs it differently, materials should be tailor-made to them. Hence, teachers must develop their own materials. We all know that this is a recipe for disaster as it brings about permanent improvisation. Piaget is the intellectual father or grandfather of this group but he cannot be considered guilty of endorsing such silly ideas. Skills development is an area with abundant teaching materials of high quality, which are highly structured and can be used by millions of students. But as a result of this "Fifth Column" inside the SENAI, the methodical series were criticized and ridiculed. They were accused of being mind-numbing and Skinnerian. The early methodical series were conceived for students enrolling with no more than 4 years of schooling. As a result, they were a bit narrow and constraining. But as schooling levels increased, this is not an intrinsic shortcoming, just a sign that most of courseware needed updating. Sao Paulo resisted pressures to dump these models and is presently producing some splendid new versions of the methodical series, but most other states were pray to this childish criticism against structured materials. The results are quite clear: the degree of improvisation in classroom in197

creased dramatically. This is the opposite of organizational learning. It is unlearning at its worst. SENAI not only expanded to other states, but it also lead to the creation of similar institutions in various sectors including the service industry, agriculture, etc. Two of which were plain clones (e.g. SENAC for the service sector and SENAT for transportation), while another two were entirely different. SENAR (for agriculture) dropped the idea of being a private entity and was created as an institution directly managed by the government. The same was the case with SEBRAE (for small firms). In both cases, one of the most revolutionary aspects of SENAI was abandoned. Not by coincidence, both became horrendous public bureaucracies. This was the wrong policy learning on the part of the Labour Ministry. SENAR and SEBRAE failed to copy perhaps the most appealing aspects of the model. Compared to the other siblings, the shortcomings of the two public institutions were so gross that, subsequently, the government privatized them, making it like SENAI. This time, we see sound policy learning on the part of the federal government. Both institutions improved dramatically as a consequence (see Castro, 1993b). SENAI was created at the heart of the Brazilian manufacturing sector. Not only that, but Sao Paulo kept up with technological advances and SENAI thus became increasingly sophisticated in its offerings. From its early start preparing machinists and electricians, it now trains technicians in robotics and even offers postgraduate courses to engineers. It also does quite well in R&D projects and technical consulting to industry. But as it migrated to less industrialized states, Sao Paulo remained the model. Therefore, there is always the risk of copying solutions that are not so well suited to different local contexts. There is not enough research on this topic and my own first hand impression is that the SENAI course offers in less industrial states are more sophisticated and expensive than necessary. They also fail to tailor-fit the programmes to the needs of very small shops, where graduates may become employees or even future employers. However, in the last decades, some progress along these lines has been achieved. One remarkable innovation is mobile training. Let us take as an example, far in the interior of the poor state of Ceara, the town of Sobral, where there is a SENAI outpost. The mobile training facility consists of a significant parking lot area. The staff consists of a man198

ager, a salesman and a secretary. The salesman visits nearby towns, convincing mayors or businessmen to buy SENAI courses or industrial consulting services. Those who want courses have to pay part of the running costs and offer space for the installation of the shops, as well as lodging for the instructors. At the given date, a truck delivers a huge metal container that unfolds to display a full workshop for the chosen trade. The courses may last weeks or months. Afterwards, the truck picks up the crates and carries them to the next location. For towns that need consultants to set up or revamp businesses, the SENAI salesman will identify a competent person from a list. Mobile training has existed for a long time. But as time goes by, the solutions improve and past hitches are eliminated. Again, this is a good example of learning organizations. The development of those unfolding and folding boxes for each occupation is a fine example of technical innovation. In a couple of hours, a complete workshop is ready to start receiving students. It is not mobile training but rather transportable training, a much cheaper solution compared to adapted buses or trailers that are more expensive than the workshops they carry. This is a case of continuous learning along several decades. Two countries in Latin America did not imitate the SENAI model. One was Mexico, as it was too closely influenced by American styles; the other was Argentina. At the time, it was ahead of Brazil in just about every dimension of development. By contrast, all the other Latin American countries imitated the SENAI model. The decision to export the SENAI model was not a planned exercise in international cooperation. Instead, it occurred through a chance event. One of the top SENAI administrators lost his political clout to a new tribe of highend bureaucrats. Removed from power, his best option was to obtain support from the ILO and go about duplicating SENAI in other Latin American countries. His first stop was Colombia where SENA was created. The second was Costa Rica, where he helped to establish INA. Together with SENAI, these two became the leading training institutions in Latin America. Other similar institutions followed suit in many countries (Cinterfor, 1990). Yet, they all made the same mistake as the original SENAR and SEBRAE: they were created as public institutions. The advantages of being private puts SENAI and the other Brazilian counterparts onestep ahead of the others. There is less waste, less corruption and a much better targeting of courses for existing jobs. This latter argument is quite potent. Employers' associations have strong incentives 199

to ensure that the training offered by the schools they control is targeted to real jobs. Going against the grain, these countries created public instead of private institutions, which was the wrong form of learning for Latin America. Or better said, incomplete learning, since the other aspects that were copied worked adequately. Some of the Latin American counterparts were quite creative. For a while, SENA from Colombia was the leading star. In some ways, it was more creative than SENAI. However, the internal difficulties of Colombia had deleterious effects even on training initiatives. INA of Costa Rica was also quite innovative. Chile is a fascinating case. During the Pinochet tenure, neoclassical economists had ample political power to experiment with privatization schemes, in all sectors of the economy. This group displayed a mixture of pragmatism and laissezfaire ideology. There was a decision to privatize INACAP, the Chilean public training institution - very similar to its Latin American counterparts. However, compared to SENAI, it was a different model of privatization. The INACAP funds were transferred to firms so that they could use them in purchasing training programmes from the market. Firms were free to decide on how these funds were to be used and where to purchase courses. INACAP therefore had to fight for contracts and had to become "lean and mean" in order to survive. The model did not provide the institution with stable public funding, but rather forced it to compete for funds with other institutions (cf. Wilson, 1996; Cox, 1997; Martinez, 1997). The results are quite instructive. INACAP was a sound institution and was able face the market turbulence. It survived and prospered. However, initial training offers suffered. As in all industrial countries, firms are reluctant to invest in initial training. The reasons are simple: they can hire workers who have already been trained and if they choose to train their workers, they risk having these employees poached by competitors. Some observers consider that Chile still pays the price of too little initial training for those who cannot afford the costs, or specifically the poorer populations. In other words, the development of mechanisms to outsource training was an important social learning in Chile. Low and behold, the solutions did have some unwelcome consequences, particularly in the case of skilled manual occupations. In recent years, most institutions have become closer to the market, creating stronger links with firms. In addition, entrepreneurs have become more powerful actors inside these institutions. All in all, there 200

is learning taking place in several institutions that are generating fresh solutions and fixing old ailments. Surely, some have undergone serious or almost fatal crises. A few are in a sore state, crippled by inertia and politics. To a certain extent, their ability to overcome old problems and create new solutions is correlated to the country's level of economic development. That is to be expected. To sum up, training in Latin America has over half a century of innovation, learning, and learning feedbacks. Compared to other parts of the developing world, this is where most innovations have taken place. Nevertheless, the process is neither linear nor are the solutions always forthcoming. Many errors were committed and some of them were not repaired. Worse, there are a few cases of "unlearning". 7. Demand-driven training: in theory or in practice?

Starting from the late 1980s, it became obvious that many graduates from public training institutions failed to find a job that sufficiently matched the skills they had acquired. Either that or they were unable to find a job in the formal sector or, worse yet, they were unable to find any job at all. The World Bank took the lead in pointing to the absurdity of offering expensive training programmes that lead nowhere. I was at the ILO headquarters in Geneva and also reverberated the idea, promoting it as one of the leading issues in training policies (Castro and Kanawaty, 1988; World Bank, 1991). The issue was quite obvious to us: the number one problem in training was lack of targeting. Too much money was being wasted due to the fact that the training programmes offered around the world did not lead to jobs that would justify the training costs, even minimally. The WB and the ILO widely preached these ideas. Interestingly, we found little opposition from the national training institutions, regardless of the continent where they were located. They were always received as a sound policy principle. After all, why train people who are not going to be able to deploy the acquired skills? Nevertheless, the reality was a different matter. We hardly saw any attempts to better target training and the situation continued to be "business as usual", with the same huge gaps between training offered and available jobs. Not that authorities in the training sector had any obligation to please the ILO or the WB, but they owed it to 201

themselves and to the taxpayers who funded training an obligation to offer meaningful programmes. Obviously, no learning was taking place. Or, if they learned, it did not lead to action in a flagrant clash between perception, policy and action. The reasons are not hard to understand. The logic of efficiency clearly sees no sense in wasting money on training that will not be useful. However, institutions have their own logic that often overpowers the efficient use of resources. Workshops, curricula, training materials and instructors are all already deployed to offer specific courses. If the graduates fail to find jobs or are unhappy for any reason, they have very few means for effectively or forcefully complaining, and often disperse after graduation. Their ability to impose changes in the training centres is close to zero. Potential employers may be disgruntled because they cannot find the skills they need, but they are also far away from the training centre environment and are not adept at complaining about such issues. The fact of the matter is that very few are interested in what goes on in vocational training. Therefore, unhappy customers have few means to make their displeasure known and even less power to put effective pressure on training centres. In other words, the political cost of missing the target is close to zero. In contrast, closing a course that has no market, for whatever reason, can bring deep political turmoil inside the institution. What do you do with the surplus instructors? How do you resist the political pressure from unions or other such political bodies that rally in favour of those instructors being displaced? New courses will have to be planned, developed and staffed, but with what resources? Inertia is the most pervasive bureaucratic trait. Change is painful and fiercely resisted. In so many words, the political costs of redirecting public institutions away from waste and dysfunctions can be huge. They tend to largely overpower the political advantages of changing course offerings. Therefore, repeating what was said before, under such conditions, either the institutions do not learn (mostly because they do not want to learn) or they learn and do nothing about it (Castro and Andrade, 1990).

202

B. Lessons: training institutions learn, but not always After sharing my experience and case studies, and realizing that I could tell many more such stories, a more difficult task is at hand: learning lessons. While each case study provides some insights into learning or non-learning at the policy-level, the following section will attempt to glean some general patterns from these examples. 1. Hidden agendas People and institutions have hidden agendas. What they say and, above all, what they do reflect these agendas, rather than formal rules or official rhetoric. These hidden agendas are driven by self-interest and organizational culture. A proud and open agenda is one where innovating, aspiring to improvements and fixing past errors are encouraged. But these are neither the only agendas nor the most powerful ones. There are, in fact, negative agendas that are focused on avoiding conflict. In many of the cases examined, failure to learn was a means of avoiding painful or dangerous clashes with the bureaucracy. The hidden agenda can be more powerful than the official statements of institutions. 2. Organizational

culture

Individuals inside organizations act in ways that are in accordance with accepted norms. Organizations develop their mores and patterns of approved behaviour, and they frown upon those who challenge them. In some rare organizations, criticizing colleagues and denouncing inefficiencies may be acceptable. But in most, it goes against the grain of organizational culture. It is considered a lack of loyalty to fellow workers. At worst, it is viewed as treason. Prudent officers know that a disgruntled colleague can trip their careers. As an example, in the multilateral banks, we observed that while the most serious failures are in the bureaucratic machinery, this particular area is off-limits to course correction that may be quite obvious to any perceptive observer. Hence, policy learning is seriously curtailed by unspoken rules, such as the hesitation to embarrass colleagues, or the avoidance of new policy decisions that would require a major upheaval for implementation. The French AFPA is not bound by the strong culture of the German Dual System. In fact, apprenticeships in France are a minor part 203

of the training machinery. Therefore, one could speculate that designing a trimmed down system for export purposes would not be a "crime lèse magesté", as it would be in the case of German training bureaucracies. SENA of Colombia had a long stalemate, resulting from the powerful presence of its employees' labour union. It could not get away from a very dysfunctional situation because it would have meant total war, and no executive wanted that, no matter how well informed they may have been about possible solutions. Decades elapsed before some of the most serious stalemates could be sorted out. 3. The critical issue is not learning but using what was learned In examining a fair number of cases, we came across many situations where mistakes persisted and were not corrected, with some persisting over a long period of time. Perhaps the institutions failed to learn from experience and did not realize that there could be other ways of doing things that were better. But perhaps this is the wrong way of approaching the question: while many may have known what was wrong and also knew of the right solution, in a cost benefit analysis they would have quickly realized that denouncing the error would be perilous or foolhardy. Heads of training institutions pay lip service to studies that trace the professional life of former graduates. But in practice, rarely if ever are these types of studies are performed. The WB or the IDB have imposed the few existing studies. But institutions do not want concrete evidence that training does not help graduates get jobs. With or without the quantitative evidence, the incentives to do nothing are paramount. Hence, we never know for sure whether individuals or institutions learn and do not use what they learned, or if they never had the incentive to learn. Perhaps this is a moot point but the end result remains the same. 4. The chances of learning are greater if the payoffs are positive In situations where professionals may "get medals" for learning and proposing new solutions, there are good chances that such behaviour will happen. Of course, alert and creative collaborators are required to come up with good ideas. But chances are that if there are personal 204

gains from learning, this will prompt people to invent solutions or identify problems. Typically, the design and engineering departments of firms are rewarded when they have good ideas. The greater the rewards, the more ideas they have. Bank officers who develop training projects are, in effect, Research & Development workers, in charge of creation. Some are not able to innovate and end up producing what bank jargon calls "cookie cutter" projects. But those who come up with better ideas get points for that. In fact, they may even get real money prizes. In so many words, there are incentives for being innovative in project design. Surely there are some limitations, in particular, from the host countries. It is easy to be creative but it is not so easy to be a Minister trying to implement a complex project, full of never-tried solutions. This paper reviewed a set of innovations inside training institutions. It also examined institutions that expanded or were replicated in different versions or in different places. In some situations, people who designed new projects were rewarded. On the contrary, learning can reveal errors and create tensions and conflicts inside institutions. Therefore, those denouncing errors will face opposition and, eventually, reprisals. In a typical situation and for most people, the best course of action is not to rock the boat. If that is the case, either no learning takes place or the solutions are not seriously considered. Let me illustrate this point with the example of a loan to a poor South American country, which consisted in two parts. The first element was to fix the official training institution that was in deep trouble: due to mismanagement, there were no funds to pay teachers or instructors, only the managerial and service staff, and as a result it did not have one single student. The public policy argument for fixing this institution stands to reason yet, several years after the loan was signed, nothing had happened even though the planned repairs were crystal clear and simple. Politically, the costs of fixing the problem were astronomical. The second part of the loan was an elaborate system of vouchers distributed to small firms. With them, the firm owners could buy training for their employees in any of the accredited private vocational centres. Against all odds, this complicated segment of the project worked quite well. The contrast is stark. For the first part of the loan, the least important repair would unleash an avalanche of powerful complaints. In the second, everybody stood to gain from the new 205

scheme. Even the bureaucrats were complimented for doing the right thing.

5. Power inside institutions matter The two Maghreb countries previously mentioned should have learned that the AFPA programmes were far more successful and less expensive. However, people pre-sold on the classroom-based formula monopolized decision-making. By the same token, the Dual System defenders must have had enduring power inside German bureaucracies. The mediocre results that accumulated all along the years took a long time to yield policy changes. IDB project developers corrected errors in new projects, because they had learned from previous versions, but they never touched the serious shortcomings in project execution because the powers to be were too threatening if they did. The aloof attitude of most training institutions when faced with the mismatch between training and demand can be observed everywhere. What counts, in practice, are the political costs of stepping on the toes of colleagues. This is because former students and employers are far away and have few means to voice their complaints in an effective manner. In other words, power matters. Those who yield power inside institutions, to a very large degree, control the decisionmaking process. Reason, logic and efficiency remain on the back burner if the learning gained challenges the comfort or peace of people with some power inside the institution.

206

Can Multilateral Banks Educate the World? 1 This paper can be seen as a primer to educators who know little about the World Bank (henceforth WB) and the Inter-American Development Bank (henceforth IDB).2 It tries to draw a broad picture of these somewhat mysterious and awe-inspiring organizations. While both banks lend for a wide variety of sectors and the rules and principles are approximately the same, the paper focuses on education loans. One consequence of this narrower focus is to bypass the painful and inconclusive discussions about the choice of sectors where these banks should operate. This essay is not the result of reading books and papers - even though some writings are helpful to better conceptualize the issues.3 The author has been an officer of both banks, as well as a close observer from the outside. This brings to the narrative a first-hand perspective and an effort to make sense out of a personal experience as an insider. There has been an earnest attempt to prevent personal interests and perspectives from contaminating his views. However, to claim neutrality in such matters is naive, at best. A. How Development Banks Work The logic behind a multilateral development bank needs to be understood clearly. They are neither commercial banks, nor are they agencies funded and under administrative control of the rich countries (even though these countries wield considerable power in major decisions. Their goal is to promote the economic and social development. After World War II it helped countries ravaged by wars. After the 1950s they focused their lending in developing countries. Perhaps at some moment in their history, economists thought they knew how to promote development. But history proved they were wrong. These days, all certainties about development recipes are gone. Therefore, 1

2 3

My gratitude to Ricardo Santiago, also a former IDB officer who read the manuscript and made useful suggestions. However, he does not necessarily endorse all ideas presented here. While other multilateral banks are similar, the author is not familiar with them. Some of the sources come from internal memoranda and other papers that were not meant to be published.

207

such doubts compound the quest to find out whether they have been successful in their endeavours. As should become clear along the text, no extreme views are warranted. Banks are neither a disaster nor an immaculate success; neither evil nor saints. They fail and they succeed. The gist of the presentation is to indicate some common denominators that have been observed. 1. Economic rationale: low interest borrowing

Let's start with a simple question. What is the financial magic performed by a development bank? In other words, in purely financial terms, what has justified their existence? Several decades ago, a number of countries got together and decided to create a bank of their own so that some of them could borrow at lower interest rates. But even more important than that, borrow for long periods of time, such as 15-30 years. Hence, from day one they were banks owned by governments. Naive comments about banks not being "democratic institutions" completely miss the target. They are democratic in the sense that their governance is exerted by a Board chosen by the member countries. Perhaps some countries have more power than others in this governance, but that is another matter. Technically, they are not banks. They are more like credit cooperatives - where only members can borrow. Or, more precisely, credit cooperatives of participating countries. How do they work? Whenever the United States wanted to borrow money, it issued bonds that pay the lowest interest rate in the market. Why? Because the country was rated "triple A" in the financial markets, due to the very low risk of default. Less developed countries have a much lower financial rating. Therefore, if they decide to float bonds, they will have to pay a rate of interest much higher than that paid by the United States, because borrowers are afraid the country will default on its obligations. If the average Latin American country tried to borrow money today, it would pay interest that would include a considerable spread, to cover the risk of default. When the World Bank or the IDB float bonds in financial markets, they can do so at the lowest interest rate of the market. Hence, the whole idea of a "development bank" is little more than a market trick to allow poor countries to borrow money at rates that can be as low as half of what they could obtain on their own and with much shorter time to pay back. But in order to break 208

even, the development banks have to add another percentage point in order to pay the salaries of the staff. And there is lots of staff, because this is a very complicated business. But since multilateral banks do not make profits, this in itself further lowers the interest rate.1 Therefore, the trick is to create a bank that includes the poorer countries that want to borrow and also the rich countries that lend their reputation and reliability. The credit worthiness of the big countries - the United States, Germany, Japan, the UK, and so on - permit the bank to have a "triple A" rating. This rating allows the bank to borrow at the lowest possible interest rate found anywhere in the world. In contrast to what some people think, Banks do not lend their own money. They lend the funds they borrow. In fact, development banks do not lend rich countries' money. They lend the money they borrow from financial markets. Therefore neither the United States nor any other country is subsidizing the World Bank loans. The same goes with the IDB, or any other development bank. The United States, like others, contributed a modest amount of money in the beginning, to build initial capital reserves. Bankers know that just a little bit of capital leverages significant borrowing in financial markets. Hence, in this respect, development banks are just like commercial operations; there are no subsidies involved. They borrow money by issuing bonds, and then they lend this money at interest rates that allow them to break even. The downside of this operation is that banks have very little funds to make grants. In fact, it is easier for them to lend $200 million than to give $20,000 as a grant. Why? Because banks have as the only source of money what they borrow from the market. And if they borrow, they have to pay back. To be more accurate, they receive a modest amount of grant money from rich countries (lately, Japan has been the most generous donor) to selectively support a few projects. But the grand total of grant money is a very small proportion of what they lend. The good news is that the banks have a lot of money. They can make a difference in many countries. They can add to the funds that

1

To simplify the discussion, this chapter ignores what are called IDA operations, geared to the poorest countries in the regions. These loans use money directly provided by donor countries. In these loans, the countries providing the capital have a much stronger voice.

209

would be available in public budgets an amount of money that a country would not otherwise have. In the absence of a loan, such funds would have to be subtracted from the salaries of civil servants or the provision of basic services. 2. Countries have to repay their loans, no matter what It is instructive to find that most professionals in the banks don't know much about the banking business. Ask anybody at the IDB or the World Bank about interest rates. Chances are they do not know. And there are good reasons for that. Banks don't worry about repayment of the loans either. This is because if the country doesn't pay, it is put on the IMF's "black list." And being in this list is one of the most uncomfortable situations a country could be. It means that the country will not be able to borrow from any bank, multinational or private. Perhaps even more important, the countries lose their ability to have both the IMF and the WB as bankers of last resource, i.e., if worse comes to worst, they still have access to the funds from these two agencies. This stiff penalty allows the banks to completely dedicate themselves to the lending end - or development objectives - of the operation. Bank officers worry about developing and implementing projects. In simple terms, there are no credit risks to worry about. Therefore, they do not behave like banks in the true sense of the word. They don't need to know about interest rates, repayments, or the like. In many ways, project officers are neither the bleeding hearts who give money away nor the bankers who think only about interest payments and how to make money off someone else. Bank officers deal with development projects. They were recruited because their professional profiles suit them for those tasks. 3. Loan preparation takes a long time Before development banks lend, they have to ensure that the money will go to projects that are fully spelled out, both in conception and in plans for implementation. There are good reasons for that. Improvisation is avoided and the ideal conditions for implementation are planned, including all the precautions against corruption. The first step before a loan is a careful diagnosis of the sector where the funds will be spent. The downside of this has to do with the fears of bureaucrats and their attempts to protect themselves against accusations of laxity and 210

lack of control. In too many cases, the bureaucratic requirements go much beyond what seems like reasonable precautions. Whatever the reasons, it takes at least one year to negotiate a loan and prepare the requisite contracts and documents. If there are disagreements between the Washington staff that is preparing the loan and the representatives of the recipient countries, it may take much longer. In the case of problematic loans or countries with ineffective governments, it may take several years to get it approved. 4. Strict contracts protect loans from political vagaries The good side of all the bureaucratic formalities is the stability and continuity that it brings to the programmes funded by such loans. A minister can announce a programme to be initiated the next day - to give milk to the children or to build schools. In three months' time, he may have been replaced by another minister announcing another programme. Guess where the money comes from? Of course, from the programme the previous minister created that a few months earlier. But change is not so easy with projects funded by a development bank. The country is committed. Contracts are signed. In other words, if both sides go through all the troubles to prepare and sign a project document, a significant amount of work is needed to get the project cancelled. This is not a good idea. In addition to losing face, the loan funds cannot be used for alternative uses, because it is locked in this operation. For decades, for every dollar included in the loan, the country's Treasure had to put another (the exact proportion depended on the level of income of the borrowing country). These are the counterpart funds. One clear implication is that not only the loan contract was binding for the funds to be disbursed, but it also froze another chunk of money of the same magnitude. Compared to the average government programme, this inertia gives a much greater stability and coherence to the bank loans. They are more robust and have far fewer chances of being killed at the political whim of a new minister or a new administration. As a result, continuity, structure, discipline and technical support have a greater probability of presenting themselves. Loan design reflects ample experience with similar loans in other countries. It represents the best practices known in the world, given the broad perspective and knowledge basis of technical staff. 211

However, there is also a downside of this inertia. A bad project is difficult to change. Evolving conditions and circumstance would suggest new strategies that are not easy to implement. For good and for evil, loans have a great amount of inertia. Another advantage is that loans set aside an amount of money that allows a country to hire the very best minds in the world, which are expensive. Usually it is politically difficult for a country to hire such expensive consultants, particularly when local salaries are so low. But well-chosen experts can make a difference. B. Why Development Banks are accused of the worst sins Gone are the glorious days when the World Bank was merrily prescribing policies and lending, rather than stirring controversies. Today, some accuse it of being the satanic tool of "neoliberalism" while others complain that it is plainly ineffective. Bank officers ponder how they can be both at the same time. These days, being a multilateral bank implies that you are criticized, no matter what you do or don't do. "Fifty years is enough," said the critics at the fiftieth anniversary of the World Bank. The United Nation specialized agencies have been paralysed by fear of displeasing member countries. In contrast, the World Bank always came forward with strong views and outright preaching on just about everything, from structural adjustment to contraceptives. As much as this was a relief from the wishy-washy positions of the United Nations, the collisions with governments and independent critics were inevitable. Could this ideological rejection come as a surprise? Well, development banks are indeed strange animals. As a former employee of these institutions, we did not know whether we were gods, devils, preachers, bankers, arrogant civil servants with Ph.D.'s, or just hopeless and pathetic bureaucrats. Actually, we may very well be a mixture of all these. /. Values and beliefs The accusations against the World Bank - and to a lesser extent to the IDB - have been persistent and, if anything, receive ever more press coverage. The days when they proudly announced their goals of wiping out poverty and backwardness are gone. In many ways, they now pay for their past arrogance, their hubris, and the obvious fact that 212

they promised more than they could deliver. But that is not to say that the critics got it right. The banks are often accused of being "neoliberal" - the worst insult in the lexicon of the left. Of course, it would be naive to imagine that it is possible to have an institution operating in such central and delicate areas without an ideology. Neutrality and agnosticism are not options. To have any meaning, policies have to accept, deny, or qualify all-encompassing concepts such as free markets, private property, and public intervention. Therefore, implicitly or explicitly, banks do have their own ideology. The question is how acceptable this ideology is to the borrowing governments and to its critics. The World Bank and the IDB still believe in some mild version of the Washington Consensus. But it is worth noting that this consensus is an ever-shifting notion. In addition, there is a huge difference in the strengths of such beliefs in either bank. The true believers have been in the World Bank, the IDB being far less religious and more pragmatic. However, stripped of all that has been added to this concept - by friends and foes - principles such as balanced budgets, paying your bills, and controlling inflation are goals that most countries accept these days. Notice that the original Washington consensus of John Williamson was not much more than that (Kuczinski & Williamson, Chapter 10). Below is a list of the principles current in the World Bank that, in a way or another, influence education lending (in a toned down version, they are the same in the IDB). • Evidence-driven policies (controlled evidence is the reality check) • Cost accounting (one must know how much it costs) • Quantification (the more one can quantify, the better) • Efficiency (getting more done with less resources is a permanent goal) • Meritocracy (the best are to be chosen and rewarded) • Competition (belief in the "invisible hand") • Market incentives (rewards and penalties are the most powerful incentives) • No subsidies (prices must reflect true costs) • Privatization (private ownership improves efficiency) Complying with some version of these principles was mandatory on most projects, in order for the loan document to have smooth sailing inside the World Bank. At the same time, while the IDB was also 213

driven by such principles, their defense was substantially less vehement. Nevertheless, this is only the beginning of the story. Thinking in either bank is far from being monolithic and compliance with the ideological impositions is more formal than real. Many loan officers added to their proposals all the mantras of the moment, knowing full well that they were not enforceable. Why fight for ideas that will never get out of the paper? It was more expedient to get the project approved quickly and then proceed to implementation, where realism prevails. The idea that bank professionals form a disciplined army, thinking alike and implementing official policies is naive. This could not happen in an institution that hires the top graduates of the world's best universities - with analytical skills honed in debates and controversies. In fact, attending technical meetings in the WB one cannot escape the conclusion that, more than anything else, professionals enjoy showing off and fencing with one another about theories and policies. There is a wide range of opinions and policy orientation among professionals and a lively debate goes on all the time. But in matters considered as serious, there is far more than intellectual legerdemain. In fact, as reported by Heyneman and witnessed personally by this author, during the first half of the 1990s, the doctrinaire clashes between education professionals, division chiefs and the management of the bank were nothing short of formidable (Heyneman, 2003). 2. Decision making: How the world influences the banks

To understand the logic of multilateral banks, it is useful to have a brief discussion of how decisions are made. Bank policies are generated in the interplay of three main actors: (i) management and staff, (ii) the Board, and (iii) the outside political forces that have a say in these matters. Notice that banks are well staffed at all levels. It has been said again and again that no other agency in the world has such an intellectually powerful and experienced staff in matters of development. In fact, no other contender stands up at the same league. The mindset of the banks is a natural consequence of the academic origins of its staff. It comes predominantly from the very best universities, with a strong participation of graduates of leading AngloSaxon institutions. Therefore, ideas spontaneously reflect mainstream 214

economic traditions. Not surprisingly, intellectual arrogance and hubris often comes in the same package. Ideological and theoretical beliefs derive directly from these academic origins. Bank practices acquire a style of their own but they rarely collide with the tenets of classical economics. Banks operate in a political world and cannot ignore it. The WB members include just about all countries in the world. The IDB includes Latin American countries plus most rich and large countries as mentioned, their presence ensures low risks and low interests for the funds it must borrow to make available to in the form of loans to member countries. In practical terms, the United States (via the Treasure Department) has a disproportionate power. Being the largest shareholder, by the constitution of the Bank, it has more voting power. Presidencies or vice-presidencies are tacitly allocated to Americans. In the past, there have been the tugs of war between the United States and other countries. Today, in the majority of the cases, American positions do not go against the grain of most other member countries. But there are exceptions. For instance, indirectly, the United States has been able to prevent the entry of Cuba, both in the WB and the IDB. Yet, it failed in its attempts to save the presidency of Paul Wolfowitz, when accusations of nepotism surfaced. North European countries take a strong stand for the environment, minorities, Indians, and the very poor. At times, it clashes with borrowing countries that may want to remain reckless in these matters. In other situations, borrowing countries find that they are too poor to impose strict Scandinavian standards. And they may be right. NGOs' themes are similar to those of the Scandinavians but tend to be more strident and radical. At first, their whistle blowing played a useful role. But they seem to have gone too far, after banks got their act together in those matters. In addition, there is an increasing perception that some horror stories were not true. Particularly in the case of the WB, NGOs have put much pressure to have a greater say on the lending operations. In fact, some of them are clearly antagonistic to the very existence of the banks. The "Fifty years is enough" campaign was very troubling to the WB, to the point that the commemorations for its half-century anniversary were severely curtailed. Washington is a sounding board for NGOs and they are able to create much noise. This is a very contentious situation, as the noise created by hundreds such institutions bring a bad press for the banks. 215

Finally, borrowing countries are entitled to have a voice. Why not? At first, the rationale was much in the line of the "white men's burden." Poor countries were thought to be disorganized, have ineffective and corrupt governments and lack good judgement on what is good for them. But progressively, it has become politically inappropriate to pursue this line. In addition, many countries have a clear and sound idea about their needs and preferences. In more recent years they do have a louder voice in deciding on their loans. It took some time to realize that when borrowers take ownership of the project, the chances of success are much higher. These political constituencies have had an oscillating role in influencing the policy orientations of the banks. For instance, specific projects to deal with one aspect of development or broad, all-encompassing development policies? Resulting from the interplay of all these constituencies, the banks developed their policies and styles. 3. Leftwing educators battle "neoliberal" values

"The supra-national agencies such as the IMF and the World Bankfare] like a kind of economical and political ministries of transitional capital... They are the lords of the world or the de facto power in the world." (Frigotto, 1995, p. 82) "...all this [education development] is being orchestrated by the World Bank with its concept of education systems linked to vocationalization, aiming at preparing individuals to meet the requirement of productive activities and market circulation..." (Ianni, 1996) "There is a growing participation of outside economic agents - the World Bank and the Interamerican Development Bank - as well as local entrepreneurs." (Monteiro, 2000, p. 48) "Sector Analysis documents of the World Bank, in general, display a lack of understanding and inadequate knowledge of education and accumulated research in the field... The dismantling of Education Ministries is, to a large extent, the direct and indirect result of the World Bank and IMF packages... The World Bank ... package presents serious shortcomings ... and instead of contributing to change in the proposed direction ... reinforces inefficiency, poor quality and inequality in school systems." (Torres) The above quotes illustrate the kind of opposition met by multilateral banks. Most of them have one feature in common. They reflect little knowledge of institutions and result from very cavalier research. 216

Let us look at a very instructive example. A book by a well-known researcher in Brazil bitterly denounced the World Bank for having masterminded the reform of technical education, by means of a loan to the Ministry of Education. She went on to describe how this foreign institution was interfering in domestic educational policies (Kuenzer, 2000). As it turns out, she did not take the time to verify that the loan was from the IDB not the World Bank! In fact, the World Bank never lent for technical education in Brazil. Furthermore, it was denouncing policy changes originally proposed by the author of this essay while he lived in Brazil, ten years before he joined the World Bank and the IDB. Subsequently, when the loan to Brazil was being designed, this idea was introduced in the loan by this author who was part of the project team. Therefore, the supposed interference by a foreign power was merely an old Brazilian idea proposed by a Brazilian staff. 4. NGOs fight for influence in

decision-making

From the 1990s on, NGOs claimed increasingly greater power to influence and to veto loans to developing countries. This has been particularly true in loans where environment issues were at stake. It is a documented fact that some dams proposed by the World Bank were environmental disasters. And it is to the credit of watchdog NGOs that such reckless projects were denounced. In part due to such criticism but also as a result of a greater overall awareness towards environmental issues, the Boards reacted by imposing a severe environmental screening, since the early 1990s. Notwithstanding, the NGOs toned up their criticism. The whole issue became somewhat muddled, as the NGOs kept repeating the criticism levied against old projects, since there was little to criticize in the more recent crop. This lead to bitter confrontations, even after bank projects became very careful in dealing with the environment. In fact, some loans imposed unreasonable increases in expenditures on the projects, in order to avoid some minor problems (for instance, to protect a small monkey from a road development in Brazil). Ultimately, in deciding the fate of a large loan, angry activists can wield more power than some major countries. The whole issue seems to have gone overboard. One ridiculous example was a small project by this author to conduct student-achievement tests in ten Latin American countries. The preparation of the project was delayed by requests of one participant in the environ217

mental clearance board. This person wanted further assurances that applying tests to a few thousand students committed no sins against nature. C. How Development Banks design and negotiate loans In order to understand the operation and the logic of development banks, it is useful to examine how loans are prepared and negotiated. 1. What governments want, what countries need, and what the banks want to sell When the bank sends a project officer to negotiate a loan, this action mobilizes several actors and lead to the discussion of a complex and sometimes contradictory set of goals. It is useful to try to understand the typical scenarios. Bank officers have development goals in their minds. Typically, they honestly try to analyse the situation and derive the policies that would be most helpful to the country. Critics may discount such lofty goals and accuse officers of hypocrisy or fake idealism. My own experience of ten years inside both banks leads me to believe that my colleagues meant well and did know their line of business. They try to design projects that respond to what they perceive as the legitimate needs of the countries. Of course, they can be dead wrong, but that is not the issue. In other cases, they may be indulging in what is called "cookie-cutter projects" that repeat solutions tried elsewhere. But that is not the predominant situation. In fact, professionals tend to be moved by an earnest sense of duty in the search for the best policies. In principle, officers propose what they see as the ideal solutions to the problems faced by the country. Their most egregious sin is to ignore institutional barriers to implementation, such as the presence of politically strong opposition to some dimensions of the project. In this sense, the project may be unrealistic, because it proposes to do more than is politically possible. But the staff means well. If they err, it tends to be due to naive idealism, not evil intentions. Programme officers do not have "the country" as counterpart. Instead, they deal with ministers and those representing them. The Minister may be easily persuaded that what is proposed by the Bank is what his country needs. Low and behold, Ministers are politicians who have as their number one priority, survival in the position. As 218

mentioned by the former French Minister of Education, Luc Ferry, many appointed ministers believe that they have been given a horse, in order to go somewhere. But it takes little time to realize that they are in a different sport. They are in a rodeo and the goal is not to fall from the bucking horse. Therefore, what is good for the country may be politically impossible, may cost his job or erode his political capital. What most impressed this author in many missions to discuss direction and content of loans was this ludicrous situation in which a foreign representing a bank defends the real interests of the country while the Minister is concerned with the arithmetic of power. But there is nothing evil in the Minister's behaviour. It is just the game of politics. If there is a frank dialogue, the Minister may be candid about his predicament. In common accord, political means to make the project more palatable to its adversaries may be sought. Very often, it is agreed that the blame for including this or that item will be put on the Bank, the Minister playing the role of a passive victim. 2. Economic realities or political feasibility?

Practical economists and bank officers very often face a severe tradeoff between what is the best economic solution and what is politically feasible in a given situation. In order to understand it, one needs to follow the ups and downs of how decisions were taken by banks, along their history. The World Bank was born as an institution to fund the reconstruction of a Europe ravaged by World War II. It was a brick and mortar bank and it did not need be more than that. All that Europe required was to rebuild its physical infrastructure. After having done that, the WB moved on to lend to less developed countries, all around the world. Sooner or later it became clear that these countries needed more than buildings and roads. Their institutions were in severe state of disrepair. In the 1970s, the World Bank added institutional reform components to its loans. The idea was perfectly sound. But its implementation turned out to be far more ridden with problems than imagined. The results were not very good. The relations with clients became very confrontational and, in many cases, little reform was achieved. At the same time, the IDB went on with simple loans to construct and equip. Little was asked in the line of reforming institutions. These loans were far more successful. They were easier to implement, even 219

though the serious problems - mostly in dysfunctional institutional settings - remained untouched. In the 1990s, the IDB modernized and became more concerned with reforms. At the same time, the World Bank softened its demands for institutional changes - after having seen so many projects stalled. Progressively, the two banks became more similar in the way they operate. In a few extreme cases, the IDB became more reform-oriented than the World Bank. But overall, they are still different banks. Let us construct scenarios that are far too exaggerated. The World Bank approaches the countries with elegant solutions to their problems. It does its homework and produces sophisticated papers analysing the issues and proposing creative and well spelledout schemes. When the time to negotiate comes, it tends to be prescriptive and to insist on its own elegant views. Quite often, the project officers do not know well the country, its politics, its bureaucracy, and power structures. Countries that are less able to defend their ideas succumb to the pressures of the World Bank and accept the loans as proposed. Several outcomes are possible. In one scenario, the loans are not implemented. In others, little in terms of reform takes place, despite great consumption of energy on both sides. Another scenario is when the Bank staff meets well-prepared and assertive government staff on the other side. Sharp disagreements may ensue. Such countries are forthcoming in what they want and stand behind their ideas. But once agreement is reached, after considerable give and take, there is a much better chance of successful implementation. IDB style tends to be different. It goes to the country and asks the Minister what he wants. Its officers know better the country and its institutions. And they always speak the language. Several outcomes are possible. In one scenario, a very simple loan is proposed, foregoing the potential to achieve something else in reforming institutions. In more favourable scenarios, politically viable reforms are negotiated with the Minister. The above is a simplistic and exaggerated description of either bank. But for this reason, it helps one to understand the tone of the negotiations and the differences between the two banks. But in addition to all those considerations, purely personal elements and circumstances - on both sides - play a role that is far greater than anyone wants to admit. Project officers end up having 220

much leeway in negotiations. They are relatively free to fine-tune the projects or to make blunders. A skilful officer has a refined perception that allows him to go as far as possible, but not farther, in insisting that this or that feature is included. He knows whom to trust and he knows where the real power is. A clumsy or arrogant officer may clash with his counterparts and insist on what is not politically possible. Bad choices can be made. By the same token, local staff makes an enormous difference. Here is one example. The IDB and the WB prepared two similar education loans to Ecuador. The IDB went to the rural sector and the WB to urban marginal schools. The WB was a great success. The IDB project never really functioned properly. The reason had to do with the excellent local manager of one loan and the unfortunate succession of mediocre managers of the other project. 3. Conditionalities: powerful, but dangerous Let us now examine what is the most delicate part of development banking. This is something called conditionalities. It reminds us of Teddy Roosevelt, when he proposed to speak softly and carry a big stick. Loans may have strings attached. Conditionality thus means that the country must fulfill a certain number of requirements in order to get the loan. For instance: reforming teacher statutes, reducing deficits, charging full costs of utilities, preserving the environment, or whatever is fashionable at the time. In the majority of the cases, conditionalities will have two key features: (i) They are good for the country - at least in the long run (ii) and they carry political costs to incumbents that must implement the reforms. Ideally, a conditionality uses the clout of the Bank with the massive volume of funds it has to offer to push through reforms which, most informed citizens agree, are welcome but politically costly. Therefore, they probably would not be undertaken if not backed up by bank pressure. This puts in the hands of the banks a most formidable weapon for pushing reform. It allows a Ministry to overcome political opposition or inertia. Conditionalities give the banks the power to be a catalyst for reform. But it is a dangerous weapon and becomes the favourite target of ideological opponents. Past experiences with conditionalities have 221

taught very humbling lessons. First, things can go wrong. Second, to be quite frank, the banks cannot impose reform. This was the painful lesson the World Bank learned in the 1980s, when as much as one third of its portfolio was underperforming. Why? Because the Bank tried to impose loans and reforms by brute force. We need to be very clear about this: Banks cannot reform. They are pitifully weak in this regard. The picturesque expression "herding cats" has been used to describe those failed efforts (a paper on this subject can be found ahead in this book). However, banks can be catalysts for reforms that are close to happening. In order to achieve that, they should have good antennas to find where reforms are brewing. Along these lines, the best banks can do is to find the good guys at the right moment and support them in their efforts to make reform happen. The education reform of the State of Minas Gerais (Brazil), in the 1980s, illustrates this successful strategy. A strong leader with considerable experience in education became State Secretary of Education and pushed for reform. A W B loan made implementation much easier.

4. Conditionalities in action (or inaction?) Since conditionalities are the most critical and contentious element of a loan, it is worth offering some concrete examples. Cases were chosen due to the author's proximity to them and to illustrate the variety of possible outcomes. Let us start with one of the worst scenarios. The political price of reform may be excessive and the advice ill conceived. Bolivia accepted a reform that imposed tests for teachers and eliminated the tax that funded their unions. When the government tried to implement these policies, this created such a political turmoil that the country went into state of siege. These conditionalities were a bad idea. Paraguay contracted a very elaborate loan with the IDB, in order to reform and support vocational training. The loan had two parts. One went to an executing agency that set up a competitive fund to train employees of small firms. The firms received training vouchers for their staff that could be used in any of the approved private schools. This segment of the loan was a great success. The other part of the loan, much larger, was directed to the national training authority. But it was contingent on several structural reforms in an institu-

222

tion that was undergoing a severe crisis. Years elapsed and the second part could never be implemented. Morocco took a WB loan to construct technical schools and to undertake some structural reforms in their system. The technical schools were built and became an instant success. But the loan also prescribed that part of the budget of the Office de la Formation Professionelle was to be redirected to fund competitive bids from private training institutions. Years went by and, in every visit, the officer of the WB would meet the Minister and ask about the creation of the fund. The answer would be predictably evasive. Of course, it goes against the grain of civil servants to take away from their own budget to fund private schools. The condition imposed was unrealistic to start with. Brazil wanted a loan to fund technical schools. But the predicament of these schools was that they offered an academic education of the highest quality, parallel to the occupational training. As a result, upper class students would bid out the more modest students in the competition to enroll in the technical school. But all the selected students were interested in was, once graduated, to get approved in the most competitive examinations for higher education. The other students who could have an interest in the occupations taught would be left out. For all practical purposes, the expensive occupational training was a complete waste. The IDB imposed as a condition for the loan that the technical schools were to be split into two separate programmes: the academic secondary school and the occupational training. Therefore, upper class students would not be interested in enrolling in the vocational track. They had all the good reasons to choose only the academic programme, leaving vacancies in the occupational track to students who really wanted this type of courses. The Minister welcomed this condition. For him, this was the right thing to do and he had the power to face the political opposition. However, when an IDB executive tried to propose that the academic track be closed, this was a non-starter. The Minister simply said no. In this example, a first condition that was agreeable to the Minister and was implemented. The second condition was unacceptable and was discarded during negotiations.

223

D. Are loans effective? The ultimate criterion in judging the role of multilateral banks is to check the result of the loans. This is the real goal. Everything else is the means. 1. Who benefits? By looking at the performance of loans, one should be able to derive some generalizations about the profile of those countries that benefit the most and those who get little out of them. It is curious to notice that despite an incredible volume of papers and documents produced by banks, relatively little is done to gauge the effectiveness of loans and their ultimate impact on development. Matters have improved considerably in the last five years. But it does not seem that the present situation is entirely satisfactory. Loan procedures follow very predictable and spelled-out procedures, both to prepare and to manage during execution. Nothing is left to chance. And to wit, on both stages, this is a very complex and demanding work, putting a very heavy burden on public bureaucracies. In fact, first-hand experience shows how painfully slow and inadequate is their delivery of services. Given the shortcomings of most public bureaucracies, we know full well how painful is the burden created by these requirements. Therefore, it is not a surprise that the weaker the civil service, the more excruciatingly slow is their operation. It follows that the execution of the loan will be harmed. It must be noted that even in the same country, some ministries are alert and efficient while others are ponderous and unable to deliver. Hence, wise bank officers must evaluate each bureaucracy and gauge its ability to handle the loan that is being proposed. 2. What can go wrong with loans? Projects in the social area can go wrong in many possible ways. Unless they are mere physical infrastructure operations, social projects are complicated and depend on the adequate performance of a multitude of agents who cannot be directly controlled. From the point of view of this paper, let us examine some typical problems. Disbursement delays are the most conspicuous difficulty and the easiest to understand and measure. The cash flows, automatically monitored by bank accounting, capture many of the difficulties and shortcomings that may suffer a project. In other words, anything that 224

goes wrong usually delays disbursements. It could be waiting for approval in a slow-moving Congress. Until some years ago, delays in counterpart funds were a most recurrent problem (more recent loans dispense with them). Or, the executing group is not efficient. These are problems that tend to be purely administrative and reappear frequently in projects that have very little in common. All such snags converge to a single result: money is not disbursed and the timetables are not met. And the more complex the loan, the more difficult it is to implement. While this author does not know any systematic research on this subject, first-hand observation suggests that countries known for their weak bureaucracies tend to have loans that do not perform or are delayed. The same applies to different ministries. For that reason, loans that perform well in one ministry tend to be followed by other successful loans in the same ministry. Of course, there are accidents, such as the loss of critical officers managing the loan, on either side. Low and behold, problem-loans can be fixed and fine-tuned to perform splendidly. An extreme case is an education loan to the north-east of Brazil that took 10 years to really get it working well. Reform-intensive loans are the most vulnerable. Even in poor countries, banks often do well in building infrastructure. They overcome the bureaucratic weaknesses by hiring international firms that know the business of building bridges and railroads. School or hospital construction is usually subcontracted to the private sector. But problems appear when banks try to tackle more complex development problems, such as reforms that change the way people and public institutions work. When banks try to go beyond heavy infrastructure and help the poor in poor countries, this may turn out to be impossible. Such projects require a level of bureaucratic robustness on the receiving end that is rarely - if ever - found. Programmes to help the poor must break down the resources in very small parcels that will go to a multitude of small institutions. Controlling this complex web of transactions is a nightmare. This is the case with rural schools, health posts, distribution of milk, or whatever requires capillarity to reach the bottom of society. Sometimes, bank staff cannot even prepare loans because the bureaucracy on the other end is administratively weak and cannot present background data and statistics. When the loans are made un225

der such conditions, they often end up being failures or they limp along for a long time. To sum it up, banks can help the very poor in middle-income countries. But they have too often failed in trying to help the poorest of the poor in poor countries. This is an unfortunate contingency. In recent years, both banks have become less arrogant and more humble, learning to be more careful in adjusting the projects to prevailing conditions within countries. The World Bank has tried to sell reform time and again, particularly in the 1980s. More often than not, it has not worked. It used to be that designing a modern reform project was a challenge for bank staff. Now, staff in both banks knows how to prepare such projects. The real challenge is to understand the country and to know how far it can go in reforming its institutions. In conclusion, development banks do have something to offer, even to poorer countries. But it has to be tailor-made for each case. The good news is that they are learning their lessons. However, banks cannot escape from a serious dilemma. The better off the countries, the more effective and reform-oriented can be the loans. And the poorer the countries the more they need reform and the more difficult it is to help them. Let us consider, for instance, that some of the recipient countries have fewer phones than the offices of the WB. Or, countries in which horse-drawn ploughs represent a technological innovation (Mallaby, 2004: 342). In other words, those who need the most are the ones who can benefit the less from the multilateral banks - at least, the way they are organized nowadays. Worse, even at the design stage, no better solutions have been found for funding development in these poorer countries. Ultimately, too much depends on the internal dynamics of the country. Banks are unable to alter it significantly. If the country cannot get its act together, loans are likely to perform poorly and have little impact on development. 3. Managing loans: a nightmare of paperwork

Banks are not as evil and incompetent as their nemesis denounce. That, however, is not the critical issue. In fact, most accusations miss the target. They focus on policies and policy controversies and usually fail to notice the most critical weakness of multilateral banks: the implementation stage. 226

Outside observers, both right and left, focus their comments on policies, policy debates and the intrinsic merits of what the banks are proposing to do. Yet, to my understanding, this is not where the real problems are. How useful is it to discuss endlessly the ideological soundness of a proposal to reform teacher regulations if all the country can implement is the construction of schools? Past experience has taught the banks very humbling lessons. First, everything can go wrong - and things do go wrong. Banks do not advertise much their failings at the implementation stage. Some projects die at the preparation level, due to the inability of the country teams to respond to the requests for information or due to basic disagreements about what the project should look like. A few are approved but never disburse, due to legal snags or bureaucratic inertia. Most loans actually disburse, but slower than predicted. In fact, very few loans come to completion within the originally scheduled programme. A significant proportion gets stuck somewhere along the line, due to many different reasons - the most egregious one used to be delays in counterpart funding. A cycle of 15 years is not uncommon. The fact of the matter is that loans have very complex disbursement mechanisms and accounting, since banks and the local government share costs and responsibilities. Bureaucratic controls to avoid corruption and embezzlements slow down everything. 4. Evaluation has unclear impact on design and procedures

In all fairness, banks have departments that have as their main task to evaluate past performance of their loans. Some of the papers produced are well prepared and candid in their appraisals. In fact, there are splendid and perceptive pieces that are able to grasp the difficulties in implementing loans. However, banks are not too eager to disseminate statistics showing which projects fall victim of each of the problems already mentioned. All parties concerned stand to lose. Perhaps this is the weakest link in the entire system. Low and behold, the results, while not tragic, tend not to be very flattering. Probing a bit deeper into the realities of implementation, there is one recurring result. The more physically concrete the line of activities, the greater the chance of being implemented. Typically, all schools are built, most teachers are trained and computers are purchased. Also typical, the reform component is not implemented.

227

5. Ideas or funds? Late in the 1990s, the W B management floated the theory that it should go into the business of helping countries with ideas, papers, and policy suggestions, instead of lending money. In parallel to making loans, it would become a huge consulting firm. After all, the technical competence of the W B is peerless. This proposal seems a good launching pad to discuss the reform components of loans. Are there the good ideas that the banks could provide to countries? Would countries pay for them? Would they implement them? We cannot go beyond conjectures and subjective perceptions. But for what they are worth, it seems to this author that these ideas will never fly. Ministries do not seem particularly motivated to contract expensive consultants to propose reforms that are politically painful. And most reforms are painful. Let us not forget that other international agencies, such as U N ESCO and ILO, are prodigal in advice to member countries. And it is quite clear that countries do not pay much attention to what they propose, even though the ideas come free of charge. Experience seems to suggest that one big and real motivation to engage in reforms is the large sums of money that come with the loans. Therefore, what sets banks apart from other international agencies is exactly the fact that the funds provided by the loans, somehow, force the Ministries to pay attention to the policy suggestions. In fact, the conditionalities - mentioned before - are just the mechanism to push the implementation of the policies proposed. Small countries - or those that are very anxious to get a loan, in order to have more slack in their budgets - are the ones that tend to take the advice more seriously, particularly when they come as conditions for the loans. More affluent countries are in a better position to resist conditionalities that have a heavier political price tag. Take the case of Brazil, where the national development bank lends three times as much as does the W B to the entire world. Therefore, W B loans in Brazil will not buy a reform that is needed but politically uncomfortable. If the country undertakes a reform, it is because the moment is ripe, not because of the loan contracted with one of the banks. Having said that, it is also necessary to recognize that banks are powerful and creative proponents of new and valuable ideas and 228

policies. Hundreds of professionals and an even greater number of distinguished academics are engaged to do serious policy research on most areas in which the Bank lends. The close links with the academic world give a strong credibility to the ideas disseminated by the W B - and to a lesser extent, by the IDB. In fact, banks originally proposed many ideas that acquired currency and legitimacy. This is the case of testing students and using the results to evaluate progress in education. The same can be said of the emphasis on high-quality textbooks and teaching materials, as well as computers. Also, ideas on health and social security reform originated in the WB. They were at first rejected but eventually, were successfully implemented. In this respect, it is interesting to notice that new ideas take long to mature. The first paper on health reform was published more than 10 years before the first country adopted it - to wit, Colombia. Another wild card in this game is that ideas change. Cynics may claim that they are subjected to the same vagaries that determine feminine fashion. The World Bank and the IDB pushed higher education vigorously. Then, they decided that there was over investment in universities and the crisis area was basic education. Then, no more lending to higher education for over one decade and a barrage of papers denouncing the mistake of putting money in universities. Now higher education is back in fashion. The question that comes to mind is the inevitable doubt about the impact of new ideas versus the impact of the loans in the long run. One could argue that good ideas eventually become acceptable and implemented, while loans tend to merely build schools or equip them. As much as this is a critical issue, it does not seem that we have good evidence to further clarify matters. There has been much discussion on matters of ideas versus money, particularly in the WB. In part, they have been prompted by outside critics that tend to exaggerate the power of multilateral banks. Yet, some observers have a much more sobering or even pessimistic view. For instance: "World Bank funding exerts little influence on social policy.... High concentrations of World Bank funding have virtually no impact on the share of educational resources devoted to primary education.... Although the World Bank has the financial resources and the technocratic allies to buttress the transmission of its ideas, bureaucratic and political forces often get in the way... In the final analysis, the World 229

Bank can pressure but cannot force the Brazilian government to adopt its recommendations... Domestic political forces prevail over international technocratic linkages when it comes to redistributive social policy making." (Hunter & Browne, 2000, pp. 126, 129, 135, 138) Be that as it may, in the long run, ideas and a better understanding of problems may very well be the most important contribution of development banks. But they take long to mature. E. Lessons? There has been mounting criticism levied against the Banks (as always, the WB is the main target). Much of it is heavily ideological. Some is just the visible manifestation of interest groups, in their efforts to influence the bank lending and policies. Our central contention is that most criticism is misdirected. The egregious shortcoming of the banks is the fact that great ideas succumb in the process of implementation. Countries just cannot implement the projects funded by the loan. But the banks are just as guilty, because they misjudged the implementability of the loan. At first, life was easy, countries badly needed the loans and were not ready to take an adversarial stand. But more and more, strident critics from the left began to criticize the banks, both in the advanced and in the less developed countries. Banks were accused of being an arm of imperialism and to impose their ideology on borrowing countries. These indictments often have a grain of truth but are largely unwarranted, as attempts to disqualify almost everything the banks do. The argument here is not that banks are entirely innocent. They are not, but the critics tend to miss the target. The ideology that stands behind bank loans reflects what Western mainstream economists believe. Surely, there are dilemmas. For instance: more growth with larger deficits and inflation? Or a more conservative financial management? More reform or more bricks and mortar? Be that as it may, only a few of the bank loans are contentious. It seems that minimally rational critics could find little to complain about the average loan. Who can be against roads or hospitals, in the case they are truly needed? As someone who has had significant contact with those designing and implementing loans, the impression one gets is that most officers are eminently competent, have ample experience in what they are do230

ing and mean well. Perhaps idealism is too strong a word, but very few officers are cynical or jaded about their work. They try to get to know the country they are dealing with and produce competent analysis, leading to the policy lines they think would be most appropriate. Of course, they have made many mistakes along the way. But they are doing their best and nobody really knows how to promote the development of a country. It was an easy bet to rebuild factories or roads in Germany. But what should they do in Chad or Haiti? A most relevant aspect of their work has to do with new ideas for reforming policies and institutions. No other agency in the world has generated so many innovations in these matters, most of the time, backed by solid empirical research. In fact, no other agency has several thousands well-trained professionals in their staff, with ample experience in the field. Long gone are the days when UNESCO had a leading role in spearheading new ideas in education. The WB has taken up this role on earnest. It is instructive to remember that the Jomtien meeting of Education for All had the strong leadership of the WB. UNESCO had no choice but to participate, and did not have a strong voice in the ideas and organization. In education alone, the banks have championed the use of student's evaluations, carefully prepared textbooks, accountability of programmes and agencies, merit pay for teachers, competitive bids for training contracts, new schemes for vocational schools, student loans, vouchers, and many other policies. It is instructive to notice that some of these ideas were initially rejected with vehemence but, after so many years, became acceptable practices. A case in point is students' evaluation that careful research shows to be the best tool to monitor education. The main problem is that the speed of acceptance of new ideas often does not coincide with the life cycle of a loan. It may take much longer for an idea to be accepted. Loans are bureaucratic procedures with timetables and budgets. Ideas have a life of their own, float around, and may find strong rejection for a much longer time than the loan lasts. Policy and ideological issues have become the object of live discussion and sometimes bitter arguments by outside critics. Yet, it is the contention of this paper that most of these invectives miss the point. 231

The Achilles' heel of the entire lending process is not ideology or imperialism. The real issue is whether the loans deliver what they are supposed to deliver, ultimately, development. As suggested before, implementation is a big problem. Banks staff and rules are a lot less enlightened in delivery than in conception. Bank procedures to execute a loan are ponderous and inflexible. Worse, there are no powerful forces to seriously repair projects that do not seem to be performing well. There are no prizes for whistle blowing. In fact, along the chain of design and execution, everybody may be guilty of some error if the project does not perform. For that reason alone, there is a strong reluctance to denounce poor performance and to redress procedures. For almost all concerned, hiding weaknesses may be a more expedite line of action. This is also true for the Ministry that receives the loan. It is highly probable that its performance has been less than stellar in managing it. The last thing the Minister wants to do is to recognize those failures and do something about them. Loans get delayed, disbursement stalls for bureaucratic snags, and abrupt changes in staff create havoc with execution. And above all, those reform components that happen to carry a greater political price fall behind and may never be implemented. That is not to say that all or most of the loans are ultimately ineffective. This is not the case. However, despite the effort of evaluation offices, we know little about their impact on development. F. The new scenarios

The world is changing, perhaps faster than one would expect or like. The situation of multilateral banks is presently quite different from what it was 10 years ago. Many of the implementation problems presented in this essay reflect recurring issues facing the banks. However, while most of these problems have not been solved, new ones emerge. Banks have no problems to raise capital. All they need is to float titles in the world's financial markets. However, potential borrowers are increasingly less interested in taking loans. In the 1990s, many countries were having difficulties in continuing to take loans from development banks. Several were indebted beyond the agreed limits of the IMF. They just could not borrow more. 232

In more recent years, surfaced another equally serious problem for the banks. It has to do with the better creditworthiness of some countries (in Latin America, Brazil, Chile, and Mexico are classical cases) and the huge amount of private capital migrating to "emerging markets." The implications are clear. Countries now can borrow from the private banking system at favourable interest rates. And with private banks, the hassle of preparing a loan and executing the project is drastically reduced. Private banks do not ask many questions on how the funds will be spent. Once satisfied that the client will pay back, the rest is not their problem. Therefore, some of the operations that previously were going through multilateral banks are diverted to private banks. This has been a major setback for the multilateral banks and matters may become worse in the future. In Latin America, the largest countries, borrowing the bulk of the available funds, have slowed down and reduced their business with the WB and the IDB. In smaller countries, even when they are permitted to borrow, the total of their operations amounts to a small proportion of the funds available for lending. In addition, the fixed costs of preparing a loan hardly justify the amount borrowed. In the last several years, both the WB and the IDB have initiated programmes to lend to the private sector of their client countries. These operations are growing and this is an area in which considerable expansion may be possible. However, it does not seem reasonable to think that they can expand to the point where they make up for the loss of business resulting from the limitations imposed on countries to borrow more. After all, the same abundance of private credit makes the multilateral banks less competitive, given their ponderous and time-consuming bureaucratic procedures. Perhaps the most promising responses by the bank, in order to face these new and difficult predicaments, lie in the development of new financial products. While both banks always had a variety of modes of lending, the conventional loans to fund a well spelled-out set of projects always predominated. The new trend is to expand loans under which the country agrees on a broad set of objectives and receives a lump sum of money to pursue them. These are outcome-based loans. Since countries freely choose the project, they thereby acquire greater "ownership."

233

But there are also variations on the conventional loans. An interesting alternative is to negotiate a step-by-step loan. Instead of making a loan and repeating the entire bureaucratic rites for the next one, along the same lines, there is an agreement that once the first is successfully completed, it can lead almost automatically to the next. All things considered, multilateral banks are not in a comfortable position. Living in a buyers' market is far more difficult. For a long time, countries could not change fast enough to satisfy their goals of reform. Now the world seems to be changing too fast for them to catch up.

234

Can Development Banks Promote Social Reform?

Development banks are committed to social reform. They consider it part of their mandates and make an earnest effort to promote it. But can they really achieve success? Is it feasible for international organizations of this nature to get countries to reform their social sectors? The answer seems to be negative. Development banks cannot promote reform or orchestrate it, even if it is mandated by the conditionalities approved by the Bank and the country authorities. Attempts to violate this principle lead to unrealistic projects that are wasteful and frustrating. Assuming that to be the case, how to get out of this dilemma? The answer is more obvious than it might seem at first. If they cannot create reform, they have to join forces with institutions that are warming up to reform. Banks have to hitchhike on the reforms that are brewing, as a result of the local fermentation of ideas and events. When a Ministry or a sector is ready for reform, banks can be most helpful. They can supply the resources to engage in activities that go beyond the routines. They can provide technical expertise to support the conceptual or intellectual end of the reform. They can provide administrative and organizational structures that have permanence (as a Minister of Health recently said, the World Bank loan is the only thing that survives a change of Ministers). Loans give some transparency to promises and add an international commitment. Also, when there is political will to change but not quite enough power to win a tug of war or two with the opponents of the project, the conditionalities may offer the little extra push that breaks the stalemate. It has been said that in war and love, timing is all. The same might be true of social reforms. The extra push, the helping hand has to come at the exact time when it can make a difference. Too early, it will find no ears or no takers. Too late and it will not make a difference. Hence, the challenge facing development banks' staff is to seize the windows of opportunity. Their business is not to engineer reforms but to chase the fleeting moment, when enthusiasm runs high, when the good guys take over, when the process is ripe. Their skills are that of a hunter who stalks the game and moves at the right moment. 235

The critical skills are good antennas, good feeling for the process of change. It requires the educated intuition to bet on winning horses. There are no reasons to underestimate the importance of good technical skills to generate good ideas and to borrow from other experiments. But this is not enough. Understanding the local scene is just as important. Who are the actors? Who wields power? Who can be mobilized to help? Which local actors are in a position to sabotage the project and need to be neutralized - by elegant bribes, by political legerdemain or whatever works. Enemies might become ambassadors in far away countries. Unions can be co-opted or outmanoeuvred. Someone must be listening to the gossips in the corridors. And as Lenin used to say, reading the daily papers is essential. The human geography of the bureaucracies has to be well understood. In the case of the Inter American Development Bank, which has strong local offices, this can be one of the most valuable assets in designing an intelligent project. Local staff can tell that the highly ingenious scheme that would solve all administrative snags may take three years to be approved in congress. They can also tell that so and so does not speak to so and so, defeating the best strategies since the cooperation of two enemies is unlikely. How competent is the bureaucracy to handle the complicated procedures proposed in the project? Clearly, the ambition of the project has to be in line with the past performance of that bureaucracy to deal with complexity. In addition, sometimes the problem is elsewhere. Weak ministries cannot ensure the counterpart funds or the congressional support to have the loans approved. Of course, local offices create an additional set of actors to be dealt with. Not only the personnel of the bureaucracy has to have ownership but also the staff in the local offices have to perceive the projects as theirs. This is more work but an added burden that pays off. Close cooperation between headquarters and local offices is essential. And this type of cooperation requires good links between all those involved in the process, rather than formal relations based on rules and regulations. The timely identification of a window of opportunity is essential but not enough. To the awareness should follow action at the required speed. Sometimes, careful and patient plodding is better. Sometimes, lightening speed is required. It should be the project needs that determine speed, not the glorified rites of the bank's bureaucracy. Tech236

nical timing should not have precedence over political timing. This is what being client-oriented should mean. Competent bureaucracies that have done well in other projects or demonstrated efficiency in other ways should be left alone to do their jobs with minimal presence of the banks. Weak and problematic ministries should be put in the equivalent of an intensive care unit to manage crisis. If this game is played right, development banks can be proud of their effective support to social reform. But what if timing is not right and reform conditions have not matured? Should the banks give up? Yes and no. No, they should not lend for reform that is not ready. Yes, there is work to be done. Withdrawal is not necessarily the best solution. Banks can be present and discretely subversive. They have legitimacy and the means to disseminate ideas. They can sow the seeds of change and patiently wait for the harvest, whenever it comes, even if it may take a decade. They can propose smaller, less demanding and safer projects, that will keep the banks close to the client and able to help prepare the pre-conditions for future action.

237

Can Training Bring about Organizational Change?

In the early eighties, microcomputers started appearing everywhere in Brazil. The Ministry of Planning adhered to the trend and commissioned its in-house training center, CENDEC, to offer courses to all secretaries. Computers were purchased and instructors began training secretaries in the arcane mysteries of word processing, spreadsheets and data banks. Ironically, CENDEC itself never offered courses to its own staff. However, the CEO and its number one advisor were avid users of Apple II computers. Soon enough, their secretaries saw that they were being left behind and managed to learn on their own. One taught another and requests started pouring in for more computers. Months later, the Ministry of Planning computers were sitting idle, as secretaries preferred their electric typewriters and saw no clear role models to imitate. In contrast, all CENDEC secretaries used computers. Lessons? Courses do not change behaviour. Role models and environment do. The bosses at CENDEC were busy at their keyboards all the time. Secretaries found on their own that "important people" used computers. At the Planning Ministry, no boss used computers. This essay analyses the relationship between training programmes and change in organizations. The main hypothesis can be stated in simple terms: Training alone does not induce changes in the way people work or do things. Yet, when an effective implementation plan is in place, it has an important role to play in institutional change. Readers should be aware that this is not an academic paper, even though it deals with researchable ideas. The paper merely proposes a few hypotheses about the relationship between change and training. These relationships are then illustrated with concrete examples observed by the author. A. Training does not bring about organizational changes As a division chief with the ILO in the late 80s, I came into regular contact with the group in charge of management training. After lengthy experience conducting training for individuals in the civil service and public enterprises, it became apparent to the ILO professionals that training alone does not bring about changes in the way organizations operate. In other words, initiatives to train the staff of an or238

ganization were the wrong way to achieve the productivity increases that everybody was looking for. Simply put, if only training was offered, nothing changed. However, as part of a policy of change, with its implementation plan, training is an essential tool to achieve the desired results. The ideas for this paper are based on that group's experiences. In what follows, I will explore them, by means of real life examples. 1. In-service training for teachers In-service training for teachers consumes an unduly high share of education budgets. Yet, teachers return to their classrooms and there is no evidence that what they learned is being put to good use. Research with Brazilian achievement tests of students shows no association whatsoever between in-service courses taken by teachers and students' improvement. This is not an uncommon finding. Granted, perhaps the courses are bad or lack focus. Too many of those programmes are lofty, floating in the stratosphere of theories that are hard to apply. Others are just bad. Nevertheless, it seems that even in the cases of appropriate training, individual teachers will not "rock the boat", once they return to their schools. There are no incentives to be different, to be more creative or to challenge the established routines. This argument militates in favour of training groups of people from the same institution, rather than isolated individuals who will not have the clout to do much by themselves. But let us not discuss solutions at this moment. Lack of effectiveness is pervasive when the courses do not belong to a well-conceived and concerted effort to improve teaching. They are stand-alone initiatives, promoted independently of any such efforts inside the schools. In fact, often administrative groups do not coordinate with each other to create the right environment to make teachers want to use what they learned in the courses. If training programmes are coupled to follow up and subsequent feedback, the chances of implementation of the new ideas learned improve considerably. This, incidentally, is the core message of the present essay. As the author will try to elaborate, the closer the course is to the real decision-making centres, the greater the chance of being effective.

239

2. Industrial training in Russia The decision to use the skills learned depends on the incentive structure. We always have to ask: is it in the self-interest of people to do things differently? Do they stand to gain if they change? Do they have anything to lose? Convincing people to take courses is usually not a problem. Most workers have good reasons to get training. Sometimes they are required to. Other times, travel and perks, promotions and other career rewards are offered to employees earning certificates. Unfortunately, in too many cases, workers do not have good reasons to put to use what they learned. Let me take as example the extreme case of the Russian factories I visited in the early 90s. All workers had at least three years of serious vocational or technical schools. Russian vocational schools were full-time institutions, with highly respectable academic preparation and plenty of time in workshops. However, during that period, leadership inside factories was weak and lacked legitimacy. Morale was low. Alcoholism was reported to be high. Workers were highly dissatisfied, as revealed by surveys conducted by a Russian researcher and this author. Nobody cared for what happened in the factories. Worse, increasing productivity was a serious breach of loyalty, as per-piece pay could decrease, if workers started producing more. Under those conditions, training could not possibly make any difference. There were no incentives to work differently or to become more productive. In all the factories visited, there was a stark contrast between low productivity and the high levels of training of the entire staff. 3. SUDENE versus EMBRAPA The case of the Brazilian SUDENE (the agency for the development of the Northeast) is quite striking. The agency was created in the early 60s, to boost the development of the poorest region of Brazil. It was supposed to be an icon of modernity. Therefore, its growing staff had to be trained into the ways of modernity. To achieve that, thousands of courses were offered to its employees. Furthermore, this was the heyday of the Alliance for Progress, a plan created by the Kennedy administration to support Latin America. Under this initiative funds for training abroad were in large supply. After a few decades, the agency was overwhelmingly considered a disaster and at one point the government decided to close it down. 240

Despite the fact that its staff took so many high level courses, for whatever reasons, training did not create efficient behaviour. Roughly at the same period, a public enterprise for agricultural research was created: EMBRAPA. From the beginning, it adopted a very aggressive policy of recruiting talented college graduates and sending them to premier universities in the United States. By the time they returned, the research laboratories were fully functional and ready for them to work with. Today, EMBRAPA hires more than 1500 PhDs. and is credited as being the main engine of the Brazilian Green Revolution. By the same token, EMBRAPA is credited to be one of the major players in the Brazilian lead in sugar cane ethanol technology These are extreme examples. SUDENE spent a fortune on training and came to a stalemate. EMBRAPA also spent a fortune on training and is a great success. In one case, excellent training could not avert the unfortunate trajectory of SUDENE, whose management was always contaminated by spoils politics. Decision making was too tangled up with political considerations and good performance was not rewarded. In contrast, EMBRAPA has always been a serious and well-managed institution, with clear goals and dedicated staff. Training was used to qualify world-class researchers. Upon their return, their work was lined up and, at least, minimum conditions for creative pursuits were ensured. Without such expensive training, EMBRAPA could not exist. But training only yielded good results because everything else was right. The lessons are clear. Training on its own does not spur change. But if a policy of change is put in place by management, training offers the tools for implementing it and increasing productivity. 4. Training in World Bank and IDB loans

The World Bank (WB) and the Interamerican Development Bank (IDB) loans follow certain rules which have all but become tradition. It is assumed that as new machines and new procedures are put in place, as a result of new loans, there is a need to prepare staff. In fact, this is a perfectly reasonable assumption. Therefore, most loans have a training component (around two thirds, as estimated from a sample taken from all loans made by the IDB). As a percent of the total loan, the training budget tends to be modest. For the IDB, around 2.5% of the loan is reserved for training. However, some loans reach billions of dollars and banks make dozens 241

of loans every year. Therefore, adding up the modest training budgets, one comes to extraordinary numbers. It was estimated that the IDB on-going loans in the mid 90s budgeted over a billion dollars for training. This is a lot of money. But the fate of those expenditures remains shrouded in a thick fog. It is not bad form to discuss them, but very few professionals or managers do. Rarely does it occur to anybody to ask about the outcomes of the training that accompanies loans (technically called Project Related Training or PRT). I tried to ask several times, during the period I worked for either bank. The answers were evasive but, above all, uninterested. From what I could piece together, the training components are added to the project as an automatic response to the traditions and expectations. They are supposed to be there. Country counterparts rarely dispute or get into the details of the training agendas. After the loans get approved, there is a rush to start disbursing. The priority is to decide how to use the bigger chunks of money of the loan. And this takes a lot of work. Experience shows how hard it is to overcome the inertia of the local bureaucracies. Disbursement procedures are tortuous. The local team has to be put together, another administrative nightmare. Legal hurdles have to be overcome. And little accidents threaten to escalate into scandalous sources of delays. Analysis of past loans of the IDB show that the training components are not spelled out in the initial loan document. In practice, they are worked out later on. As it frequently happens, training gets assigned to lower ranks in the Ministry. Often, it is assigned to teams who maintain little communication with the groups in charge of the core business of the loan. Individuals in charge rarely talk to the Minister or Vice-Minister. Just as frequently, the training teams lack implementation capacity. Top managers hardly every worry about this item of the loans. It is an afterthought. Therefore, the training gets separated from the main activities promoted by the loan. Banks have chronic problems with poorly performing loans. This is a serious matter, jeopardizing its mission and image. As it turns out, evaluation of performance is a mixed bag. Everybody agrees that it is "very important". Yet, it tends to happen less frequently than expected and results only appear when it is too late to do anything. Critical evaluations put managers into a bad perspective and internal "esprit de corps" tends to minimize all harsh comments. 242

Under those conditions, it is no surprise that evaluations of PRT within loans tend to be so rare. After all, these are small expenditures and nobody has paid any attention to them during execution. In the years I worked for the WB, only one such evaluation was performed. And it attracted little attention inside or outside the Bank. In the early 90s, Richard Johanson prepared a survey paper evaluating the training components of World Bank loans. Unfortunately, this paper was not available for reference. But quoting from memory, Johanson mentioned results that were quite striking. First of all, very little is known about the performance of the training in each loan. There is no tradition of asking whether it made sense, whether it brought any significant benefits. He concluded that an alarming share of Project Related Training was ineffective. A similar paper was prepared for the IDB in the late 90s. During the period the study considered, 463 loans were in execution. Sixty percent of them had training components. For methodological reasons, education loans were excluded, since, in them, education/training is the core purpose of the loans. Training is expected to "support the systematic development of the attitudes, knowledge and skills required... to ensure that the project objectives are not constrained by lack of trained manpower. Training ...is designed to: Ensure availability of qualified manpower...improve planning, finance and other management functions and... create a continuing capability... to identify manpower development needs1..." This quote suggests a purely mechanical conception of PRT. It is a service that eventually has to be delivered. The description does not hint at the complexity and complementarities of the processes involved. Indeed, it suggests a simplistic understanding of training and change. As mentioned at the initial paragraphs of this paper, the Management Branch of the ILO, one decade earlier, had a much more realistic view of such training. At that time, I was working for the ILO and had frequent contacts with its officers. It was clear for the team that unless training was fully integrated into the process of change it would not fulfill its objectives. Many technical cooperation projects had been undertaken by the Branch and it became clear that standalone training did not accomplish the desired improvements. 1

"Project Related Training: The current portfolio", IDB, unpublished, no dates, p. 1. 243

Unfortunately, in both banks, PRT was seen as a stand-alone component. Or, was seen as something else, but in the rush to disburse, ended up stranded from the mainstream activities of the loan. And often, living in an underworld of the bureaucracy, it was removed from the real action. The point we are trying to drive is that training removed from the real core business of the loan is ineffective. The IDB paper also noticed that the PRT components "do not have clearly defined Bank-wide objectives... and that few if any guidelines have been adopted... The failure to fully integrate [PRT] in the process of project design necessarily increases the gap between project and training objectives. Any positive impact of training on overall project success thus is potentially reduced and perhaps more alarming, any negative impact, potentially increased."1 The paper also notices that few loans include training needs assessments in the planning of the training components. Indeed, "the overwhelming majority of such projects do not contain training needs assessment in any sense of the term."2 In other words, there is no systematic attempt to find out who needs to be trained in what. When one asks why training produces such modest results, the answers seem to follow the same pattern. First of all, training is disconnected from the process of change. By the same token, it is disconnected from the leitmotif of the loan. To sum up, training comes too late or it does not come at all. Those who manage training are not the same who manage the core business of the loan and the latter are unable or uninterested in coordinating their work with training activities. 5. Computers in schools Computers in schools are a case in point. Millions of computers are installed in schools all over the world. Training programmes almost always include computers and they are duly delivered. But such attempts tend to be a universal failure. Why? The reason is because some critical conditions for success are lacking. What are the teachers expected to gain if they use the computers? They risk embarrassment in front of computer-sawy students. It takes time to become truly familiar with the machines. It also takes a lot more time to prepare the classes. Syllabuses are chronically too 1 2

Ibid, p. 7. Ibid, p. 9.

244

ambitious and computer activities subtract from the time to deliver it. In addition to all that, rewards for using the computers are non-existent. Therefore, teachers discreetly leave the computers aside and proceed to teach as they always did. We need to understand that change has to be the result of a decision taken by those who have the power to transform rules, to impose compliance, to orchestrate the sequencing of activities and to introduce training activities in full sync with everything else. Training cannot come before the right time, or after. This is not to downgrade the importance of training but merely to insist that it cannot happen independently of many other changes. In addition, rarely is the power to plan and execute this plot is in the hands of training departments. In the success cases, training and implementation form a seamless process. It often becomes hard to tell where training ends and where implementation starts. B. Training works when it is part of a decision to change In order for it to be effective, training needs to be part and parcel of the core strategy of change and needs to be implemented according to a schedule that is determined by this strategy. If training is to help in the process, it must occur within this framework. Sometimes, organizations change in the directions that are expected. Understanding the conditions under which change is possible and can be successfully engineered is an ambitious goal, way beyond the scope of the present paper. Yet, unless we try to grasp the nature of this process, we will have little to say about the role of training. Some people or some groups have the power to induce changes in organizations. Surely, not all changes, not all the time. Depending on what is proposed, there might be a strong resistance. For instance, attempts to evaluate teachers are vehemently protested by unions. Charging tuition at public universities is taboo in several countries and no wise Minister tries to challenge this practice. However, the point here is different. Unless attempts to change come from levels and nodes in the bureaucracy that wield the requisite power, not much is likely to happen. Personnel or human resource offices often may have the power and autonomy to offer training. The catch is that they usually lack the clout to engineer the change that is expected, as a result of people acquiring new skills. 245

In the multilateral bank examples offered previously, training components are allocated to offices in charge of training within the Ministries. Following schedules established in the loan documents, these officers would proceed to organize and deliver training programmes - often with significant delays, since few on top seem much concerned with this component. Repeating what was said before, the catch is that unless those people who truly wield power are fully committed to the transformations that are supposed to be induced by training, nothing will happen. The offices dealing with training are unable to muster the power to lead people to change, as well as to alter the structure of the incentives that would make people want to change. The international consulting firm Caliper has accumulated a long experience on the circumstances under which training is effective. According to its observation, in most cases, Human Resources departments do not sit at the same levels as other executives. Worse, many executives delegate training, with too often little concern for how it will handled. In those cases, classroom training is ineffective or even detrimental to performance. People may learn the techniques, but behaviour does not change. The problem with classroom training is that it can take place quite independently of change processes. Therefore, it can easily become detached from real action. Serious consulting firms have mastered the art of merging training with implementation of change. They diagnose problems and engineer change. Training comes at the exact moment when the need for new skills or new attitudes is felt. This is the only way to make training effective. 1. Training professionals to use computers in the ILO Upon becoming the Chief of the Training Policies Branch of the ILO, by the mid 80s, I found that practically none of the professionals in the group were using computers. Since I preferred typing my own papers, I suggested to my staff that this would be a more efficient w a y of working, considering the unfavorable ratio of professionals to secretaries in the organization. But nobody was ordered to go to the keyboard or offered training. There was a bit of reluctance on the part of the staff. One French professional prepared a display of his impressive collection of fountain pens on top of his desk. Nevertheless, he taught himself how to 246

use the computers and proceeded to buy a Macintosh for his home use. In less than a year, almost all professionals in the Division were using computers to write their papers. They all taught themselves how to use word-processing software, impelled by a chief who gave the example. It was simple and painless. And secretaries were happy to see the reduction in their workload, as professionals typed their own papers. 2. Pitágoras and the Integrated Management

System

To remain in the education field, let me discuss the case of the Pitágoras Foundation that has developed an integrated management system (SGI) for municipal school systems. The rationale behind the system is a modified version of Total Quality Management. When a municipality begins to implement SGI in its schools, consultants try first to coach the system to develop priorities. The municipal Secretary of Education has to provide her set of priorities, the same with principals, teachers and students. An elaborate procedure is required to ensure that all these priorities are made compatible. Once all actors have their priorities and goals spelled out, a system to monitor the operations, pointing to strengths and deficiencies, is put in place. The implementation of this system requires a lot of training, since none of the people involved have any notion of Total Quality procedures (it is interesting to notice that the expression "total quality" is avoided, because many teachers consider it "neo-liberal"). But the activities are not identified as training. They are considered meetings to implement the system. There are books, handouts and written instructions. It is all seen as part of understanding and implementing the ideas. Nobody is really aware that there is a training programme to implement it SGI. Teachers and principals are exposed to the system and go through the process of learning how to operate it. Training and implementation are merged. There is a high degree of compliance, since the system is not implemented unless the Mayor and the Secretary of Education are totally convinced. Usually, there are a few recalcitrant schools - for whatever reasons - but their presence does not detract from the overall compliance. The results in terms of improving the performance of education are quite impressive. But exploring them is beyond the purpose of this paper.

247

3. Preparing technical staffai INEP (Brazil) INEP is the research, evaluation and statistics office of the Brazilian Minister of Education. As a semi-independent agency, it has had a roller-coaster performance throughout the years. When Paulo Renato de Souza took up the Ministry of Education, in 1994, the institution was in shambles. But a strong woman with powerful leadership was appointed as its President. INEP came back to life, becoming the most successful office in that Ministry. In order to deliver serious statistics and evaluation, the technical preparation of its staff had to be strengthened. Many professionals were sent back to universities to refresh or to develop their statistical skills. In frequent cases, they were sent abroad, to institutions like the Educational Testing Service. Upon their return, they were allocated to the most appropriate slots in the office. This was a period in which several student achievement tests were being created and statistical procedures subjected to a major overhaul. The fact of becoming a star performer in the Ministry attests to the success of the training programme. Ironically, INEP never had any formal "training programme". The decisions to send someone to be trained were taken on an ad hoc basis by the President, in order to fill gaps in the technical capabilities of the institution, as they appeared. In other words, training decisions were taken at the same level of all other major decisions. Training was part and parcel of the core strategy of the institution. It was a collection of responses to perceived needs here and there. 4. Training in the petroleum industry The present author and Joao Batista de Oliveira prepared a paper in the late eighties, focusing on training practices in the petroleum industry - a highly consolidated industry in which a few large players dominate. These firms are rich, efficient and well managed, and their practices reflect world standards in policies to prepare human resources. It was found that training departments (or whatever the names given) had a more or less predictable trajectory. With the acceleration of technical change, sooner or later, companies decided that they needed more training and better training. Usually, firms had programmes scattered all over the place, with little guidance and uneven effectiveness. The first move was to create 248

strong training centres on top of the organizations. Being close to decision-making foci gave more clout to the new trainers and a better chance of synchronizing training with planned changes. Voice from the training centre was seen as the voice of the top management. In fact, management gave legitimacy to training. CEOs often became its main proponents and advocates. A typical pattern in modern firms is to plan an overhaul of a certain department or process. With it, comes the need to buy new and different machinery and to prepare the workers to operate them. But usually, this is also the right time to change management styles. If nothing else, new machines deploy workers differently. Therefore, training and the replacement of equipment must go hand in hand. Since we are referring to highly competent firms, this entire process tends to work out properly. As top-loaded training departments become stronger, they can plan training, together with the high management. Being close to decision-making makes it easier to work together with the engineering departments that replace the equipment. However, our research discovered that firms were not too happy with this scheme. It could be improved. The tendency for firms having this centralized system was to delegate the training back to operational levels. Training decisions landed in the hands of those supervisors directly in charge of operations. The same engineers that chose and ordered the new equipment were also in charge of commissioning training. The decision to train or not was discussed at the same level as decision to invest, modernize or alter the productive process in any way. They decided on technology and they decided who would deliver the training programme. The high up training centre got out of the picture. Training and other technical decisions were moved down to the same staff that runs the department or unit. Training budgets merged with expansion and investment resources. There was still a central training authority sitting on top and close to decision foci. This ensured a smooth operation at the operational levels. But this central office lost power. It acquired the role of oversight of the entire process, instead of delivering training. The centralized centre progressively became little more than a source of technical assistance in training to the many operational departments. It re249

tained its leadership and proximity to the top management. But it did not have as many operational roles. In fact, it did not train. The rationale is perfectly coherent. Training moves up to conquer higher legitimacy and to emanate from higher levels of the bureaucracy. Without the clout permitted by the higher position, training remains disconnected from the real action. However, as effective training becomes a routine, in some firms, it moved down, closer to the locus of day-to-day decisions taken in the implementation of change. In other words, it never gets stranded from decision making in the firm. It was also observed that, as much as possible, training is done on the job, by the supervisors themselves. Classroom training is minimized. This way, work, training and change processes merge into a seamless whole. And it stands to reason, supervisors do not train just for the fun of it or to show statistics at the end of the year. They train to get things done differently and better. C. Tentative conclusions

As stated previously, the present paper presents hypotheses but is not a research paper. It postulates that stand-alone training efforts accomplish little. Training, if it aims to impart skills to be deployed to further productivity and catalyze a change in behavior, requires certain pre-conditions to be in place. Individuals need incentives to utilize their newly acquired skills. Training needs to be seen as advantageous to one's career. But above all, training must be in sync with a forceful decision to change, taken by those who are in a position to implement it. Unfortunately, we observe too many cases where training is ineffective. Sometimes the timing is wrong. In other occasions, the incentives to apply new skills are not there. In addition, those who decide on training sometimes lack the political clout to create the environment where it could be put to good use. To conclude, it seems that one of the main culprits of the ineffectiveness of training is its distance and lack of synchronization and coordination with what happens at the levels where serious decisions take place.

250

Monitoring Training Projects: A phylogenetic Approach 1

This paper describes an attempt to appraise the performance of education loans from the Interamerican Development Bank (IDB). There is a dire need to better understand how well multilateral bank loans perform on grounds other than disbursements. This paper defends the idea that there is a built-in tendency to underestimate problems and to define performance too narrowly, only on the basis of financial indicators. Dollar figures may say a lot in terms of short-run bank priorities, but they often say little about the impact a given loan has on the institutions and populations it was intended to benefit. Unless we know more about the consequences of past and present of loans, it becomes difficult to say whether it was worth the effort for the borrowing country. The need for better project monitoring thus is crystal clear. A. Are banks learning from experience? Do projects improve? At the time of this study, around the year 2000, the IDB had dozens of education projects in execution or in the pipeline for approval. In fact, it has or has had projects in all of its borrowing member countries. The breadth of such investments has an obvious relation to country priorities and politics, factors that are beyond the immediate control of the Bank. Much closer to home, however, we would expect that each new project bear an undeniable relation to its predecessor. Indeed, if a cardinal law ever existed in the social sciences it would be that the best predictor of new policy is past policy. This holds true whether we are interested in content, quality or both. In the simplest of terms, new projects are a function of how much we have learned from our past experience. Diligence, foresight and creativity play a part, but only a part. Keeping up with the technical literature and exploring technological breakthroughs brings new ideas, expanding the realm of possibilities and framing what it is we 1

Authored with Aimee Verdisco. 251

may like to do. Past experience from similar projects frames what it is we can realistically do. To sum up, projects have ancestors. It therefore makes sense to study what experience the Bank has accumulated in designing and implementing projects. What has worked? What hasn't? How and why have the realities of poverty and backwardness butchered good ideas? How is it that these same ideas survive the worst of all odds in some projects but not in others? Above all, have project designers learned from past experience? Do projects avoid mistakes made in the past? Are they intentionally designed to do so? Do new projects take older ideas to new heights? Or to new lows? This paper attempts to shed some light on these questions. There are two polar approaches for looking at existing projects: en masse or individually. The first considers all projects together and attempts to derive some conclusion or observation valid for all or most. The second approach examines one project at a time, attempting to understand the context in which it was designed and implemented, its results and impact, then comparing these finding with the next project, the results of these projects with the next, so on and so forth. Neither is entirely appropriate for our purposes. The first approach is barren in the sense that common denominators of a pile of projects, looked at as a group, tend to be banalities. Such commonalities will, almost by definition, be too abstract, too general or too obvious to be interesting or insightful. The second approach is full of life-size insights and realities but falls short on the side of replicability or general conclusions. It is difficult to take the mini-conclusions arising from each individual project and add them into a coherent whole - the classic predicament of anthropologists. What we are proposing in this paper is an approach that falls between the two extremes. It is based neither on the whole set of projects nor on each individual case. Instead, it builds from two analytical criteria. We start by sorting projects into groups that are sufficiently similar to be studied together. The following three categories of training projects were identified: technical education, training centres, and the projects targeted to unemployed youth. Such disaggregation is sufficient to avoid the banalities of an en masse type of analysis. But dealing with groups, we can eliminate idiosyncratic features of individual projects.

252

Can these smaller clusters of projects be analysed with metaphors borrowed from biology? The idea is to consider each project as an individual organism and assume that organisms (within the same phylum or class) are subject to mutations. Projects in the same class share the same general features but, over time, they mutate, thus the metaphor from biology. Mutations arise from any number of causes. The same person may be responsible for designing the next project and may want to try some new ideas. Or, the new project manager may come from a different background and want to conform the project to his preferences. Perhaps none of the above. Our approach intentionally has chosen groups of relatively homogeneous projects and looks at each as a mutation from the previous project. As projects acquire features that the previous ones did not possess, their phylum evolves. We would expect these features to be responses to new environments or to enable a project to better adapt to its existing environment. However, as in biology, not all mutations have survival value. Nature makes mistakes and project managers make them even more frequently. Therefore, new generations of projects should improve on previous one, although this is not necessarily the case in every project. Projects can mutate in the wrong direction, exacerbating previous deficiencies and, like animals and plants that cannot adapt to their surroundings, become extinct. Of course the phylogenetic metaphor should not be taken too literally, lest we get bogged down in the Darwin vs. Lamarck controversies on whether acquired traits are transmitted to offspring or the less fit just die off more often. Our approach is just a metaphor. Applying it to the Bank's portfolio, we start by sorting projects into categories that are sufficiently homogeneous to permit analysis and meaningful conclusions (criterion one, above). These blocks of relatively similar projects become the relevant units of analysis, our phyla, so to speak (criterion two). Once defined, the different phyla of projects will be mapped on a time axis. From this mapping, several layers of analysis will be undertaken to bring out patterns in project design. Can we discern a pattern of evolution? Do the projects mutate and differentiate in an adaptive response to the environment - like species in the biological world? Can we observe the phylogenesis or the development of different phyla as the environment changes and creates problems for older individuals? Or, contrarily, what mutation fails 253

take place, potentially hurting the project (e.g. naïve institution-building components, discussed later on)? What projects don't respond to the environment? Once these questions are answered, the next step is to find out why. The analysis that follows is divided into two parts. The first looks at issues of project design - that is, the expected responses to the substantive needs of clients. This first part, therefore, is about training per se. Does the training deliver what it is supposed to deliver? Is the training solution offered an appropriate response to market and society needs? The second part is process-oriented. It examines the loans from the viewpoint of the managerial and administrative solutions they adopt. Are the projects well managed? Is their implementation effective? Are the institutional solutions adequate? In cases of shortcomings, are better administrative solutions introduced in the next project? These analyses should give some insight into whether or not the IDB is learning from its past experience. B y answering the questions outlined above, we will have some indication of if and how information from the field and otherwise related to the implementation of a given project is fed back into the project design process. Have steps been taken in new projects to avoid the pitfalls and mistakes found in previous projects? If not, could such difficulties have been avoided by mutation from the original formula or at least predicted at the design stage? Ultimately, avoiding errors up front is much more efficient than having to fix them ex post facto. The purpose of this paper is to identify design errors, i.e., problems that could have been avoided with a better design (as opposed to problems arising during implementation that neither could have been foreseen nor avoided through project design). At the same time, however, the paper has a larger objective of generating a broader evaluation of training projects.

B. Mutation and adaptation: the search for solutions A loan design is a response to a development need. In the case of the family of training loans to be examined in this paper, design responds to a training need. But the identification of this need has to be correct, as does the relevance of its response. Anything can go wrong at any time. Perhaps there is no market for the training offered. Perhaps there is a market but it can't be reached with the proposed solution. 254

What training solutions were proposed in early loans? How did they evolve with time? Did they put forth successful, adaptive strategies? Did they avoid known pitfalls? Looking at the set of training projects past and present, three categories emerge: technical schools, centre-based training and funds to contract training (the so-called "checkbook approach") geared to the youth. These are clearly distinct clusters, or phyla. Each will be examined separately, although some interaction between categories does exist. Some projects mutate, moving closer to and taking on aspects of other phylum. Other projects are hybrids of different categories, taking ideas from each and, as it were, creating new sub-categories. Or, as a biologist would say, developing new species. Looking at the set of projects one can see the evolution over time in the preferences of the Bank and its clients. Technical schools and centre-based training are the older generation of projects. Both have met serious problems and been adapted, in attempts to avoid these hurdles. The checkbook approach is newer, although more recent projects also show some mutation within this category. In what follows we explore the three clusters of projects, as well as the deviations and mutations observed. We start with the older line of support for training centres and with the support for technical schools. Last, we examine the projects where training is contracted, rather than directly managed by the training institutions. 1. Supporting

training

institutions

In 1967, the IDB began its trajectory of lending to training. Resources from one single loan (Technical Education Programme in Brazil) went into two categories: the training institution model and the technical schools.1 Loan resources were used to both beef up the Brazilian SENAI and to develop the network of federal technical schools. We first consider the evolution of the training institutions. These institutions had been in existence long before the IDB entered the picture. Indeed, Latin America pioneered the creation of training systems independent from the academic education systems. Brazil led the way, creating SENAI in the early 1940s. SENA from Colombia followed. Within a relatively short time, most countries in the region had adopted the model as a response to the post World War II boom. 1

The third line of training projects, funds to contract training competitively, only begins in 1993 in Chile. 255

These were immensely successful institutions, independent, market-driven, and funded though payroll levies. Their teaching methods (the methodical series) were effective and their graduates found jobs. Thus, when the Bank decided to broaden its portfolio to include training, it was only natural that it directed its efforts toward the expansion and further development of these institutions. A sequence of loans followed from the original one made to Brazil. From 1967 to 1978, 13 loans were made to these institutions, many of which had names that started with as "Servidos" or "Institutos" (hence the term used henceforth: S&I). As alluded to above and is examined in detail elsewhere (see de Moura Castro, 1999, 2000), these were solid institutions. They benefited from well-paid and competent staff, high morale and esprit de corps. They supplied Latin America with the workers required to jumpstart the import-substitution industrialization that swept the region at the time. Bank loans reflected these realities. The rapid expansion of industrialization had serious and direct implications for training: it was the supply of training that mattered. The more, the better. The Bank responded. Its resources supported the construction of new training institutions, the training of teachers, and the production of teaching materials. Loans favoured the modern sectors, especially as time went on. Rural education gave way to technical training in industrial arts (Ecuador, Uruguay, Costa Rica, Guatemala, Chile). These projects largely met or exceeded their objectives, at least in the short-term. This is not surprising, since implementation often required little more than retaining good architects and contractors. Many of the intangibles, including issues of governance, institutional capacities, delivery mechanisms, were left pretty much untouched. This supply-driven mix of infrastructure, equipment and materials continued to offer a seemingly good response to the region's training for several decades. By the late 1970s, however, the model began to wear. Economies contracted and import substitution stalled. The S&Is, still stuck to their supply-driven ways, found themselves unable, even unwilling, to adapt to new realities. Many institutions were graduating workers unable to find a market for their skills.1 It was perhaps this institutional stalemate that led to a hiatus in IDB lending to the S&Is. While governments could not force these in1

Indeed, a summary of ex-post evaluations carried out in 1985 found maintenance of loan-financed equipment to be largely lacking.

256

stitutions to change and adapt, they could deprive them of fresh capital for expansion. This gap lasted for 12 years. Meanwhile, support to the technical schools became dominant. Lending to the training institutes was reinstated with the explicit purpose of introducing a more market-oriented approach to training. The 1990 loan to ITCA (Instituto Tecnológico Centroamericano) in El Salvador specifically sought to bring the supply of training in line with market demand. Curricula were revised, infrastructure improvements made and organizational and institutional mechanisms restructured. The phyla thus mutated. The model took a turn from supply- to demand-oriented. Three subsequent projects (Tertiary Non-University Education in Argentina, Technical and Vocational Education Reform in Brazil, and Reform of SNPP in Paraguay) can trace their progeny to the El Salvador project. Each is demand-oriented in its approach and includes some serious investments for the reform and restructuring of delivery mechanisms. Yet, two of these projects (Argentina and Brazil) fall under the technical school approach. The other (Paraguay), at least as it was originally conceived, is less about the restructuring of an S&I per se and much more about the transition of an S&I into something completely new: a contract institution, one among many contenders in the training arena, along the lines of the "checkbook model". Inspiration for the Paraguayan loan came from Chile. Unfortunately, its performance has been less than spectacular. After an encouraging stage, some years earlier, the irrefutable decadence of the SNPP led the IDB to emulate the example of the Chilean INACAP. Under Pinochet, INACAP's budget was reduced 20% a year until it was left to fend on its own. These resources were turned over to enterprises, "liberating" them to contract training, as they deemed appropriate. If INACAP hoped to provide training and have access the resources it had previously enjoyed, it had to compete with other providers. Since INACAP survived the transition, this could be considered a successful model. But this was Chile, not Paraguay. An institution as weak as the SNPP, stuck in a country undergoing a rocky transition to democracy simply would not have either the wherewithal or the political clout to withstand such a radical reform. It balked. Five years later, the loan had not begun disbursing. A new Bank mission returned to the country in 1999 to try to get things moving. It restructured the loan to allow the SNPP to keep half of its budget - a com257

promise that, above and beyond solving the stalemate, has served to drive the project squarely back into its phylum: the reform of a training institute as a training institute, although one that would be expected to compete for some of its resources.1 To summarize, projects supporting the S&Is have undergone considerable mutation. Supply-oriented approaches have given way to more competitive environments. These mutations are likely to stick, especially if the support given serves to rescue - not kill - the battleworn and weary S&Is. 2. Support to technical schools

Technical schools, although entering the Bank's portfolio with the same loan as the training institutes (Technical Education in Brazil, 1967), have shown a different trajectory altogether. Whereas support to the S&I model has been stop and go, support to the technical schools has been sustained. Indeed, over the 30 years of its implementation (1967-1997), this model has come to be the dominant mode of Bank lending to the training subsector. The model we refer to here as the technical schools is a combination of vocational training, technological training, and academic programmes. Emulating the French experience, practical training is offered side-by-side with or embedded to academic training. And, insofar as institutional responsibilities fall to the ministries of education, graduates receive diplomas legally equivalent to secondary school degrees. As time went on, it became increasingly clear that, at middle schools, students were far too young for occupational training. Therefore, technical programmes were progressively pushed to secondary schools. Regardless of the trade to be learned, students needed a set of basic, durable skills, many of which were more academic than practical in nature. But demographics complicated the picture. With more kids entering the formal education system, access had to be rapidly expanded, especially in low-income neighbourhoods (Ecuador, Honduras, 1

It is also interesting to note that this same loan earmarked a small fund for a voucher scheme to train staff in micro-enterprises. This activity proved highly successful, thus calling attention to issues related to timing and prevailing conditions. Despite the political climate in Paraguay, voucher schemes to support private training were implemented. Their implementation, however, complemented - not replaced - training provided by the SNPP.

258

Paraguay, Uruguay). Training options were introduced throughout the basic cycle with the objective of making it possible for students to obtain employment at different stages of their technical training (Paraguay, Panama). Such mutations were not particularly effective responses to the difficulties of technical schools. They may have hurt both education and training. They also weakened the Bank's implementation record. The technical schools spread their efforts too thinly.1 They were ambiguous and ambitious institutions: secondary academic schools preparing students for higher education, schools for technical training, and institutions preparing skilled workers. They often failed on all three accounts. Many schools became plagued by the chronic predicament of trying to do too much and ending up doing nothing well. Others mutated into comprehensive schools, with performance heading down hill (Ecuador). Curricula were too crowded in that they demanded student time be spread between academics, technological subjects, and hands-on activities. And, with the ministries of education at the helm, participation from employers in course design and structure of courses was insufficient. Training became devalued across the board and its supply fell out of line with demand. But where did the problem lay? The model? The country? Implementation? While producing admirable results on the other side of the Atlantic, the model rarely presented an appropriate response to training needs in the region. Its concern with quantity (e.g„ enrolment) overrode any commitment to quality. Many schools purposely catered to students of more modest means than found in the regular, academic schools. In the end, however, it was the quality of education that was poor. And it was poor on all accounts: academically poor, technologically poor, and oc1

The comparison with the first loan made in this phylum is instructive. This loan (Brazil) recognized and maintained a clear separation between the training institute and technical schools. Training delivered through one type of institution was not substituted or replaced by training delivered through the other. Nor did responsibilities overlap. Labour trained at the technical schools was assumed to be "intermediate"; labour trained at the secondary schools was assumed to be "skilled." One complemented the other. In building on and accommodating the complementary nature of intermediate and skilled labour, at least in terms of the relevance of the training response, the loan was more successful than later projects.

259

cupationally poor. Technical education neither prepared students for further study nor gave them the practical skills-base they needed to compete in the labour market. Those schools offering quality education (e.g., federal schools in Brazil) gradually became co-opted by the upper middle classes, thus eroding the effectiveness of training even further. These kids went on to further study at elite universities. It wasn't until the 1990s that the model started mutating back towards its original intentions: technical education for technicians; regular education for everyone else. The Uruguayan loan (1993) started the trend, supporting the reform of CEPT schools. In essence, the loan provided for the removal of technical subjects for which there was no market demand and, with the creation of industry-led advisory councils, focused courses and schools at the post-basic level. These reforms were taken further in subsequent loans (Argentina and Brazil), where positive mutations can be seen along a number of dimensions. More so than in Uruguay, these subsequent loans paid attention to the importance of timing and sequencing in training. Technical education was pushed up to the post-secondary level. Equity also received fair play. The separation of vocational from academic tracks went a long way in vaccinating systems against the extremes of either bad schools with real occupational training or good schools converted de facto into elite high schools. In summary, at first, the performance of the technical school model paled in comparison with the S&Is. But they have improved, while the S&I have faced a stalemate in their ability to adjust to new times. Be that as it may, the fact that the Bank's support to the technical schools has been sustained over a 30-year period has opened the way for more experimentation and mutation. Errors made in earlier loans have been corrected in later loans. The evolution of this phylum serves to illustrate a key point. New projects are a function of how much we learn from past experience. Whereas we may be awestruck at the potential of new ideas and experiences exported from abroad, the feasibility of concrete investments and projects on the ground depends on what has been done in the past. In the end, there are no optimal solutions, only second- or third-best alternatives. The legacies of past policies reduce the maneuvering room for project implementation, mutating - quite by definition - the outcome of what was originally intended.

260

3. The checkbook approach: short training with competitive funds With the financing of Chile Joven in 1992, a completely different species of projects appeared in the Bank's portfolio. This project was novel and seemingly radical for its time. Rather than modernizing, reforming or otherwise strengthening existing institutions or creating new institutions, the idea behind Chile Joven was to create a market for training by recasting the very mechanisms through which it was funded and delivered. This response was appropriate and relevant to the context at hand. As alluded to above, Pinochet all but did away with Chile's public training institution. What remained of INACAP after privatization migrated upwards, both in terms of clienteles and markets. Its focus on the working classes and manual labour gave way to higher-end training for the middle classes. Chile Joven responded to the gap left by INACAP's conversion. Targeting unemployed youth and low-income sectors of society, the project created a fund to contract training under competitive rules. All training providers - public, private and non-governmental - were eligible contenders. Those awarded training contracts had to present a cost-effective proposal in which the training offered was aligned with market demand. Herein laid the innovation. The supply-demand match had to be substantiated. Bidders were required to find either jobs or internships for their trainees. This innovation turned out to be a powerful engine to bring training closer to demand. It also served to generate a basic and forceful rule for all projects to come: no demand, no training. The success of Chile Joven spawned projects in Argentina and Venezuela.1 There was little mutation in the project design of either. Proyecto Joven (Argentina) was quite successful in targeting training to demand but revealed some of the weaknesses of the model. 1

The Educational Quality Improvement Project in Peru (PE-0116) also provides for the reform and modernization of technical education and calls for close collaboration with the private sector when undertaking such activities. In essence, the activities included in the project (e.g., studies for the rationalization and strengthening of selected institutos superiores tecnológicos) could lead to initiatives along the lines of the checkbook model in the future. The implementation of the model, however, is not included in the scope of the project. An interesting point of comparison is the Planfor programme in Brazil. Financed through the country's unemployment fund (FAT), the programme replicates many of the same principles found in Chile Joven and Proyecto Joven, although its underlying rationale differs (social, as opposed to economic). See CMC and AEV.

261

First of all, economic growth seemed to be a prerequisite, a factor that undoubtedly contributed to the success of Chile Joven. Secondly, there was only so much a country could do with a checkbook to buy training. As the Chilean and Argentinean experiences showed, checkbooks can prod suppliers to find market niches. They can catalyse suppliers, stimulate demand for training and even structure the rules of the game so that training reaches populations with limited abilities to pay and clienteles with less than desirable profiles. But checkbooks are clumsy instruments for covering fixed investments, sunk-costs and the like. Training requires good quality materials, and well-trained trainers. In other words, requires heavy investments in people and courseware. These outlays were not contemplated in either Chile or Argentina. At the time Chile Joven was implemented, INACAP was already serving a different clientele. And in Argentina, the deterioration of the technical schools had reached such dramatic levels that the nation's public training system existed in name only. Thus, in neither case, was there a pool of high quality trainers or a stock of infrastructure that could be hired by means of the checkbook. The quality of training suffered and its delivery somewhat improvised.1 The new model presented some mutations, as a result of learning from past lessons of Argentina and Chile. The new version of Projecto Joven (AR-0169) maintained the "no demand, no training rule", but seems to have taken measures to avoid the pitfalls encountered in its last incarnation. The project allocates resources to develop training materials and train trainers. Whether or not such steps have been sufficient remains to be seen. Budget cuts - factors beyond the control of the project - have slowed implementation to the point that the loan may be cancelled. To recap, the checkbook model is the newest variant in the Bank's portfolio of vocational-technical training loans. Despite its youth, the model has shown considerable promise. By opening the training arena up to competition from multiple suppliers and by conditioning contracts to concrete identification of jobs or internship offers, it has 1

In Argentina, in particular, the quality of the training offered left much to be desired. Costs were relatively high due to the small scale under which the contractors operate and the short time horizon of the contracts led to improvisation. See CMC, Proyecto Joven: "Nuevas soluciones y algunas sorpresas." Boletín Técnico Interamericano de Formación Profesional. ILO. Abril/Setembro 1997, No. 139-140.

262

ushered in a new era of reform. Training is moved out of the ministries and into an open field, an economist's dream. But quality has not been satisfactory and costs have been high, as compared, for instance, with those of the SENAI, a training institution well known for the quality of training it delivers. 4. Survival of the

fittest

When we look at more recent projects, it becomes clear that the original phyla are mutating. Projects are adapting the innovations spawned elsewhere. In some cases, sound ideas are migrating and surviving in new environments. For example, although the move towards the checkbook model collapsed in Paraguay, significant progress along these lines has been made in Brazil. The PLANFOR programme, funded through the nation's unemployment fund (FAT), replicates basic principles of the checkbook model, and the Brazilian training institutions have been found to be especially good competitors. In other cases, projects are straddling phyla (e.g., Uruguay, Paraguay). After a less than stellar record, there is positive news coming from the technical school projects. By pushing training to the post-secondary level, these projects have moved beyond the chronic stalemate secondary-level technical schools faced in the past. This trend is likely to continue, especially given the failure of universities to offer short and practical programmes at this level.

C. The contrast: stalemate in management Mutation drives evolution. But not all mutations have survival value. Survival is a complicated process, a function of timing, politics, error and luck. A good idea is not enough. Good ideas may lead to good project design but, unless good design leads to good implementation, good projects may fail to produce good results because not all ideas are equally easy to implement. Three preliminary conclusions are anticipated here: (i) There is a need to redefine "good" projects as those yielding good results in the end, not for being elegant, (ii) Implementation is what matters, (iii) The Bank has not learned its lessons in matter of implementation, little progress is to be observed. Project design and implementation are intimately related. Good design, as we understand it, is implementable design. Non-implementable design is bad design. The difference is not merely abstract. 263

For example, the troubles encountered during the reform of the SNPP in Paraguay (Programa de Mejoramiento de Educación Técnica) can be now recognized as faults in project design. The project was not an appropriate response to the situation at hand. The outright conversion of SNPP into an institution without a public budget proved politically infeasible. In hindsight we could say that a bit of critical thinking could have shown that the political will on the part of the government to impose such a painful reform was not there and the ability of SNPP to resist it could have been foreseen. Despite the above comments, issues of design and implementation can be separated, at least analytically. In this section, we focus on those factors that link design to implementation. We examine those factors that, if misinterpreted or altogether disregarded, come back to haunt project implementation. Of these, factors related to management and institutions weigh heavily. In the end of the day, design mutations survive only when success in disbursements parallels success along other dimensions, including institutions and management. /. Predictable and unpredictable accidents In examining the roadblocks faced by projects, it is important to ask whether they could have been predicted or not. Predictable accidents are design flaws that could be avoided. In contrast, when the unpredictable happens, designers are forgiven. Some problems appear only as events unfold, they cannot be anticipated. Catalysed by accidents, politics, economics or some mixture of both, there are problems that could neither be foreseen nor accounted for at the time of project preparation. For example, the reform of the Uruguayan UTU was frustrated by a political coup that all but rendered the project irrelevant. The project for the expansion and improvement of intermediate level vocational-technical education in Panama similarly was frustrated when the reform upon which it was structured and dimensioned was repealed. In Trinidad and Tobago, to cite yet another example, a political decision taken after loan approval significantly reduced the number of buildings to be supported by Bank resources. The loan was cut to 17% of its original amount. In other cases, problems can be traced to economic factors. Scarcities of cement and other imported materials needed for construction and an increase in the basic salaries for workers delayed the restructuring of the Escuela Politécnica Nacional in Ecuador. In a project in264

volving secondary schools and technical training - not included in this essay - a spotless project was delayed more than a year in the Brazilian Congress, because a senator wanted to boycott the state governor who was his long-time foe. This pitfall and several others could hardly have been predicted by designers. 2. Buildings get built, institutions don't change

In general, every Bank project includes three investment categories: infrastructure, quality inputs and institutions. The relative weight each receives varies between projects and models. For example, the checkbook projects place much less emphasis on infrastructure than do those projects supporting the development of technical schools and S&I institutions. From a different perspective, whereas all projects regardless of the model - contain activities targeted towards institutional strengthening, these measures appear to predominate in the reform-oriented projects. It is without a doubt that the institutional strengthening components have shown the greatest degree of evolution. In fact, the great revolution in social-sector projects in the IDB has been the emphasis on reforming institutions. The efforts have evolved from supporting a given ministry to supporting the overall reform of the sector. Yet, despite this evolution in function and purpose, an analysis of all projects in the area of training, since the first one in 1967, shows that the interventions used by the Bank to support, reform, and otherwise restructure institutions have been slow in evolving. Mutations [i.e., variations in the tools to support reform] have been far and few in between. It thus should come as little surprise that, by the Bank's own account, institutional problems are endemic and that, if implementation of a given project is difficult, the cause of such difficulty is likely to be found in institutional factors. As projects have evolved from an emphasis on infrastructure to one on reform, institutional aspects have all but stood still. The Banks seems to have neither learned how to fix institutions nor realized that its medicines are ineffective. Even a quick read of projects implemented in support of vocational-technical training reveals the similarities and lack of creativity between institutional strengthening activities implemented more than 30 years ago and those found in the current portfolio.1 1

Actually, this holds true for all Bank investments. Institutional strengthening 265

These activities routinely include some mix of training activities for management and staff on issues assumed to be associated with the efficient execution of a Bank loan (e.g., informatics, accounting). This has been a one-size-fits-all approach, varying little in scope and design and across projects. The experience of the IDB and that of other international agencies seems to confirm the ineffectiveness of training as a tool to boost institutions. As suggested in the previous paper in this book, staff lacks training, because the institutions are weak, not vice-versa. Hence, adding training to weak institutions is either creating unused capacity or creating capacity for another institution, as trained staff moves on to greener pastures, where their skills can be used. Such inertia bears a close relationship to the lack of evaluation (discussed later in this section). What evaluation does exist has been all but crude. Neither objectives nor activities are defined in concretely operational terms. Although these activities are assumed to increase the capacity of the institutions, the input-result relation is never demonstrated. Nor has there been much rigour in evaluation.1 Now as before, an institution is considered strengthened when its staff attends seminars or training sessions or new equipment is installed. "Successful" attainment of institutional goals, then, becomes a function of adding up the numbers of seminars held, staff trained or hired2 and equipment installed. Since relatively little is known about the impact of such "institutional strengthening activities", there has been little from which subsequent components can evolve. The fact that institutional strengthening activities in the present portfolio bear an incredibly close re-

1

2

components have undergone extremely little evolution over the Bank's entire lending history (more than 40 years). In past portfolios, issues of institutional strengthening were addressed through parallel technical cooperations, not as components proper of a given loan. Despite what was perhaps a more focussed preparation (given both the smaller scale of operations and their more narrow objectives), focus on the results has been considerably weak. Technical cooperations rarely are evaluated with the same rigour as projects. In the current portfolio, indicators to measure progress along these lines, if specified in project documents, leave something to be desired. Such indicators merely serve to count the number of seminars or training sessions held, the number of staff attending, and the number of computers or other equipment installed. This can get out of hand. For example, the executing unit of PROMEET II employed, at one time, 66 people. 266

semblance those financed early on is extremely telling. It suggests that either little is understood about institutional requisites of project implementation and sustainability or that the financing of project executing units, justified on the basis of "dedicated" and highly paid staff, is tantamount to addressing such requisites. A vicious circle is created: the very lack of institutional capacities stifles any attempt to innovate. Such assumptions are not pulled out of the blue. They are supported by evidence from past and present portfolios, ex-ante and ex-post evaluations. Ex-ante analyses of institutions are routinely included in loan documents. Yet their raison d'être appears to be a justification for the creation and financing of executing units as opposed to an inquiry into the institutional arrangements, leadership and professional capacities required for project implementation and sustainability. Ex-post evaluations routinely recognize the limits of these activities (For instance, in Guatemala, Paraguay, Brazil, Trinidad and Tobago, Ecuador and Honduras). These limitations, however, seem to somehow be forgotten in terms of project objectives. The same evaluations go to great length to show that, despite some problems on the institutional side (e.g., turnover in the executing units, delays in disbursements), other, presumably more important project goals - e.g., construction and provision of books and other material - have been reached. More detailed and concrete examples can be used to further illustrate this point. In one case (Expansion and Improvement of Technical Agricultural Education in Argentina), reprogramming led to a 60% decrease in funds originally allocated for institutional strengthening. These funds were reallocated to physical infrastructure (increasing total resources allocated to these activities by 26%). In another (tertiary non-university education project in Argentina), the number of institutions benefiting from the project has been reduced. The root of this problem can be located in the failure to adequately analyse the capacities of beneficiary institutions - that is, of those institutions targeted to receive project support. Institutional considerations thus do not seem to weigh into the overall project analysis or do so only tangentially. Their importance is minimized, especially insofar as priorities are reflected in budgetary allocations. Institutional strengthening is only considered problematic if construction is jeopardized. Ultimately, above and beyond equipping ministries and training their staff, institutional strengthening re267

mains operationally vague and ambiguous. No indicators are provided to measure progress or determine whether if, at the end of a project, a given ministry has been effectively strengthened. In so many words, no learning and, therefore, no improvements in dealing with institutions. 3. Weak institutions, weak

management

Institutions are both the target of Bank investments as well as the instruments through which these investments are to be implemented. Regardless of how tangential the actual impact of Bank loans on institutions may be (benefits tend to accrue mainly to the executing units), institutions and institutional capacity, to a very large extent, determine the success of implementation. Where institutional capacities are low, implementation suffers. Where management is weak, implementation suffers. This paper hopes to make the case for taking institutions and capacities into consideration up front. Failure to adequately assess institutional capacities at the earliest stages of design is of paramount importance and a likely cause of failures. The implications of weak institutions reverberate both within countries and the Bank. Weak institutions are inefficient. Delivery mechanisms leak and resources are wasted throughout the system. Many studies have found that how much is spent in education is not a primary determinant of performance. Rather, it is the efficiency of delivery systems that matter - that is, the institutional arrangements through which services are delivered. Repeating, weak institutions lead to weak project management. And, despite the volume of resources invested in institutional strengthening (roughly estimated at more than US$135 million, or more than 8% of all resources invested in vocational-technical training, over the course of the portfolio), our review of project-related documents suggest that management problems continue to plague the implementation of loans. The PROMEET projects in Ecuador provide particularly fertile ground for exploring the consequences of institutional weaknesses perpetuated by poorly designed and implemented institutional strengthening components. The Departamento de Educación Técnica (DET) of the Ministry of Education executed both phases of this project. The project document for phase II clearly recognizes the weakness of DET in implementing phase I. DET proved unable to carry out its responsibilities, thus prompting the Bank to require proof of additional 268

strengthening measures prior to its execution of phase II. Such measures included additional staff and staff training. The institutional strengthening included in PROMEET II was insufficient, and learning from previous experience - in the same country and in the same subsector - was next to zero. The project was hurt by a variety of factors, including the implementation of a technical school model not particularly well suited to the realities of the country, the outright abandonment of plans to rationalize schools, and the failure to increase the quality of technical school graduates. An evaluation locates these failures and problems in DET's lack of capacities - an institution that, "despite efforts at strengthening, remain[ed] weak." The PROMEET example speaks to weaknesses in line ministries. The fact that Bank projects often fail to strengthen them comes as no surprise. Indeed, it is these weaknesses that are used to justify the use of consultants to prepare loans and the creation of project executing units to implement loans. Project teams routinely hire consultants to undertake technical analyses during processes of loan preparation. These experts work in collaboration with ministry officials and other specialists in the country. In most cases, such arrangements work well. Detailed and to-thepoint analyses can be undertaken in a timely manner and their results are used to structure the loan at hand. Yet, there are some risks involved. For example, according to some reports, the design of the loan to improve technical education in Uruguay drew too heavily from consultant reports. Input from the executing agency was minimized. As a result, buy-in from the executing agency was limited. The loan was later restructured (1996), with significant participation from the executing agency. 4. The dilemma of the executing units The rationale for creating an executing agency is simple enough. Ministerial bureaucracies are weak, incompetent, slow and poorly paid. Reforming an entire Ministry is a task beyond the capabilities of a Bank loan. Therefore, creating an enclave of competent, well-paid and honest staff is a quick fix that would allow the project to start operating fast and efficiently. However, the problems associated with executing units are widespread. The purpose of these units is to execute Bank loans in an efficient manner. Despite the salaries of their staff (far in excess of regu269

lar civil servants), the dedication of their time solely to the project, the training they receive, and the technical support and assistance they receive from the representations, project-executing units suffer from many of the same weaknesses as line ministries. Contrary to the perception of some, the documents reviewed suggest that the creation of project-executing units guarantees neither efficient nor smooth execution. Often, the creation of an executing unit is a safe bet the opposite will occur. From Argentina to Brazil, Ecuador, Honduras, and Trinidad and Tobago, high staff turnover in executing units complicated and delayed project implementation. Such delay and complication stem from myriad of causes. Purposely recruited as "outsiders", staff lacked both an understanding of government procedures and political protocol, including a reluctance to enter into necessary alliance with other public agencies (Ecuador, Uruguay).1 In Panama, the "educational background" of staff in the executing unit was less than expected, slowing down implementation. In Paraguay, resources were misallocated and in Chile, Brazil and Trinidad and Tobago, resources were altogether lost, as the lack of capacities led to the cancelling of loans. Again, in Chile, the "passive ness" of the executing unit (Universidad Tecnica del Estado) led to deficient coordination, administration and financial management. These weaknesses were manifested in delays in export licenses from Canada and delays in processing disbursements. According to one account, it took more than a year to take inventory of 260 boxes of merchandise that had arrived from Canada, a situation that jeopardized the project's success in attaining those objectives related to materials and equipment. The unit executing the project to expand and improve intermediate-level vocational-technical education in Panama was similarly "passive" and "noticeably lax" in its administration of schools. This lassitude was manifested in a lack of forward planning. A failure to define the role the project was to play in the education sector, and an evident lack of control, organization and administration of individual schools. Project executing units have high costs, especially in terms of

1

In this regard, the counter example of Brazil is instructive. An evaluation of the Technical Education Program, the first project approved by the Bank in the subsector, suggests that the efficiency of the executing unit was directly related to the fact that staff came from the inside, were kept to a minimum, and that operating procedures were kept simple and streamlined. 270

missed opportunities to strengthen line ministries, since these ministries are notoriously weak. And, whereas a given line ministry may supervise an executing unit and ultimately remain accountable for the success or failure of the project, ministry staff are not likely to be involved in the day-to-day operations of project implementation. It is the executing units that are involved in these daily activities.1 Thus, if the spotty record - and high costs - of the executing units is factored in, the opportunity costs of not strengthening line ministries increase even further. It increase still more if this line of reasoning is taken a step further, to issues of reform and modernization of the state, because at the end of the project, the executing unit disappears. In this regard, the Bank should take heed of an observation made in an evaluation of PROMEET II: "the project was too ambitious and complicated to be executed well in an environment where there is a clear lack of institutional capacity." To sum up, creating an executing agency is a troublesome idea. The rationale is that they allow the bypass of morose and dysfunctional bureaucracies, permitting a quick start of the project. The repair of this bureaucracy would be postponed or avoided altogether, a clear loss, but the results of the loan would materialize faster, a clear gain. However, examining the available documents, this theory fails to be confirmed: the fast gains are not to be observed. Executing agencies are plagued by instability and isolation when not by the same inefficiency they tried to avoid. Perhaps the situation would have been worse without them - a counterfactual statement that cannot be demonstrated one way or another. But we are also faced with the unavoidable conclusion that too many executing agencies fail to perform. To sum up, executing agencies do not have a good track record. They are mired in problems and they bypass the ultimately more important task of fixing the bureaucracies of public agencies. But the alternative of working without them has not been shown to be better. This remains an open question, awaiting further efforts to yield a clear answer.

1

Above and beyond the opportunity costs of not strengthening line ministries are issues of project "ownership." For example, a sector report of Ecuador suggests that the ministry's "ownership" of PROMEET II was limited due to the fact that all daily operations were in the hands of the executing unit. 271

5. The procurement and contracting

nightmare

For the Bank, weak institutions complicate both project administration and country-Bank relations. The administrative requirements of Bank loans are often at odds with the institutional capacities of ministries and executing units. In evaluation after evaluation, mention is made of overly complicated procurement rules (Paraguay, Honduras, Trinidad and Tobago, Argentina) and excessive formalism in contracting regulations (Uruguay, Panama, Argentina). The Project Completion Report for the project to expand and improve technical agricultural education in Argentina (AR-0147) even goes so far as to suggest that if Bank regulations had been strictly followed, the project would not have met its objectives. Recommendations stemming from the evaluations of this and other projects speak to the need for simplified and streamlined procedures. This mismatch between administrative requirements and the ability of the local bureaucracies to meet them should be a cause for concern. The Bank, as a development agency, insists in asking from its clients an administrative performance that is beyond their ability to provide. This happens, exactly, because they are poor and backward. First and foremost, assuming that the training package of a given project is appropriate to the needs at hand, the very success of this project could be jeopardized by the inability to comply with Bank regulations. This is a point clearly made in the evaluation of the project in Argentina referred to above. The Project Completion Report states in no ambiguous terms that the fact that project met its objectives owed much to the "creativity and initiative" of the sector specialists in handling procurement and contracting arrangements in such a way that they did not interfere with the course of the project. The reader can use his own imagination to decide what "creativity and initiative" mean. Second, the inability to comply with Bank regulations may squelch any initiative on the part of ministry officials or other stakeholders to take risks or otherwise innovate. At a bare minimum, even if the initiative isn't stifled, failure to comply with Bank regulations is likely to delay project implementation. Existing data reveal an all but direct correlation between delays in execution and problems with procurement and/or contracting. For the overwhelming majority of projects with delays, bureaucratic snags from either procurement or con272

tracting are to blame (Honduras, Brazil, Paraguay, Argentina, Ecuador, Uruguay, Trinidad and Tobago). Third, as discussed above, if institutions are weak, it is likely that management and administration, without which there is neither procurement nor contracts, will be equally weak. As a telling example, failure to submit the necessary papers led to a partial cancellation of loan resources in Brazil (project for mid-level technicians). In highlighting the difficulties procurement and contracting regulations often pose for projects, we call attention to two different, albeit related, sets of issues: one at the country-level, the other, internal to the Bank. The first, as discussed above, relates to institutional capacities. As we argued there and will continue to argue here, project objectives and deliverables need to be aligned with prevailing institutional capacities and arrangements. This needs to be done as a routine practice during processes of project preparation. Failure to do so leads to design errors. The second issue is that of Bank regulations and procedures. Its bureaucratic requirements are heavy and overly burdensome. While the need to ensure accurate accounting and transparency and to avoid corruption is clear, it is also clear that, in some cases, such bureaucratic burdens contradict the underlying logic of lending modalities designed for quick disbursement and flexible implementation. In this respect, the issues to be resolved are beyond the reach of project managers. They are issues requiring action and decision on the part of Bank management. Yet, there is one dimension that is within the control of project managers: execution periods. It has long been known that projects take longer to implement than is originally planned.1 According to some reports (e.g., Wilson, 1996), vocational and technical education projects take, on average, 7.5-8 years to complete. Yet, with few exceptions, Bank projects have execution schedules that hover around four years. There may be some room for extending these schedules or being more realistic in estimating time frames, particularly when institutional reforms are at hand. What is gained by approving four-year projects that take eight? 1

See, for example, Summary of Ex-Post Evaluations of Secondary TechnicalVocational and University Education Projects (GN-1543), 30 September 1985, which finds that no project under consideration was completed within its original timeframe. 273

Above and beyond the additional realism such extensions may bring, there may be other benefits as well. Insofar as the Bank-country link stays in place longer, countries may be able to do more. For example, Bank support may help insulate tentative or incipient reforms from hostile political interests, allowing governments to push these reforms further and faster than would otherwise be possible. Notably, an evaluation of the first loan the Bank made in support of vocational training (Brazil) points to these factors. It suggests that delays in execution (due to procurement and contractual issues) actually allowed the government to exceed the goals set out in the project. By keeping the direct link with the Bank in place longer, delay was a key factor leading to the success in the execution of the project. To sum up and return to the core issue in this item, just about all has already been said by Saint Mathews when he stated: "he who has, more will be given". Development banks ask their clients for a degree of administrative skills for handling money that tends to lie beyond the capabilities of its less affluent member countries. 6. Counterpart

financing

Delays due to counterpart financing are and have been endemic in the Bank's experience with vocational-technical training. Such delays occur regardless of level of socio-economic development: they have been as prevalent in Guatemala and Paraguay as in Chile, Uruguay, and Argentina. And they occur prior to first disbursements, as well as after execution starts. Under normal procedures, the amount of counterpart financing varies by country group. In most cases, the local offices insist that governments meet their commitments prior to first disbursement. Delays in meeting such commitments, obviously, lead to delays in declaring a given project eligible for disbursement. Once declared eligible, a project may very well be implemented smoothly (e.g., Guatemala). For the most part, there is little that can be done at the time of project preparation to avoid problems with counterpart financing. Counterpart financing, by definition, remains a matter subject to internal politics, political whims, and economic conditions. Regardless of the priority a given executing agency may attach to a project, issues of counterpart financing are likely to be beyond its immediate control. The agency may even face competition for counterpart re274

sources if loans in different sectors are under preparation at the same time. In fact, the executing agency may lack the power to ensure counterpart funds at the appropriate moment. However, outcomes in this particular issue should not come as a surprise to the Bank officers in charge of planning and managing the loan. Countries that underperform in this area are likely to continue doing so. The only progress in this area comes from the elimination of the problem. Indeed, a new generation of loans do not require counterpart funding. D. Evaluation, feedback and institutional learning In the first part of this essay, it became clear that there is considerable progress in adjusting projects to environments that change - some times quite fast. In our biology metaphors, there is mutation and evolution of the phyla. In the second section, we examined the issues of management and implementation. By contrast to the previous situation, while there may be mutation, there is no discernible evolution. Decades elapse and the problems remain the same and the responses are not any more creative or effective than they were before. The present section deals with the points of contact between projects and their implementation. As a background for all that will be discussed ahead are the following questions: Why processes and implementation have not improved? What is wrong with the way projects are monitored and evaluated? Why the feedback between information and action to correct course is effective at project design but not at the implementation level? In other words, why does design improve and implementation does not? 1. Information

and analysis: a weak link

Of the pressures project officers face, time pressures ranks near or at the top. No one has either the time or the motivation to engage in serious and realistic analyses of all project outcomes or implications. This is understandable. In headquarters, there is pressure to prepare loans in timely a manner. Narrow deadlines are ever present. In the representations, there is pressure to close the projects out as quickly as possible. Consultants need to be hired to support loan preparation and routinely are retained to do project evaluation. Fair enough. But, in some cases, the pressure to push a loan through the Bank's hierar275

chy has meant that the rigor and critical thinking needed to prepare good, implementable projects has been missing. There is evidence to suggest that had data available ex-ante been taken into consideration, solutions proposed by projects would have been deemed infeasible from the start. For example, an early evaluation of a Guatemalan project questioned the validity of the project analysis, especially its calculation of returns to scale. The Uruguayan example alluded to above (expansion and improvement of technical education) similarly found that if the situation had been adequately analysed, Bank resources would had been better used to address issues of internal efficiency. In much the same vein, PROMEET II in Ecuador was doomed from the start in that it was based on neither "reliable nor timely information" and thus failed to attack basic ailments plaguing the sector (Uribe, 1994). This project was further hurt, insofar as it benefited little from insights gained from phase I: the expost evaluation of phase I was scheduled to begin after preparation from phase II had begun. Survival of mutations and overall success of projects depends on myriad factors, among them timing, luck, and circumstance. Success is not, nor can it be, data-driven. But, the case can be made for longer preparation periods in difficult projects. Not all projects require a lengthy preparation. In fact, many don't. Loan preparation should vary in accordance with the complexity of the project at hand: relatively simple projects (e.g., expansion of infrastructure) can be prepared within a matter of months; more complex projects - particularly those that are reform-oriented and data-intensive - will require a longer time to design and negotiate. And these are the ones justifying a much greater effort to understand institutions. Such flexibility is needed. It not only would relieve some of the time pressures currently faced by project officers, but would likely reduce the likelihood of error due to shoddy analysis. To sum up, initial despite the long cycle of project preparation, analysis tends to be superficial, particularly in the case of the institutional set up. There is too little effort to understand power structures, communication networks, effectiveness of command and execution and motivation and quality of staff. Weak bureaucracies are treated just like robust ones. Too much time is spent spelling out details of projects in bureaucracies with excellent track record and too little is spent trying to understand how weaker bureaucracies work - or fail to work. 276

2. Evaluation is a mutating species (is it mutating

backward?)

The issue of evaluation is of paramount importance. How does the Bank find out about the performance of its projects? Is it learning from its past experience of ex post evaluations? Unfortunately, the answers appear to be negative. Take the institutional strengthening components. The fact that these activities have changed little throughout the Bank's entire investment history is directly related to the lack of evaluation, ex-ante and ex-post. Ex-ante evaluation, it appears, falls victim to the pressures of project preparation, It is much easier to create a project-ex ecuting unit. Yet, there seems to be no similar excuse for shoddy or superficial ex-post evaluation. Unfortunately, the level of analysis and rigour of evaluation is disappointing. Worse, the calibre of evaluations seems to have deteriorated. In part, this deterioration reflects a change in how evaluations are conducted. Evaluations of early projects benefited from the work of consultants, who were mostly experts in education or in vocational-technical training. This practice, however, was gradually replaced by the use of in-house "expertise", where many of those appointed may be specialists, but in sectors other than education. For example, the evaluation of the Technical Vocational Education Project in Honduras was written by specialists in the transportation sector; the evaluation of the Expansion and Improvement of Technical Education in Uruguay, by agricultural and sanitation specialists. To some extent, this situation reflects the lack of education specialists in the representations. It also reflects the unwillingness to increase the costs of project executing units. These units, not sector specialists in the representations, often are responsible for collecting data on project impact and monitoring benchmarks and project progress. Yet, executing units cannot perform all duties required for good project monitoring. Not only are the time horizons required to monitoring performance and feed results back into the project cycle longer than the life spans of executing units, but some distance and neutrality is also needed. Learning from past experience requires that projects be analysed and compared with others in the same phylum. Executing units are simply too close to the project at hand and are often unable to draw the necessary connections between other projects in execution, preparation or under evaluation. 277

In other words, evaluation of Bank projects has always been a second thought. But, what is even more worrisome, its quality is deteriorating. Yet, with projects becoming increasingly complex and involving more ambitious reform components, it makes little sense to allow the quality of the evaluation to deteriorate. 3. Strategic retreat on objectives: implementation or design faults? In an ideal world, project objectives emerge independently of resources. In the real world, this logic often is reversed. Objectives change so that they can be met, while interventions often remain the same. The distinction is not merely academic. It refers to the "strategic retreat on objectives", an expression coined by A. Wildavsky nearly a quarter of a century ago.1 Under this line of thought, an organization that fails to attain a given objective will seek to change not its methods but the objectives themselves. The bottom line is clear and simple: if you cannot do what you are supposed to do, make what you are supposed to do easier. Retreat the objectives instead of extending you performance. This somewhat perverse logic has been observed in the Bank's portfolio of vocational-technical training loans. In general terms, project components have remained stable. Each project contains some mix of infrastructure, quality inputs, and institutional strengthening. It is the objectives that change, or strategically retreat. In some instances, retreat emerges as result of unplanned or unforeseen circumstances, such as a political coup in Uruguay, student protests in Chile or shortages of cement in Honduras. This is an expedient decision, driven by unpredictable events. In others, however, the retreat on objectives result from failures of design, namely poor project conception (i.e. SNPP in Paraguay; PROMEETII in Ecuador). Let us leave aside those hurdles that cannot be blamed on the Bank and focus on those that were engendered in house. In those, it is relevant to decide whether the retreat was the result of incompetence in implementation or was provoked by a design flaw that led the project to promise too much. The Bank has an obvious interest in making loans. Countries have an obvious interest in making good on their commitments and implement the loans. Project officers have an obvious interest on both ac1

See Aaron Wildavsky, "The Strategic Retreat on Objectives" Policy (1976) 2:2, 500-526.

278

Analysis

counts. Clearly, no one has an interest in being quixotic. When implementation sends signals that originally planned objectives may be not attainable, projects are adapted accordingly. In other words, objectives are redefined to make the project look good. But nobody asks why did this had to happen. Was it unrealistic design or poor implementation? The difference between strategic retreats emerging as a result of poor implementation and those prompted by fundamental flaws in project design is significant and worth pinpointing. The first teaches lessons about the implementation process. The second is telling that designers made mistakes. Unfortunately, due to inadequate evaluation, the Bank is shortchanged in these answers. E. Do institutions learn from the past?

This review of training projects started with a metaphor from development biology: as new generations of project mutate, do they evolve in a direction that makes them better? The results seem quite straightforward. (i) The conception of projects clearly evolves. Designers tend to learn from experience and avoid the errors of the past. They respond to new challenges from society and avoid pitfalls met by previous projects. (ii) In contrast, the mechanisms for implementation rarely, if ever, change, even though the same problems keep reappearing for several decades. No matter how evaluated, there are far too many hurdles along the way and little is done to avoid them. In fact, some procedures fail to produce results, project after project. (iii) Therefore, it is easy to conclude that there is learning only on the design and the forms of intervention of the project. The machinery to implement these ideas does not seem to improve. (iv) Perhaps one of the reason there is no institutional learning is that the overall effort of evaluation is inadequate. Worse, it has deteriorated. The Bank has shown little inclination to examine more carefully the performance of its projects and to seriously fix the implementation problems.

279

References

Agudelo, S. (1980). Formation individuelle: système appliqué' par le SENAI de Rio de Janeiro. Cinterfor, Montevideo. Banco Interamericano de Desarrollo, Oficina de Evaluación, BID/EVO (1998). Análisis de Ejecución de Proyectos. Chile: Programa de Capacitación Laboral. Préstamo 686/OC-CH. Washington, DC: BID/EVO. Barbosa de Araujo, A. (1991). O Governo Brasileiro, o BIRD e o BID: Cooperaçâo e Confronto. Ipea. Barros, Alexandre R. et al. (n/d) "Acompanhamento de Egressos do Programa Estadual de Qualificaçâo Profissional do Estado de Pernambuco em 1997" Recife: FADE-UFPE. 1998. Becker, Gary (1993). Human Capital : A Theoretical and Empirical Analysis, With Special Reference to Education. Chicago: University of Chicago Press. Boltho, Andrea and Andrew Glyn (1995). "¿Hay alguna política macroeconómica capaz de crear empleos? Revista Inernacional del Trabajo" 114:4-5 (507-528). Carnevale A, Gainer L, Meltzer, A. (1988). Workplace basics: The skills employers want.. Alexandria, Virginia, The American Society for Training & Development. Castro C. M. (1988). "The right courses for the wrong jobs? Or viceversa?" in Managing Human Development. Edited by Khadifer Haq and Uner Kirdar. Islamabad, North South Round Table _ _ _ _ _ (1993a). "Algeria: Training for Jobs or Training to Keep the Youth out of the Streets." Unpublished Paper, World Bank, Washington. (1993b). "SENAI: The Modern Solution of 50 Years Ago." DSE, Manheim. (1994). "Training in Tunisia". Unpublished Paper, World Bank, Washington. (1995). "Training Policies for the End of the Century". Paris: UNESCO, International Institute for Educational Training. (1997). "Proyecto Joven: New Solutions and Some Surprises". Washington, DC: Inter-American Development Bank. (1998). "The Stubborn Trainers vs the Neoliberal Economists: Will Training Survive the Battle?" No. EDU-106. Washington, DC: Inter-American Development Bank. (2002). "The World Bank Polices: damned if you do, damned if you don't", Comparative Education 38 (4), 387-399. 280

(2004). "Brazilian technical education: the chronicle of a turbulent Marriage", In: Brock, C., Schwartzman, S. (Eds.) The Challenges of Education in Brazil. Symposium, Oxford. (n/d). "Senai: The Modern Solution of 50 Years Ago". Mimeo. and Alfthan, T. (1994). "Cuts in education: policy or politics?" In: Samoff, J. (Ed.), Coping with Crisis: Austerity, Adjustment and Human Resources. Cassell/ UNESCO, London/Paris. and Andrade, A.C. (1990). "Supply and demand mismatches in training: can anything be done?" International Labour Review 129, 3. and Carnoy, Martin (n/d). "Secondary Education and the Transition to Work," background paper to the Primary and Secondary Education Strategy. Mimeo. and Kanawaty, G. (1988). "Report for the Governing Body Committee on Employment, Training, Retraining and Labour Mobility." 240th Session of the ILO Governing Body, May-June 1988. and Kempner, K„ Bas, D. (1993). "Apprenticeship: the perilous journey from Germany to Togo." The International Review of Education 39. and Oliveira J B A (1991). "Education for all and the roles of training". Paper prepared for the Regional UNESCO Conference on Educational For All, Cairo, Feb. 1991. and Verdisco, A. (2000). "Training unemployed youth in Latin America: same old sad story?" In: Atchoarena, D. (Ed.), Transition From Youth From School To Work. HEP, Paris. Cinterfor, 1990. La formación professional en el umbral de los 90. Cinterfor, Montevideo. Cognition and Technology Group at Vanderbilt (1991). "Anchored Instruction and its Relationship to Situated Cognition". Educational Researcher. 19(6) 2-10. Costa Couto, R. (2000). A historia viva do BID e o Brasil. Edi?ao BID. Cox, C., 1997. "La reforma de la educación chilena", PREAL, Documento 8. Dougherty, Christopher (1989). "The Cost-Effectiveness of National Training Systems in Developing Countries". Washington, DC: World Bank. Ducci, María Angélica (1990). La Formación Profesional en el Umbral del Siglo XXI. Montevideo: CINTERFOR. Dybowsky, G., Hanf, G., Walter, M. (2007). "BIBB International: Strategy Paper on the Internationalization of German Vocational Education and Training." BIBB, Bonn. Elan M. (1989). "A critical introduction to the post-Fordist debate: Technology, markets and institutions". Working Paper No. 45, Linkoping 281

(Sweden), Linkoping University, Department of Technology and Social Change. Eliasson G (1987). The knowledge base of an industrial economy. Stockholm, Almqvist and Wiksell International. Favennec-Hery (1996). "Work and Training: A Blurring of the Edges". International Labor Review 135:6 (665-674). Federal Ministry of Education and Research (2007). Dual Training at a Glance. BMBF, Berlin. Frigotto, G. (1995). "Os delirios da razao." In P. Gentili (Ed.), Pedagogia da Exclusao. Petropolis: Editora Vozes. Greinert, W. (2002). The Dual System of Vocational Training in the Federal Republic of Germany. GTZ, Eschborn. Heyneman, S. (2003). "The history and problems in the making of education policy at the World Bank: 1960-2000." In D. Baker & D. Gustafson (Eds.), International Perspectives on Education and Society). Oxford: Elsevier Science.. pp. 315-337 Hollenbeck, Kevin M. (1996). "An Evaluation of the Manufacturing Technology Partnership (MTP) Program". Kalamazoo: W.E. Upjohn Institute for Employment Research. (1997). "School to Work: Promise and Effectiveness". Kalamazoo: W.E. Upjohn Institute for Employment Research. Hunter, W. & Browne, D. (2000) "World Bank directives, domestic interest and the politics of human capital investment in Latin America." Comparative Political Studies, 33(1). Ianni, O. (1996). "Teorias da Globalizagao." Cadernos de Problemas Brasileiros, No 318. Inter-American Development Bank, Board of Governors (1994). "Report on the Eighth General Increase in the Resources of the Inter-American Development Bank" (AB-1704). Washington, DC: Inter-American Development Bank. Social Programs and Sustainable Development Department, Social Programs Division (1996). "Supporting Reform in the Delivery of Social Services: A Strategy". Washington, DC: Inter-American Development Bank. (1997). Brazil: Technical and Vocational Education Reform Program (BR-0247). Loan Proposal. Washington, DC: Inter-American Development Bank. "Vocational and Technical Training: A Strategy for the IDB". (Washington: Interamerican Development Bank - SDS. (Working Paper,12/00,EDU,E,S) 282

International Labor Organization/Caribbean Office, ILO "The Challenge of Youth Unemployment in the Caribbean: The Role of Youth Employment Training Programs". Port of Spain: International Labor Organization. (1997). International Labor Organization/CINTERFOR (1998). Juventud, educación y empleo. Montevideo: ILO/CINTERFOR. International Labor Organization, ILO (various). International Labor Statistics Yearbook. Geneva, ILO. International Monetary Fund, IMF (1999). International Financial Statistics, 1998. Washington, DC: IMF. Jacinto, Claudia (1996). "Políticas Públicas de Capacitación Laboral de Jóvenes: Un Análisis desde las Expectativas y las Estrategias de los Actores". Buenos Aires: CEIL-CONICET. Johanson, Richard (1994). "World Labor Report: Retraining Workers Displaced by Structural Adjustment". Mimeo. Jones, P. W. (1992). World bank financing of education. London and New York: Routledge. Kopp, Hillary and Richard Kazis, with Andrew Churchill (1995). Promising Pratices: A Study of Ten School-to-Career Programs. Boston: Jobs for the Future. Kuenzer, A. (2000). Ensino Médio e Profissional. Cortez Editores. Lee, Eddy (1995). "Exposición general" Revista Internacional del Trabajo 114:4-5 (495-506). (1996). "Mundialización y empleo: ¿Se justifican los temores?" Revista Internacional del Trabajo 115:5 (527-541). Lindely, Robert M. (1996). "The School-to-Work Transition in the United Kingdom" International Labor Review 135:2 (159-180). Lopes, S. (1982). "Urna saga da criatividade brasileira." SENAI, Rio de Janeiro. Mallaby, S. (2004). The world banker. Penguin Books. Martinez, E. (1997). Chile: "Experiences in a Market-Oriented Training System." ILO, Geneva. Middleton, John. (1993). Skills for Productivity: Vocational Education and Training in Developing Countries. Washington, DC: World Bank. Ministerio de Trabajo y Previsión Social (1997). Creciendo Juntos: Chile Joven. Santiago: Ministerio de Trabajo y Previsión Social. Miranda-Ribeiro, Paula (1997). "Relatório Final: Grupos de Discussáo. Projeto "Metodología de Avaliagáo do Plano Estadual de Qualificagáo." Volume I. Belo Horizonte.

283

Mitchell, Ayse G. (1998). Strategie Training Partnerships between the State and Enterprises. Geneva: International Labor Organization. Moisés Nairn (April 1994). "The World Bank: Its role, governance and organizational culture", Carnegie. Monteiro, M. A. (2000). "Aprender a aprender: Garantía de uma educaijäo de qualidade?", Presenqa Pedagógica, 6 National Commission on Secondary Vocational Education (1985). The unfinished agenda. Business Education Forum. Nubler, I. (1997). "Evolutionary theory in education and training: a comparative analysis of France and Germany." Vierteljahrshefte zur Wirtschaftsforschung/Quarterly Journal of Economic Research 66 (2), 215-229. O'Leary, Christopher J. (1995). "Performance Management of Active Labor Programs in Eastern Europe." Kalamazoo: W.E. Upjohn Institute for Employment Research. Oliveira, J., Castro, C.M. (1991). "Individualised learning systems for vocational technical training: the SENAI system. A case study of the Euvaldo Lodi School in Rio de Janeiro", Brazil. In: Oliveira, J., Castro, C.M., Wilson, D.N. (Eds.), Innovations in Technical, Distance and Computer-Delivered Education and Training. Organization for Economic Cooperation and Development, OECD (1996). Enhancing the Effectiveness of Active Labor Market Policies: Evidence from Program Evaluations in OECD Countries. No. 56. Paris: OECD. Paredes, Ricardo, Ramón Figueroa, Guillermo Campero, and Guillermo del Campo (1996). Evaluación Programa Chile Joven. Santiago. Perkins, David (1992). Smart Schools, New York: The Free Press. Raizen, S (1991). "Learning and work: The research base". Paper presented at the United States/OECD Seminar on Linkages in VocationalTechnical Education and Training: Challenges, Responses, Actors, Phoenix, Arizona, 19-22 Mar. 1991. Rios-Neto, Eduardo and Ana María Hermeto Camilo de Oliveira (1998). Uma Metodología de Avaliagäo do Plano Estadual de QualiflgäoProfissional (PEQ): O Caso de Minas Gerais em 1996. Belo Horizonte. Schaack, K. (2008). "Why Do German Companies Invest in Apprenticeships?" Klaus Schwarz Verlag, Berlin. Schwartz, L. (1983). Pour sauver l'universite'. Seuil, Paris. Silverberg, Marsha and Joshua Haimson and Alan M. Hershey (1998). "Building Blocks for a Future School-to-Work System: Early National

284

Implementation Results. Executive Summary." Princeton: Mathematica Policy Research Inc.. Soffer, Benson and Manuel Zymelman (1993). A Global Overview of Youth Employment Programs. Washington, DC: World Bank, Africa Technical Department, Human Resources Division. Stern, David, and Neal Finkelstein, James Stone III, John Latting, and Carolyn Dornsife (1994). Research on School-to-Work Transition Programs in the United States. Berkeley: National Center for Research in Vocational Education. Torres, M. R. (....) "Melhorar a qualidade da educagáo básica? As estratégias do Banco Mundial." In O Banco Mundial e as Políticas Educacionais (pp 125-193). Sao Paulo, Editora Cortez. Unidad Coordinadora del Programa (n/d). "Programa de Capacitación Laboral de Jóvenes: Chile Joven." Santiago. Mimeo. United States Department of Labor 1990. The bottom line: Basic skills in the workplace. Washington DC. Williamson, J. (2004). "Nova agenda e o consenso de Washington." In Kuczinski, Pedro-Pablo and J. Williamson. Depois do consenso de Washington. Editora Saraiva. Wilson, D.N. (1996). Reform of Vocational and Technical Education in Latin America. PREAL, Occasional Paper Series 2. World Bank (1980). Central Projects Note 5.01: Policy and Guidelines on Training in Bank/IDA Projects. Washington, DC: World Bank. , Education and Employment Division, Population and Human Resources Department. (1988). Annual Operation Review: Fiscal 1987 Education and Training. Washington, DC: World Bank. (1991). Vocational and Technical Education and Training. World Bank, Washington.

285

Full Table of Content Preface

7

Chapter I The Art and Science of Training

9

How Trainers Train - How Educators Teach

11

Training Ideas from Oklahoma A. The academic warranty B. Teachers accountable for the employment of their students? C. Advisory Boards: telling schools what employers want D. Courses mutate: cobblers become pedorthic technicians E. Teachers with diplomas or teachers with experience? F. From roadside mechanics to automotive engineers G. Do we still need lectures in the classrooms? H. What if your teacher is in another town?

17 17

Schools with Technology or Schools versus Technologies? A. What do we expect technology to do for us? 1. To make serious education even better 2. Technology to reach farther with respectable quality B. Test of replicability: the pilot project works ok, but 1. The cash flow test 2. Doctor Frankenstein's test 3. Institutional sabotage test 4. Integrity test C. The technological tools Graph 1 D. Choice of technology is constrained by the numbers of students E. Computers are devious fellows, they don't always teach 1. Tutorials and Computer-Aided Instruction (CAI) 2. Drill and Practice 3. Simulation and Games 4. Development of critical skills 5. Basic skills and the preparation for the labour market 6. E-Learning and virtual classes 7. The digital textbook 286

18 19 19 20 20 21 23 25 25 25 26 27 28 28 29 29 30 31 33 35 37 39 39 40 40 41 42

8. Computers to manage schools F. Image Technologies: Some poorer countries do much better 1. Talking heads: a teacher, a camera and a tripod 2. Recycling live courses 3. Teacher styles 4. Demonstrations and live examples 5. Dramaturgy: soap opera migrates to education G. Do machines know the science and art of teaching? 1. Contextualization H. The sociological wars: schools versus computers

43 43 45 45 46 46 46 47 47 48

Skills Training: Where Simulations are at Home

50

Chapter II Training Policies that Make Sense

55

Training in the Developing World: Issues and Policies 57 A. Economic development without training? 57 B. Training policies: general principles 58 1. Training requires an enabling environment 58 2. Training does not replace a good education 59 3. Good training is also good education 60 4. Training pays 61 5. No demand, no training 63 C. Those who benefit should pay 65 D. Finetuning the delivery of training 68 1. Improving the performance of existing providers of training...68 2. New modes of apprenticeships 70 3. Promoting lifelong learning 71 4. New forms of delivery for forgotten clienteles 72 5. Upgrading training for the modern economy 73 6. Developing materials and training trainers 74 7. Training as social policy but not to create jobs 75 Have Training Policies Changed? A. General principles are the same B. New priorities are emerging 1. Reforming training institutions is the number one priority. 2. Programmes to reach the poor 3. Improvements in classical vocational training 4. Funding the private sector

78 78 79 79 80 81 82

Training Dilemmas and False Dilemmas

84

287

A. Training the few or training the masses? 84 85 B. Cohibitation of the old crafts with high tech C. Structured training of teachers "creating" their own materials?..86 D. Structured learning curbs creativity? 87 E. What if workers lack basic skills? 90 The Cost of Equity and the Cost of Efficiency

93

Training the Poor When There Are Not Enough Jobs A. Education with a bit of training or training with a bit of education B. Wrong education and wrong training C. Promising solutions 1. Search the job market and target the poor: the"Joven Projects" 2. "Practical" education in academic schools?

96

100 100

Is Training in Developing Countries Different from Training in Developed Countries? A. Many hits, many misses: heterogeneity is the norm B. Weak targeting, worse correction of fire C. The mind and the hand do not learn together D. Poor students, under-funded schools E. The logistics are more vulnerable F. The ever-elusive informal sector G. One-Stop training for all H. Conclusions?

102 102 103 104 105 107 108 108 109

Chapter III The Latin American Panorama

Ill

Community Colleges and Latin America: Clone, Reject or Inspire? A. An overview of Community Colleges 1. Origins: Junior Colleges 2. Financing 3. Economic benefits for students 4. The controversial role in social mobility B. The changing functions of Community Colleges 1. From Junior Colleges to Multi-Purpose Institutions 2. The new path to vocational education 3. Contract training 4. Adult Education

113 116 116 118 120 121 123 123 124 127 128

288

96 98 99

C. Good for the United States, good for Latin America? 128 1. Secondary education for the poor: similarities with Latin America 129 130 2. Most people take small steps in social mobility 3. The financial equation must accommodate the less affluent. .133 4. Pedagogical innovation: different students, different classrooms 134 5. The market imperative: no demand, no training 136 138 6. Titles and status do matter 7. Certification and accreditation in Latin America 139 8. Bridges to Four-Year Programmes: are they possible? 140 9. What should we expect from the public sectors? 142 D. Community Colleges should inspire Latin America 144 When Brazilian Employers Control Training, Many Things Can Happen A. SENAI: industrial training B. SENAC:The trainee-driven solution for services C. SENAR: Why run schools when you can buy training? D. SEBRAE: Employers' support to micro-enterprise E. SENAT: Distance education to truck drivers F. Lessons: Ample variety, the same legal formula

146 148 149 150 151 151 152

Brazilian Technical Education: The Chronicle of a Turbulent Marriage A. The Catch 22: The more you improve, the more dysfunctional it becomes B. World trends in technical education C. Finally, divorce! D. Implementation and confrontation E. New confrontations and growth F. The elusive balance: the staircase theory G. The rough ride of technical education

154 158 160 161 166 168 172

Chapter IV Multilateral Agencies: Understanding the Beasts

175

Does Skills Development get Short-Changed in Post-Jomtien Days?

177

Do Training Institutions Learn from Experience? A. Do institutions learn? 1. The holy German Dual System

180 181 181

289

154

2. The French AFPA exports "alternance training" 3. How multilateral banks learn and teach (preach?) 4. The World Bank and its comprehensive high schools 5. IDB training: new project ideas and administrative stalemate 6. Is SENAI a learning organization? 7. Demand-driven training: in theory or in practice? B. Lessons: training institutions learn, but not always 1. Hidden agendas ~ 2. Organizational culture 3. The critical issue is not learning but using what was learned 4. The chances of learning are greater if the payoffs are positive 5. Power inside institutions matter Can Multilateral Banks Educate the World? A. How Development Banks Work 1. Economic rationale: low interest borrowing 2. Countries have to repay their loans, no matter what 3. Loan preparation takes a long time 4. Strict contracts protect loans from political vagaries B. Why Development Banks are accused of the worst sins 1. Values and beliefs 2. Decision making: How the world influences the banks 3. Leftwing educators battle "neoliberal" values 4. NGOs fight for influence in decision-making C. How Development Banks design and negotiate loans 1. What governments want, what countries need, and what the banks want to sell 2. Economic realities or political feasibility? 3. Conditionalities: powerful, but dangerous 4. Conditionalities in action (or inaction?) D. Are loans effective? 1. Who benefits? 2. What can go wrong with loans? 3. Managing loans: a nightmare of paperwork 4. Evaluation has unclear impact on design and procedures 5. Ideas or funds? E. Lessons? F. The new scenarios 290

183 186 188 190 194 201 203 203 203 204 204 206 207 207 208 210 210 211 212 212 214 216 217 218 218 219 221 222 224 224 224 226 227 228 230 232

Can Development Banks Promote Social Reform?

235

Can Training Bring about Organizational Change? 238 A. Training does not bring about organizational changes 238 1. In-service training for teachers 239 2. Industrial training in Russia 240 3. SUDENE versus EMBRAPA 240 241 4. Training in World Bank and IDB loans 5. Computers in schools 244 B. Training works when it is part of a decision to change 245 1. Training professionals to use computers in the ILO 246 2. Pitágoras and the Integrated Management System 247 3. Preparing technical staff at INEP (Brazil) 248 4. Training in the petroleum industry 248 C. Tentative conclusions 250 Monitoring Training Projects: A phylogenetic Approach 251 A. Are banks learning from experience? Do projects improve? 251 B. Mutation and adaptation: the search for solutions 254 1. Supporting training institutions 255 2. Support to technical schools 258 3. The checkbook approach: short training with competitive funds 261 4. Survival of the fittest 263 C. The contrast: stalemate in management 263 1. Predictable and unpredictable accidents 264 2. Buildings get built, institutions don't change 265 3. Weak institutions, weak management 268 4. The dilemma of the executing units 269 5. The procurement and contracting nightmare 272 6. Counterpart financing 274 D. Evaluation, feedback and institutional learning 275 1. Information and analysis: a weak link 275 2. Evaluation is a mutating species (is it mutating backward?). .277 3. Strategic retreat on objectives: implementation or design faults? 278 E. Do institutions learn from the past? 279 References

280

291

Melanie Krebs Zwischen Handwerkstradition und globalem Markt Kunsthandwerker in Usbekistan und Kirgistan ISBN 9 7 8 - 3 - 8 7 9 9 7 - 3 7 9 - 8

Seit Zerfall der Sowjetunion durchlaufen die unabhängigen Staaten Zentralasiens vielfaltige politische, wirtschaftliche und kulturelle Transformationsprozesse, die sich auf alle Lebensbereiche der Bevölkerung auswirken. Internationale Organisationen der Entwicklungszusammenarbeit nehmen Einfluss auf diese Prozesse, indem sie einerseits lokale Ressourcen zu stärken versuchen, andererseits die Implementierung internationaler Normen und die Öffnung auf den globalen Markt fördern wollen.

Sonja Hegasy/ Elke Kaschl (eds.) Changing Values among Youth Examples from the Arab World and Germany ISBN 9 7 8 - 3 - 8 7 9 9 7 - 6 3 7 - 9

It is crucial, not merely for Arab societies, to know more about the attitude of almost two-thirds of the Arab population to social and political issues. This volume takes examples from Morocco, Egypt, Palestine, Iraq, Syria and Germany to demonstrate the potential and the limitations of youth research in the Arab world and beyond.

Carsten Wieland Im Grenzland Reportagen aus drei Kontinenten ISBN 9 7 8 - 3 - 8 7 9 9 7 - 6 4 4 - 7

Dieses Buch ist ein Plädoyer, den Respekt vor Grenzen zu verlieren. Es verbindet die Erlebnisse eines Reisenden und Journalisten mit Einsichten in Politik, Kultur und Gesellschaft von Grenzregionen. Ob im Nahen Osten, in Lateinamerika, auf dem indischen Subkontinent oder auf dem Balkan: Carsten Wieland erzählt die Eindrücke eines Reporters jenseits der Tagesberichterstattung, Kuriositäten am Rande von Konflikten, menschliche Tragik, aber auch Freude. Im Grenzland unterscheiden sich die politische und die menschliche Perspektive oft deutlich voneinander. Ein beeindruckendes Werk. Deutschland Radio Kultur Gesamtprogramm: www.klaus-sehwarz-verlag.com