Contracts and Pay: Work in London Construction 1660–1785 (Palgrave Studies in Economic History) 3319575074, 9783319575070

Table of contents :
Acknowledgements
Contents
List of Figures
List of Tables
1 A Short History of Builders’ Wages in Economic History
The Creation of Statistical Wage Series
Nominal London Wage Series
Gilboy’s Sources, Methodology, and Findings
Schwarz’s Sources, Methodology, and Findings
Boulton’s Sources, Methodology, and Findings
Associated Series
London in the Sixteenth Century
Beyond London: Northern England
South East England, Real and Silver Wages
‘Nominal’ Sources and Their Effect on ‘Real’ Wages
Comparisons with Other Regions and Cities
Market and Business Organisation
2 The Market for Building
Construction Market Segmentation
Development Before the Fire
Rebuilding and the Market for Construction Services
Beyond Rebuilding and into the Eighteenth Century
Organisation of Contracts and Pay at the Sites and Sources
Westminster Abbey
Greenwich Hospital
Middlesex Sessions
Middle Temple
St Paul’s Cathedral
Westminster Bridge
Bridge House
The Office of the King’s Works
Summary
3 Contractors
4 Contracts and Ways of Working in the Building Trades
Three Ways of Working
Contracts
Contracting Practice
The System in Practice and Its Influence on ‘Wages’
5 What did Bosses (in London Construction) do?
The Jenings Case
Management and Operating Costs
Search and Information Costs
Bargaining Costs
Enforcement Costs
The Costs of Operating: Managing Sites and Labour
Administration, Accounting and Auditing
Rent
Carriage
Tools
The Costs of Finance
Deductions
Profit and Overall Margin
Summary of Operating Costs and Profit, and Ordinary Work
Early Modern Bosses
6 Contracts and Pay at St Paul’s Cathedral, and at the Office of the King’s Works
Work and Pay in Practice at St Paul’s
Hierarchy, Skill and Pay
The Number of Days Worked and Income
Westminster Abbey
The Office of the King’s Works
A Summary of New Findings
7 Pay at Bridge House, Westminster Bridge, and Middle Temple
Bridge House Pay 1660–1685
Bridge House Masons 1684–1780
Tide Carpenters at Bridge House 1720–1785
Land Carpenters 1706–1785
Bridge House Pay in the Long Run
Pay at Westminster Bridge and under Andrews Jelfe
Pay at Middle Temple
8 Contracts and Pay in Construction in the Long Run
Wage Formation and Non-pecuniary Pay
Skill and Wage Formation
Nominal Incomes and Real Wage Implications
9 Conclusions and Further Notes on Understanding Early Modern Contracts and Pay
The Level of Pay in London
International Comparison of Building Wages
Bibliography
Index

Citation preview

PALGRAVE STUDIES IN ECONOMIC HISTORY

CONTRACTS AND PAY Work in London Construction 1660–1785

Judy Z. Stephenson

Palgrave Studies in Economic History

Series editor Kent Deng London School of Economics London, UK

Palgrave Studies in Economic History is designed to illuminate and enrich our understanding of economies and economic phenomena of the past. The series covers a vast range of topics including financial history, labour history, development economics, commercialisation, urbanisation, industrialisation, modernisation, globalisation, and changes in world economic orders. More information about this series at http://www.palgrave.com/gp/series/14632

Judy Z. Stephenson

Contracts and Pay Work in London Construction 1660–1785

Judy Z. Stephenson Bartlett, University College London London, UK

ISSN 2662-6497 ISSN 2662-6500  (electronic) Palgrave Studies in Economic History ISBN 978-3-319-57507-0 ISBN 978-3-319-57508-7  (eBook) https://doi.org/10.1007/978-3-319-57508-7 Library of Congress Control Number: 2017943661 © The Editor(s) (if applicable) and The Author(s) 2020 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Acknowledgements

This book was researched during my doctoral studies in the Department of Economic History at the London School of Economics and Political Science 2012–2015. If it were not for the intellectual generosity and loyal stewardship of Patrick Wallis, my supervisor and mentor, the evidence and ideas within would not have got very far. Overturning established ideas in scholarship is arduous and anxiety inducing at the best of times. Through some quite hard ones his disciplined support made it possible. Gerben Bakker, Chris Minns, Oliver Volckart, Max Stefan Schulze, Michael Aldous, Tobias Vogelsang, John Gent, Pat Hudson, Stephen Broadberry, Kent Deng, Peter Sims, Tim Leunig, and Linda Sampson all provided mentorship, encouragement, camaraderie, or advice that helped a controversial idea become reality, and I am very grateful for all their help. Bruno Blonde, Heidi Deneweth, James Campbell, and my longsuffering examiners, Jane Humphries and Jeremy Boulton, did much to improve the arguments and presentation. All errors, typos and misconceptions are mine. I thank the staff of the London Metropolitan Archive, The National Archives, Christine Reynolds at Westminster Abbey, Lesley Whitelaw at Middle Temple and Jennifer Thorp at New College, Oxford for v

vi      Acknowledgements

their time and insight, and the privilege of viewing and handling their immaculately maintained and, often, very beautiful records. I am indebted to the Economic History Society who provided me with a badly needed Ph.D. bursary in 2013 and a wonderful fellowship at the IHR in 2015–2016. Colleagues at the Cambridge Group for the History of Population and Social Structure helped me improve some of the context and arguments, whilst those at Wadham College, Oxford provided a happy intellectual home in the latter stages, and I thank Carol Richards and her family for endowing the David Richards fellowship in economic history there. My children, Belle and Arthur, provided the best discipline and distraction any author could ask for. It is, of course, dedicated to them, with the wish that they may one day find as much professional fulfilment and support as I have in researching and writing it.

Contents

1 A Short History of Builders’ Wages in Economic History 1 2 The Market for Building 35 3 Contractors 65 4 Contracts and Ways of Working in the Building Trades 79 5 What did Bosses (in London Construction) do? 107 6 Contracts and Pay at St Paul’s Cathedral, and at the Office of the King’s Works 141 7 Pay at Bridge House, Westminster Bridge, and Middle Temple 173 8 Contracts and Pay in Construction in the Long Run 201

vii

viii      Contents

9 Conclusions and Further Notes on Understanding Early Modern Contracts and Pay 223 Bibliography 239 Index 253

List of Figures

Fig. 1.1 Fig. 1.2

Gilboy’s wage series for London 1700–1787 11 Schwarz’s craft and labour wage series for London 1700–1800 13 Fig. 1.3 Boulton’s wage series for London 1661–1721 16 Fig. 1.4 Comparison of three series, craft d. per day 16 Fig. 1.5 Comparison of three series, labourers d. per day 17 Fig. 1.6 Allen and Gilboy, London craftsman’s d. per day 24 Fig. 1.7 Allen and Schwarz, London craftsman’s d. per day 25 Fig. 1.8 Allen and Boulton, London craftsman’s d. per day 25 Fig. 1.9 Allen and Gilboy, London labourer’s d. per day 26 Fig. 1.10 Allen and Schwarz, London labourer’s d. per day 26 Fig. 1.11 Allen and Boulton, London labourer’s d. per day 27 Fig. 2.1 Middlesex deeds registered by year 48 Fig. 5.1 Likely roles under large mason contractor off-site 117 Fig. 5.2 Likely roles under large mason contractor on-site 117 Fig. 5.3 Discount of Edward Strong’s Bills Greenwich Hospital 1698–1708 showing actual percentage discount on each bill and 10 bill MAV 133 Fig. 6.1 Amount of day bills and task/measure/other bills at Westminster Abbey 1713–1715 in £ 164 ix

x      List of Figures

Fig. 7.1

Wrought with the gin, first team number of men and tides worked October 1745–January 1749 Fig. 7.2 Wrought with the gin, second team, number of men and tides worked October 1745–January 1749 Fig. 7.3a ‘Gin men’s team’s tide work at Bridge House April 1755–April 1766 Fig. 7.3b ‘Gin mens’s tide work at Bridge House April 1755–April 1756 Fig. 7.3c ‘Gin mens’s tide work at Bridge House April 1755–April 1756

187 187 189 189 190

List of Tables

Table 1.1 Table 3.1 Table 5.1 Table 5.2 Table 6.1 Table 6.2 Table 6.3 Table 6.4 Table 6.5 Table 6.6 Table 6.7 Table 6.8

Eighteenth century London sources for wages in London Mason’s contracts at St Paul’s Cathedral Calculation of operating margin for ‘extraordinary’ day work Operating margin for contractors supplying weekly labour at Bridge House—ordinary work St Paul’s charge out rates for day work in d. Day rates Kempster team St Paul’s 1700–1702 Day rates 1708, Kempster team, by hierarchy found in books Mason’s day rates on the columns (from “An account of masons time it was at the columns”) Levels of skill and ranges of day rates paid in Kempster team Percentage difference between Boulton series and Kempster pay per day for craftsmen Carpenters on day work, quarterly accounts, St Paul’s, 1696–1700 (quarterly averages of total numbers of men observed and modal days worked) Average number of days worked for all men

8 69 136 138 146 150 152 153 154 157 160 160 xi

xii      List of Tables

Table 6.9 Table 6.10 Table 6.11 Table 6.12 Table 6.13 Table 7.1 Table 7.2 Table 7.3 Table 7.4 Table 7.5 Table 8.1 Table 8.2 Table 8.3

Average number of days worked per annum by day rate paid Average number of days worked by craftsmen in a week when on site Average numbers of days worked by length of employment with Kempster Labourers days worked per week 52 weeks Sept 1712–1713, and 40 weeks March 1714–January 1715 Office of King’s works, charge out rate ranges for craft and labour in d. per day Pay at Bridge House in 1661 Land Carpenter’s bill for 3rd October 1685 Thomas Wise, Mason’s bill for 3rd October 1685 Sample of Sparruck’s summer and winter bills 1722 Mason’s team day rates found in Jelfe letter book, May 1734–March 1735 Nominal annual wage income in £ based on pay per day or other unit of measurement and varying units per annum Craftsmen’s revised Welfare ratio’s London, where the nominal wage is deflated by 20% Labourer’s revised Welfare ratios for London, where the nominal wage has been deflated by 30%

161 161 162 164 168 177 180 181 185 197 218 220 220

1 A Short History of Builders’ Wages in Economic History

Throughout history, construction workers have not only built many of the landmarks of modern day London but, unknowingly, given us the building blocks for the methods most widely used to evaluate the economic performance of the past. Economic historians have a long been interested in the pay of construction workers. Just as masons, carpenters, and plumbers proliferate in the account books of old institutions and among wage regulations and statutes of boroughs and sessions across England, building workers and their labourers appear regularly in all discussions of work and wages, in urban and rural history. From the end of the nineteenth century, economic and social historians began to use builder’s recorded ‘day wages’ to represent an average non-agricultural worker’s pay in studies of long run wage trends. By converting this ‘average wage’ into a long-run data series, they created influential estimates of living standards, ‘welfare ratios’, GDP and national output, economic growth rates, human capital or skill levels, and associated labour supply and demand.1 1By

way of current example see; Van Zanden, ‘The Skill Premium and the Great Divergence’, pp. 121–153; Allen, ‘The Great Divergence in European Wages and Prices’, pp. 411–447; Clark, ‘The Condition of the Working Class in England, 1209–2004’, pp. 1307–1340.

© The Author(s) 2020 J. Z. Stephenson, Contracts and Pay, Palgrave Studies in Economic History, https://doi.org/10.1007/978-3-319-57508-7_1

1

2     J. Z. Stephenson

Our knowledge about the trends and comparative features of early modern European economies may rely on builder’s wages, but those wages were gathered without much reference to how the construction industry actually developed throughout the early modern period.2 This book will show, with hard evidence, that the empirical base for our understanding of early modern wage earning and the organisation of the construction industry in England should be revised. It argues that builders’ wages are a poor proxy or data set for estimating living standards and real wages in the very long run, but that there is much to be learned from a closer look at the construction industry in this period. It offers a new narrative for economic historians to understand and interpret the construction industry in early modern England that draws on established architectural history and on a broadly conceived transaction cost tradition in economics and business history. Whilst firmly rooted in broad economic history narratives, it brings new case studies from new sources to light which will be of interest to construction, business and social historians. For some, this new narrative will appear unusual. It does not talk much about guilds, corporations, customs, apprentices, and trades. The unit of analysis is, for the most part, the contract, the ‘firm’, and the worker within it. The analysis challenges some established narratives, ideologies, and practices, but the method is wholly rooted in archival evidence.

The Creation of Statistical Wage Series How did builder’s pay come to be the measure of the historical ‘average’ wage? One of the primary reasons is the lack of any other sources. Students of early modern economic and social history are frequently told about the scarcity of wage records. The records of most organisations before the mid nineteenth century do not contain wage books. They are heavy with bills, vouchers, acquittances, and payments.

2Broadberry et al., British Economic Growth 1270–1870 is the first output based (as opposed to real wage based) study of long-run economic growth in England.

1  A Short History of Builders’ Wages in Economic History     3

Any references to wage payments tend to be irregular and fragmentary. Court and sessions records, the sources of so much rich economic and social information, often refer to the payment or non-payment of ‘wages’, but rarely tell us how much those wages were. In England, a seminal contribution to historical wages and prices was made by J.E. Thorold Rogers in the late nineteenth century.3 Rogers’ aim was a definitive history, or a full account of prices and wages. The introduction to the final volume highlights how huge an undertaking this was. The task may have been long and arduous, but the methodology was simple. Rogers sourced the recorded prices of goods, commodities, materials, produce, and labour across the United Kingdom over a span of six centuries. Identifying institutions that were likely to have very long-run records—such as Oxford and Cambridge Colleges, Westminster Abbey, the Inns of Court etc.—he noted the units and prices at each location at various dates in the long run. The results were displayed in list form, categorized by type of good, in seven volumes, with the last volume made up of two parts. Old organizations have both old buildings that need maintenance and an occasional requirement for new accommodation, and so tend to have a lot of building accounts. Rogers’ data set had almost ­continuous records of carpenters, masons, bricklayers, and labourers, and many for plumbers and glaziers too. Rogers’ records are acknowledged, rightly, as completely accurate, but it is easy for the modern researcher to not appreciate his life’s work as much as they might. The information given is listed by year, by type of good, and then place, but the observations are given in note form with idiosyncratic abbreviations for the quantities and other information. Anyone who wants to use the information must re-transcribe the data, and then convert the measures or units into standard form before comparing or utilizing the data further. As no names or other account information are usually given, it cannot be ascertained whether long-term contractual considerations affected prices. For instance, when bread or coal was purchased by Westminster Abbey at the same price per loaf or cauldron in two separate years, it is 3Rogers,

A History of Agriculture and Prices in England.

4     J. Z. Stephenson

not clear if it was sold to them by one or various vendors.4 Regarding wages, Rogers’ unequivocal finding was that from 1563 to 1824 the authorities and employers conspired to ‘cheat the English workman of his wages’, and ‘were engaged in grinding the English workman down to the lowest pittance’.5 At the end of the nineteenth century, the role of wages in ­understanding the economy more broadly continued to be developed in a statistical, rather than historical or theoretical, manner. Arthur Bowley is considered the father of modern statistics by many. We may also consider him father of the modern historical wage series. Bowley was concerned with wages and real wages as expressions of national output, and to this end he assigned huge importance to the calculation of a national ‘average’ wage over time. The key barrier to this was, as it remains, lack of data but Bowley proposed the solutions which defined the approach of economists and historians for another century. The most natural method, he said, would be to calculate the total amount paid in wages and divide by the number of workers. Since this data was not readily available, he advocated calculation of the mode, the observation of the most common or prevailing wage rate, which he acknowledged could be misrepresentative, or the median, the wage earned by the man where equal numbers of men are paid at ‘above and below that wage’. He acknowledged that the distribution around an average might be difficult to estimate.6 Crucial to either method is the issue of its ‘representativeness’ of the average. In this regard, Bowley relied on two assumptions, both of which were to become influential. The first was theoretical—the law of one price, or in this case, wage: …in spite of this apparent want of connection between the wages of one class of men and another there are very distinct causes which make the following law hold:- at the same time and in the same place the wages for equal effort 4Rogers, Six Centuries of Work and Wages: The History of English Labour, pp. 159–197, 326–357 discusses labour and wages in the long run in context. 5Ibid., p. 398. 6Bowley, Wages in the United Kingdom in the Nineteenth Century, pp. 21–23.

1  A Short History of Builders’ Wages in Economic History     5

of men of the same capacity are equal to one another; or more generally, the wages throughout the country of equal degrees of skill are equal at any given time. If this is so we shall find it useful to watch the change of the rate paid for a certain degree of skill even though the number of persons paid at this wage may be a very small proportion of the total number doing similar work.7

In other words, if the law of one wage or price holds, as a simple neoclassical perspective in economics asserts it does, then we can assume that persons of the same or equivalent skill, or productivity, in the economy will be paid the same amount. If this is assumed, then it is not necessary to collect a data set of all wage earners in order to understand or model all earners. All one needs to know is a person’s wage, and his place within the distribution of skill or wages. The law of one price remains to the present day the bedrock of many authors’ work on wages of the past, although it is rarely explicitly stated. The second assumption has been just as important. Although Bowley’s numbers for the nineteenth century utilised sources from agriculture, shipyards, textile, cotton, and iron industries for the long run, after some empirical testing of existing data taken from the likes of Eden’s ‘State of the Poor’ and McCulloch’s statistics, Bowley decided that the rate of change of pay in the building trades, specifically bricklaying, would ‘serve as an indication’ of rates of change elsewhere. This was justified on the grounds that there had been no technological change in the business of bricklaying over the centuries.8 From 1770, he took rates from Rogers and the McCulloch Trade Directory (which reported Greenwich Hospital price rates), and from the later nineteenth building builders’ price books for London, and declared them representative of a worker for whom technological change could not account for wage or productivity increases.9 Subsequent authors have mostly taken

7Ibid.,

p. 18. pp. 59–63. Bowley cautioned that rates in building trades had been steady where other trades were rising in 1877–1891, but still felt that bricklayers and other building trades were indicative in the long run, p. 63. 9See also Eden, The State of the Poor; McCulloch, A Dictionary, Practical, Theoretical and Historical of Commerce and Commercial Navigation. 8Ibid.,

6     J. Z. Stephenson

this to mean representative of the average worker. Since Bowley, because the records of large building projects are usually well archived, and the skill and activities of builders have been assumed to be reasonably uniform across Europe, builders’ wages became the proxy for the average, representative urban, skilled and unskilled manual wage throughout the ages. In 1956, Henry Phelps Brown and Sheila Hopkins used such nominal wage series to compile probably the most influential data-set ever in economic history: a long-run analysis of ‘the Prices of Consumables, Compared with Builders’ Wage-Rates’.10 Although they cautioned that their series was not that of a real wage it has always been treated as such.11 The series, still used and cited today by anyone who wants to show insight into the living standards of the past, enshrined the role of builders in the determination of the wages and living ­standards of the past. Both the methods used to calculate a basket of goods, and the prices of goods within it have been revised and added to since 1956, but the wage figures have changed surprisingly little. The most up to date London real wage series was constructed in 2001 by Robert C. Allen as a part of a comparative study of the divergence of wages and prices across Europe and the globe in the eighteenth century. Gregory Clark has compiled long run real wages for both urban, and rural, agricultural sectors on a composite basis for the whole of England from similar sources.12 The nominal wage figures that both Allen and Clark have used come from a surprisingly small set of sources.

10Phelps

Brown and Hopkins, ‘Seven Centuries of the Prices of Consumables, Compared with Builders’ Wage-Rates’, pp. 296–314, based on ‘Seven Centuries of Building Wages’, pp. 195–206. They credit the work of Gustav Steffen (1901) in identifying long run trends, although did not draw on his work for the eighteenth century, Ueber die Kaufkraft der Löhne erwachsener männlicher Arbeiter in England während des Durchbruches des Fabriksystemes (1760–1830). 11Ibid., p. 296. 12Allen, ‘The Great Divergence in European Wages and Prices’, pp. 411–147; Clark, ‘The Long March of History’, pp. 33–38; ‘The Condition of the Working Class’, pp. 1307–1340.

1  A Short History of Builders’ Wages in Economic History     7

Nominal London Wage Series For eighteenth century London, only four series of nominal wages have ever been produced: those of B.L. Hutchins; Elizabeth Gilboy; Leonard Schwarz; and Jeremy Boulton.13 Hutchins’ figures, published in 1899–1900, have never been widely used, although they were from a large-scale, important source—the Office of the King’s Works at the Tower of London. The other three have been used by multiple authors and they form the bulk of the series constructed by Allen for international comparison.14 For the period prior to 1609, the series comes from Steve Rappaport’s study of sixteenth century social structure in the city.15 The main sources and chronological coverage of the three best known and most used archival-based series created for London 1650–1800 are listed in Table 1.1. They are all studies of building craftsmen and labourers in London and they have some considerable overlap, commonalities, and gaps or omissions. For instance, Schwarz’s sources replicate one part of Gilboy’s (Greenwich Hospital). Boulton’s data provides the only sources on London earning between 1609 and 1700. All subsequent authors draw on these data. For instance, in his 2001 comparative wage study, Allen used Schwarz, Boulton, and Gilboy for London wage figures.16 For south-east England, he used Phelps Brown and Hopkins.17 Phelps Brown and Hopkins had used Gilboy’s suburban figures and some from Thorold Rogers, who in London had pretty much the same sources as Gilboy—Westminster Abbey and Greenwich. A few years after them, Deane & Cole’s seminal ‘British Economic Growth’ also used Gilboy. Botham and Hunt’s 1987 paper on wages used Gilboy, 13Hutchins,

‘Notes Towards the History of London Wages’ (1899), pp. 599–605; ‘Notes Towards the History of London Wages’ (1900), pp. 103–104; Gilboy, Wages in Eighteenth Century England; Schwarz, ‘The Standard of Living in the Long Run’, pp. 23–41; Boulton, ‘Wage Labour in Seventeenth-Century London’, pp. 268–290. 14See Allen’s compiled data set at https://www.nuffield.ox.ac.uk/people/sites/allen-research-pages/, http://www.iisg.nl/hpw/data.php#europe. 15Rappaport, Worlds Within Worlds: The Structures of Life in Sixteenth-Century London, pp. 123–160. 16Allen, ‘The Great Divergence in European Wages and Prices’, Appendix 1, p. 345. 17Phelps Brown and Hopkins, ‘Seven Centuries of Building Wages’, pp. 195–206.

8     J. Z. Stephenson Table 1.1  Eighteenth century London sources for wages in London Author

Archival sources

Elizabeth Gilboy 1934

1700–1787 Westminster Abbey, Greenwich Hospital, Southwark and Middlesex sessions Middlesex session, Greenwich Hospital 1700–1860 Middle Temple, Carpenters Company, 1574–1721 Charterhouse

Leonard Schwarz 1985 Jeremy Boulton 1996

Years covered

while referring briefly to Schwarz, as did Feinstein in his 1998 paper on living standards.18 Lindert and Williamson’s earlier work on living ­standards also used Gilboy. J.L. Van Zanden’s influential papers on comparative wages and skill use wages taken from John Chartres 1986 study on living standards, which is in turn, solely a moving average of Gilboy’s 1934 data.19 An influential paper by Broadberry and Gupta from 2006 comparing wage levels and productivity between England and Asia used Gilboy (as well as Allen’s 2001 revision of Gilboy’s material).20 The dominance of these data series, particularly that created by Elizabeth Gilboy, is clear. This might not have been the case if authors had known more about the context and background behind the data sets, and their relative methodological underpinnings.

Gilboy’s Sources, Methodology, and Findings Elizabeth Gilboy was a Harvard economist. She is known for early work on a theoretical and historical perspective of demand and consumption as causal factors in economic growth.21 In 1930 or thereabouts, 18Deane and Cole, British Economic Growth, 1688–1959: Trends and Structure; Feinstein, ‘Pessimism Perpetuated: Real Wages and the Standard of Living in Britain’, pp. 625–658; Botham and Hunt, ‘Wages in Britain During the Industrial Revolution’, pp. 380–399. 19Lindert and Williamson, ‘English Workers’ Living Standards during the Industrial Revolution’, pp. 1–25; Van Zanden, ‘Wages and the Standard of Living in Europe, 1500–1800’, pp. 175–197; Chartres, ‘Food Consumption and Internal Trade’, pp. 168–195. 20Broadberry and Gupta, ‘The Early Modern Great Divergence’, p. 12. 21Gilboy, ‘The Propensity to Consume’, pp. 120–140; ‘Demand as a Factor in the Industrial Revolution’.

1  A Short History of Builders’ Wages in Economic History     9

she came to England to gather material for her doctoral research into ‘Wages in England in the Eighteenth-century’. Given the technology and resources available to her at the time, her achievement in just a couple of years—a comprehensive list of day rates from sites throughout England—is considerable. She toured the whole country and gathered data for building craftsmen and labourers. Her London sources were Westminster Abbey, Greenwich Hospital, and the Middlesex and Surrey Sessions, and her data were taken from account books and bills. From these, she transcribed day rates for craftsmen and labourers in all the building trades in London from 1700 to 1787. Published in 1934, the book was for many decades the definitive study of eighteenth-century wages in England. The discussion within which the data were presented focused on considerations of the consumption and aspiration of eighteenth-century workers. Gilboy was interested in demand-side explanations of growth in the eighteenth century, and the idea that the factory system needs a market sufficiently developed to consume what it produces was one of the guiding themes of the text.22 Gilboy’s series display notable features that have since been associated with all discussions about eighteenth-century wages. First, her published data showed a very high degree of uniformity in the day wage figures presented.23 The day rates were virtually all the same and seemed to show that all crafts, no matter what level of skill, were paid virtually the identical rates for decades without any change. When one realises that many of the small variations in the figures are actually the product of the calculation of an average of two figures 2d. apart, this becomes even more striking. She noted that “the stability of wage rates over a considerable number of years was shown by the figures for any one place throughout the century, and consequently the lack of continuity does not invalidate the course of the rates”.24 In other words, she viewed the wage rates as so set or stable that if she did not have data for any given years she felt she could confidently interpolate the same rate as before

22Gilboy,

Wages in Eighteenth Century England, p. 239. See the discussion pp. 236–243. pp. 8, 254–292. 24Ibid., p. 250. 23Ibid.,

10     J. Z. Stephenson

or after. By interpolating or repeating data based on the previous year, Gilboy calculated what she called “a simple arithmetic average” of the figures she found for each year, and in some cases the mode.25 Given its uniformity, the method of calculating the average probably did not bias the series.26 The trends that she noted for the course of wages for labourers was a gradual increase to 1734 and then a stable position until the last decade of the century.27 Craft wages did not rise at this time but varied before 1734. She wrote about craftsmen as a cohesive group and the wage rates as something they received from a single market for their skill where a given rate prevailed, for instance: Paviours… rates increased to 3s. 4d. in 1763. The plumbers got 2s 6d from 1700 to 1707 without a break. The bricklayer’s rates rose more sharply on the whole. Although there was a sharp temporary increase to 3s. in 1704 the rate dropped back to 2s. 8d. in 1706, and then went permanently up to 3s. in 1718.28

She commented on the organization of workers and their wages. She explicitly noted that workers were employed by contractors, some of whom had long relationships with their institutional clients, but she dismissed the idea that the wages she had transcribed represented anything except the pay a worker would have received.29 She noted that the building trades were among the ‘better paid trades’.30 What is less clearly stated in the book is that Gilboy’s sources give her data a very large bias to projects designed and managed by Sir Christopher Wren. These were large, fine, stone-built, publicly funded

25Ibid. 26Ibid. Also see pp. 10–11, and pp. 18–19, 250, and note p. 13, “There can be little error in the assumption that the rates remained the same when the data are missing. The stability of the rates is evident from the Abbey Figures”. 27Ibid., p. 9. 28Ibid., p. 9 and see whole of discussion pp. 9–13. 29Ibid., pp. 16–17. 30Ibid., p. 19.

1  A Short History of Builders’ Wages in Economic History     11 40 38

wage in d. per day

36 34 32 30

Lab

28

Cra

26 24 22 20 1700 1704 1708 1712 1716 1720 1724 1728 1732 1736 1740 1744 1748 1752 1756 1760 1764 1768 1772 1776 1780 1784

18

Fig. 1.1  Gilboy’s wage series for London 1700–1787 (Source Gilboy [1934])

projects, with a high degree of specialist craft skill such as carving. Given this, it is hardly surprising that the wages she found look comparably high when set against other sources. She found a craftsman’s rate of predominantly 3s. (or 36d.) a day from 1718 to 1776, as can be seen in Fig. 1.1. Gilboy was clear that the wage series that she reported was for journeymen’s wages, yet she did not discuss levels of skill that such a categorization must imply, declaring that the issue could not be worked out through the account books themselves.31 Her appendix, however, fully listed craftsmen’s and labourers’ day rates by trade. The series showed that a prevailing rate of 36d. a day for craftsmen in 1720 declined somewhat in the late 1730s but returned to previous levels and remained broadly constant until 1778. Finding labourers at 20d. a day at the turn of the century, Gilboy reported a 24d. per day rate for them from 1736 to the end of her series in 1787.

31Ibid.,

pp. 15–17.

12     J. Z. Stephenson

Schwarz’s Sources, Methodology, and Findings Leonard Schwarz, known for his research on all aspects of work in London in the eighteenth century, published a London nominal and real wage series in 1985 under the auspices of a discussion of longrun living standards. In the context of an argument that sought to assert that wage rates in cities with international trade would have been similar, comparable, or correlated, he raised the question of representativeness of builders again, but reassured his readers that contemporary literature put them in an average position.32 This rationale is often used by historians to justify the use of builders’ rates—citing other economic historians use of them as representative and noting a contemporary source. The ‘London Tradesman’, written in the 1740s as a career guide for middle class parents, is frequently this corroborating source, but it quotes rates 15–20% lower than Gilboy for craftsmen.33 Schwarz proposed that the movement of wages in this period should be viewed as related to that of other large cities with international trade. But in presenting such a long run set of data he found, in a more marked manner than Gilboy, that living standards in London had fallen significantly over the period under question. Much of his attention and analysis is given over to the difference between the price series of Phelps Brown and Hopkins and Lindert and Williamson.34 He also rehearsed Woodward’s argument about income from allotments and other employment.35 He showed there was a rapid increase in the nominal wage after 1792. Overall, like Gilboy, Schwarz observed stability or stagnation in nominal wage rates over the long term, with no meaningful increase in day rates between 1736 and the late 1770s. From 1700 to the late 1780s, he showed an overall increase of roughly 20% for craft and labour. However, on considering the

32Schwarz,

‘The Standard of Living in the Long Run’, pp. 26, n. 14, 15. ‘The London Tradesman’. 34Schwarz, ‘The Standard of Living in London’, pp. 27–28. 35Ibid., p. 33; Woodward, ‘Wage Rates and Living Standards in Pre-industrial England’. 33Campbell,

1  A Short History of Builders’ Wages in Economic History     13 40 38

wage in d. per day

36 34 32 30

Cra

28

Labour

26 24 22 20 1800

1795

1790

1780

1785

1775

1770

1765

1760

1755

1750

1740

1745

1735

1730

1725

1720

1715

1710

1705

1700

18

Fig. 1.2  Schwarz’s craft and labour wage series for London 1700–1800 (Source Schwarz [1985])

price data, he found a 9.5% decrease in a bricklayer’s real wage from 1750 to 1760, and a 14.2% decrease from 1760 to 1770. Schwarz’s declared source was bricklayers’ and carpenters’ bills retained in the Middlesex sessions papers, and in a separate note, McCulloch’s series on Greenwich.36 He did not state whether his figures were means, medians, or modes but the text suggests that they may have been modes.37 He found that London rates differed somewhat from Greenwich ones.38 No specific bills or sites or names were given by him with the data. In fact, he presented two series; one for carpenters and one for bricklayers. He did not give the number of observations, and my own review of the MJ/SP Middlesex sessions papers he referred to found a very small number of bills with recorded wages, and no accounts.39

36McCulloch,

A Dictionary, Practical, Theoretical and Historical of Commerce and Commercial Navigation. 37Schwarz, ‘Standard of Living’, p. 26, I surmise they are modal figures from the following— which is all that is given on methodology; “Even for the years before 1820 bills could not always be found, necessitating careful judgement when filling the gaps. Figures for these years have been interpolated, using previous trends and builders’ price books as a basis for estimation and are incorporated in the series presented”. 38Ibid., p. 26. 39Ibid., n. 10, 11, 12.

14     J. Z. Stephenson

This suggests that most of the data came from Greenwich. When combined, in places they are sometimes more than 10% lower than Gilboy’s composite series, which supports the idea that they are mostly Greenwich figures. I have represented it here by simply calculating a mean average of the bricklayers’ and carpenters’ rates Schwarz gives, see Fig. 1.2.

Boulton’s Sources, Methodology, and Findings In 1996 Jeremy Boulton offered an entirely new series for craftsmen’s and labourer’s wages for the seventeenth century.40 Boulton noted the existing paucity of knowledge of the topic and, in contrast to the previous authors, he was wholly explicit about methodology. His sources (bills again) were from a wider range of locations and organisations than either Gilboy or Schwarz had used. Boulton drew on material from the Carpenters’, Stationers’, Tallowchandlers’, Drapers’, Grocers’, Bakers’, and Cooper’s Companies’ warden’s accounts, as well as from Charterhouse, St Thomas’s Hospital, and St Martin in the Fields. The greatest proportion of his data was from Middle Temple. Boulton is the only scholar who acknowledged the significant variation in rates recorded. He included scatter diagrams to show that there was a wide range of day wages found in his sources.41 His methodological approach for dealing with this variance, or range of rates was to use the modal daily wage rate for each year in the series: The approach adopted here has been to eliminate, as far as possible, subjective judgment as to the prevailing rate and for every year, where possible, the modal rate has been selected as the wage prevailing in any one year. Such a treatment has the advantage that stability is not built into the system (the result of selecting a ‘prevailing’ rate) but is clearly still at the mercy of the distribution of rates encountered in the manuscript material, of skill and status differences within crafts, and of changing rates between summer and winter.42 40Boulton,

‘Wage Labour in Seventeenth-century London’, pp. 268–290. pp. 276, 277. 42Ibid., p. 275. 41Ibid.,

1  A Short History of Builders’ Wages in Economic History     15

Boulton’s data set showed a narrowing of the range of day rates for both labour and craft in the 1700s and 1710s. His series also shows a narrowing of the skill premium (the difference between labourers and craftsmen wages), between the 1690s and the mid-1710s. Rates for those labelled as apprentices, journeymen, workmen, lads, and boys were excluded from the data set. For this reason, the Boulton craft series could be seen as one of master’s rates. Although his raw data had substantial variations within it, when presented as a modal average the wage rates found were not much different from that implied by Gilboy and Schwarz (or shown by Phelps Brown Hopkins). Boulton’s diagrams show a clear trend of steady increase in the labourers’ day rates from 1560 through 1670, and a slower increase thereafter, which looks to have slightly flattened after 1700. For craftsmen, the rise is perhaps less pronounced, and there is possibly no clear trend after 1670. In the 1640s the range of day rates was from 14d. to 36d. per day for labourers, and from 15d. to 35d. for craftsmen. In the 1700s the range was from 12d. to 26d. for labourers and 30d. to 36d. for craftsmen. The use of modes in the final series obscured the key finding of the range of wages. For instance, the modal rate given for most of the latter years of the 1660s is 36d. for craftsmen, yet a mean average of the discernible figures in the plot would be 28d. Boulton cautioned against using the series as a proxy for income for a variety of reasons. He asserted that masters earned not only their day rate, but their apprentice’s pay and a mark-up on materials.43 He cited a carpenter’s bill that showed a mark-up being taken on labourers’ pay.44 He also speculated that perquisites increased the take-home income of some men.45 Boulton warned that institutional rates might be lower than the bills for private work that he collected.46 Despite these

43Ibid.,

p. 271. p. 275. 45Ibid., p. 271, n. 23. 46Ibid., p. 274. 44Ibid.,

40 38 36 34 32 30 28 26 24 22 20 18 16

Lab

1721

1718

1715

1712

1709

1706

1703

1700

1697

1694

1691

1688

1685

1682

1679

1676

1673

1670

1667

1664

Cra

1661

wage in d. per day

16     J. Z. Stephenson

Fig. 1.3  Boulton’s wage series for London 1661–1721 (Source Boulton [1996])

40 38

d. per day.

36 Gilboy

34

Boulton

32

Schwarz

30

1661 1667 1673 1679 1685 1691 1697 1703 1709 1715 1721 1727 1733 1739 1745 1751 1757 1763 1769 1775 1781 1787

28

Fig. 1.4  Comparison of three series, craft d. per day (Sources Boulton [1996], Gilboy [1934], Schwarz [1985] as per text)

comments and cautions, it is only the day rates that Boulton recorded that have made their way into the real wage calculations of subsequent historians. His rates are shown in Fig. 1.3. The rates of all three series are compared in Figs. 1.4 and 1.5.

1  A Short History of Builders’ Wages in Economic History     17 28

wage in d. per day.

26 24 Gilboy

22

Boulton Schwarz

20 18

1661 1667 1673 1679 1685 1691 1697 1703 1709 1715 1721 1727 1733 1739 1745 1751 1757 1763 1769 1775 1781 1787

16

Fig. 1.5  Comparison of three series, labourers d. per day (Sources Boulton [1996], Gilboy [1934], Schwarz [1985] as per text)

Associated Series London in the Sixteenth Century The wages used for London for the period prior to the start of Boulton’s data series come largely from Rappaport’s real wage and price series for the period 1490–1609.47 Although not directly relevant to the period under review in this study, as his data are used alongside the others the issue of comparability should be discussed. Rappaport’s builder’s wage data was extracted from London livery companies’ bills and books. In terms of nominal wages, Rappaport found a doubling of day rates from the mid-1540s to the mid-1570s and another 25% increase at the end of the century to 1609. As prices increased faster this translated into a decline in real wages over the sixteenth century of over 30%. The decrease was less than that found by Phelps Brown Hopkins, and Rappaport maintained that living standards did not decline by anything like as much as a third due to increased family income, and effective

47Rappaport,

Worlds Within Worlds, pp. 123–161.

18     J. Z. Stephenson

substitution of goods.48 Rappaport found that some men’s wages were supplemented by food and drink, but it was not frequent enough for his series to include it.49 Rappaport also found an oversupply of skill and consequent unemployment in London throughout the Tudor period, and of course, his discussion of earning is in the context of the profound structural change of the period.50 The failure of builders’ income and living standards to keep up with price inflation in the sixteenth century is a phenomenon associated with changes in social structure in another important study of the same period for the Low Countries. Blonde and Hanus used tax and household records from s-Hertogenbosch to show the income and relative status of building craftsmen declined relative to other groups.51 Despite the obvious implications that builders are not representative of the average worker in the long run, their strong substantive evidence has not altered the use of building craftsmen for international comparisons of living standards. Rappaport’s methodological approach and analysis differed somewhat from the later series described above. Noting that variation in wages presented a methodological challenge, in his series “with the exception of a few years … when fractional wages were used in the series to express what clearly was a gradual increase from one level of wages to the next, median wages were used in most years.” Since the highest and lowest wages “differ by little more than pence” an arithmetic mean would probably not have been very different.52 Most significantly, Rappaport’s sources were for skilled and semi-skilled craftsmen, not unskilled men. His series included assistants and servants and only ‘sometimes’ labourers.53 He noted, “It would be incorrect to infer from

48Ibid.,

pp. 123–161. Rappaport only provided nominal data in index form. p. 129. 50Ibid., p. 110. 51Blonde and Hanus, ‘Beyond Building Craftsmen’, pp. 179–207. 52Rappaport, Worlds Within Worlds, p. 129. 53Ibid., p. 128. 49Ibid.,

1  A Short History of Builders’ Wages in Economic History     19

the accounts that semi-skilled and unskilled workers were paid the same wage”.54 It is, therefore, highly likely that Rappaport’s median wages for semi-skilled workers are higher than wages would have been for ‘labourers’, and in fact his series is not of unskilled men’s pay at all.

Beyond London: Northern England Outside of London, the widest ranging study of craftsmen and labouring men in England is Donald Woodward’s Men at Work. It examines the work of building craftsmen and labourers in Northern English towns over the 300 years to 1750. The bulk of the data he produced was from the sixteenth and seventeenth centuries. Woodward’s work on builders is the significant product of a lifetime’s scholarship, and any attempt to summarize it in a section such as this would miss the finer points. However, for the purposes of this study, two of Woodward’s consistently argued themes should be treated as highly significant. First, Woodward in all his work, repeatedly challenged the idea of building workers as representative of early modern ‘wage labour’. He showed that many had subsistence from outside the construction industry, and suggested that they gained income from selling the services of others. He showed that many workmen owned tools and materials. He highlighted the seasonality of building work. Whilst his data have been faithfully aggregated by other historians who use it in wage series, his cautions seem to have been largely ignored. Secondly, Woodward showed that labourers in Northern towns were a standalone group with their own work, and their own organization and teams, as well as being assistant and auxiliary to craftsmen on building sites.55 The implications of this categorization challenge the notion of labourers as unskilled, but also the practice of using craftsmen’s and

54Ibid.,

p. 128. Men at Work, p. 3, and Introduction generally. Chapter 4 describes the work of labourers outside of the building industry, pp. 116–167. 55Woodward,

20     J. Z. Stephenson

labourer’s wages to calculate a ‘skill premium’. If the market for ­labourers and the market for craftsmen are sometimes different markets, then the premium for skill will not be directly attributable to the market in which they both operate. Again, this evidence has not been taken up by ­historians who compile and use real wages series. Woodward showed large variations in wages and day rates in the towns he studied. Rates paid in larger towns were up to 50% higher than in smaller ones, yet as he acknowledged, the question of how skilled the relative workers were could not be discerned from the sources. Woodward’s day rates for the fifteenth to seventeenth centuries have been aggregated into current long-run data sets.56

South East England, Real and Silver Wages In 1955, compiling a wage series for south-east England Henry Phelps Brown and Sheila Hopkins described their use of sources as follows: Until 1700 we have depended on Thorold Rogers. He drew chiefly on the college and estate accounts in the muniment rooms of some of the Oxford and Cambridge colleges, and the farm bailiffs’ rolls and monastic accounts in the Public Record Office. Of his builders’ wages, some 40 or 50 per cent come from Oxford itself, down to 1620; the other main region of origin is the southern counties, of which fourteen provide entries, the most represented being Bucks, Hants, Sussex, and Kent. Cambridge contributes substantially throughout, and when the entries from Oxford fall off after 1620 it is Cambridge, and Eton, that take up the running. For the most part, the entries for craftsmen come at the rate of fifteen or more a year; but except for 1580–1620, those for building labourers are much fewer-more like three a year. There is a great falling off in entries of all kinds after 1660.57

56At https://www.nuffield.ox.ac.uk/people/sites/allen-research-pages/; http://www.iisg.nl/hpw/ data.php#europe. 57Phelps Brown and Hopkins, ‘Seven Centuries of Building Wages’, pp. 196–197.

1  A Short History of Builders’ Wages in Economic History     21

From 1700 we could look ahead to the 1890s, when the wage reports of the Board of Trade begin. Our guides through the years between are Dr. Elizabeth Gilboy and Sir Arthur Bowley: but Dr. Gilboy’s entries from Oxford are not continuous, and fade out in the 1770s, and in the nineteenth century Bowley’s best series is for London. Dr. Gilboy, however, has a long run for Maidstone, and we know that the Maidstone rates were the same as the Oxford rates in the first half of the eighteenth century and again in the 1890s: so we took the Maidstone series as representative of Oxford from the 1730s till it ends with the century.58

In other words, Phelps Brown and Hopkins’ nominal wage rates for south-east England in the eighteenth century come from an interpolated, rather haphazard combination of Maidstone and Oxford day rates recorded by Elizabeth Gilboy. Phelps Brown and Hopkins’ were explicit about their choice of data because they knew they were not ideal for the purpose intended. Gilboy placed Maidstone in the ‘London metropolitan area’, but Oxford well outside it. It should be noted that the Maidstone wage figures were taken from the ‘Kent session rolls’ and were 10% higher than those recorded for Oxford for many of the years.59 A year after publishing the long-run wage series, Henry Phelps Brown and Sheila Hopkins created a basket of consumable goods and ‘compared’ it with the found wages. Whilst the authors’ explicit caution that they had ‘not tried to construct a measure of real wages in the modern sense’ has been repeatedly overlooked and the series is commonly used and referred to as one of real wages, the text remains a seminal essay on the difficulties of and strategies for constructing an effective basket of substituted consumption of goods over time.60 Neither of their series utilise data from London, as they and other researchers have frequently avoided London sources as unrepresentative of the rest of the country, although the series is treated as a long run wage and cost of living index which can universally apply to south-east England.

58Ibid.,

p. 200. Wages in Eighteenth Century England, p. 55 and Appendix II, Tables V and VII. 60Phelps Brown and Hopkins, ‘Seven Centuries of the Prices of Consumables’, p. 296. 59Gilboy,

22     J. Z. Stephenson

‘Nominal’ Sources and Their Effect on ‘Real’ Wages There are a number of obvious biases to the three original, nominal, series for London described above, and they will probably have inclined the data upwards. Firstly, there is a consistent reliance on large, highly skilled projects built by Crown and City, and no representation of private housebuilding.61 Gilboy’s data was biased to prestigious projects with high level skill inputs, whilst Boulton’s data might have captured more everyday kinds of workmanship on maintenance. Secondly, there is a lack of consistency of skill levels between the series. It is hard to believe that men fitting expensive ashlar on Westminster Abbey were paid the same as those doing routine maintenance work for the Carpenters’ company, yet we have little information about relative skill levels. Boulton recorded only the day rates of craftsmen and he excluded journeymen. Gilboy understood herself to have recorded journeymen. Schwarz did not elucidate on the matter of skill level but just gave two categories, craft and labour. In other words, it’s possible that the underlying skill level for craftsmen excludes ‘journeymen’ up to 1721 and assumes them to be the subject from then on. As mentioned, there is a particularly glaring issue for ‘labourers’. Rappaport recorded the wages of those he described as being semiskilled, assisting craftsmen. But Boulton and Gilboy both recorded general labourers and labourers assisting craftsmen. Schwarz only recorded the wages of bricklayers’ labourers, so his figures might be comparable to Rappaport’s data. It is clear that the level of skill of those referred to under the category of ‘labourers’ is highly variable but includes many semi-skilled people. Yet those who derive real wage series from the data always use the term ‘unskilled’.62 It seems likely that existing series overstate unskilled men’s wages.

61Data gathered for areas outside London by Gilboy frequently included labourers rates for municipal work such as road maintenance, but the figures are far more sparse, and the contractual basis for the work unclear. 62Lindert and Williamson, ‘English Workers’ Living Standards during the Industrial Revolution’, p. 17; Allen, ‘Why the Industrial Revolution Was British’, p. 363; Allen, The British Industrial Revolution, pp. 44–45.

1  A Short History of Builders’ Wages in Economic History     23

Thirdly, one of the concerns that Phelps Brown and Hopkins had about Rogers’s figures was that he had always selected the highest or summer rates. However, none of these three series give seasonal variations of day rate.63 If pay was seasonally adjusted the current series are likely to represent pay higher than that received. Fourthly, as will become apparent, recording only ‘day rates’ biases the series to particular kinds of work. Although economic historians usually assume that the ‘day rate’ was something that could be earned all year—or at least 250 days of it—none of the series have ever provided any information about how much work, or how many days, men had. These biases matter. As used currently, a builder’s income is taken to be representative of an average—skilled or unskilled—wage, and so, to indicate wage earning more generally. If builder’s incomes are not representative of this, then they may have given us a misleading picture of the living standards of working Londoners in the critical period before industrialization. This matters, because living standards are thought to be an indicator of the relative advancement, or development, of an economy.64 Our understanding of the relative advancement of London compared to other European capitals and cities before 1800 rests on comparisons of nominal and real wages. The most comprehensive current data-set of prices and wages for London, compiled by Robert C. Allen, is an important contribution to long-run economic history.65 Although it utilises records from a variety of sources for prices, for wages it is wholly reliant on the nominal

63Phelps

Brown and Hopkins, ‘Seven Centuries of Building Wages’, pp. 196–197. Vries, ‘An Inquiry into the Behaviour of Wages in the Dutch Republic’, pp. 79–93; Van der Wee, ‘Prices and Wages as Development Variables’, pp. 58–78 (both later compiled in J. Jan de Vries, and Woude, The First Modern Economy, pp. 607–664); Hanauer, Études économiques sur l’Alsace ancienne et moderne (denrées et salaires); Mocarelli, ‘Wages and the Labour Market in the Building Trade in 18th Century Milan’, pp. 61–82. 65https://www.nuffield.ox.ac.uk/people/sites/allen-research-pages/. 64De

24     J. Z. Stephenson 41

wage in d. per day

39 37 35 Allen

33

Gilboy

31

27

1650 1655 1660 1665 1670 1675 1680 1685 1690 1695 1700 1705 1710 1715 1720 1725 1730 1735 1740 1745 1750 1755 1760 1765 1770 1775 1780 1785

29

Fig. 1.6  Allen and Gilboy, London craftsman’s d. per day (Sources Allen [2001, 2013], Gilboy [1934])

wages described above.66 There are slight differences between it and the original series. The stated averages are always on the lower end of Boulton’s figures after 1650. The large variation Boulton found in the 1690s is not shown.67 Turning points at the beginning and end of the 1660s, 1736, 1773, and 1794 emerge as significant in ways that are not as pronounced in the archival series. For labourers, the figures are more similar to the original series, and follow the more gradual trends Gilboy and Schwarz show (although again Boulton’s 1690s rates are smoothed out). Figures 1.6, 1.7, 1.8, 1.9, 1.10, 1.11, show how the series described above relate to the long-run data currently compiled for London as a part of an international comparison of wages.

66Without making any changes for the differences in skill and sources. The sources and treatments are similar to that of Clark, ‘The Long March of History’, pp. 33–38, and ‘The condition of the working class’, p. 1334, although Clark makes an allowance of 10% for mark up before 1800. 67For the high price of labour in the 1690s see Hatcher, ‘Labour, Leisure and Economic Thought Before the Nineteenth Century’, pp. 64–115.

1  A Short History of Builders’ Wages in Economic History     25 41

wage in d. per day

39 37 35 33

Allen

31

Schwarz

29 1650 1656 1662 1668 1674 1680 1686 1692 1698 1704 1710 1716 1722 1728 1734 1740 1746 1752 1758 1764 1770 1776 1782

27

Fig. 1.7  Allen and Schwarz, London craftsman’s d. per day (Sources Allen [2001, 2013], Schwarz [1985])

37

wage in d per day

36 35 34 33 Allen

32

Boulton

31 30 1650 1653 1656 1659 1662 1665 1668 1671 1674 1677 1680 1683 1686 1689 1692 1695 1698 1701 1704 1707 1710 1713 1716 1719 1722

29

Fig. 1.8  Allen and Boulton, London craftsman’s d. per day (Sources Allen [2001, 2013], Boulton [1996])

26     J. Z. Stephenson 27 26

wage in d. per day

25 24 23 22 21

Gilboy

20

Allen Lab

19 1650 1656 1662 1668 1674 1680 1686 1692 1698 1704 1710 1716 1722 1728 1734 1740 1746 1752 1758 1764 1770 1776 1782

18

Fig. 1.9  Allen and Gilboy, London labourer’s d. per day (Sources Allen [2001, 2013], Gilboy [1934])

28 27

wage in d. per day

26 25 24 23 22

Schwarz

21

Allen

20 19 1650 1656 1662 1668 1674 1680 1686 1692 1698 1704 1710 1716 1722 1728 1734 1740 1746 1752 1758 1764 1770 1776 1782

18

Fig. 1.10  Allen and Schwarz, London labourer’s d. per day (Sources Allen [2001, 2013], Schwarz [1985])

1  A Short History of Builders’ Wages in Economic History     27 Ϯϳ Ϯϲ ǁĂŐĞŝŶĚ͘ƉĞƌĚĂLJ

Ϯϱ Ϯϰ Ϯϯ ϮϮ Ϯϭ

ŽƵůƚŽŶ

ϮϬ

ůůĞŶ

ϭϵ ϭϴ ϭϲϱϬ ϭϲϱϯ ϭϲϱϲ ϭϲϱϵ ϭϲϲϮ ϭϲϲϱ ϭϲϲϴ ϭϲϳϭ ϭϲϳϰ ϭϲϳϳ ϭϲϴϬ ϭϲϴϯ ϭϲϴϲ ϭϲϴϵ ϭϲϵϮ ϭϲϵϱ ϭϲϵϴ ϭϳϬϭ ϭϳϬϰ ϭϳϬϳ ϭϳϭϬ ϭϳϭϯ ϭϳϭϲ ϭϳϭϵ ϭϳϮϮ

ϭϳ

Fig. 1.11  Allen and Boulton, London labourer’s d. per day (Sources Allen [2001, 2013], Boulton [1996])

Comparisons with Other Regions and Cities If wages and prices are, as Van Zanden put it, the ‘DNA’ of the economies of the past, the core purpose of gathering nominal wage and price figures is to compare economies and markets across space and time.68 The challenges to effective comparison are much more complex than is often realised, however. It has frequently been asserted that a varying employment year, access to poor relief, and non-pecuniary elements of pay are not reflected in day wages, and these factors were driven by regional and local customs and influences. From a consumption point of view, housing, access to land, the substitution of goods, household composition, and other welfare priorities all compromise the accuracy of ‘real wages’ as a proxy for living standards.69 Indeed, the composition of the basket of goods and its relative welfare index are currently still hotly debated; most recently Humphries has shown the ‘male breadwinner’ subsistence calculations to be woefully inadequate for the working women and children of the Industrial Revolution.70 It is only in 2015 that long run GDP data 68Van

Zanden, ‘The Skill Premium and the Great Divergence’, p. 121. Annals of the Labouring Poor, pp. 3–6; Schwarz, ‘The Formation of the Wage: Some Problems’. 70Humphries, ‘The Lure of Aggregates and the Pitfalls of the Patriarchal Perspective’, pp. 693–710. 69Snell,

28     J. Z. Stephenson

based on output started to inform great divergence debates, but even if real wage calculations are not the only indicator of economic performance any more, the analysis of comparative living standards remains crucial.71 They remain the most empirical and robust way of measuring welfare across time and space. While substitution of consumption goods and prices over time presents one problem, the substitution across regions is quite another, given the difference in relative prices in each region. Aside from the capacity to store and deal with large data sets which computing power has conferred, the methodology for compiling and comparing wage and real wage series to use for macroeconomic calculations has not changed dramatically since Phelps Brown and Hopkins. Subsequent contributions for England have mostly improved the price data, which has confirmed that there was a decline in living standards for builders.72 Purchasing power comparisons for workers thousands of miles apart where factor prices were dramatically different have proven both methodologically challenging, and controversial right up to the current day. Current practice is to compare welfare ratios, but many comparisons rely on the development and widespread use of the ‘silver’ wage, created by converting nominal wages of any currency to an international standard based on long-run silver prices, with which to calculate comparable purchasing power of grain and other goods.73 The silver wage allows a direct comparison of the purchasing power of traded goods in different economies, and the results can differ, sometimes drastically, from other evaluations. For instance, it has been shown that Indian workers in the eighteenth century had a relatively high grain wage, easily comparable to that of a London worker, due to agricultural prices, but the advantage in traded or market goods in silver terms was European.74 71Broadberry et al., British Economic Growth 1270–1870, is the first output based (as opposed to real wage based) approach to long run economic growth. Also see Broadberry, Custodis and Gupta, ‘India and the Great Divergence’, pp. 58–75. 72See Schwarz’s discussion of the relative price indices in ‘The Standard of Living in London’, pp. 26–28; Feinstein, ‘Pessimism Perpetuated’, pp. 625–658, Allen’s basket is the most advanced and takes into account all this work on price levels, https://www.nuffield.ox.ac.uk/people/sites/ allen-research-pages/; http://www.iisg.nl/hpw/data.php#europe. 73Van Zanden, ‘Wages and the Standard of Living in Europe’, pp. 175–197. 74Parthasarathai, ‘Rethinking Wages and Competitiveness in the Eighteenth Century’, pp. 82–85; Broadberry and Gupta, ‘The Early Modern Great Divergence’, pp. 2–31.

1  A Short History of Builders’ Wages in Economic History     29

This relative values problem can only be viewed to have been resolved by the silver wage method if one accepts that the silver valuations used reflect actual purchasing power for consumption goods in each place at a given time, and it is far from clear that they do. Silver was a commodity with its own fluctuations in price and availability in the early modern world.75 The basic requirements of the nominal series—representativeness of the ‘average’ or known category of skill, drawn from similar samples or sources and levels of skill in each region—remain. In this respect, silver wages do nothing to address the substitution and index problems of prices, nor the biases of nominal wages.76 For this reason, and for ease of reference, the wages and other prices dealt with in this book are presented as they were paid—in pounds (£), shillings (s.) and (d.). This book does not offer new material for other European cities but, even without new research, some observations can be made. Craftsmen’s wages for Amsterdam for the period which came from the de Vries and Van der Woude’s 1978 studies are mostly for building craftsmen working at wharves and shipyards all over the Dutch Republic, unlike Gilboy’s subjects, who were working on carved stone on important architectural monuments right in the centre of the nation’s biggest city.77 The labourer’s equivalent in ‘Amsterdam’ data sets are for not for assistants but ‘onge-school’ (unskilled). Related concerns for comparability apply for Paris where Baulant described how in Paris, just as in London, contractors charged out one uniform rate for day work, but took a margin of 25–30% on wages through the seventeenth century on craftsmen’s and labourer’s wages, and also that a full years work was not available to all men.78 However, the ‘wages’ published are rebased daily equivalents, the annual amount

75See

Mayhew, ‘Money in England from the Middle Ages to the Nineteenth Century’. The Great Divergence, pp. 84–100. 77De Vries, ‘An Inquiry into the Behaviour of Wages in the Dutch Republic’, pp. 79–93; Van der Wee, ‘Prices and Wages as Development Variables’, pp. 58–78. 78Baulant Micheline, Le salaire des ouvriers du bâtiment à Paris, de 1400 à 1726, Annales. Économies, Sociétés, Civilisations. 26e année, N. 2, 1971. pp. 463–483, especially 464, 470. 76Pomeranz,

30     J. Z. Stephenson

received divided by days worked, and so already take into account such a contracting system. Mocarelli has shown the day rates for Milan or Northern Italian city wages have also been understated. In the Italian case, the contract system allowed extra payments to men to retain them, which are not included in the ‘day rate’.79

Market and Business Organisation Economic historians have frequently questioned whether builders’ wage rates, and their fluctuations or changes, have been determined by supply and demand. Their considerations have usually focused on the extent of the labour market, non-competing groups, custom wages and optimist and pessimist narratives.80 In the long run, most discussions, with the exception of those of Woodward and Boulton, have excluded contemporary scholarship on the organisation of the building industry.81 More recently, they have avoided important developments in labour and construction economics and business history. Economic historians have tended towards interpreting building craftsmen as both ‘artisans’ and wage earners, and have assumed, like Arthur Bowley, that they were directly paid by employers. The long run evidence is a bit more complex. In the medieval period some building craft skills, particularly those of stone masons, were sufficiently rare at the time of requirement for large projects to necessitate impressment by the Crown. Whether pressed or not, building craftsmen were frequently hired as gangs or individuals directly by the clerk-of-works or surveyors until the fifteenth century.82 Due to demand, craftsmen in carpentry, plastering, and woodwork tended mostly towards to local needs,

79Mocarelli,

‘Wages and the Labour Market in the Building Trade in 18th Century Milan’, pp. 61–82. 80Boulton, ‘Wage Labour’, pp. 283–284; Lindert and Williamson, ‘The Structure of Pay in Britain’, pp. 1–54; Schwarz, ‘Custom Wage and Workload’, pp. 143–176. 81Woodward, ‘The Determination of Wage Rates in the Early Modern North of England’, p. 23. 82Knoop and Jones, ‘The Impressment of Masons in the Middle Ages’, pp. 57–67, see note 4 page 62 on contracting masons.

1  A Short History of Builders’ Wages in Economic History     31

and rural and small-town carpenters would probably have had access to skills to build and finish houses. More highly skilled work was, by its very nature, less localised and providers of it travelled.83 Through the medieval period, masons were mobile specialists of sorts, and shared many skills with bricklayers. Knoop and Jones’s definitive reviews of medieval and early modern stone masonry assert that, from the mid seventeenth century, the direct method of employment was rare, and they identified seventeenth century mason contractors as ‘firms’.84 They also showed that by this time the trade guild in London was used as networking trade club for the well-to-do-businessman, rather than a body which regulated employment, the quality of output, materials or prices. The level of intermeddling among the contractors at St Paul’s and other major sites in this study bears this out. Like Woodward, Knoop and Jones found that the system of career development from apprenticeship to master did not operate as mostly understood and that most men in the masonry and other trades remained journeymen most of their careers.85 Since Knoop and Jones a number of architectural historians and researchers have shown the organisation of building industry contracting to have been highly capitalised and complex. Howard Colvin wrote of the importance of credit and contracts to Crown projects, and of the decline of the artisan and rise of the architect.86 John Summerson noted credit and subcontracting in the speculative development of the West End.87 Elizabeth McKellar’s landmark Birth of Modern London dealt a blow to the artisan craftsman myth by exposing practices of leveraged finance, chains of subcontracts, and pre-fabrication of crafts and architectural features in the rebuilding of the City after the Great Fire.88 By 83Saltmarsh,

Kings College Chapel: A History and Commentary, Chap. 7, ‘The Input of Labour’. and Jones, The London Mason, p. 47. 85Knoop and Jones, ‘Masons and Apprenticeship’, p. 351; Woodward, Men at Work, pp. 64–72. 86Colvin, A Biographical Dictionary of English Architects, pp. 4–11; History of the King’s Works, Vol. V, pp. 42–46. 87Summerson, Georgian London, pp. 50–64. 88McKellar, The Birth of Modern London. 84Knoop

32     J. Z. Stephenson

contrast, James Ayres’s painstaking work demonstrates the high and varied level of craft skill at work on the built environment through the Georgian period, and Peter Guillery has contributed an in depth study of London’s eighteenth century housing.89 James Campbell has shown that the finances of the large carpentry, stonemason and bricklaying businesses involved with the rebuilding relied on significant margins from both materials and labour.90 By contrast, whilst the contracting methods and organisation of the industry are described by some architectural historians as remaining in place until the early nineteenth century,91 Clarke’s important thesis used a Marxist framework to firmly identify builders as at the vanguard of the development of wage labour in a new capitalist mode of production that broke with artisan traditions from the 1780s onwards.92 Such conditions would suggest that there were changes over and above the cycle of supply and demand in the market for labour in the building industry throughout this period. However, wage historians have generally avoided further examination of all of them and maintained a view of labour markets which was, for many years, also at odds with much social and economic historiography of employer wage setting and early modern labour regulation.93 Generally, the assumptions that have underpinned the collection of wage rates have been that the rates represent expansion and contraction, or supply and demand, in the labour market.94 Such interpretations rely on a view of labour market that is to a greater degree characterized by competition between employers for workers, where there is a symmetry of information,

89Ayres,

Building the Georgian City; Guillery, The Small House. ‘Seventeenth-Century Bricklayers’ Contracts: Wren’s City Churches’; ‘The Finances of the Carpenter in England 1660–1710’, pp. 313–346; ‘Building a Fortune: The Finances of the Stonemasons Working on the Rebuilding of St Paul’s Cathedral 1675–1720’. 91Satoh, Building in Britain: The Origins of a Modern Industry, Ch 1; Cooney, ‘The Origins of the Victorian Master Builders’, pp. 167–176. 92Clarke, Building Capitalism, pp. 53–79, 241–244. 93See Wallis, ‘Labour Markets and Training’, pp. 178–210; Kelsall, Wage Regulation Under the Statute of Artificers. 94See Allen, ‘The High Wage Economy and the Industrial Revolution: A Restatement’, p. 14. 90Campbell,

1  A Short History of Builders’ Wages in Economic History     33

and relatively costless transacting. This neoclassical view of supply and demand has been overturned by those who research labour markets in capitalist or market modes of production. It is widely acknowledged that the law of one wage does not hold.95 In fact, we know that wage dispersion is found even in markets where skill is homogeneous. The price at which labour is exchanged in markets is determined by substantial frictions and search costs which affect outcomes in a way not predicted entirely by the laws of supply and demand.96 Empirical and theoretical studies show that wages for the unskilled may tend towards homogeneity, but wages for more skilled people— at the same level of skill—are usually more dispersed, due to the idiosyncrasies of bargaining and productivity.97 The builder’s wage series described in this chapter present characteristics of wage homogeneity and uniform bargaining at odds with the way we know labour markets usually behave. A finding such as Gilboy’s, that skilled men carrying out different tasks were all paid exactly the same by the day should infer caution or further analysis. Of course, if all men paid the uniform rate were not equally skilled, and it’s very unlikely they were, the assumptions of the law of one wage would be utterly undermined. Most historians have mentioned something about ‘custom wages’ when this matter has been raised, but this does nothing to solve the problem. In fact, using custom to explain the phenomenon implies that there was not a ‘market’ for labour, and so undermines the validity or meaning of a wage series. The persistence of uniform nominal rates over time—decades— without any deviation for new projects, lay-offs, cycles, or change at all would be remarkable in any organisation or labour market. It is odd, or

95Stigler,

‘The Economics of Information’, pp. 213–225; ‘Information in the Labour Market’, pp. 94–105. 96Phelps, Microeconomic Foundations of Employment and Inflation Theory, also most notably Burdett, ‘A Theory of Employee Job Search and Quit Rates’, pp. 212–220; Burdett and Mortensen, ‘Labor Supply Under Uncertainty’. For a fuller accessible description of these developments, see http://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/2010/pissarides-lecture.pdf, and Manning, ‘Monopsony in Motion’, pp. 1–20. 97Burdett and Mortensen, ‘Wage Differentials, Employer Size, and Unemployment’; Manning, Monopsony in Motion, pp. 12, 132–134.

34     J. Z. Stephenson

at least questionable, especially because this is the case not just for one level of skill or class of workers, but for the skilled and unskilled figures in the series for long eighteenth-century London. It would be expected that, given the premium for skill, more workers would want to train, increasing the supply of skilled workers and affecting the supply of unskilled workers. This would vary the premium (and indeed the premium does narrow over time, but much slower than would be expected). There is, of course, a much better explanation. As the evidence in this book will show, the ‘wages’ collected and published by previous historians are mostly not wages at all. The figures meticulously transcribed by Gilboy, Schwarz, Boulton, and others for the middle of the seventeenth century onwards were the rates at which large building firms charged out workers of various skill for certain construction services to large clients. They represent just one type of construction service. What men were actually paid for their work was a completely separate matter, which the contractor or businessman alone was responsible for. As this book will show, the organisation of the building trades in the late seventeenth and early eighteenth centuries was based on a system of contracting which managed risk, finance and production according to their relative costs. Large client institutions contracted out both new and maintenance construction work, and the responsibility for almost all labour relations rested with their contractors. The nominal wage series for eighteenth century London, on which so many real wage and living standards calculations rely, are no such thing. They are series of prices for services, just one part of complex contracts that managed and mitigated the risks of new and costly construction and engineering projects in an ambitious architectural age. To understand the nature of such a misapprehension, its implications, and its opportunities, we need to understand the business of construction in Restoration England.

2 The Market for Building

From all accounts, the period from after the Restoration to the 1780s was one of significantly high investment, both public and private, in the built environment in London. ‘Towards the end of the seventeenth century London become the largest city in Europe.’1 The population rose from 200,000 in 1600 to nearly half a million by 1670; by 1700 it was 575,000; by 1750, 675,000; and almost a million by 1801.2 Such massive population growth and associated consumption made London an ‘engine for growth’, and our knowledge of the city’s infrastructure and housing innovations in the period certainly conveys and confirms that conclusion.3 This was the period in which London’s two centres—city and port, and Westminster—were ‘joined together’.4 It was also a period of huge architectural and engineering ambition, with many landmarks planned and constructed such as the first new bridge over the Thames for half

1Wrigley,

‘A Simple Model of London’s Importance’, p. 44. ‘Urban Growth and Agricultural Change’, p. 686. 3Broadberry et al., British Economic Growth, pp. 271–272. 4Schwarz, London in the Age of Industrialisation, p. 7. 2Wrigley,

© The Author(s) 2020 J. Z. Stephenson, Contracts and Pay, Palgrave Studies in Economic History, https://doi.org/10.1007/978-3-319-57508-7_2

35

36     J. Z. Stephenson

a millennium. London experienced population, economic, and infrastructure growth throughout the seventeenth and eighteenth centuries, but the picture is dominated by one seismic event: the Great Fire of London. It completely destroyed the old city but, it seems, most of the reconstruction had been undertaken and completed by the mid1670s, suggesting that not only had London the necessary resources of capital, materials, and labour, but that its institutions and organizations ­managed the process in an effective and productive way. In fact, in the mid-twentieth century, as London lay partially ruined after the Blitz, and rebuilding was slow and disorganized, historians looked enviably at the seventeenth-century rebuilding programme.5 As I will discuss, any efficiency of the 1670s was probably possible because the building industry was already busily deployed across the city, and the requirement was an increase in output, rather than a crisis to be managed from a low base. The question of the state, or size, of the market for construction services is important because it has been traditionally understood that demand for building will affect wages for building workers. If the labour supply could not meet the requirement of ongoing projects, then employers that needed certain types of skill might have raised the rate at which they were willing to hire it. In a stagnant or declining market for building, we would expect less work at the usual rate, or declining wages for skilled and unskilled men, depending on how many men were looking for work, their ease in finding it, and the other options open to them. In the following three chapters, I give details of how the market for construction was organised as a background to the labour market and worker’s pay that is explored in the latter chapters. In this chapter, I look at the size and significance of the market for construction services in Restoration London and the sites that provide wage and price information. I highlight their relative position in the market to give an indication of how representative they were for the market for construction services as a whole. In the next chapter, I examine contractors—the key 5See

the preface and introduction to Reddaway, The Rebuilding of London after the Great Fire.

2  The Market for Building     37

players in the organisation of the market and how they and clients dealt with each other and assigned work. In Chap. 4, I look at the way contractors organized their firms, supply chains, and production.

Construction Market Segmentation There are mostly two types of work for which eighteenth-century building records survive: new projects and maintenance. The words ‘extraordinary’ and ‘ordinary’, turn up in many records and accounts, and are reflective of accounting procedures in the Treasury, not of architectural practice, but they correspond broadly to our understanding of ‘new building’ and ‘maintenance’. Work that was referred to as ‘extraordinary’ for Crown and City, was large new building while work accounted for as ‘ordinary’ work involved expected or routine maintenance or repair. By way of example, the planned costs of maintaining, and possibly improving, existing buildings and structures at Whitehall would have been ‘ordinary work’; the construction of a new building, the Banqueting Hall, was ‘extraordinary’, because the budget was not one that was regularly accounted for. These terms may help one navigate institutional records, but there was a wider picture beyond these types of accounts. In trying to evaluate the relative importance of different sectors of the building market, we can surmise that in the period under review, London had, broadly, seven types of building work being carried out. 1. Architecturally ambitious public or institutional projects (legacy projects such as the City Churches, Whitehall, Somerset House, Livery halls, etc.) commissioned by Crown and City, many still standing today. 2. Infrastructure work such as wharfing the Fleet Ditch or building Westminster Bridge. This would include work on dockyards, and the Ordnance for example. It had much in common with the above and was frequently designed and contracted by the same people. 3. House building; this was probably the largest portion of the market and included the reconstruction of the 13,000 houses destroyed in

38     J. Z. Stephenson

the fire. We can think about this market for housing construction as being comprised of broadly two sectors. a. The first, which provided large terraces and town houses, mostly built on a speculative model, for wealthy middle-class and merchant families. In the middle of the seventeenth century, these were a new kind of residential development and were substantial in number and scale. b. Smaller-scale residential development on a more modest scale frequently involving subdivision and extension of properties as well as new building.6 4. A fourth type of construction would have been that of commercial premises, which we know very little about, but which may have been similar in scale and organisation to 2. above. 5. The general repair and maintenance of public and commercial ­buildings, and churches (which would have been recorded as ‘ordinary’). 6. The repair and maintenance of private residential property. 7. Interior and decorative architecture and its associated arts and crafts, a sector that has attracted a great deal of attention from scholars.7 In order to be truly representative, any survey of building workers, their forms of employment, and their wage should really come from all of these sectors. However, as should be plain from the discussion of sources in Chap. 1, there is a considerable source bias in the records that are available to us. We currently know very little about contracts, prices, organization, or pay in the residential sector, or the commercial one; likewise, some infrastructure projects are well documented,

6An authoritative review of housing stock in London throughout the eighteenth century is Peter Guillery’s The Small House in Eighteenth Century London. For the classic account of seventeenth-century London speculative development, see McKellar, The Birth of Modern London; for smaller scale domestic development see Baer, ‘Housing for the Lesser Sort in Stuart London’, pp. 61–88; ‘Using Housing Quality to Track Change in the Standard of Living and Poverty for Seventeenth-Century London’, pp. 1–18; Power, ‘Shadwell: The Development of a London Suburban Community’, pp. 29–46. 7For crafts see Beard, Craftsmen and Interior Decoration in England.

2  The Market for Building     39

others less so. Detailed wage or employment records come mostly from public or Crown institutions, although the majority of the work may have been for private clients. Even if we did have more records, there are no reliable estimates of the relative size of each part of the market. This matters, because we need to know how representative masons and carpenters at St Paul’s, Westminster Abbey and London Bridge were in order to discuss findings on their wages later in the book. Therefore, the next part of this chapter attempts to chart the scale of development throughout the seventeenth century in relative terms; the following section discusses the sites and sources within that context.

Development Before the Fire Whilst population growth always brings about the need for extra housing, in the first half of the seventeenth century, even before the Fire, or the interregnum for that matter, there was a profound change in the built environment, and in the scale and scope of public and private building. There had been a visible expansion in building in London from the 1620s. An increasingly wealthy population desired better buildings and private individuals, particularly merchants, landowners, and office-holders began to invest in housing in new areas of the city. New style townhouses and terraces grew up in the area north and west of the Inns of Court. Covent Garden was developed in the 1630s, and Lincolns Inn’s Fields in the 1640s and 1650s, whilst areas to the north and east of the city grew from hamlets to thriving suburbs of the metropolis. The effusion of the gentry and merchant classes, their predilection for London residence, and new building, have been cited as causes of the development of the West End.8 Inigo Jones, appointed Surveyor to the King in 1613, is usually credited as the first architect in the modern sense in England and, through his influence, the Crown

8Stone,

‘The Residential Development of the West End of London in the Seventeenth Century’, pp. 167–212; Allen, The British Industrial Revolution, pp. 92–93.

40     J. Z. Stephenson

invested  heavily  in new architectural projects such as redevelopment at Whitehall and Greenwich in the earlier seventeenth century.9 His influence on the King’s Works carried beyond the Stuart era. Although the Office of the King’s Works was suspended, there is evidence that there was building in London during the interregnum. It appears that there was a sharp downturn in construction at the beginning of the 1640s, but indicators suggest that the industry quickly recovered most of its output.10 Despite the dramatic changes in the built environment there are no accurate or reliable estimates of how many new buildings were put up in the period, nor any for the expenditure on them. Later surveys of London show large development in both east and west of the city and Westminster.11 Strype’s 1720 survey (now accessible online), also shows that there was considerable development in the east of the city, at Whitechapel, and Spitalfields, and north of the Tower before the 1660s as well as during the rebuilding.12 Some construction was carried out by small-scale carpenters and bricklayers speculating, often against Crown or City prohibitions. Although larger scale investors and speculators probably had a larger part of the market, Baer has shown that most investor developers were not builders by trade, and the scale of development businesses varied greatly.13 The scale of pre-Fire development could possibly be estimated by a quick reference to the population growth of the period. The most respected estimates show London’s population doubling from c. 200,000 to c. 400,000 between 1600 and 1650. Combining this with a figure from Baer’s work on subdivision of between four and five persons per dwelling, this implies almost 900 dwellings built per annum

9Summerson,

Georgian London, p. 11. Cyclopaedia of London, 1851, pp. 18, 373, 789; Coates, The Impact of the English Civil War on the Economy of London, pp. 209–211; Baer, ‘The House-Building Sector of London’s Economy, 1550–1650’, pp. 409–430. 11Guillery, The Small House, pp. 18–34; McKellar, The Birth of Modern London, pp. 18–20. 12Strype, Survey of London (1720) [online] (hriOnline, Sheffield). Available from https://www. hrionline.ac.uk/strype/images/figures/810061-001.jpg,  https://www.hrionline.ac.uk/strype/ images/figures/810061-002.jpg, accessed 16.01.17. 13Baer, ‘The House-Building Sector of London’s Economy, 1550–1650’, pp. 412–414. 10Knights

2  The Market for Building     41

in the capital.14 The associated output figure and the demands on ­various sectors of the construction industry would depend on their quality and scale. However, what we know about this composition comes mostly from suburban developments east of the city. Power found that 72% of houses in mid-seventeenth-century Shadwell were of two or three storeys although houses in that area were modest compared to the larger structures of the West End. He also commented that the larger buildings were the newer developments.15 Other research has found that building prohibition probably curtailed development in the first half of the seventeenth century, and as much London building work may have involved subdivision as primary construction.16 Despite subdivision, house building was likely to have been running at several hundred to over a thousand units per year well before the Fire, and increased investment after the Restoration would have done nothing to dampen this output. We also have evidence that investment in churches, public building, and infrastructure were not as insubstantial before the Restoration as previously thought.17

Rebuilding and the Market for Construction Services The Fire had literally destroyed the old city; laying ‘waste to about 440 acres, destroyed over 13,200 houses and 89 churches and chapels, rendered homeless about 200,000 people and, altogether, caused losses, previously variously estimated at between £9,900,000 and

14Harding,

‘The Population of London 1550–1700, A Review of the Published Evidence’, pp. 111–128; Baer, ‘Using Housing Quality to Track Change in the Standard of Living’, pp. 1–18, esp. p. 7. 15Power, ‘Shadwell: The Development of a London Suburban Community’, pp. 29–46. 16Baer, ‘Housing for the Lesser Sort in Stuart London’, pp. 61–88. Guillery estimates that no more of 15% of the stock was subdivided, The Small House, pp. 34–35. 17Merritt, ‘Puritans, Laudians, and the Phenomenon of Church Building in Jacobean London’, pp. 935–960.

42     J. Z. Stephenson

£10,788,000’.18 The classic discussion of the great rebuilding holds that the city was not only rebuilt, but improved, with works mostly finished by the early 1670s.19 This almost incredible achievement has always slightly puzzled historians, but if one thinks that the industry had already been engaged fully in the city at the sort of rates discussed above the problem seems less insurmountable. The implication is that an active and already growing industry doubled output (or went from 800–1000 houses per annum to 1800–2000—still a considerable increase), rather than suddenly, from standstill, creating new supply chains and skill pools. Indeed, it was not just the industry practitioners, but the institutions which were able to respond quickly. Reddaway asserted that damaged areas had been fully surveyed and plans and claims ratified by the Fire Court by April 1667.20 The City’s swift and clear regulations to stem opportunism in the face of disaster and ensure that those willing to reconstruct had access to materials and labour are credited with enabling such reconstruction. By December 1666, the ‘Rebuilding Act’ enabling assignment of landlords and tenant rights and regulating construction materials for the minimisation of fire risk, had been passed, and the Fire Courts established.21 Despite Reddaway’s description of the rebuilding as effective and speedy, it is obvious that the City Churches and St Paul’s were not completed in the 1670s, nor the 1680s in many cases. Spires were still being erected in the 1710s. Therefore, one should think of the rebuilding as the continuation of a longer process of improvement of the city, and this process continued through the early eighteenth century for both residential and other buildings. The stipulated improvements for houses after the Fire were, in the main, a more uniform width and specification for streets, and a predominance of fireproof brick and tile as building materials. 18As described by Knoop and Jones, The London Mason in the Seventeenth Century, p. 5. Their figures are taken from Strype, 1720, in n. 12 above, Chap. 28. 19Reddaway, The Rebuilding of London after the Great Fire, pp. 72–76. 20Ibid., p. 112. 2119 Cha. II. c.8.; 18 & 19 Cha. II c.7. A 1670 act provided for the rebuilding of St Paul’s, 22 Cha. II c.11.

2  The Market for Building     43

If before the Fire, much London housing had been timber, afterwards most was only internally constructed of timber with external walls of brick.22 In the wider built environment, the improvements were the widening and regulating of streets and thoroughfares, the wharfing of the north bank of the Thames, the reconstruction of the Fleet ditch towards Farringdon, the construction of the new Churches and the new Cathedral. The rebuilding of the Churches and St Paul’s required large quantities of stone, a material commonly used in some West End development.23 Although the demise of the wholly timber-built structure might lead to conclusions that carpenters lost some market power or work, stone building required the input of a great deal of carpentry, both in scaffolding and in centring for arches and vaults, and contemporary sources have been interpreted as bemoaning a decline in mason’s fortunes.24 The likely reality is that the rebuilding boom was one for all trades. Brett James reckoned that approximately 35,000 new houses were erected in London in the 40 years to 1708, bringing the total to between 67,000 and 78,000, and the total buildings to over 100,000.25 Field shows that the rebuilding was often financed by reduction in rents for leaseholders who could afford to build, and although it was swift it displaced many members of the community. Wealthier residents were more likely to return to the City, whilst poorer households were more likely to stay away.26 The Fire followed shortly after the Restoration of Charles II and punctuated a period distinguished by its increasing architectural ambition. The Crown had generally increased its expenditure on building since the beginning of the seventeenth century. Expenditure under James I had already risen to four times the level under Elizabeth.27 22Summerson,

Georgian London, pp. 65–66. The Rebuilding of London after the Great Fire, pp. 122–123. 24Campbell, ‘The Finances of the Carpenter in England’, pp. 314–316; McKellar, Birth of Modern London, pp. 79–80. 25Brett James, The Growth of Stuart London, p. 509. 26Field, London, Londoners and the Great Fire of 1666, Chap. 3. 27Colvin et al., The History of the King’s Works, Vol. 5, p. 32. 23Reddaway,

44     J. Z. Stephenson

The re-establishment of the Office of the King’s Works at the Restoration, in a manner echoing Inigo Jones’s organization of the same before 1642, was a strong signal of continued interest in building investment, and between 1670 and 1686 expenditure at the Office of King’s Works increased from £20,000 per annum to £45,000 per annum.28 The scope of Crown enterprise extended beyond new palaces to institutional spaces and public facilities, such as the Hospitals of Greenwich and Chelsea. Likewise, although with perhaps more limited financial resource, the City of London acknowledged the need for the improvement of public buildings. Many of the guilds had spent heavily on new buildings and halls in the first half of the century. The importance attached to investment in public buildings is demonstrated by Parliament’s assignation of a portion of the coal tax to cover some of the costs of rebuilding the City Churches. In terms of infrastructure, before construction of Westminster Bridge began in 1738 there were substantial investments, such as Wren’s fine new Customs House which cost in excess of £10,000 in 1668–1669, and the work carried out by John Fitch and Thomas Cartwright on the Holborn Bridge and the Fleet River.29 The City also wharved the Fleet ditch, where the work was led by Thomas Fitch in 1671–1674, at a cost of £56,000, and designed new Quays on the north side of the river.30 Knoop and Jones estimated the value of mason’s contracts alone on ‘municipal’ projects to have been £24,482.31 Close to London, the dockyards of Deptford and Woolwich and their surrounding areas were continuously being expanded and improved. The part of public investment in reconstruction after the Fire that we have most evidence of is the replacement of destroyed churches. Of the fifty-one new City Churches needed, forty-two were completed well before 1690. The Wren Society put the total costs at £263,786, 10s. 4.5d. spent between 1670 and 1694, but accounts for five churches were

28Ibid.,

pp. 40–42. House Quay and the Old Custom House’, in Survey of London: Vol. 15, pp. 35–40. 30Skempton, A Biographical Dictionary of Civil Engineers, pp. 228–229. 31Knoop and Jones, The London Mason, p. 6. 29‘Custom

2  The Market for Building     45

missing. If we assume they were built at the average cost of £5367, this implies the total price of the rebuilding programme was £290,622.32 The final total may even have been higher. Peter Jeffrey, who examined individual parish records put the total costs at £362,793 as this was the amount paid out by the coal tax. The difference may be accounted for by interior decoration costs.33 A further fifty churches were planned from 1711, but only twelve were actually completed.34 The expenditure before finance costs on St Paul’s Cathedral alone was £804,758 to 1716.35 What did the rebuilding mean in economic terms? The implications of the figures that Reddaway gives, that the fire caused £10m of damage, replaced within a decade, are that the rebuilding programme involved expenditure of at least £1,000,000 per annum from 1667 to 1677. Although any per capita figures for the whole country will hugely underestimate output for the city, the most recent estimates of output for Britain in this period come from Broadberry et al. They estimate that total annual GDP in England in the 1660s was £42.59m, rising to more than £52m in the 1670s, and that building output was equivalent to about 7.5% of gross domestic product in the 1670s, or about £3.8m. Given an English population of 5.2 million, and a London population of 575,000 at the end of the seventeenth century, if London’s market for building were in line with it’s share of population, this would mean that annual building expenditure or output through these decades would have been in the region of £400,000. At £1,000,000 per annum the capital was outspending on building by a factor of 2.5. However, just as today, the valuation of any building may vary greatly from its reconstruction cost. Brett James’s figures are taken from Strype’s London survey of 1720, and Strype’s valuation of £300 per house is actually the value of loss of an average of 25l. rents over 12 years. It is difficult to know how much of that £300 would have been taken up or exceeded 32The

Wren Society, Vol. X, pp. 45–53. The City Churches of Sir Christopher Wren, p. 356. 34Yeomans, ‘Managing Eighteenth-Century Building’, p. 5; Summerson, Georgian London, p. 69. 35Campbell, Building St Paul’s. The amount raised or borrowed by the Cathedral was £1,157,782 10s. 2½d. and only £721,552, 7s. 7¼d. was spent on direct building costs to 1711, the difference was in interest and other associated cost of building, pp. 8, 68. 33Jeffrey,

46     J. Z. Stephenson

by construction cost. McKellar also settled on an average figure of £300, also showing examples at £250, and she presented some cases showing West End houses built for as little as £100. Guillery priced the smallest sort of post-fire house at £50–100.36 This price may be considered for the exterior shell of the building without any further fitting out. By contrast in the 1760s the cost of building a gentleman’s house was given by William Paine as approximately £1600, excluding the costs of carriage.37 The late seventeenth-century private housing building boom was characterized by speculation by financiers, such as Nicholas Barbon, who acquired building leases of newly released property from major landowners, and resold them to building contractors (in various trades) who took on the financial costs of building, before selling on the residential lease.38 Those taking up the building leases were frequently established entrepreneurs and city office-holders who speculated on the profit from the sale of the completed houses. Many of these contracts involved a complex system of mortgage credit, shares, work for work and subcontracts, rather than cash payments.39 McKellar’s research firmly implies that pricing of all materials, work (including labour) in privately contracted work included cost of credit and operating costs, but it also explains why accurate estimates of costs or private building accounts are so hard to come by.40 There does not seem to be enough consistent information to form guiding assumptions to work out the value for the rebuilding.41 What is more relevant to this study is the likely proportion of the total market that the legacy projects, for which we have wage and price figures, represented.

36McKellar, The Birth of Modern London, pp. 64–65, and p. 73 for varying brick values that obscure accurate estimation of building costs; Guillery, The Small House, p. 50. 37Paine, The Builder’s Companion, p. 10. Between 1677 and 1757 the CPI varied between 1.2 and 1.8 in London, although the 10-year average was declining for most of those decades, implying building materials appreciated faster than other goods. In other words, Paine was writing for a very distinguished audience, or McKellar’s prices are low estimates. 38McKellar, The Birth of Modern London, Chaps. 2 and 3 for a discussion around costs and contracts generally. 39Ibid., Chaps. 2 and 5, and see London Metropolitan Archives List of City masons, Bricklayers, Carpenters compiled by H.M. Colvin, loose leaf in COL/SJ/09/004, MSS/133/21. 40Ibid., pp. 87–89. 41Ibid., pp. 58–65, 72–73; Baer, ‘The Housebuilding Sector’, pp. 412–416 makes this clear.

2  The Market for Building     47

On this, Strype might give us more information.42 Using his figures the 13,200 houses represent £3.9m, the public edifices £939,000, and St Paul’s £2m. The implication is that the public edifices, the Cathedral, and Churches represent three-sevenths or somewhere in the region of 40% of the market. In actual fact, St Paul’s only cost c. £1m, however, Strype did not include the infrastructure for improvement rather than replacement. Nevertheless, the indication is that the legacy projects for which we have worker and wage data were a substantial proportion of London building, at least one third, and due to quality of output, possibly more.

Beyond Rebuilding and into the Eighteenth Century If the late seventeenth and early eighteenth century were characterised by a building boom the mid eighteenth century is usually characterised as a construction slump, as dates of the Building Acts (1667, 1724, 1764, 1772, 1774) indicate.43 Given the importance of the housebuilding sector Middlesex deeds registered are possibly the best proxy for building activity. They show a peak of over 3200 registrations per year in the mid 1720s, falling by up to 20% before a slight recovery in the mid 1730s, and a decade long decline from then until the mid1740s. Housebuilding did not exceed its 1725 peak until after 1765, as shown in Fig. 2.1.44 However Crown and City continued to invest sums in public projects. After fire at Whitehall in 1698 there was an ongoing expenditure and redevelopment on the site including modest estimates of under £10,000 for William Kent’s Treasury building in the early 1730s.45 The development of the area combined planned works by the Office of the King’s Works, and the release of lands for residential

42https://www.dhi.ac.uk/strype/. 43Guillery,

The Small House, pp. 282–284; Barras, Building Cycles, pp. 7–8, 180. Belcher, and Cottrell, ‘The Middlesex and Yorkshire Deeds Registries and the Study of Building Fluctuations’, pp. 176–217; Guillery, The Small House, p. 25. 45Survey of London: Vol. 14, St Margaret, Westminster, Part III: Whitehall II, pp. 23–36. 44Sheppard,

48     J. Z. Stephenson

1709 1712 1715 1718 1721 1724 1727 1730 1733 1736 1739 1742 1745 1748 1751 1754 1757 1760 1763 1766 1769 1772 1775 1778 1781 1784

5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0

Middlesex deeds registered

Fig. 2.1  Middlesex deeds registered by year (Source Sheppard, Belcher, and Cottrell, The Middlesex and Yorkshire deeds, pp. 176–217)

development. The Customs House was rebuilt in 1721. Estimates for the completion of Greenwich in the late 1720s were over £130,000.46 A timber bridge over the Thames between Fulham and Putney was built in 1729, at a cost of £12,000.47 The mason’s contract alone for Westminster Bridge (1738–1750) was £155,000, and the project was undertaken by Andrews Jelfe and Samuel Tuffnell, whose contracts in the first year were for £27,000. The total cost of the bridge has been estimated at between £198,000 and £397,000.48 Later in the century, Blackfriars Bridge cost £166,000 in 1760–1769, whilst the costs for Somerset House were approximately £330,000 to 1794.49 Other notable works executed by others through the period included Montague House and the new Bedlam Hospital. 46Sir

John Soane Museum, SM volume 109/2 and 3. Biographical Dictionary, p. 824. 48Skempton, Biographical Dictionary, p. 824 gives the first figure. TNA Work 5/46 does not have complete accounts. Also see London Metropolitan Archive, B/CWB/I/1, ‘Westminster bridge Original Manuscripts, Notes on’, p. 4. 49Colvin, History of the King’s Works, Vol. 5, p. 467. 47Skempton,

2  The Market for Building     49

Can the projects for which we do have records be taken to be representative of the market as a whole? Given that the working practices described in the following chapters applied to all City contracts for maintenance or new work, to all Crown projects for maintenance or new work, and that those who contracted with those organizations had private work as well, it seems very likely that the working practices described here represent the way the market predominantly operated. As is still the case today, and as will be discussed in Chaps. 4 and 5, construction work involves much idiosyncrasy in contracting, but there are no grounds to think that ecclesiastical, Crown and City projects were organised or procured along vastly different principles to that of the rest of the market. However, Crown and City projects were some of the most expensive and experimental of the day. In using them for gathering ‘wages’ we are examining the earning of some of the most skilled building workers in the market. What of the guilds associated with the building trades? Most of the men and women written about in the following chapters were members of their trades’ Company. Some were members of another Company, or none. The Companies were notably absent from debates or issues about supply of materials, arrangements of contracts, or regulation of labour. Knoop and Jones found that the Mason’s Company was a nothing more than a well-to-do business association over a century before the period under investigation here.50 Carpenters Company minutes show that the most pressing concerns of the Company’s business were to do with apprenticeship, quarterage, rents and benevolence.51 When the guilds conducted searches of sites and businesses in construction it was to collect quarterages and fees, not to evaluate plans, vaulting, or the execution of tuck and pat work. Indeed, the most up-to-date literature is clear that wider City institutions, both formal and informal, rather than Companies were the means by which the rebuilding was achieved.52

50Knoop

and Jones, The London Masons’ Company, pp. 157–166. Carpenters Company Court Minutes, London Lives, 1690–1800, www.londonlives.org, GLCCMC251040001, January 1721 to October 1727. 52Field, London, Londoners and the Great Fire, Chap. 3. 51See

50     J. Z. Stephenson

The guilds were active throughout the period studied in this book, and their members involved in construction, surveying, technological innovation, regulation, and but they were not responsible for, nor it seems interested in, regulating the output, nor the organisation of the industry, although they remained strongly involved in the apprenticeship and training of skilled and specialist workers.

Organisation of Contracts and Pay at the Sites and Sources Despite the impression often given in the legion of literature about the paucity of evidence of wages in early modern London, there are actually voluminous ‘day rates’ listed in the building accounts of London’s large institutions. However, each has limitations for the researcher of building wages and incomes. As discussed in Chap. 1, the sources have been used by previous historians without much consideration given to the type of work that was being carried out, or the terms or contracts under which it was done. When data from these sources have been used until now it has been assumed and implied that the monies recorded were the day rates that the worker (craftsman or labourer) received. In fact, as becomes apparent on viewing account books and bills in situ, account books cited tend not to be journals or day-books of a busy building site, but important documents that were produced with an eye on posterity, written in careful ink on parchment and, at Westminster Abbey particularly, arranged to justify the revenue from the coal tax that was being used to fund the project. By way of example, the first entry in the ‘Christopher Wren Fabric book’ at Westminster Abbey reads ‘To Edward Tufnell Mason for work done at the Chapel of St Peter Westminster, commencing Michaelmas 1712 and ended Michaelmas 1713’, this followed by a calculation of a ‘Bill’, with detailed descriptions of work done, but no labour or day rates and a ‘Day Bill’, for a smaller amount with day rates included. The rates are as Gilboy recorded them, but what differs from expectations is the amount of work that has no day rate recorded.

2  The Market for Building     51

Tufnell’s bills were for a total of £1971 in the first year (1712–1713), yet they recorded only 699 mason days, the equivalent of just two and a half masons’ annual work, with no given names and they are all at exactly the same rate, 2s. 6d.53 The work was ‘measured’ on 29 October 1713, but Tuffnell was not paid the large sum until 3 May 1714, which implies that he waited between 8 and 20 months to recoup the monies he had outlaid for materials and labour brought to the Abbey and worked on. The value of the work is equivalent to a labour value of over £3m today.54 Could it be possible that work of that sort required just a couple of men for a year? How did Tufnell manage such a long payment period? There is an explanation of such bills. Work that was charged to a client ‘by the day’ in early modern London building was just one kind of work, and it was usually work for large and rich institutions with complex requirements where estimating a price was difficult. Most work was carried out ‘by the measure’—a sort of piece rate, or ‘by the task’ or ‘great’—as part of a single bid for a project or part of it. In these contracts labour costs were incorporated into the end price of a fixed length or volume of carving, brickwork, or carpentry. The types of work and contract are discussed in detail in Chap. 4. Measured work and task work did not specify the labour costs within them, although labour was part of what made up their cost. ‘Day rates,’ assumed by previous authors to be ‘day wages’, are no such thing. They are prices for a particular kind of contract work, and their relationship to earnings is complex. Workers who were not regularly employed or position-holders were mostly, by and large, paid by contractors.55 Men who were paid by the day by contractors earned various amounts, which the recorded, charged-out day rate is not an accurate indication of. That the responsibility for paying labour rested with contractors has been asserted and described in some detail before by Campbell, who also provided a brief but highly informative discussion 53Westminster

Abbey Muniments, Christopher Wren Fabric Books, 34513, 34514. This is the Westminster source for Gilboy, Wages in Eighteenth Century England, Appendix. 54Calculated at http://www.measuringworth.com/ppoweruk/. 55Labourers at St Paul’s Cathedral are a notable exception up to and throughout the early part of the eighteenth century. See Chap. 6.

52     J. Z. Stephenson

of the contracting system.56 The following reviews what can be known of the buildings, work and payments systems at each of the sources used for wages.

Westminster Abbey Many of the building account sources used in this study and elsewhere are ecclesiastical, and the most westward of the sources used here for London is Westminster Abbey, as referred to above. The church is a ‘Royal Peculiar’, solely responsible to the sovereign, since the late sixteenth century. The Abbey has a floor area of 32,000 sq ft and is up to 156 m in length, but the surrounding buildings and estates are a significant part of Westminster. The Abbey dates from the tenth century but the current building from the reign of Henry III. Major work was undertaken during the reign of Henry VII, and the Abbey had a school attached to it from early years, and the administration of both are intertwined in the accounts. With such provenance buildings maintenance has been a feature of the accounts throughout the centuries, and in the early eighteenth century, after the construction of the City Churches, it received major repairs and new stonework under the direction of Christopher Wren, and from 1722 to 1745 its famous towers, designed by Nicholas Hawksmoor, were erected. The work between 1712 and 1719 was under the auspices of the Act for the Building of 50 new churches, and thus, funded by the coal tax.57 The accounts that Elizabeth Gilboy used for her research in the early 1930s were the Christopher Wren Fabric books. The books give totals paid to all contractors providing materials or labour to the Abbey on the repair in biannual accounts audited in September and March (Michaelmas and Lady day). In this decade the Abbey contracted with twelve contractors on this project alone, including the aforementioned

56Campbell,

‘The Finances of the Carpenter in England’. Georgian London, p. 69; view details at http://www.westminster-abbey.org/__data/ assets/pdf_file/0009/86076/ABBEY-Dimensions-for-web-.pdf. 57Summerson,

2  The Market for Building     53

Tufnell, Elizabeth Gregory, carpenter, and Sarah Spoore, smith. Names included those which were also contracting for the Office of the King’s works and the City, such as Richard Adams, carter; George Mortimer, plumber; and Edward Drew, glazier. The Abbey, as a large institution, appointed master craftsmen to consult and manage the appointment of contractors and workers, and Tufnell’s father before him held that position, and judging by their large house on Milbank, may have profited. Both were Burghers of Westminster, but the Abbey seems not to have paid an allowance to its master craftsmen. There are no payments for retention or salary that can be found in the accounts, in fact, Edward Stanton, who succeeded Tufnell on his death in 1719, testified to the committee in 1722 that he had put up a bond of £2000 for the contract.58 This despite the fact that Christopher Cass, a prolific and well regarded mason contractor had appealed to the Abbey for the appointment, promising that he could carry out work for twenty percent less than others.59 Ralph Sims is named in some secondary sources as clerk-of-the-works at the Abbey, but, it seems he had little input to the management of the Wren works as William Dickinson, one of Wren’s protégé surveyors audited, measured, and approved contractor’s work and bills.60 Sims does seem to have operated as contractor for labourers and other services, as a purveyor. Abbey expenditure in the years 1712–1719 was between £2500 and £3500 annually. The accounts were signed by a Samuel Barton; Thos. Nots, Laurence Broderick of the Abbey; John Battley, paymaster; William Dickinson, surveyor, and Christopher Wren (Junior). A committee to oversee the works comprised the Lord Bishops of Rochester and Carlisle, Dr Barker and Dr Mandeville in 1722.61 Throughout the period under review, the Abbey provides examples of major construction,

58Westminster

Abbey Muniments 34517, minutes of 11 October 1722 Subcommittee for the Repairing of the Abbey. 59Westminster Abbey Muniments 65615. 60http://www.westminster-abbey.org/our-history/people/henry-spoore. 61Ibid.

54     J. Z. Stephenson

and repair, and maintenance, but all of a highly specialised and skilled kind. There are bills in the Abbey’s voucher collection for repair work, the format of which contains a brief description of the work, a number of men employed if day work was involved, and the rate at which they were charged out (as per Gilboy), but there are no call books or names of men employed.

Greenwich Hospital The ‘day rates’ for labourers and craftsman at Greenwich Hospital have been published or used many times, by McCulloch, Gilboy, Bowley, and others.62 The site, over six miles from Westminster, and five miles from St Paul’s was property of the Crown. It was chosen as the site for the new Queen’s House, designed by Inigo Jones in 1616 and thus commenced a century-long period of Crown development there. Originally the site of the Hospital was planned as a new palace. In the 1660s a large building, now incorporated to the current scheme, was erected. The site was allocated as the new Hospital for Seamen for the Royal Navy in the 1690s. Summerson wrote of Greenwich as one of the last substantial projects of ‘royal enterprise in building’.63 Its vast quadrants were planned and designed by Wren, but both Nicholas Hawksmoor and John Vanbrugh designed parts as showcases for dramatic colonnades, elevations, and arches. The majority of the Wren phase of building was carried out between 1698 and 1709. Accounts of the Fabric Committee survive at the National Archives.64 From 1696 Nicholas Hawksmoor, as surveyor, and clerk-of-works took instructions from the Fabric committee, drew up contracts, and considered tenders. The contractors hired were solely responsible for the labour deployed on-site.

62Gilboy, Wages in Eighteenth Century England; McCulloch, A Dictionary, Practical, Theoretical and Historical, pp. 952–953; Bowley, Wages in the United Kingdom, pp. 81–82. 63Summerson, Architecture in Britain 1530–1830, pp. 236–275. 64TNA ADM 67, and ADM 64/2, 80/4, Greenwich Hospital Fabric Committee. Gilboy’s source is given as ‘accounts of ’, under ‘Admiralty’ only, in her list of sources.

2  The Market for Building     55

The bill book of the main mason contractor for this early phase, Edward Strong, is held at the London Metropolitan Archives.65 Strong’s book is a copy of bills submitted to the Hospital, and it shows the amounts and rough dates of his bills with notes on portions disallowed by his clients. There are no daybooks nor call books for construction work in the Hospital archive, as they would have been the responsibility of the operating contractors, although some men are named in bills in the household accounts.66 The rates recorded by previous authors are the rates that appear in bills as agreed charge out rates for man days.

Middlesex Sessions Gilboy’s third source for London was the Middlesex session papers collection. These were also possibly the sole source for Schwarz’s 1985 series. Schwarz gave his source as solely “MJ/SP” and cited the work of McCulloch for Greenwich.67 The Middlesex session papers are preserved at London Metropolitan Archives and they have been re-catalogued since the 1980s. They refer to details of building work done and administrative functions such as maintenance of gaols, asylums, bridges, and other session’s houses. On reviewing the session’s catalogues and papers at the LMA, for MJ/SP55 I found fewer bills, and far fewer work descriptions in MJ/SP than one would expect for such a long series as Schwarz’s. The records for the building of the New sessions house, completed in 1782, only contain contract and payment information, but no records of wages, likewise records for maintenance, repair and other works.68

65LMA

CLC/B/227-175. 68/670. 67Schwarz, ‘The Standard of Living’, n. 11, 12. 68MJ/SP/1796/A/017. 66ADM

56     J. Z. Stephenson

Middle Temple Middle Temple, adjacent to the Inner Temple, just west of the City and a liberty of it, was developed piecemeal throughout the centuries to the Great Fire, when it avoided large-scale damage. However, a fire in Middle Temple Lane in 1679 destroyed the courts and buildings. The Temple archive holds fully documented bills and accounts from the late sixteenth century. In the mid-1990s Jeremy Boulton logged over 1350 entries of seventeenth-century day rates from Middle Temple accounts. Although much of the work billed there was ‘ordinary’, for repairs and alterations, Middle Temple employed many of the same contractors found at St Paul’s and Westminster Abbey. There are bills from Edward Stanton, mason, and City Mason Christopher Horsenaile, who measured and approved work there.69 The Middle Temple seems to have employed some men directly to build a garden wall in 1614.70 But a review of the rest of the seventeenth- and eighteenth-century accounts found no other direct payments. The bills in Middle Temple are mostly for smaller work than the large sites of Westminster Abbey and St Paul’s. A bill for plasterer’s works in 1682 shows the nature of the organization of work, and the way in which clients could discount or reject finished work. Mr Philips, the plasterer, carried out (or had some of his men carry out) the plastering work detailed “for 797 yards washing stopping whiting, colouring …” and billed the presumably agreed amount, 7d. for the unit yard, which came to, £23 4s. 11d. The work was measured (and approved as done) by Tho. Proudlove on 8 August 1682. A note was made at some point that a sum of £2 10s. was paid for use of the scaffold. The bill however was then passed to Sir Christopher Wren who wrote: “I have perused this bill and find it may be allowed as follows… for whitening stopping sizing and colouring in part of stonework in part on render 2d, the yrd, for making the scaffolding £7, for use of scaffolding £1, 5s., for new rendering sixty yards that were fallen down as the workmen assert £1 10s… in all demands 69Middle Temple 70Middle Temple

Archive, Treasurers books MT.2/TRB/1682. Archive, ACCVOUBI MT/ TOT/3/2 discussed in Chap. 5.

2  The Market for Building     57

for the Middle Temple £16 7s. 10d”. Presumably, Philips would have been out of pocket if he was charging materials and labour at cost. The difference between 7d. and 2d. a yard, suggests he was not, although the shortfall of nearly £7 must have hurt. The bill gives no detail of how many days of work were involved, nor whom was paid for labour.71

St Paul’s Cathedral Boulton may have noted that pay at St Paul’s Cathedral was lower that he found elsewhere, but wages from it have not been used by economic historians before. In one form or another, the Cathedral has been standing in London since Roman times. In the late sixteenth and early seventeenth century it fell into an advanced state of disrepair. Inigo Jones designed and oversaw a large programme of repair and rebuilding, including a new west front, in the 1630s. In the Great Fire it was virtually destroyed, however. The new Cathedral was planned and designed by Sir Christopher Wren as the most important, experimental, and beautiful building of its age, on warrant from the King for the design in 1675. Work began soon after and took at least 35 years. The building was announced complete in 1711 yet some works continued for years afterwards. The scale of the building is still notable: the exterior, wholly of ashlar masonry, is of two storeys; the building is over 175 m in length, and the Dome reaches a height of 365 ft, or 111 m. This made it the tallest building in London until the twentieth century. The most up-to-date and succinct account of such a large and important project was published by James Campbell in 2007.72 Campbell, and Lang before him, described the system at the Cathedral’s Office of Works. Contractors were appointed by the surveyor (Wren) and committee and were responsible for their own men and their own pay with the exception of the Cathedral labourers, who were

71MT.2/TRB/1682, 72Campbell,

Christopher Wren bill, August 11, 1682. Building St Paul’s.

58     J. Z. Stephenson

paid directly by the clerk-of-the-works.73 Contractors operated a system whereby they bid for some work by the day, and some work by the measure or task. If they had contracted day work the Cathedral’s accounting system required that they submit the names of those working by the day. As such, the Cathedral’s accounts contain the names of men working by the day, but they omit those who would have been working under other contracts, so they are not a complete record of all men on site. In the 1920s and 30s, the Wren Society transcribed, compiled and published most of the accounts of the rebuilding, and some correspondence and minutes for the years 1675–1720.74 One of the mason contractors described in the accounts was William Kempster, who was responsible for work on the south west of the Cathedral 1700–1709. A box of his papers is held at the National Archives.75 They include his personal records of the amounts he directly paid his men in 1700–1702 and 1706–1709. These day books have not previously been identified or used by wage historians. They form the most important part of the new evidence I present in this book.

Westminster Bridge Westminster Bridge was one of the most complex and challenging engineering feats of its age. The first new stone bridge across the Thames since London Bridge, it was funded by lottery authorised by an act of Parliament. The ambition of the commissioners, all 175 of them, sought a breakthrough, highly innovative, project. Consequently the decision to build a stone bridge was contested and the engineering techniques were controversial.76 The challenges were focused on both sinking the 73Ibid.,

Chap. 6; Lang, Rebuilding St Paul’s after the Great Fire of London, pp. 79–81. original accounts moved from the Guildhall to the London Metropolitan archives between 2009 and 2011. Original account books can be consulted at LMA CLC/313/I/B but the Wren Society’s transcription of the account is accurate and aided by knowledgeable narrative. Wren Society Vols. XIII, XV, XVI., index in Vol. XX. 75TNA C 106/146. 76http://www.arct.cam.ac.uk/Downloads/ichs/vol-1-887-902-cross-rudkin.pdf; Skempton, A Biographical Dictionary of Civil Engineers, p. 589. 74The

2  The Market for Building     59

piers and centring the arches in the water, and they created both design and organizational disputes.77 The mason contractors were Andrews Jelfe and Samuel Tufnell, who as Jelfe’s letter book attests struggled with both contracts and supply chains.78 The book has copies of the correspondence to his agent throughout the period of the building of the bridge. It also has his own account of costs and payments made to masons working for him in 1734–1735 whilst doing work in London and Cambridgeshire (4 years prior to work on the Bridge). The Office of the King’s Works maintained accounts of the tradesmen’s bills examined and allowed for Westminster Bridge from 1736 to 1749 showing that contracts ranged in value from just under thirty pounds, to the tens of thousands.79 There are also accounts of bills in the Office of the King’s Works records which include day rates for carpenters and masons charged to the bridge by Etheridge, master carpenter, and masons, Jelfe and Tuffnell.80

Bridge House One of the most voluminous sources for wages and prices in London throughout the mediaeval and early modern period is Bridge House. The institution that owned, ran, and maintained London Bridge had a wealth of property throughout the city and Southwark. If London Bridge was falling down, it was the Bridge House masters and workmen who were responsible, and possibly liable. At the Bridge, office holders of artisan positions were responsible for building work, and they in turn employed teams to carry it out. They presented weekly bills to the Bridge Master, and in turn they were paid for the labour costs. The Bridge House archives hold bills and payment details for workmen conducting building work there from the fifteenth century right through to

77Ibid.,

p. 890. MS 27587. 79TNA Work 6/ 44, 45, 46. 80TNA Work 6/46, 48. 78BL

60     J. Z. Stephenson

the middle of the nineteenth (with breaks).81 No historian of wages has tapped this rich seam in any depth before. Jeremy Boulton mentioned Bridge House in his 1996 wage study but did not include the figures in his series. The records may be under researched due to perceptions about the institutional bias of the city, but another factor in their relative neglect may be because the accounting systems are complex and not always comparable with other sources. This study draws on the bill books, remittance books, weekly payment records and accounts to examine solely the construction work carried out by contractors on the Bridge. Construction work at Bridge House was managed through a complex system of days and tides. ‘Days’ were worked by some and ‘tides’ by others, and a small number got paid for both. The system of tide rates was not constant, but rather changed as the requirement for men and scope of work changed over time. Mark Latham has studied early eighteenth-century Bridge House organization and management in depth and asserted that a key feature of administration there was corruption.82 Some of his research implies that the accounts of workmen cannot be trusted as accurate records of working hours, days, pay, or scope. A key piece of evidence is a deposition from a labourer employed at 7s. a week, that position holders were using Bridge funds and materials for their own private business, that they were not present on-site, and that they were overcharging— that wages and days which were paid for were not worked.83 The contractors at Bridge House were also contractors elsewhere. Whilst they employed up to 50 or 60 men on the Bridge each year they also held short-term contracts for new or maintenance work at other sites at the same time. They paid a great deal for the positions at Bridge House, and they expected to profit from them, therefore the findings of such ‘corruption’ should not surprise.84 The implication for the purposes

81All at LMA; Receipt books at CLA/007/FN/05, bill books at CLA/007/FN/04, payments at CLA/007/FN/03, sundry papers at CLC/215/MS31022, 3, and COL/CC/BHC/10 03-10. 82Latham, The London Bridge Improvement Act of 1756, p. 168. 83London Metropolitan Archives CLA/007/AD/01/007. 84Mark Latham, ‘The City Has Been Wronged and Abused!’ p. 1044.

2  The Market for Building     61

of studying work and wages is the Bridge might have its own further upwards bias for figures in its records.

The Office of the King’s Works The Office of the King’s Works—variously referred to in the literature as the Board of Works or the Office of Works—is another source of large numbers of day rates and bills throughout the period from 1661 to the end of the nineteenth century. Held at the National Archives in the series Work 5 and parts of Work 6, the records have not been used for wage analysis since B.L. Hutchins in 1900, possibly due to a notion that Royal works payments were not representative of the broader market. The organization of the Office, and the many changes in scope and finances over the period under discussion, have most famously been chronicled by Colvin.85 The predominant features of the organisational constraints in this case were not corruption on the part of contractors, but failure to pay on the part of the Crown. Contractors waited sometimes many years to be paid for work carried out. Colvin offered that this late payment system ultimately cost the Crown dear as over time contractors raised prices to meet the costs of providing such credit. Accounts of the ‘ordinary’ works were kept to approve the expenditure of various clerks-of-works at various sites. They comprise bills, abstracts, and monthly accounts for routine ordinary maintenance work at a number of palaces and sites around London, including The Tower of London, Kensington Palace, Denmark (Somerset) House, and Whitehall. In this study, I have utilized data from the Tower of London and Whitehall as the most central sites in the city. The accounts are generally not presented with day rates for large extraordinary work as these

85Colvin,

‘Financial Stress’ in The History of the King’s Works, Vol. 5, pp. 42–46.

62     J. Z. Stephenson

were bid for by large contractors who worked under the types of contracts described in Chap. 4.

Summary From the Restoration to the Revolutionary Wars the market for construction in London expanded to deal with unprecedented population growth. It built new housing, neighbourhoods, and infrastructure and public amenities for what was to become the largest city in the world. The organisational, contract and pay records for the period have a significant bias to large, architecturally ambitious institutional projects. Despite my inclusion of Middle Temple and London Bridge, the sources in this study maintain a predominant bias in the early part towards the work of Wren. This is not surprising given Wren’s output and influence. Even the Middle Temple, which never commissioned a large project from him, shows substantial influence from his working practices. In the later period the Office of the King’s Works and Bridge House dominate the sources, offering a combination of ordinary and extraordinary work. There are no residential project sources, nor is there a wage source for residential maintenance, other than what is found at Bridge House for maintenance of its estates. As the earlier figures and discussion described, the project types discussed in this study may have been about 40% of work being carried out by value, but they represent the majority of non-private or non-residential work carried out. Furthermore, they represent a substantial if not dominant proportion of the stonework, masonry, and other highly skilled work. By including the maintenance contracts of London Bridge, there is a plausible control for skill in this study, but it should be reiterated that the most skilled, most specialist labour available in the market through the eighteenth century in London is likely to be found among these records. Notwithstanding that skill bias, the sample is likely to be representative of the working practices and organisation in the market generally. All evidence is that those who carried out such work operated at other levels of the market. The artificers, artisans, tradesmen

2  The Market for Building     63

or contractors who worked for the Crown were the same as those who worked on elite residential projects, and who carried out and speculated on residential projects in the growing parts of the capital.86 It is to them that the analysis of the next chapter turns.

86By

way of example, see Connor, ‘Flitcroft, Henry (1697–1769), architect’. Oxford Dictionary of National Biography.

3 Contractors

Who were the men and women who undertook the delivery of these large-scale seventeenth and eighteenth-century construction projects? A body of research in architectural history has sought to detail the backgrounds and biographies of the ‘master craftsmen’, or ‘artisans’, who built St Paul’s, the City Churches, Greenwich Hospital, and who carried out work at other sites—particularly those who worked for Sir Christopher Wren. In this chapter I argue that they are best understood as well capitalised-businessmen. In some of the extant literature, they are considered as architects themselves, while elsewhere they have sometimes been depicted as craftsmen, and sometimes as sculptors or artists. Knoop and Jones argued that due to the length of their partnerships and the scale of their business, we should view them as heads of firms.1 To some extent, this is the perspective that I adopt here; I will show that they were, on the whole, entrepreneurs who managed large teams of skilled men, complex supply chains, and profited richly from financing contracts with Crown and City. For this reason, the term I use

1Knoop

and Jones, The London Mason, p. 47.

© The Author(s) 2020 J. Z. Stephenson, Contracts and Pay, Palgrave Studies in Economic History, https://doi.org/10.1007/978-3-319-57508-7_3

65

66     J. Z. Stephenson

throughout for those responsible for building contracts is ‘contractors’ rather than ‘master craftsmen’, as contracting was their primary activity. Architectural historians agree that well before the late seventeenth century, contractors were capable of providing design and engineering solutions, as well as managing the business of construction; they sold not just building services, but design to the Crown and other important institutions and clients.2 After the mid-seventeenth century, they worked increasingly with and for architects, rather than directly with clients. Much twentieth-century architectural literature sees this as somehow marking a step-down for the trade, a loss of craft status and aesthetic agency.3 Another strand ignores that lament and just writes about masons and plasterers as artists. Entries for this period in the Biographical Dictionary of Sculptors in Britain 1660–1851 provide a case in point. The entry for Joshua Marshall who laid some of the foundations of St Paul’s begins with a long description of his art: He was the eldest son of Edward Marshall, from whom he probably received his training. …Like his father, Marshall held office in the Masons’ Company…[…]. He became master in 1670 […] He carried out much work at the royal palaces. […] Only ten of his memorials from the pre-Fire period are signed or documented…[…]. [In] The earliest identified Marshall made use of one his father’s compositional formulae, a half-length figure in a swagged architectural niche. Sir John Cullum, writing in 1774 … noted that the bust had admirable vitality and the hands and draperies were finely rendered: ‘in short it would be no discredit to any of our most modern artists’ (Nichols 1812–15, VIII, 674, quoted by Gunnis 1968, 255). The monument to John Whatton …also owes debts to his father’s work, has three frontal busts …similar in style to fifteen other monuments attributed by White to Joshua Marshall. Another recurring formula makes use of a panel architrave with a segmental top

2Knoop

and Jones, The Mediaeval Mason, pp. 15–43; Ayres, Building the Georgian City, Chap. 1. A Biographical Dictionary of English Architects, pp. 4–11, the introduction to this edition is almost a lament for a master artisan; Summerson, Architecture in Britain, pp. 142–156, created a whole new oeuvre of seventeenth century architecture—the ‘artisan style’, to justify the design capability and skill of English building contractors. 3Colvin,

3 Contractors     67

surmounted by diminutive pediments and indented along the lower edge to make room for a winged cherub’s head. The first recorded instance of its use is on the monument to Sir John Gore…

However, the entry goes on to note that: He must have employed a large work-force, for he was extensively involved in the City’s rebuilding programme and handled large sums over a period of 17 years. He rebuilt or helped rebuild six of the City churches, largely under the supervision of Sir Christopher Wren: they were St Mary, Aldermanbury, 1670–86 (£3190), St Stephen, Coleman Street, 1674–81 (£2160), St Peter, Cornhill, 1677–87 (£741), St Mary-at-Hill, 1670–76(£1928), the tower of St Clement Dane’s, 1682 (£2525) and St Bride’s, Fleet Street, 1670–84. His work at St Bride’s brought in £8964 between 1670 and 1684…[…] …Marshall was the master mason for the Monument to the Great Fire, working under Wren and the architect Robert Hooke, a project that was delayed, according to Wren’s son, by the difficulty Marshall experienced in obtaining stone of an appropriate size and shape. Over a four year period Marshall handled payments of more than £11,000…4

Marshall was probably capable of working stone, but, as should be apparent, his management commitments and likely time dedicated to contracting and accounting should raise serious questions about the provenance of statuary work ascribed to him. It would more likely be accurate to say that his statuary work is ‘from the workshop of ’, and not by his own hand. This is generally acknowledged by architectural historians but less widely understood. Marshall was not alone in having many valuable and large contracts and sites under management, combined with public office. For example, in the period from 1690 to 1710, Edward Strong had large operations at St Paul’s, Winchester Palace, and Greenwich Hospital and was master of the Mason’s Company in 1696. His work at Greenwich and St Paul’s was concurrent. Likewise,

4Roscoe,

Hardy, and Sullivan, et al., A Biographical Dictionary of Sculptors in Britain, entry for Joshua Marshall 1628–1678.

68     J. Z. Stephenson

Christopher Kempster had operations at St Paul’s, St James Garlick Hill, Tom Tower in Oxford, and was master of the Mason’s’ Company in 1691 and again in 1700. Thomas Wise was a major contractor at St Paul’s, Portland, Winchester Palace, and London Bridge. Although some historians’ treat these contractors as ‘masters’, the scale of their business stretches to breaking point the usual associations of an artisanal workshop setting. Some sense of this can be obtained by looking again at the sums involved in these contractors’ operations. The accounts at Westminster Abbey show that Edward Tuffnell, mason, billed the Abbey on average just under £2000 annually from 1712 to his death in 1719. In 2016, the equivalent in labour cost terms would have been c. £3.82m.5 At St Mary Le Bow Church, one of the more famous Wren Churches, the masons’ charges from Thomas Cartwright and John Thompson for 1671–1677 were £9279 15s. 14d. out of a total of £15,421.6 As the Wren Society accounts, and Office of King’s Works records show, there were many projects and contracts of similar financial scale, and many firms or businesses competing for them. Bill Addis described Robert Hooke and Christopher Wren as operating the world’s first architectural engineering consultancy in London after the Great Fire.7 Like any consultancy they had suppliers and specialist technicians who they used or recommended repeatedly. Wren’s master masons and carpenters were contractors sought not just for their fulfilment of building designs but also for their input to particular and specific engineering design or process problems, and many of them worked for him over and over again. Table 3.1 lists the mason contractors at St Paul’s Cathedral and the value of their contracts. Joshua Marshall may have got the first contract at St Paul’s, but the largest of the mason’s contracts went to Thomas Strong. In the following

5Calculated

at http://www.measuringworth.com/ukcompare/relativevalue.php. Society, Vol. X, p. 89. 7Addis, Building: 3,000 Years of Design, Engineering and Construction, p. 198. 6Wren

3 Contractors     69 Table 3.1  Mason’s contracts at St Paul’s Cathedrala Contractor

Total bills of work

Dates

Also contracted at

Joshua Marshall

£5543 14s. 5 ½ d.

August 1675–1678

Thomas Strong

£7918 10s. 11 ¼ d.

1675–1681

Edward Strong

£46,446 19s. 3d.

1679–1708

Office of King’s Works, City Churches Trinity College, Oxford Greenwich Hospital, Blenheim, Winchester Palace, St Stephen Walbrook

Edward Pearce Thomas Wise Thomas Wise Junior and Thomas Hill Jaspar Latham John Thompson Samuel Fulkes Nathaniel Rawlins Christopher Kempster and Ephraim Beauchamp William Kempster

£13,494 17s. 8 ¼ d. £5616 18s. 10d. £24,509 9s. 1 d.

1678 1678–1686 1687–

£10,537 14 8 ¾ d. £8089 4s. 11 ¼ d. £23,115 17s. 10d. £15,751 5s. 2 ¼ d. £15,132 11s. 10 ½ d.

1678–1693 1688–1700 1688– 1693– 1693–

£9019 3s. 0 ¼ d.

1700–1708

Total

£185,196 7s. 10 ½ d.

aWren

Bridge House, City Churches City Churches City Churches City Churches (St James Garlickhythe) City Churches (St James Garlickhythe)

Society, Vol. XV, pp. xiii–xiv

paragraphs I summarize what is known of the training, work, and wealth of some of these contractors.8 The Strongs were a family and firm that dominated Wren’s network and stone building in England for at least three generations. Thomas Strong (1634–1681) came from an established family of quarry owners in Taynton, Oxfordshire. The family have been traced to Timothy Strong, who acquired freestone quarries in Little Barrington in

8Where

specific notes are not given the information comes from Colvin, A Biographical Dictionary of English Architects, and Roscoe et al., A Biographical Dictionary of Sculptors in Britain, 1660–1851.

70     J. Z. Stephenson

Gloucestershire in the early seventeenth century, and operated at Oxford in the 1630s, establishing the family at the premium end of the market. His son Valentine continued the business, and he in turn had six sons including Thomas, who all worked as masons. Thomas carried on the business and quarry and is credited with work at Longleat, among other places. Thomas apprenticed John Strong, son of his brother Valentine in 1672. His famous working association with Christopher Wren may have started before the contract at Trinity College, Oxford. But it was not until after the Fire, in 1670, that he, and his brother Edward, took freedom by redemption of the Mason’s Company in London. So close did he become to Wren that he reputedly laid the first stone of St Paul’s.9 He died in 1681, leaving the business to Edward, who hung the lantern of the Cathedral as a symbolic completion. The Strongs feature prominently in the St Paul’s accounts and provided stone and other materials as well as mason’s ‘work’.10 In the same period, Edward Strong partnered Christopher Kempster in the construction of St Stephen’s Walbrook, where the total value of the mason’s contract was £4423 18s. 6d. between 1674 and 1682. Edward also contracted for work at St Clement East Cheap, (£2661), St Mary Magdalen Old Fish Street, (£2776), St Mildred, Bread St, (£872), St Michael College Hill (£4766), and St Benet Thames St (£1866). Later, the Strongs shared contracts for the Queen Anne Churches with Edward Tuffnell.11 Edward, in particular, was one of the most prominent masons of the age.12 He married Martha Beauchamp, the sister of Ephraim, a wealthy London mason.13 He bound his own son (Edward II) as apprentice in 1691, and another relation from Taynton in Oxfordshire in 1717.14 He contracted work at Winchester Palace, the majority of the masons’ work at Greenwich Hospital, and he was the primary or largest contractor at Blenheim Palace, and successfully petitioned for the payment tens 9Campbell,

Building St Paul’s, p. 75. A Biographical Dictionary of English Architects, p. 996. 11Lambeth Palace Library MS 2703. 12Campbell, Building St Paul’s, p. 76. 13Beauchamp was registered for the £600 surcharge in the Marriage Duty Assessment of 1695. 14Campbell, ‘Finances of the Carpenter’, p. 343. 10Colvin,

3 Contractors     71

of thousands owed to him for that project.15 Unsurprisingly, he also became an extremely rich man who purchased lands near St Albans in later life. His son Edward II succeeded him to the business.16 The family’s portrait, a fashionable conversation piece, was painted in 1732 and is in the collection of the Metropolitan Museum of Art.17 It shows the family taking tea, at gracefully laden tables, with women and children finely dressed in silks, in an elegantly panelled room, worlds away from chisels, combs, and stone yards. Often associated with the Strongs, the Kempsters also hailed from Oxfordshire. Christopher Kempster (1627–1715) was also from Burford, where he owned a quarry inherited from his father William. In 1681, Wren put him forward to John Fell, Bishop of Oxford to supervise the build of Tom Tower at Christ Church, Oxford, in very strong terms as “a very able man, modest honest and Treatable”.18 He seems to have acted as Wren’s overseer or agent on the Tower project 1681– 1683.19 He was a partner of Strong at St Stephen’s Walbrook and also held independent contracts on St Mary Abchurch (£1695) and St Mary Somerset (£4140).20 He was in partnership from 1691 to 1707 with Ephraim Beauchamp at St Paul’s. We know little of his or any of these men’s lifestyles, but Kempster’s pocketbook, surviving at the National Archives, details payment for a silver watch at £1 13s.21 His son William, worked closely with John Thompson a prominent city mason, before taking over Thompson’s contracts at St Paul’s.22

15The

National Archives, E 134/8Geo1/Hil8. A Biographical Dictionary of English Architects, pp. 577–579. 17View at http://www.metmuseum.org/collection/the-collection-online/search, accessed 29 June 2018. 18Wren Society, Vol. V, p. 18. 19Ibid., pp. 20–21. 20Strong and Kempster are referred to as dining with Robert Hooke and Wren during work on St Stephen’s Walbroke by Colvin, A Biographical Dictionary of English Architects, p. 996. See Campbell, Building St Paul’s, Chap. 8 for further discussion of wealth also. 21The National Archives C 106/145. 22Roscoe et al., A Biographical Dictionary of Sculptors in Britain, entry for John Thompson c. 1655–1700. 16Colvin,

72     J. Z. Stephenson

William, who shall feature prominently in this book, was master of the Mason’s Company in 1705. He had a son named William also, who was made free of the masons by patrimony in 1714, although he was working with his father at St Paul’s well before that. The Dictionary of Sculptors records that William senior “spent four days modelling the scroll for a pilaster in the library at a rate of 2/6 per day and in 1703 he was paid £10 for making a stone cistern to hold water for making mortar. In 1704 he made a model for the geometrical staircase in the south-west tower; and he received just over £463 in June 1705 for the staircase itself ”.23 Chapter 6 of this book gives details of the men working for him on the project. They include Joshua Fletcher, who seems to have worked as foreman for him. Fletcher went on to be a prominent contractor in his own right and worked at Blenheim and Burlington House.24 Kempster continued to work at St Paul’s after its declared completion and was made ‘Master’ there by 1714. He died in early 1719. In his will of 27 March 1718, he left his wife £300, his son William £300, and “the stock of stone about the yard belonging to my now dwelling house” at St James Garlickhythe. He also left his son rights to a debt of £60 that Edward Francis, painter-stainer owed him since 1707, and houses and cottages at Norwood in Middlesex.25 Those who worked at sites other than St Paul’s have similar profiles, very possibly because they were involved with those who did. Edward Tuffnell, whose work at Westminster Abbey was discussed in the previous chapter, was the son of a burgher of Westminster, John Tuffnell, who was Master Mason of Westminster Abbey. He was himself was a burgher. He was apprenticed to Christopher Kempster in 1692, and very possibly worked at St Paul’s before he returned to the family business. He worked extensively with Edward Strong on a number of the Queen Anne Churches, including St Anne, Limehouse and St John,

23Roscoe et al., A Biographical Dictionary of Sculptors in Britain, entry for William Kempster 1651 to 1719. It is likely that work on the staircase continued into 1706 as it is referred to in the books discussed in Chap. 6. 24Roscoe et al., A Biographical Dictionary of Sculptors in Britain, entry for Joshua Fletcher I, d1729. 25LMA MS 9172/113, Will Number: 100.

3 Contractors     73

Westminster (now Smith Square). He was buried in the Abbey in 1719 and there is a large memorial to him. He owned a large house on Milbank, but little else is known of his life, however, the Tuffnell family continued to be involved in construction. Samuel, son of Edward (?–1765), was also mason at the Abbey and most notably was a partner on the important masonry contract at Westminster Bridge, as well as known for fine statuary work and fireplaces. Edward Stanton put up a bond of £2000 to succeed Tuffnell as mason to Westminster Abbey. Stanton was also part of a well-­ established family business. His (possibly) grand-uncle Thomas Stanton had apparently established workshops at Holborn by 1635. His father William contracted with Edward Pearce for the reconstruction of St Andrew’s Holborn, and he was also known for building large country houses. William Stanton took on much work in the City—he contracted for the gate at St Bartholomew Hospital—and supplied a very similar quote to Middle Temple for £550 for the gate there, and he became one of the most established London masons of the mid-seventeenth century.26 As well as his work at Westminster Abbey and School, Edward Stanton continued this tradition, although he seems to have concentrated on monumental masonry. He was appointed City Mason in 1708. The Stantons were not the only large masonry business at Holborn, as Joshua Marshall’s yard was at St Andrew Holborn in the 1670s and 1680s, and Samuel Fulkes had three men on his premises at Fetter Lane in the Mason’s Company search in 1694.27 It is worth considering, given the seeming strength of family networks described so far, whether ‘outsiders’ could gain access to the business and achieve success. The work of Andrews Jelfe is possibly such a story. Although he is frequently cited as having been apprenticed to Strong there is no evidence of this, rather the apprenticeship records show him bound to a little-known mason, John Keene, in 1704 when Keene also took on another apprentice. Jelfe obviously learned more than how to 26Fisher, ‘Stanton, William (1639–1705)’, Oxford Dictionary of National Biography, http:// ezproxy-prd.bodleian.ox.ac.uk:2167/view/article/38004, accessed 19 July 2017. 27Knoop and Jones, The London Mason, p. 78. The details of the two searches in 1678 and 1694 are transcribed, pp. 67–80.

74     J. Z. Stephenson

cut stone with Keene, because on gaining freedom it is thought he very quickly entered into partnership on projects with Christopher Cass and Edward Strong. Cass, possibly the son of a City mason was subsequently the main contractor at St Martin in the Fields in 1721 and worked closely with Joshua Fletcher around the same time.28 Cass and Jelfe seem to have benefitted from Strong’s expertise, training and networks, and they continued a partnership which lasted until Cass’s death in 1734. Jelfe held positions as architect at the Board of Ordnance from 1716–1727.29 He contracted for carving and sculpture, as well as taking on advanced engineering projects such as Westminster Bridge which he did in partnership with Samuel Tuffnell from 1738 to 1745. Jelfe died a rich man with property worth £30,000.30 All of these examples of contractors are tied to ‘extraordinary’ work. These were the heads of the firms who took on large new projects. However, the contractors on ordinary maintenance work for major institutions were not obviously different. All of the men discussed above developed private and residential work as well as large Crown or City contracts. Those engaged on maintenance, such as contractors at Bridge House worked in the same manner, employing large and small teams, contracting for both labour and materials, and submitting audited accounts, including billing up to £1000 per annum just for labour costs.31 Individuals operating at Bridge House also took on extraordinary projects. In 1681 the salaried contractors at the Bridge included Joseph Cartwright, mason, and William Gray, carpenter. Cartwright was the son of Thomas Cartwright, one of the most prominent sculptor and masonry contractors of mid-seventeenth-century London whose work spanned fireplaces and carving at Whitehall, Edward Jerman’s Royal Exchange and the new St Thomas’s Hospital until 1702. Gray was also a contractor to the City Churches.32 For their position at

28Roscoe

et al., A Biographical Dictionary of Sculptors in Britain, entry for Joshua Fletcher I d1729. Parnell, The Buildings and Works at the Office of Ordnance, pp. 94–95; BL MS27,587. 30Colvin, A Biographical Dictionary of English Architects, p. 3. 31London Metropolitan Archives CLA/007/FN/04/001-10. 32Wren Society, Vol. X, pp. 45–53. 29See

3 Contractors     75

Bridge House, they were paid £4 per annum each, and could bill up to 11s. a week for their own time. They presented bills for their men based on days worked and materials used. When Thomas Wise Jr. succeeded Cartwright as mason in 1684–1685, he maintained the same salary of just £4 per annum plus took another £1 for ‘edge tools’ until 1720.33 His father, Thomas Wise had taken his freedom from the Mason’s Company in 1672 by redemption, and by 1678 had risen to Master Mason to the Office of the King’s Works. In addition to his work at Bridge House, Thomas Jr. was the contracting supplier of Portland stone to St Paul’s, in partnership with Thomas Gilbert. He was also a contractor on the City Churches, receiving £1019 for work at St Michael Wood Street 1670–1687, £3141 at St Nicholas Cole Abbey 1671–1681, and £2658 at St Benet Gracechurch, 1681–1687.34 The Wise family reflect the norm, not the exception, throughout London. As is plain, in London in this period contractors working in the same environments were frequently related by marriage or connected through apprenticeship. A study of the financial strategies to survive late payment of Crown and City in the seventeenth and eighteenth century cited family or dynastic capital as a major factor in their connections, but it seems that non-familial links arising through partnership and apprenticeship also thrived. There were many partnerships between these known successful contracting firms, families or individuals.35 A good indication of the significance of partnerships can be found in the way in which the reconstruction of London’s churches was organized. Of the fifty-four City Churches, eighteen of the carpentry contracts were awarded to partnerships. Some of these contracts had as many as four partners. John Longland, Master Carpenter at St Paul’s, undertook the carpentry as a sole contractor on nine churches, but he was partner in the contract at another six churches, all with different

33See

London Metropolitan Archive COL/CC/BHC/10/001-6. Hardy, and Sullivan, et al., A Biographical Dictionary of Sculptors in Britain, entry for Thomas Wise II; Wren Society, Vol. X, pp. 45–53. 35Mobus, ‘Surviving Late Payments’, pp. 273–280. 34Roscoe,

76     J. Z. Stephenson

combinations of contractors. Matthew Banks was involved in twelve contracts, of which five were awarded to him solely.36 William Gray contracted on one in partnership with Israel Knowles, who partnered Longland at St Paul’s. Knowles, was involved in nine, just two solely. The masons’ contracts for the churches show partnerships too, but not quite as many or as complex. There was one partnership of three men, Joshua Marshall, Abraham Storey, and Thos. Humphreys at St Peter Cornhill. Marshall also partnered Samuel Fulkes on two churches, and Fulkes in turn also took on four churches solely. Edward Strong partnered Christopher Kempster at St Stephen Walbrook, as well as undertaking six solely, whilst Christopher Kempster contracted as sole mason at St James, Garlick Hill.37 Some of these partnership arrangements were extremely long lasting, whilst others were flexible or brief. At St Paul’s Cathedral, the long-term partnership of Thomas Hill and Thomas Wise survived for 21 years, that between Christopher Kempster and Ephraim Beauchamp 15 years. Edward Strong and Thomas Hill originally partnered on the masonry contract at Greenwich, in 1696, although by 1699 it seems only Strong was billing for new building.38 Ephraim Beauchamp, is mentioned in the minutes but his involvement seems to have gone no further.39 As mentioned, Andrews Jelfe and Samuel Tuffnell partnered on the stone and masonry construction of Westminster Bridge, an association that survived more than a decade and many business challenges. Jelfe’s letter book contains correspondence to do with his early partnership with Christopher Cass, showing that in 1741, seven years after Cass’s death in 1734, Jelfe was still dealing with finances and bills arising from that partnership.40 Of only some surprise is the number of women found contracting on the books of large organizations such as Westminster Abbey, London 36Wren Society, Vol. X, pp. 46–53. Banks is discussed by Campbell in ‘The Finances of the Carpenter’, pp. 313–346. 37Wren Society, Vol. XIX, pp. 46–52, Table of the Fifty-Four Churches with Trades and Costs. The masons may also have been operating independently at the same site but the indication from known partnership is otherwise. 38The National Archives ADM 67/2. 39Ibid. 40Andrews Jelfe, letter Book British Library MS 27,587, p. 38 dealing with bills on Hawksmoor residence.

3 Contractors     77

Bridge, and the Office of the King’s Works. Elizabeth Gregory, was the widow of John Gregory who had been carpenter to Westminster Abbey before his death in 1708. Gregory carried on contracting with the Abbey for almost a decade after this, providing both specialist materials and labour, and signing her own name against payments and measurements. Her bills and signatures can be found alongside records of contracts with Sarah Spoore. Spoore was the contracting Smith, and the mother of Ralph Sims who was acting clerk and contracting to supply labourers for the Abbey. Despite the fact that Sims seems to have been of age to manage the family’s smith business, Spoore remained in charge. Widows were signatories to important contracts and large bills and seem to have been running and managing the business day-to-day, not just legally ‘holding’ the construction business, in the absence of a male family member of age. This short prosopography of the leading contractors suggests that the characteristics of masons, carpenters, plumbers, painters, glaziers, carters and other contractors responsible for this type of building work were of those of a ‘merchant’ class more than that of ‘master craftsman’.41 Furthermore, the following features were prevalent. • The businesses can be typified as dynastic: They depended on access to financial and social capital that could be transferred across generations to deal with late payment, fluctuating prices, and other uncertainties.42 It can be seen from the list above that the Strongs, Kempsters, Tuffnells, and Stantons all worked as dynastic businesses. • They were well capitalised: These were wealthy entrepreneurs who had capital resources to fund the growth of their business activities. The Kempsters and Strongs had quarries, the Stantons and Marshalls had large statuary businesses with premises. They required wellestablished supplier networks and capital to finance projects carried out under the Office of the King’s Work’s and City’s notorious

41A

typology referred to in Unwin, Industrial Organisation in the Sixteenth and Seventeenth Centuries, pp. 64–65. 42Mobus, ‘Surviving Late Payments’.

78     J. Z. Stephenson

long payment schedules. That finance was a product of dynastic accumulation and earned profit.43 • Entrepreneurs combined regional specialism with use of London networks: Many came from outside London and used the livery company to naturalize and establish themselves in London work. Strong and Kempster bought freedom by redemption and Strong also purchased it for others in his employ from elsewhere.44 • Varying co-ordination mechanisms: As well as family ties, they made heavy use of partnership and other non-familial links in contracting. In partnerships and joint ventures there seem to have been varying and idiosyncratic co-ordination mechanisms and relationships, suggesting that the boundaries of the firm were fluid, and that formal or informal institutions and social structures supported such varying contracts. This brief description of the work and backgrounds of what has previously been written about ‘master craftsmen’, has used evidence of the scale of their businesses and personal backgrounds of wealth and connections to cast them as businessmen, investors, entrepreneurs, and contractors, as distinguished from a tradition within economic history of viewing them as ‘artisans’.45 The following chapters describe how they worked and managed the business of construction.

43See Colvin’s remarks on Fulkes and Jelfe, Biographical Dictionary, p. 3; Also, Hewlings, ‘Jelfe, Andrews (c.1690–1759)’, Oxford Dictionary of National Biography. 44Knoop and Jones, The London Mason, pp. 13–14. 45By way of example see Farr, Artisans in Europe, definition, p. 3, does acknowledge those who ran enterprises, but the bulk of his cases and research is those who performed craft functions manually.

4 Contracts and Ways of Working in the Building Trades

How did the large institutional clients and contractors discussed in the previous chapters organise the logistics and manage the risks of b­ uilding such innovative and expensive projects as St Paul’s Cathedral? This chapter will argue that seventeenth-century businessmen in construction understood the costs, incentives, and benefits of different organisational and contractual arrangements, and that they used such varying arrangements purposefully to mitigate risk. The central device or co-ordination mechanism I will examine in this chapter is the contract for construction and how it was executed in practice, because, as will become apparent, clients protected their interests through the use of varied contract types to meet specific needs and incentives, and the form of the contract agreed determined prices, the form of any subcontracts, and the organisation under which work was carried out. Craftsmen’s and labourer’s experience of work and welfare were governed by these contractual developments, but they had neither part in nor influence over them. There were various kinds of contracts used in building, and each had respective advantages and disadvantages for contracting parties. Each had associated costs of bargaining and enforcement. The contracting systems shed a great deal of light on © The Author(s) 2020 J. Z. Stephenson, Contracts and Pay, Palgrave Studies in Economic History, https://doi.org/10.1007/978-3-319-57508-7_4

79

80     J. Z. Stephenson

the role and skills required of a building contractor, and they provide the conditions under which wage rates and pay developed. The chapter also shows that the types of contracts used explain one of the fundamental limitations of our understanding of building wages—the limited ­number of ‘day wage’ observations. In the seventeenth and eighteenth centuries the problem of quality control and timely progress of building work was just the same as it is for clients today. Building sites are and always have been dangerous, messy, and confusing places, whilst the process of construction exposes the client to financial risk.1 Designs on paper are just that, and rarely visualise with any sort of realism or accuracy what the threedimensional space and finished product will look and feel like. Even the ‘Warrant’ design of St Paul’s Cathedral was in no way an accurate representation of the finished building.2 Monitoring building work is difficult. It presents problems of agency, asymmetries of information positively correlated with technical complexity, associated asset specificity, sunk costs and moral hazard, especially in the kind of idiosyncratic work discussed here.3 The quality and craftsmanship of builder’s work is hard to evaluate except through use. You do not know how well the new roof is fitted until it rains; if the pointing on the brickwork is solid until it crumbles; and you cannot admire glazing, joinery or ironwork until it is done and fitted, and the materials and labour are spent. Only prior performance or very good information may indicate whether money will be best spent with one contractor or another, and even that is not a guarantee, as risks cannot be assessed equally or entirely by all parties. Moreover, as some of the industry-specific literature attests, the process of building compounds rather than resolves the problems.4 The associated competitive and collaborative relations between parties are complicated further by finance, and the

1Ayres,

Building the Georgian City, p. 2; Nisbet, A Proper Price, pp. ix–x. Building St Paul’s, pp. 31–32. 3In modern times too: Reve and Levitt, ‘Organization and Governance in Construction’. 4There is a good summary schema of where problems usually arise in the construction process in Vrijhoef, ‘Supply Chain Integration in the Building Industry’, p. 7. 2Campbell,

4  Contracts and Ways of Working in the Building Trades     81

relative capital-intensity of building.5 Colvin put it plainly: “War and building are two of the most expensive activities in which governments can engage”. He also said, “For neither did the Stuarts have adequate means, either from their own resources, or from the supplies grudgingly granted by parliament”.6 Moreover, a problem in large-scale infrastructure or architectural development more generally, especially in the period under review, is idiosyncrasy. Large projects are by definition, one-off ’s (most of Wren’s projects were unique and untried designs) which brings a further complexity to organisation. Varying aspects of technical development and design will call for the expert advice and management services of architects, surveyors and engineers. These agents act for and on behalf of clients or owners, and their interests may compound agency problems whilst their behaviours with contractors may display clan relationship features.7 Information and other assets developed by them specific to one development may not always be used in another or may be of little benefit to other clients. Economic and business theory sees the problems of managing such situations as those of transaction costs, and it is well recognised that some of these costs are peculiar to the building industry.8 Contracting for building work is broadly within the sphere of contingent claims contracting, but it is acknowledged that particular risks and hierarchies will arise.9 “Contracts that give the client the right and obligation to supervise and monitor the work of a construction contract neither meet the requirements of a pure market, nor have all the characteristics of

5For

a description of how the relations may develop in a modern context see Smyth, ‘Competition’, pp. 22–38. 6Colvin et al., History of the King’s Works Vol. V, p. 41. 7Reve and Levitt, ‘Organization and Governance in Construction’. 8An empiric and theoretic rejection of another models is made by Eccles, ‘Bureaucratic Versus Craft Administration: The Relationship of Market Structure to the Construction Firm’; See Ive and Gruneberg, The Economics of the Construction Sector for a TCE based description of the contracting in the sector. 9A good summary of transaction costs in a building contract, (without a detailed consideration of the role of consultants), is made by Li, Arditi, and Wang, ‘Determinants of Transaction Costs in Construction Projects’.

82     J. Z. Stephenson

a hierarchy”.10 They are closest in some regard to firms within firms.11 This chapter will assert that transaction costs were recognized and understood by those responsible for commissioning and supplying large construction work in the seventeenth and eighteenth centuries, and that their use of market and hierarchy reflected such costs.12 There was no shortage of advice for those wishing to build in the seventeenth and eighteenth centuries, and, in the body of literature that offered it, authors and advisors repeatedly counselled that the client’s money and reputation were always at risk. To protect themselves, they needed expert help. As Balthazar Gerbier, known for his design of Hampstead Marshall, put it: Whoever is disposed to build ought in the first place to take a skilful choice of surveyor from whose direction the several workmen may receive instructions by way of draughts models frames etc. For the better managing their intended work since an ill built palace leaves a perpetual selection of ignorance on the Builder.13

Gerbier gave full advice on reducing risk in the building process by testing the surveyor’s capability, by buying one’s own materials to avoid profiteering workmen, and by only appointing master workmen who would be present on site.14 He also gave advice on getting the best value from labour: The chosen master workmen must be bound to a presixt time for the performance of their undertaking, to observe exactly the model and moulds held fourth to them by the chosen surveyor and to make good at their own cost what they do amiss. They are to manage the paying of their own workmen, on such a contract as they have made with the proprietor of 10Reve

and Levitt, ‘Organization and Governance in Construction’, p. 17. ‘The Quasifirm in the Construction Industry’. Also see Williamson, The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting, pp. 217–218. 12See Williamson, The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting; Williamson, The Mechanisms of Governance; Coase, ‘The Nature of the Firm’. 13Gerbier, Counsel and Advice to All Builders, p. 5. 14Ibid., pp. 61–64. 11Eccles,

4  Contracts and Ways of Working in the Building Trades     83

the buildings for the master workmen must keep the workmen under a certain regular proportion of pay to hinder them form spending their wages to salt and to run to other works as many (to slight occasion) doe.15

A burgeoning literature on managing the building process included advice on choosing the ground and designs of pleasing elevations alongside advice on managing workmen to avoid malfeasance and loss. Their authors were frequently architects or surveyors who wished to persuade rich clients of the need to hire a consultant such as themselves.16 They contained endless information about materials, types of contract, prices, rates, and design. The titles of the volumes speak clearly. For example, William Paine’s Builders companion, and workmen’s general assistant, demonstrating after the most easy and practical method all the principal rules of architecture from the plan to the ornamental finish…. Being not only useful to all masons bricklayers plasterers, carpenters, joiners, and others concerned with the several branches of building &c but also necessary for the gentlemen who will be humbly enabled to know the exact expense of any building, alteration or repair. Published for the first time in 1758, it was reprinted many times, and naturally gives a detailed guide to quantities and prices, and estimates of materials and workmanship.17 Battey Langley was another such author, his best known work entitled The London prices of bricklayers materials and works : both of new buildings and repairs, justly ascertained, and the common exactions and abuses therein detected, interspersed with rules for estimating, performing and measuring all kinds of … brick-works … : also illustrated with a great variety of designs for plain and rusticated piers, for gates, piazzas, &c., in thirty-two curious copper plates.18 It was not commonplace in the seventeenth or eighteenth century to accurately estimate the costs of building, and building in stages was seen as having advantages in managing financial and engineering and 15Ibid.,

p. 60. also Pain, A Builder’s Companion; Neve, The City and Country Purchaser and Builder’s Dictionary. 17Pain, A Builder’s Companion. 18Second Edition, 1750, London, Adams, Dryle’s Head. 16See

84     J. Z. Stephenson

aesthetic risk.19 Architectural historians believe that accurate estimation remained elusive into the late eighteenth century.20 As more and more investment was made into large building projects, participants mitigated this problem through applying varied ‘ways of working’, which were, to all intents and purposes, organisational responses to the limitations of contracts.

Three Ways of Working There are 3 ways of working: by the Day, by Measure and by Great; if by day it tells me when they are Lazy. If by measure it gives me light on every particular and tells me what I am to provide. If by the Great I can make a sure bargain neither to be overreached nor to hurt the undertaker; for in things they are not every day used to, they doe often injure themselves and when they begin to find it, they shuffle and slight the work to save themselves. I think the best way in this business is to worke by measure: according to the prices in the estimate or lower if you can and measure the work in 3 or 4 measurements as it rises. But you must have a trusty measurer, there are few that are skilled in measuring stone worke, I have bred up 2 or 3.21

Sir Christopher Wren wrote the above to John Fell, Bishop of Oxford, when he was undertaking the erection of Tom Tower at Christchurch College, Oxford, in 1681. In further correspondence, to assuage Fell’s worries about managing the undertaking of the work, he recommended, then contracted, Christopher Kempster, mason, to be his overseer or agent at the site, advising that Kempster was both trustworthy and

19Campbell, Building St Paul’s, Chap. 4, p. 33; Akira Satoh, Building in Britain: The Origins of a Modern Industry, Chap. 1. 20Arguably it remains so today. James Nisbet, A Proper Price, throughout and p. 39; Campbell, Building St Paul’s, Chap. 10. 21Wren, written correspondence to Bishop of Oxford John Fell in 1681, as quoted in Colvin, A Biographical Dictionary of English Architects, 1660–1840.

4  Contracts and Ways of Working in the Building Trades     85

knowledgeable.22 The arrangement gave Fell peace of mind in having an experienced expert manager of such projects on site, but also allowed Wren to monitor the works. The ‘three ways of working’ are oft quoted, but rarely considered to be a statement of management strategy rather than just a description of seventeenth century custom.23 Wren’s consideration of the incentives and risks of these ways of working shows his understanding of the use of organisational responses to varying risk and quality control issues; moreover it shows a strategic use of organisational form. As Wren recognised, each way of contracting had different incentives associated with it. In ‘day work’, where the client paid for labour on an ongoing basis, the lack of an incentive to finish quickly is obvious. The risk was that the client would be paying workers to not be productive, with few monitoring opportunities that allowed them to prevent this. In contracting ‘by the great’ the risk was that a contractor would mis-estimate or underbid, risking sunk costs and unfinished work or poor workmanship. In ‘measured work’, however, money would not be paid out until the work was certified as satisfactory. Wren advised negotiating a price lower than the estimate for finished work, then evaluating the work, and paying out in stages. He was advocating a sort of ‘piece rate’, and a ‘stop option’ at the same time. What was measured work? Put simply, it was a system of buying fixed amounts of building work, quite like a piece rate, with self-enforcement mechanisms, and some associated costs in the bureaucracy and skilled monitoring involved. In working under a measured contract, contractors built the required end product in given units—a rod of brickwork, a foot of carving, and so on—and clients only paid for the work when the units had been ‘measured’ or surveyed complete as per the contract by a qualified measurer.24 If all was as specified they paid the price agreed, likely after a lengthy approval process. If, however, the work was 22Wren

Society, Volume V, pp. 22–25. fact, the three ways of working were strategised in other industries also. See Schwarz, ‘Custom, wages and workload’, p. 146. 24A rod was 272.25 square feet, a lineal rod 16.5 feet. According to Nisbet, ibid., p. 9. In practice bills and estimates or large projects tended to specify items idiosyncratic to its requirements. 23In

86     J. Z. Stephenson

found dissatisfactory, the client reserved the right to make deductions from the price, or have the work made good at the contractor’s expense. The practice of measuring was the father of and precursor to quantity surveying.25 Measurement for the client was carried out by surveyors, or clerks-of-the-works.26 Given the risk to their profit, contractors hired their own measurers too.27 Guidebooks to effective measurement were available to accompany the guides to building and design that appeared in the eighteenth century. Indeed, a number of guides to measurement were being published from the 1660s onwards: The Carpenter’s Rule,28 The Mechanic Exercises,29 The Marrow of Measuring30 all give practical guidelines for how to measure and evaluate and pay for building works of all kinds accurately.31 Measurers were surveyors, but it seems that there were independent measurers, who did not engage in design or building alongside as early as the 1660s, and probably before.32 For the client, the benefit of working through a measured contract was that the system was essentially self-monitoring.33 The risk of having bills discounted ensured quality workmanship from the contractor, and this self-enforcement was made possible and reiterated by the possibility and ensuing process of contracting further work. In breaking up large contracts or tasks into measured pieces the contracts also allowed the transference of information between the parties about performance and enforcement costs.34 The contractor working under a measured contract

25Nisbet,

A Proper Price, p. 1. pp. 24–26. 27Ibid., p. 26. 28Darling, The Carpenter’s Rule. 29Moxon, The Mechanic Exercises. 30Venterus Mandey, Mellificum Mensionis, or the Marrow of Measuring. 31Occasionally contractors well known for their skills were measurers. Ayres, Building the Georgian City, pp. 4, 5, 13, 120, 35–37, 90, 218, 237. 32Nisbet, A Proper Price, pp. 1–2. 33The “the most likely and indeed empirically observable state in which contacts are self-enforcing is that in which the parties to the exchange have a great deal of knowledge about each other and are involved in repeat dealings.” North, Institutions, Institutional Change and Economic Performance, Chap. 7, p. 55. 34See North 1993, pp. 32–33 and 36–39. 26Ibid.,

4  Contracts and Ways of Working in the Building Trades     87

was incentivised to drive down the costs of inputs, but only to the point at which quality could still be assured. If the contractor found they could not keep to the terms of the contract and still profit, then theoretically they could choose not to complete further measures—essentially stopping a loss that would not be possible if they were tied into a contract by the great, or task. This categorisation of day work, measured work and work by the great is but a guide, however, because contracts came in many variants and with multiple idiosyncrasies. In fact, in the most instructive review of working methods to date, James Campbell has identified six types of contracts: 1. Day rates: the client paying a rate per person per day on site. All materials paid for by the employer. 2. By measure without materials: the contractor to supply labour, the materials supplied by the client, payment by measurement at preagreed rates at intervals or on completion or later. 3. By measure finding all materials: as above, but the contractor supplying all the materials and labour. 4. By measure and by day: with the contractor providing materials and labour, and the client paying a small amount to the workers directly, and on completion that amount being deducted from the measured total. 5. By the great without materials: the contractor supplying labour only, with payment in stages or on completion. 6. By great with labour and materials: contractor supplying all, with payment in stages or on completion.35 For the client, the benefits of measuring (2, 3, 4 above) in managing risk and monitoring performance can be clearly perceived. In (1) above the client bore the search and bargaining costs of obtaining materials, and they faced a skill-matching risk that the day workers they hired may not be up to scratch, so they risked paying out without having any means of ensuring quality, or the costs of suitable monitoring such as

35Campbell,

‘The Finances of the Carpenter in England 1660–1710: A Case Study on the Implications of the Change from Craft to Designer-Based Construction’, pp. 331–332.

88     J. Z. Stephenson

a foreman. The costs of obtaining materials might have been advantageous if the workmen or contractor could not be trusted to bring the best deal, or if materials were available to the client but not their suppliers, however. In (2) above the risks of the skill matching were mitigated, the costs or benefits of sourcing materials were still present, and the contractor did not get paid until the work was done and checked. In essence the contractor or day workers bore the cost of the skill-matching problem, and they would also have borne the cost of the delay in payment arising from the workmanship being checked, verified, and then paid for. One would expect that prices charged would reflect this. In (3), the classic ‘by the measure’ model, the client wholly contracted out the construction work, and contracted out all risk of it too, but the system of measurement monitored and protected him from opportunism or poor workmanship. The contractor therefore bore all the costs of information, searching, bargaining, and the risk of production, and probably credit as they paid suppliers and workers before the client finally paid out after measuring, and processing the bill. It should be expected that the prices that the contractor would quote and charge for such services would be higher than if none of this risk was present. Approach (4) allowed the client to ensure the contractor did not skimp on skill, by paying day workers too little. (5) allowed a client who had access to his own materials, and required known and specific high or very low skill to carry out specific works.36 Contract form (6) would only be a good contract if the contractor was well known to the client and could be trusted to be sufficiently capitalised to endure credit and pay suppliers, and was sufficiently experienced to estimate and strike a bargain accurately. The six types (and there were probably subsets and variations even of these) strongly indicate that contracting for high-value building work was the product of not just advanced engineering and aesthetic design, but an advanced set of organisational techniques and process design. Both clients and contractors thought in a way that we would think of

36For a description of such work see Beard, Craftsmen and Interior Decoration in England 1660–1820, pp. 14–16, 115–165.

4  Contracts and Ways of Working in the Building Trades     89

as ‘strategic’ about the best way to design the process of building which would affect the best or most profitable outcome.

Contracts To grasp the effects of this strategic thinking about building contracts more fully, it is possible to look in greater detail at the actual contracts agreed between clients and contractors. How did the oft-quoted “three ways of working” translate into contracts and what other characteristics do we observe? Historians who have studied construction have concluded that contracting became more uniform over time and that the system evolved into one of “measure and value”, which was universal on public works by the mid-eighteenth century.37 It should be said that there is not a huge body of extant sources to base analysis of this on. At St Paul’s Cathedral there are both contracts and minutes to review, and at Westminster Abbey the minutes and correspondence allow us to infer contracts. At London Bridge the contract was inherent in the position purchased. At the Office of the King’s Works all appointments and contracts were governed by Board along similar rules.38 In the networks run by Wren, his protégé’s and the Office of the King’s Works generally there was a clear predilection for working with contractors and suppliers already known or worked with before, reducing the costs of transacting by drawing on relational capital.39 The resultant relationships had features of a form of bilateral monopoly (which is discussed as a feature of construction contracting everywhere).40 Whilst it was not unusual for a contractor to enter a bond 37Ibid.,

p. 16; Nisbet, A Proper Price, pp. 26–27, 29–31. Mordaunt Crook, Chap. 6, in Colvin et al., The History of the King’s Works, Vol. V, 1660– 1782, pp. 67–117. 39Elfenbein and Zenger, ‘What Is a Relationship Worth? Repeated Exchange and the Development and Deployment of Relational Capital’; Ariño, de La Torre, and Ring, ‘Relational Quality: Managing Trust in Corporate Alliances’. 40Vrijhoef, ‘Supply Chain Integration in the Building Industry’, pp. 61–63 summarises this from a TCE perspective. 38See

90     J. Z. Stephenson

to secure a contract, particularly if it was his first contract with a client, (for instance Edward Stanton put up £2000 to become master masons to Westminster Abbey after 1719) such a mechanism should not give the impression that this was an entirely ‘open’ market.41 James Campbell posited that as working relationships developed, Wren’s contracts with his carpenters became simpler, and less was iterated, more trusted.42 This suggests features of clan-type relationship between contractors and consultants.43 We might expect that among parties that had previous contracting experience, terms would be more accurate, and indeed contracts in Wren networks are predominantly simple in form. They state the work to be carried out, the payment and measurement terms and any other expectations. They can be highly detailed in specifying quantities, quality, and other tasks or obligations they wish to pass to the contractor, but they do not specify any means for resolution of dispute other than withholding payment. Presumably informal arrangements, working practices and norms enabled these to be reasonable expectations. Not all contracting followed the simplicity of the system adhered to by Wren and his protégés, subsequently at Office of the King’s Works. Contracts for the building of the Mansion House in the City in the 1730s show a different approach, with a high specificity in listing requirements, and long recitals of expectations of obligations.44 Records are not able to tell us if such complete contracts performed or were renegotiated; the case is highlighted here to consider that outside of the networks of those working for the Office of the King’s Works or Crown contracting may have been more costly. The examples given below however, suggest a level of management skill and experience that accord with the quality of the buildings; if this was a function of the experience and shared norms of the actors rather than common practice, then the costs of doing business outside this network may have been higher. 41Westminster

Abbey Muniments 34517, October 11, 1722. ‘The Finances of the Carpenter in England 1660–1710: A Case Study on the Implications of the Change from Craft to Designer-Based Construction’, p. 330. 43Reve and Levitt, ‘Organization and Governance in Construction’. 44LMA COL/CP/02/148, Mansion House Box 8.4 Workmen’s contracts 1737–1739. 42Campbell,

4  Contracts and Ways of Working in the Building Trades     91

Taking any conclusions about the development in contract specificity, accuracy, or efficiencies any further would require further research with a larger universe of contracts. At St Paul’s, and at Greenwich Hospital, two large Wren projects, the written contracts show clearly how measurement, or risk management was put at the centre of the arrangement. Contracts gave the client the right to monitor work, to appoint an inspector of work externally if need be, and to discount, or not pay the full amount, if the quality they had stipulated was not met.45 However, it is also striking that contracts for several parts of a project might be written, specifying the contractor’s duties to supply work at specific prices until the building was completed. As St Paul’s contracts would be renewed frequently, or passed on to a subsequent contractor. This gave clients a further mechanism to discipline contractors, by turning single large projects into situations in which the need to retain trust across repeated transactions offered a counterweight to opportunism.46 The key elements of a typical contract as they were written in the period can be observed in one dated July 1675, in which Joshua Marshall undertook to lay the foundations of the new St Paul’s Cathedral. The contract specifies the stone to use, the dimensions, and that Marshall would be paid by measurement, and would be able to claim funds: when one hundredth Rod is wrought, & ye same certified by Mr Edward Woodruff & allowed by Sir Christopher Wren, Knt, so much lawful English money as ye said one hundred Rods of Work shall amount to after ye Rates foresaid, and so proportionally for every one hundred Rods of Work, as ye work goes on until measurement be made of the whole, & when the said severall works shall be well and sufficiently done, & approved of, then the said J.M., his Exors, or Assigns, shall receive so much more as the said Work upon a Just measurement thereof had made & shall justly amount to…47

45For

a surviving contract, see: LMA CLC, MS00233. Building St Paul’s. 47Wren Society, Volume XIII, p. 15. 46Campbell,

92     J. Z. Stephenson

The degree to which the authority in measured contracts was put in the hands of the client’s agents in this contract was by no means unusual. In general, the detail of the stipulations is a prevailing characteristic of the contracts. The contract with Thomas Gilbert and Thomas Wise to provide the Portland Stone to the Cathedral stipulates that they: at their own proper cost and charges, for and in consideration of the rates and prices hereafter mentioned, raise, scaple, prepare, and cause to be delivered on board such vessels as they can procure to take ye freight such and so much Portland Stone, as they shall be directed to prepare & send from the Isle of Portland to such Wharfe in the Port of London below Bridge as shall be appointed. And shall scappell the said stone according to the moulds and directions both for the fashion and the number of stones […] and give account from time to time of the Marks and Measures of the said stone therein […] and shall take care that all Stone by truelly scappelled, not wanting of ye moulds at the corners or sides, also that the said stone be well conditioned and proper for use intended, Without flinty beds or rag beds or clay holes near the faces of the stone…48

In March 1685 it was noted that Gilbert and Wise were billing the Cathedral for “mending of wayes, for Crain ropes, and other such like Charges”. The commissioners were of the opinion that they were not to be charged, that these costs were to be borne by the contractor and this was humbly submitted to by the contractors.49 At Greenwich Hospital a similar approach can be seen in a contract which clearly specified monitoring and discounting systems from the outset. On June 5, 1696 the Hospital Fabric Committee recorded that “agreement was made with Daniel Foe of the Parish of Islington, Brick maker” to “burn and supply” from time to time as the work shall require

48Wren 49Wren

Society, Volume XVI, pp. 15, 19. Society, Volume XIII, p. 51.

4  Contracts and Ways of Working in the Building Trades     93

…bricks, delivering the same to the wharf near the workes, and if a considerable part of any load… appear to be clinkered or otherwise unprofitable for the use of the work it shall be lawfull for the clarke of the works to reject or turn back such ill goods, 14 shilling the thousand for stock and 25 shillings the thousand for rubbing bricks contract in force for this summer only.50

Presumably everyone was clear about when the season ended (probably Michaelmas). The same day Thomas Hues and Richard Billinghurst, bricklayers, agreed a price of £17s per rod for new brickwork: “Rubbing and setting the straight arches being brick and half deep and one brick thick twelve pence a foot running and they shall find all workmanship making up of mortar well-tempered mixed and beaten and all scaffolding and towards scaffolding they are to be allowed £15”.51 Then on June 13 the first masons contract was made with Thomas Hill and Edward Strong “to perform that masons worke […] and the said masons do agree to keep as many workmen and labourers as the surveyor of the said work shall think reasonable for the carrying on the same with the brickwork so that they not be obstructed”.52 It is not known when Thomas Hill stepped away from the partnership, but by May 1698 Strong was contracting alone for the stone work at the wharf alongside the site.53 There was also contract with James Grove, carpenter, “that the said James Grove by himselfe servants and workmen finding all materials, workmanship and labour will performe and finish in goode, such carpenter work as shall be directed by the surveyor of the said work”.54

50TNA

ADM 80/ 2, p. 3. The London Prices of Brick layers Materials and Works, both of new building and repairs &c. Batty Langley, London, New Head, 1750 (2nd Ed.) gives the price of a thousand stock bricks as 18s. delivered, p. 11 (It also gives a retail calculation for prices to yield 25 per cent margin). McKellar found brick prices at c. 11s. per thousand for stock bricks and 20s. for rubbing bricks in the 1680s. 51TNA ADM / 80/2, p. 5. 52Ibid., p. 9. 53Ibid., p. 27. 54Ibid., p. 11; Grove is the subject of Campbell’s case study and calculations of contractor profit margin, ‘The Finances of the Carpenter in England 1660–1710: A Case Study on the Implications of the Change from Craft to Designer-Based Construction’.

94     J. Z. Stephenson

The predominant approach to contracting for building projects seems to have been for institutions to write detailed contracts for set amounts of work, and then to either renew them or add further contracts as the project moved on to new parts of the work. The records of St Paul’s show that sometimes contracts were inherited or passed from contractor to another. For example, Edward Pearce succeeded Joshua Marshall’s contract, Thomas Wise Junior and Thomas Hill succeeded Thomas Wise senior’s contract, Nathanial Rawlins succeeded Joshua Latham’s contract, William Kempster assumed John Thompson’s in 1700.55 But individual contractors also had several contracts in force at any one time for one client. Where a contract needed to be limited, it was, as with Foe’s Greenwich contract above. A detailed example, from a little later at Greenwich when Strong was contracting solely implies that the parties has both trust and experience in each other, but that risks could also be highly specified: Agreement was then made with Edward Strong Snr And Edward Strong Jnr, Masons as follows. The said Masons in consideration of such sums of money to be paid them by the Treas. Of Greenwich Hospital as that become due to them from time to time upon the several admeasurements of their work and stating the accounts for the same according to the rates and prices hereafter and according to the Rules and practice of the said Hospital, do for themselves […] and assigns covenant and agree with the said Directors to do execute and perform such Masons works in the Staircase in K. Charles his building, and pavings, chimney hearths, window soyles, and steps in the collonade and buildings adjoining to the new road and Fryers Row, as shall be be ordered from the said Directors by the Clerk of the Works and that they will perform the same well and workmanlike according to their deft skill and ability finding all materials of stone, Lime sand, making of mortar, scaffolding, hoisting, lifting and all other necessary relating thereto except iron work which shall be found by the Hospital. The prices to be as follows.

55Wren

Society XVI, pp. xiv–xvii.

4  Contracts and Ways of Working in the Building Trades     95

The contract then states the price for worked Portland stone, different marbles, paving and other specific works. It ends with a remarkable undertaking on the part of the contractors to allow their work to be evaluated by the Hospital without limit to the possible loss or deduction that might arise. And the said Masons do further agree that if any part of the work when performed shall appear to be deficient either in good [—–isfs] of materials performance of quantity they will submit to such [?] action as Sir Chris Wren and the Directors for the said Hospital shall think fit. In with & whereof the said Edward Strong Snr and Edward Strong Jun. have hereunto set their hand this 10th day of October 1706. 56

Both parties to the contract were experienced large scale builders, and there is no reason to think that this clause was in any unusual for either party. The case highlights the predominant theme of all contracts from the period: they specify perceived possible risks and costs to the client in the process of building, and they put those risks firmly in the hands of the contractor. Gilbert and Wise took on the responsibility of the (hugely problematic) transport of Portland stone, and failed in their attempt to get their clients to absorb the costs.57 Strong and Hill were to absorb their own costs in cooperating with another contractors’ work or schedule. Marshall was to await payment for work done until substantial portions were completed and then submit to quality checks before it could be approved. The Strongs agreed to a contract at Greenwich in 1706 which left them open to deductions on agreed rates for work already done, with no protective clauses to limit loss, nor ability to appeal. Nor did they know exactly when they would get paid. It is likely that only persons with large capital, strong networks and established supply chains could bear such risks, but when they successfully did, they profited.

56As

copied in Strong’s account book LMA CLC 227 /15 MS 20233. Building St Paul’s, pp. 91–95.

57Campbell,

96     J. Z. Stephenson

Contracting Practice How did the contracting process for major projects work in practice? For ‘extraordinary’ work or new building, the client and the surveyor or clerk-of-the-works agreed the requirements for the job. The client, in the case of the new City Churches, or Greenwich Hospital, or St Paul’s were commissioners, appointed by Parliament, crown or city to administer the new building.58 Once surveyors had been appointed and designs agreed, the building requirement was then posted at gates or submitted for publication to the London Gazette.59 Contractors would submit their tenders or estimates and the surveyor would meet with those that could meet the requirements.60 Meetings and presentations with commissioners or clients also took place depending on the scope of the work. Case studies of St Martin’s in the Field, or St Paul’s show that the appointment of contractors was frequently a long drawn-out process, involving meetings with a range of stakeholders.61 To succeed in securing a contract, a contractor had usually to submit the lowest price—or near lowest, but also had to demonstrate to commissioners that they could withstand the payment terms.62 This is not to say that the process always ran in the best interests of contractors or clients as the case of the railings at St Paul’s in 1710/11 attests, but the process was designed as competitive and set up in such a manner that all aspects of proposals could be considered.63 The responsibility for hiring and

58See Reddaway ‘Rebuilding’ for the formation of the Commission for the rebuilding of the city after the Fire, pp. 55–56; see Campbell, Building St Paul’s, pp. 18–21. 59Yeomans, ‘Managing Eighteenth-Century Building’, p. 11. Descriptions of the process survive in: TNA ADM 67/4 (1697 description of the Greenwich Hospital Commission’s instructions to Nicholas Hawksmoor for the procurement of masons services); Portland papers, Volume X (the appointment of contractors in 1710 at St Paul’s). This is not to say that contracts were not handed directly on occasion to specific contractors. 60Records of deliberations include: ADM 67/4, Westminster Abbey Muniments no. 34516. 61See Ayres, Building the Georgian City, Appendix II; Campbell, Building St Paul’s, Chaps. 2, 6, 8, 9, 11. 62Yeomans, ‘Managing Eighteenth-Century Building’, p. 11. 63See Campbell, Building St Paul’s, Chap. 11, Frauds & Abuses for the railings case.

4  Contracts and Ways of Working in the Building Trades     97

managing labour rested solely with the contractors who were awarded the contract.64 Once an estimate was agreed and work had started the contractor submitted bills for each measure or part of work, quarterly or biannually as agreed. At Westminster Abbey payment was bi-annual for most of the eighteenth century. At St Paul’s there was a monthly reckoning and payment system. The paymaster or treasurer would receive the bill and pass it to the surveyor or the clerk-of-the-works for approval. To approve or ‘pass’ a measured bill, the clerk had to ensure the work met the specification by appointing a surveyor to measure and evaluate the work done.65 Once assessed or measured, the bill was passed or discounted and went back to the treasurer to be signed and eventually to the paymaster to be paid. Some contracts enabled contractors to draw down small amounts of money by way of imprest between quarters. Delays in payment were endemic, and expected.66 Late payment was in fact part and parcel of the system and contracts involved. Colvin assessed that some contractors were out of pocket for decades and that accounts were never settled within two years for the latter part of the 1600s.67 To contemporary eyes such tardiness in payment seems hugely problematic but, it is well documented that credit and currency had complementary relationships in the early modern world.68 With a lack of specie, a constant problem in the late seventeenth century, the Crown was well known as a late payer for everything, not just building work. In fact, most large institutions were. As Colvin pointed out the wonder is not that everything relied on credit, the wonder is that the Crown 64Campbell,

‘Seventeenth-Century Bricklayers’ Contracts’. measure it himself if he were so qualified such as Hawksmoor at Greenwich. The final account usually states a number of dates: the rough dates the work was completed; the date the work was measured by the Clerk; the dates the work was approved and the bill passed by Committee; the date the sum was paid. See Westminster Abbey Muniments 34513 for dates of Dickinson’s measurement and latter dates for passing of bills. 66Colvin, ‘Financial Stress’, pp. 34–44; Colvin et al., A History of the Kings Works, Volume V. 67Ibid., p. 44. 68See Lucassen (ed.), Wages and Currency: Global Comparisons from Antiquity to the Twentieth Century; Muldrew and King, ‘Cash Wages and the Economy of Makeshifts in England 1650– 1800’; Muldrew, The Economy of Obligation: The Culture of Credit and Social Relations in Early Modern England. 65Or

98     J. Z. Stephenson

maintained its credit. The costs of the credit are discussed in the next chapter. What is relevant to the bargaining of contracts is that contractors knew and understood the system of late payment, so they would have accounted for it in the prices they charged. The costs of supplying and dealing with late payers was ‘in the price’. The role of the clerk-of-the-works who acted as the on-site surveyor and manager of all contracts was thus a critical one in this system.69 Yet, the role was one that varied by institution and client. Clerks-ofthe-works were usually skilled and experienced men, but that skill and experience varied. Nicholas Hawksmoor, one of Wren’s most talented protégés was clerk at Greenwich, where he procured contractors and drew up contracts himself.70 But Ralph Sims, who is sometimes described as clerk-of-the-works of Westminster, is not known as surveyor (but he presumably had some knowledge of the trades as Sarah Spoore who was paid for iron work, is described as his mother in notes in the accounts). Westminster Abbey procured the services of another of Wren protégés, William Dickinson, to act as surveyor and measurer.71 At St Martin’s in the Fields a clerk was not appointed until work was underway and contractors already on site.72 In ‘ordinary’ or maintenance work the process was streamlined somewhat, and payments were usually made regularly for labour.73 At sites managed by the Office of The Kings Works after 1714 there were clerks-of-the -works responsible for making regular payments. At Bridge House, a role similar to clerkof-the-works was undertaken by the purveyor, who paid regular bills weekly.

69Ayres,

Building The Georgian City, p. 34. ADM 67/2. 71Westminster Abbey Muniments. 34513. Dickinson signed the accounts as bills examined. 72Ayres, Building the Georgian City, p. 236. 73Bridge House contractors billed weekly for labour costs, and payment was made weekly assuming their account was not unaudited. CLA/007/FN/03/022, CLA/007/FN/03/019/A, CLA/007/ FN/03/027, by way of example. 70TNA

4  Contracts and Ways of Working in the Building Trades     99

The System in Practice and Its Influence on ‘Wages’ Against this background, it is easy to appreciate the strategic merit of contracting by the measure. Measurement was a vital tool in being able to transact highly idiosyncratic, financially costly work with a high percentage of sunk costs at outset, and it worked for both parties. The advantages for clients are clear: measurement allowed them to make terms, monitor progress, reduce negligence and agency problems, and pass many of the transaction, search and operational costs onto the contractor. Thus, the system of measurement could provide an effective monitor for the client. Although it seems like contractors took on all the risk, and covered this in their margin alone, there were also upsides to measurement for contractors. For contractors a measured contract fixed an upper bound of profitability on any one job because if the price of the goods or inputs were fixed, profit could not exceed the difference between those prices at supply and those at sale, less overhead, labour, financing and operating costs. One priced a measured contract to ensure enough margin to operate and profit either immediately, or in the long run through future contracts when relations were established. But a measured contract also had the potential to fix an upper bound on risk—a sort of stop-loss, because the potential loss was limited to the scope of that measurement alone. If one measurement lost the contractor a large sum he could adapt his prices or his measures or his inputs in the next one, or simply rescind his part of the contract. Thus, for contractors, measurement was a risk-managing strategy in the long run. It limited sunk costs. In agreeing to a measured contract the contractor would have to know that he could manage to get the inputs at a price low enough to make the measurement profitable, or less. He would also have to calculate his costs of meeting the contracts—which included his costs of measuring. Measurement was not a ‘customary’ way to account for building work, it was an advanced management tool to minimise risk and assure quality, with a minimum of monitoring.

100     J. Z. Stephenson

Measurement, of course, entailed certain costs to operate. Firstly, both parties had to be able to accurately estimate, evaluate, measure, and issue bills in the same units of account. Both parties might employ surveyors and measurers, and both sides needed someone who could estimate costs and maintain accounts.74 The contractor’s costs of working by the measure were search costs to find inputs at the right price (owning one’s own quarry minimised some of these), estimating costs, which equated to the skills of an experienced and accurate surveyor to measure and price, and accounting costs to account for and bill accurately. All of these things are highly skilled labour inputs, some of which the contractor would have done himself, some of which would have been outsourced to agents or factors. Labour costs were priced and presented differently in each of the contracting methods in use in this period. Crucially, when looking at the accounts of measured work or work by the great or task, labour costs cannot be observed or determined; only work by the day gives any insight into day rates—but those days rates include the margin that the contractor took to cover his costs of managing the work. What proportion of work was carried out under each type of contract? As Wren did not favour the system there is no major work ‘by the great’ found for any of the major projects I have examined here. However, there was plenty of task work for small project amounts. For instance Thomas Strong erected the screen wall around St Paul’s site with rubble on site for the fixed price of £25.75 Fixed prices were also agreed for statuary by piece for instance.76 Task work at smaller prices and for well-defined products would have minimised the risks of misestimating or non-performance. Where task work is found in the accounts, the amount of labour and its cost is not described in any way.77

74Nisbet, A Proper Price, p. 26, believes that practice was for both sides to deploy measurers by 1750. 75Wren Society, Volume XVI, p. 16. 76Lang, St Pauls, p. 81. 77Wren Society, Building Accounts Jan–June, 1697.

4  Contracts and Ways of Working in the Building Trades     101

The majority of work at St Paul’s, Westminster Abbey, or Greenwich, was agreed by the measure, but it was common everywhere. A typical measured bill is shown by one that Thomas Knight, paviour, submitted to the city in 1675: Thomas Knight Pavior his bill for worke done at Holborn Bridge November the 12th 1675.

For 315 yards of paving at 2d ½ per yard For 16 loads of stones at 7s. per load

£03: 05:7 ½ £05: 12: 00 £08: 17: 7½78

The bill was signed by the City Surveyor, Nicholas Duncombe, and the acquittance acknowledged by Knight below that. There are no day rates given for the labour employed on the job, although this is explicitly for ‘worke done’, as well as materials delivered, rather ‘worke’ is costed on a per-yard basis. The labour costs of paving the 315 yards are incorporated within the price per yard. In this period, Knight held the position of City Mason, but he was also an agent for St Paul’s Cathedral to the Isle of Portland, and was simulatneously contracting with the Office of the King’s Works. It seems safe to say he was not laying the stones himself, and we have no idea of what he paid to those who did. It is only day bills that include the figures that we have previously interpreted as wage rates. A day bill from Edward Strong at Greenwich Hospital (1699–1709) offers a typical example of a bill for ‘day work’. December 1700 Days work by Edward Strong pulling down the scaffolding made to raise the pedestals and trophies upon the pediment of the B[…], in cutting way into the brickwork for the corridor frames chimney pieces corner stows, windows, soyles and window stones and taking out the iron barrs of the cellar windows in the KC his building aforesaid and cutting way for the top of four staircases descending from the first floor of the B[…] the cellars or vaults of the same. 78LMA

COL CC MRK /02/01/1.

102     J. Z. Stephenson

Wm. Loggans 20 days at 2s 6d. Jb. Adams 25 days at 2s6d. Richard Nailer 8 days –

02:10:00 03:02:06 01:00:00

Edward James 20 days at 20d. 01:13:04 Wm. Derry 10 — 00:16:08 Wm. Hosslet 1 day 00:01:08 J How—1 day 00:01:08 Matt Allen 2 and half days 00:04:02 Wm. Macon 1 00:01:08 09:11:0479 Four characteristics of such day bills need to be understood if the ‘wage’ rates they report are to be interpreted correctly. First, they only indicate some of the labour employed on a project. On ‘extraordinary’ or new build sites the proportion of work carried out by day was small. Day work was approximately only 10% of the masons’ bills at Westminster Abbey in 1712–1713, zero the year after, and 30% in 1714–1715, but less than 5% in 1718.80 Only carpenters on that project had the majority of work contracted for as day work; days formed the bulk of their billable work except in 1714–1715 when there was also no day work for them either.81 Likewise, at Greenwich Hospital, Edward Strong’s bill book shows day work as approximately 15% of the work.82 Second, day bills show the price the contractor charged the client for a person’s work for a day. They are not a record of what the worker actually received. That is because, thirdly, day bills contain no separate line for the contractor’s own overheads or profit, nor do they include the labour costs of those supporting or assisting such as carters, messengers and labourers not directly billable. And finally, day bills were used for only particular kinds of work. Day work was the contract type used

79LMA CLC/B/227/ -115 MS00233. The calculations seem to be incorrect by 4d; in the books there are a number of corrections. 80Westminster Abbey Muniments, Christopher Wren Fabric book, 34513. 81Westminster Abbey Muniments 34153, 34514. 82Calculated from the bill book of Edward Strong, LMA CLC B/227/175.

4  Contracts and Ways of Working in the Building Trades     103

for jobs which could not be successfully estimated or measured, or were risky. It seems possible that a key benefit of day work was that if more or fewer men were needed, the contractor could adjust the numbers in consultation with the surveyor or clerk-of-the-works.83 Appreciating the specific role of day bills may also explain the puzzling uniformity of wage rates found in account books and bills. Rather than a ‘custom wage’ as has been a previous interpretation, the contracting context suggests that a day rate was a price set by institutions for kind of service, and the persistence of the rate was a product of contractors colluding or competing for a ‘going rate’ or acknowledged trade price. Institutions may have given guidance or set rates after comparing estimates. Nicholas Hawksmoor, surveyor and clerk-of-the-works at Greenwich was asked to conduct a review of prices for advising on the renewal of contracts in 1698.84 In fact, as we will see in Chap. 6 surviving records of direct payments show little sign of a customary wage for craftsmen, rather there is much variation, presumably for skill and productivity. How did contractors and clients determine what was billed by the day and what by task or measure? One might assume day work was for specialist skill, but it is striking that most day bills do not detail highly crafted carving or intricate levels of ashlar smoothness. Day work records describe the conditions under which work was carried out. By example: in November 1675 at St Paul’s Thomas Strong charged out four men for 50 days in total at 2s. 6d. per day “cutting holes in wall of north gable for scaffolding”. This is not especially skilled work, except that it was performed in the winter, probably at height, and the cutting tools would have had to be carried up to do so.85 At Greenwich Hospital, we can see further examples of work by the day that share this characteristic:

83Lang,

St Paul’s, p. 81. this and other evidence of institutions bargaining with suppliers, see: ADM 67/ 2. Wren Society, Volume XV, p. 51. 85Wren Society, Volume XVI, p. 73. 84For

104     J. Z. Stephenson

Days work pulling down the scaffolding made to raise the pedistals and trophies upon the pediment of the dass wing in cutting way in to the brickwork for the window frames. Chimney pieces corner stows window soyles and window stons and taking out the iron barrs of the cellar windows in KC building aforesaid and cutting way for the stops of four staircases descending from the first floor of the Bass wing to the cellars or vaults of the same

This was recorded in Dec 1700.86 Day contracts were also used when one wanted to eliminate any incentives to cut corners. According to Campbell it was only the safety considerations of careless underpaid rushed carpenters making scaffolding for carvers and masons to work at height that forced such a large amount of day work onto the accounts at St Paul’s.87 Indeed, it wasn’t the highest skilled or most specialist work which was charged by the day. Most of the carving and specialist work at Paul’s was done by measure as the Wren Society’s edition of all accounts in chronological order makes clear.88 For instance, Edward Strong’s bills in the St Paul’s accounts of 1691 record: For masons worke […] for worke and sett 480 ½ ft of outside Circle P Stone in round columns and rustiq. at 18d ft £36 0s 9d. for worke and sett 55ft circle plinth P stone upon attic cornice17 ½ d. foot sup. £58 15s 8 ½ d for carve 52 ft run large leaves in heads of […] windows10s ft run £26 0s 0d.89

86LMA

CLC/B/227-175. Society, XVI, p. 151, quoted in Campbell, ‘The Finances of the Carpenter in England 1660–1710: A Case Study on the Implications of the Change from Craft to Designer-Based Construction’, pp. 332, 337. Campbell is of the opinion that the purpose of giving day rates was to ensure adequate numbers of men on site. 88Wren Society, XXVI, p. xv. 89Wren Society, Volume XIII building accounts Nov, Dec 1691, p. 91. 87Wren

4  Contracts and Ways of Working in the Building Trades     105

On the other hand, day rates, or time rates seem to have been the norm for much maintenance work. At Bridge House all contracting carpenters and masons were paid day or tide rates for their workers, the number of days or tides determining pay. At the Office of King’s Works day work was also common on short-term work until 1668.90 In both these places bills were submitted regularly, so if there was a problem with the number of days this could be dealt with. Significantly, it is possible to see that consideration of the relative cost and risks of the ways of working seem to have determined the contract type. Day contracts were used when the benefits of monitoring for the client were worth more than the risk of potential loss through non productivity, or lazy workmen. Working by the day had disadvantages for contractors too. They had to submit to client monitoring so that accurate day numbers could be billed. Unless they were taking a substantial margin on day wages they would not make any money out of day contracts, so did they avoid them? Their assessment may have been that the benefit in day work was that you did not lose money on work that was risky, hard to estimate at the outset, or dangerous and hard to get men to do. The new evidence in Chap. 6 suggests that contractors employed men by the day even when they were working on measured contracts. In failing to understand the contracts and organisation of building work of the early modern period economic historians have mistaken components of contract pricing for ‘wages’. The scale of the misapprehension for estimates of ‘average’ or ‘real’ wages shall be discussed later. However, the misapprehension has wider significance. The evidence shows that seventeenth-century construction firms understood clearly what we would term the transaction costs of large-scale building investment and construction processes and the costs of transacting itself. The market was not one governed by custom and unexplained institutions of tradition, it was one in which businessmen and women understood the costs of information, of relations, and of finance and priced their services accordingly. In some ways Sir Christopher Wren’s management

90See

below, Chaps. 5, 7.

106     J. Z. Stephenson

ideas anticipate much that preoccupied economists such as Oliver Williamson three centuries later. If “two types of transaction for which specialised governance structures are commonly devised are recurring transactions supported by investments of the mixed and highly specific kinds”,91 then the governance response of early modern builders anticipated the costs, and indeed many of the features of such governance have persisted until today.92 Seventeenth- and eighteenth-century contracting systems were purposeful organisational and contractual arrangements that apportioned and priced risk, minimised monitoring costs, mitigated hold-up problems and the effects of bilateral monopoly, and provided finance.

91Williamson, The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting, p. 75. 92Eccles, ‘The Quasifirm in the Construction Industry’ Richard Fulford and Craig Standing, ‘Construction Industry Productivity and the Potential for Collaborative Practice’.

5 What did Bosses (in London Construction) do?

‘What do bosses do?’ Harvard economist Steve Marglin asked in 1974, in an oft-cited paper about the roots of modern capitalism.1 Marglin’s key question was, in the process of industrial development, why did the producer lose control of production, or why does capital hire and manage labour and not the other way around? He contended that the two stages in which this occurred, the development of specialisation and division of labour under the eighteenth century putting-out system, followed by centralised monitored workplaces under the nineteenth century factory system, were driven by society rather than technology, not by technological innovation, but by the desire of capital to take the lion’s share of profits.2 This question, and debate about processes of pre-industrial management are highly relevant to understanding organisation of the building industry in the long eighteenth century. As will be apparent by now, large institutions chose to pay middlemen, contractors, or entrepreneurs, a significant share of scarce public funds in order not to have to

1Marglin, 2Ibid.,

‘What Do Bosses Do? The Origins and Functions of Hierarchy in Capitalist Production’. p. 62.

© The Author(s) 2020 J. Z. Stephenson, Contracts and Pay, Palgrave Studies in Economic History, https://doi.org/10.1007/978-3-319-57508-7_5

107

108     J. Z. Stephenson

manage construction. Indeed, Pollard asserted that subcontracting in the eighteenth century was essentially a case of capital choosing a convenient substitution for management or monitoring.3 Why did the commissioners at St Paul’s not employ all crafts people directly and avoid contractors pocketing profit? Why did the City pay position holders to manage the building work on the Bridge and not assess it themselves and hire the relevant skill to do so? The implication is that building management was difficult, costly or scarce, or could not be purchased in the market any other way. I have already shown that some contractors made a great deal of profit. What did they do for it? If the previous chapters explored the nature of the clients and ­contracts, this chapter aims to describe and analyse in further detail the organisational form and financial underpinnings of early modern English construction companies or contractors. The question has been examined before, for the later period of industrialisation most n ­ otably by Linda Clark. In Building Capitalism she contends that waged labour and the capitalist mode of production developed with new scale and scope in the building trades in the late eighteenth and early nineteenth century.4 Her analysis, like Marglin’s, is heavily Marxist in influence. This analysis may pose the same question, but it does not draw on the Marxist framework to analyse it, and it finds highly capitalised organisations and varied modes of production long before the industrial revolution. The starting point in this version of events is Knoop and Jones’s 1935 assertion that seventeenth century building contractors were akin to ‘firms’. By current definitions they may have been justified in this claim, but the boundaries between firm and market were very broad.5 A great deal of production was organised outside of the firm, in a manner we would think of as ‘disintegrated’, in a system of partnerships and subcontracts, and what could be described as the ‘firm’ was defined by the

3Pollard,

Genesis, pp. 39–40. Building Capitalism. 5Knoop and Jones, The London Mason in the Seventeenth Century, p. 47; Lamoreaux Raff and Temin, ‘Beyond Markets and Hierarchies’. 4Clark,

5  What did Bosses (in London Construction) do?     109

terms of the contract, with partners, agents, craftsmen, and suppliers participating solely for the parts of the contract needed, and without any residual claims. When these sorts of mechanisms are discussed in the economics literature the usual presumption is that they are modern phenomena,6 but early modern historians will recognise a project based structure to much commercial activity before 1750. One of the purposes of the analysis in this chapter is to clear up a long-running misunderstanding of institutional accounts and price and wage data from early modern sources: the question of margin. Previous economic historians have looked at accounts of institutions such as St Paul’s or Westminster Abbey, observed records such as “Masons, 5 men, 4 days @30d. per diem”, or “32 cauldrons of coales, £2 8s.0d.” and concluded that the price of a mason per day was 30d., or the price of a cauldron of coal was a shilling. But both ‘prices’ represent not just the goods and labour described, but the costs of getting those to the client and deploying them in the client’s organisation. In the case of the coals the price of a cauldron includes the carriage and delivery costs for water and road, including the merchants operating costs of finance, administration, billing, auditing and so on and his profit. In the case of the masons the price includes the master mason’s, or contractor’s costs, of searching the right skill, organising the associated team, tools, support, and the same finance billing, auditing and administration costs. The idea that the prices and wages recorded by previous historians have a margin in them is not new but it is generally disregarded. In relation to prices everyone knows that price series are gathered from institutional records, and it is taken that all final consumers would have paid some kind of margin to a retailer, but it is assumed that the prices of wholesale and retail goods would have trended the same way. Wage historians have mostly just accepted stated ‘day rates’ as paid wages.

6Efficient

use of such structures is sometimes treated within the economics literature as entirely modern, see for example Oliver Williamson, The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting, pp. 120–122. Similarly in pp. 232–233, Williamson’s historical examples highlight the problem of embezzlement discussed by Hudson in West Riding. For the modern construction sector see Gruneberg and Ive, The Economics of the Modern Construction Firm, pp. 180–191.

110     J. Z. Stephenson

The only author who has attempted to deal with the problem is Gregory Clark, who acknowledged that wages prior to 1815 should have some overhead removed. However, he only discounted 10%.7 Given that authors dealing solely with the construction industry in the nineteenth century have been explicit about mark up and profit in charge-out rates (London day-work rates in 1826 included a 20% allowance for overhead and profit) it is surprising that the question of margin has not been applied to wages more seriously before.8 There has been only one previous attempt to calculate the likely profit margin that seventeenth or eighteenth-century building contractors would have been taking on day wages—by James Campbell in 2005. In that study, Campbell examined the finances of carpenters on seventeenth-century King’s Works projects. He mostly examined day work, of the kind where the client was providing materials. He used the cases of James Grove, chief carpenter at Hampton Court 1689–1696 and Richard Jenings at St Paul’s Cathedral to explore how much profit contractors made. Campbell proposed that the contractors took 21–41% profit margin on labour.9 Although this seems large, he justified it on the basis of the large fortunes they accumulated. Campbell argued that mark-up on materials was not guaranteed to contractors because of the predilection of clients to source their own materials, but he also thought the costs of finance and administration were impossible to calculate.10 The only evidence Campbell had to substantiate his claims and calculation was the ‘Jenings case’. In this chapter, I begin with a discussion of the Jenings case, as this has been the natural starting point for any prior debate on margin in this period for economic historians; moreover, it provides a good summary on all that is known to date on the finances of the kinds of contractors and firms involved. I then set out and pursue a somewhat 7Gregory Clark, ‘The Condition of the Working Class in England, 1209–2004’, his main agricultural source; LSE Archive, Beveridge Price History D:3/4/5, E:7b/8/9/10/24a/25 F:1/8/9 records a mixture of direct and contracted prices for labour throughout this period. 8C. G. Powell, An Economic History of the British Building Industry, p. 33. 9Campbell, ‘The Finances of the Carpenter in England’, p. 334. 10Ibid., pp. 338, 360, in diabbato.

5  What did Bosses (in London Construction) do?     111

different approach, building an estimate for contractors’ overheads from the bottom up. I start this with an examination of their management costs, before moving on to the costs of credit, discounting and other costs. Taken cumulatively, my calculations align in many ways with Campbell’s estimate; contractors could easily justify a margin of 20 to 40% on bills. As will be discussed at the end of the chapter however, the likelihood is that the level of margin taken on jobs of differing kinds varied on a large scale. The chapter will also try to explore the nature of the ‘firms’ that operated under such conditions. This question is important because research into how organisational forms or coordination mechanisms in early modern England affected productivity or outcomes is rare.11 This is partly the result of a highly limited record set, and partly the effect of the confusing nature of the legal forms of business and incorporation available before the mid-nineteenth century.12 Mokyr’s view, that English business organisations were ‘simple’ structures and that eighteenth-century growth was the product of entrepreneurs, whose command of technical competence created innovation, is a summary of a general view that the organisation of industry before 1800 had little effect on productivity.13 Historians of management view the period before 1770 as the era of “personal capitalism” (as defined by Chandler) described as that of separation between ownership and control of a commercial enterprise, and a “general unwillingness to delegate responsibility to salaried managers”.14 The substitution for management highlighted by Pollard is subcontracting, frequently depicted by its reduction or avoidance of supervisory costs—essentially passing the risk and monitoring cost onto someone else.15 The lack of organisational form or strategy inherent

11Wilson

and Thompson, The Making of Modern Management, p. 4; Also see Berg, The Age of Manufactures, 1700–1820. 12Harris, Industrializing English Law, pp. 1–13, 37–38. 13Mokyr, The Enlightened Economy, pp. 357–358. 14Wilson and Thompson, The Making of Modern Management, pp. 6, 56 quoting Chandler, Scale and Scope, p. 235 and part III. 15Pollard, The Genesis of Modern Management, pp. 38–39, 47; Mokyr, The Enlightened Economy, p. 353.

112     J. Z. Stephenson

in personal capitalism is offered as a reason for a lack of growth in late nineteenth and early twentieth century Britain.16 In this light recent output figures for the late seventeenth century present a paradox where ‘un-managed’ seventeenth century firms increased total output by half, and doubled industrial output, in forty years.17 Before moving to a detailed analysis, and point obvious as to be missed, should be clarified. If contractors had not taken a margin on the labour and goods they supplied, they would not have been paid anything. The salary of a position holder at Bridge House was £4 per annum or about one-tenth of what current estimates are for a craftsman for a year. Elsewhere there was no allowance or salary. Only rarely did contractors get paid a day rate for themselves, and when that was the case it was usually as foreman, as at St Paul’s, or a nominal amount for special attendance to ensure the attendance of the contractor, yet these men accumulated fortunes and lived well. Edward Tuffnell charged 16d. per day “for mine owne attendance” for just some of the days at Westminster Abbey.18 Presumably, the 16d. a day—the same as winter rates at St Paul’s for labourers in the same period—was not the source of funding for his fine house on Millbank. If a contractor was selling work done on any contract type—with no day rates specified for themselves—then unless they built in a substantial margin, they had no way of being remunerated for their work. The way contractors made their own pay, and fortunes, was through marking up goods and services sold to clients.

The Jenings Case The Jenings case is the only case previously dealt with by historians on the question of the margin a contractor could extract from day wages. However, its implications have also been disputed. Elizabeth Gilboy herself referred to it in her discussion of contracts, dismissing the idea

16Wilson

and Thompson, The Making of Modern Management, p. 56. et al., British Economic Growth 1270–1870, pp. 238–444. 18Westminster Abbey Muniments 34513. 17Broadberry

5  What did Bosses (in London Construction) do?     113

that employers may have taken some form of mark-up. Whilst she floated the possibility that a master may have “shaved a penny or two off wages”, she decided that this was unlikely to have been standard practice.19 She referred to Jenings to support her assertion. The outline of the case is as follows. In 1710–1711 Richard Jenings, Wren’s chief carpenter contractor at St Paul’s from 1708, admitted to a committee of commissioners tasked with evaluating whether he was guilty of fraud, that he was paid 15s. a week for his carpenters, (or 2s. 6d. a day for a 6 day week) but he paid them on average only 11s. The correspondence generated by the case is considerable.20 The accusation of fraud against Jenings was part of a complex conflict over Wren’s role, a souring of relations between him and the commissioners at St Paul’s, and a contract for railings.21 Richard Jenings had begun his career as an apprentice to the carpenter John Longland and had risen to become Longland’s partner or foreman in 1703. When Longland died, Jenings assumed the contract.22 His career trajectory, being so tied to St Paul’s, made Jenings very much Wren’s man. When a large contract (for £11,000) for ironwork railings went to an unknown contractor, Richard Jones, rather than to Jean Tijou as Wren directed in 1710, Jenings was apparently outspoken in his criticism of Jones, who was subsequently found to have been of dubious character, and who subcontracted the railings for £6000, making an enormous profit for himself.23 The commissioners, who were later suspended, may well have tried to make an example of Jenings to cover up the dubious nature of the contract with Jones. Although he may have been temporarily dismissed from work at St Paul’s, Jenings was not prosecuted since the complaints against him were found to be without foundation and he carried on working at other key sites in the city with Wren and others.24 19Gilboy,

Wages in Eighteenth Century England, p. 17. Duke of Portland’s involvement can be read in the Portland papers Volume X, the best collection is in the Wren Society, Volume XVI, under the heading of ‘Frauds and Abuses’. 21See summary in Campbell, Building St Paul’s, Chap. 11 and p. 154. 22Ibid., p. 86. 23Ibid., pp. 158–161. 24Ibid., p. 161. 20The

114     J. Z. Stephenson

Campbell concludes that “There is no reason to doubt the veracity of Jenings claim that it was normal practice at the time for overseers to deduct a portion of the money they received to cover their own costs”.25 Jenings evidence states that he was paying varying rates to different men, with the more skilled or valuable getting 12s. a week, and the lesser just 7s.26 Jenings’ defence was that it was normal practice to take margin on wages, and that skill in the team was variable, and that he had to look after old and indigent workmen who were not productive. Some men “were not worth 12d.a day”, he declared. Moreover, Jenings complained that he was discriminated against; because he lacked the financial capacity to trade on his own account, he could not take on profitable task work.27 The commissioners for the rebuilding of St Paul’s never disproved Jenings’ claim that his practice was normal. Given the figures stated, we can calculate that Jenings’ average mark-up on skilled craftsmen’s wages was approximately 27%, but without his pay books we cannot know the exact distribution of that figure. By taking the highest and lowest margin possible on the figures given, Campbell calculated that it must be between 21 and 41%.28 He proposed that this margin was all profit, as, he asserted, on day work under Wren’s stewardship, all other supplies and costs were provided.29 In fact even contractors supplying day workers had costs and overheads as I will discuss below. The issue of margin or mark-up was not a secretive rent extracted by building contractors. It was explicitly discussed by those who advised clients on new building, and who wrote manuals for surveyors who would commission work, and guidebooks for clients to the process of contracting builders. From the late seventeenth century these were commonplace and offered clients easy ways to calculate what ‘artificers’ would ‘freely

25Ibid.,

p. 160. Rebuilding St. Paul’s After the Great Fire of London, p. 80. 27This seems unlikely, as a glance as the table of contractors on the City Churches will attest, Jenings was contracting in varied contract forms elsewhere. 28Campbell, ‘The Finances of the Carpenter in England’, p. 334. 29Ibid., pp. 333–4. 26Lang,

5  What did Bosses (in London Construction) do?     115

accept’.30 By the middle of the eighteenth century Batty Langley advised that a mark-up of 25% on materials and workmanship was ‘honest’ and good practice, although margins on new building could be as little as 12.5% as the contractor would profit from the economies of scale.31 However, to stress the workmanship point he advised; “…when gentleman find, or buy in their own materials, which in most counties is commonly done, then to the prime cost of workmanship add a fourth part thereof (which is 25 per cent. Profit as aforesaid) and the sum will be its just value which the master workmen ought to be paid”. Thus, the approach I take here is rather different to Campbell’s method. Rather than trying to identify the level of profit, I begin with the costs that contractors faced in operating building firms and contracts. Chief among these were the cost of giving credit to clients, the risk of retrospective discounting of bills, and the contractor’s own management or operational costs. Together these amounted to the overhead or operating margin the contractor added to the costs of goods and services (including labour) that he billed before allowing for profit. Due to the differing levels of organisational integration and the different types of contract, I distinguish between margin calculations for extraordinary work, or new build, and ordinary, or maintenance work.

Management and Operating Costs Contractors’ margins were largely determined by their own operating costs. As is attested by the size of these projects, the contractors were not small masters. They were employing tens or hundreds of staff, some working on-site fitting and constructing, some off-site prefabricating and preparing. In the years 1698–1710, for example, the Strongs had operations and teams of persons on at least three large sites continuously, and those are just the contracts for large clients we know about. In each site or contract situation contractors paid for the carriage of

30Leybourn, 31Langley,

A Platform for Purchasers, Guide for Builders, Mate for Measurers, p. 7. The London Prices of Bricklayers Materials and Works, pp. viii–ix.

116     J. Z. Stephenson

supplies, negotiated with suppliers, and used their own surveyors and estimators to verify lengths and work done. Not all costs were detailed or billed directly to clients, rather they went into making up the prices of the products and services sold to them. Even when clients were supplying materials, contractors faced the operating or management costs involved in running complex operations. These were the administration, accounting, and fulfilment of measured contracts. To identify and estimate these costs we need to grasp the range of roles and functions that contractors’ firms needed to carry out beyond the direct application of labour on a building site. Contractors have not left us organograms, and even the rare day books, bill books and letter books that survive usually give us only one side of any correspondence and only for parts of their businesses or some contracts. As a result, working out the organisational form or hierarchy of these contractors’ firms is largely a process of induction based on some fragmentary evidence. The first step to building up a picture of the organisational form is to understand the activities required to fulfil the contacts described in the previous chapter. As a result of the way these were sought, tendered, agreed and granted, these are best understood as falling into four sets of management activity: search and information, bargaining and decision making, policing and enforcement, and directly managing and monitoring the factors of production.32 These are discussed below in the order in which they were likely to affect production. It is useful to survey the likely range of organisational roles needed within a large contracting operation such as Strong’s operation at Greenwich or St Paul’s; Jelfe’s at Westminster Bridge; or Kempster’s at St Paul’s. An implied organisational hierarchy, illustrated in Figs. 5.1 and 5.2, which illustrate some kind of organisational hierarchy based on analysis of contractor’s own books where available, are a world away from what is traditionally assumed to be that of small masters, and the artisanal or domestic system.33 All contractors would have had to deal

32As

described in Dahlman, ‘The Problem of Externality’. Library Ms 27587; The National Archives C 106/145; London Metropolitan Archives CLC – 227-15, MS00233. 33British

5  What did Bosses (in London Construction) do?     117

Master Contractor

Off site

Accountant

Surveyors Measurers

Watch men

Assistants

Subcontracted Craft

Craft

Labour

Other Agent

Quarry Agent

Yard Foreman

Sub

Craft

contractor

Labour

Subcontracted Labourers

Mess enger /carrier

Labour

Fig. 5.1  Likely roles under large mason contractor off-site

Master Contractor On site Foreman / Agent

Messenger / carriers

Directly employed Craft

Skilled craft

Journeymen Labourers

Ancilliary/ Suppliers

Subcontracted Craft

Craft Labourers

General Labourers contracted Boys

Craft/ Foremen

Labour Messenger /Boys

Fig. 5.2  Likely roles under large mason contractor on-site

with some form of this organisational hierarchy to be able to deliver what they contracted to do. When wages or day payments are listed in contractors’ accounts such as William Kempster’s or Andrews Jelfe’s,

118     J. Z. Stephenson

they tend to be for men working on a site. But payments for agents, suppliers, managers, and others were more likely to be by piece or on commission or for services rendered.

Search and Information Costs As was discussed in Chap. 4 prices for goods, labour and materials were agreed as part of the overall project contract before work began. Generally, a client took the right to discount or amend the prices paid after work was carried out if they were not satisfied with quality. So, in setting their prices for a contract, the contractor needed to know the best sources for materials and labour and reckon on how the risks of supply-chain shocks and client action could affect their margin. If a contractor owned a source of materials this process would have been to estimate the labour costs of extraction and/or the transport costs, but those that did own such resources also contracted for other kinds of materials. Edward Strong sold worked Purbeck, slate and marbles to Greenwich, not just his Burford stone, so the search and information costs of supply chain management applied to all contractors. We can see the effects of supply-chain difficulties clearly in the letter book of Andrews Jelfe. This includes a large and regular correspondence with his and Samuel Tufnell’s agent for procuring materials for Westminster Bridge, one Mr Roper.34 Jelfe faced ongoing problems with quarrymen, Mssrs. Tizard and Bryer, over the delivery of stone. Many of the letters deal with sourcing alternatives and discuss how Tizard and Bryer could be persuaded to extract and transport what was needed. Roper received instructions as often as twice a week on sizes of stone and cutting and carriage directions. Jelfe also wrote to him regularly to advise him of contract developments; in July and August 1743, it was to say that work on the Bridge was not going according to plan and to warn him to lay off men (his “many hands”): “I am sorry there is no work for 34A Thomas Roper worked as a mason at 22d. per day for Jelfe in the year 1734–1734. It is possible that the Roper acting as his agent in 1741–1743 is the same man but cannot be confirmed. See pp. 1–10 of BL MS 27587.

5  What did Bosses (in London Construction) do?     119

them here”.35 Mr Roper may have been sole agent for stone or also other labour and materials. Jelfe may have had more men who provided similar information for different parts of the process. The costs of maintaining such a network would have been the price of commission to such men, and the costs of correspondence. A reliable network of merchants, carters, and procurers meant that supply could be produced flexibly to meet contracted demand, and in other trades it has been shown that the use of middlemen was widespread in the supply chain.36 The contract between Gilbert and Wise and St Paul’s for Portland stone discussed in Chap. 4 illustrates some of the costs that would have to be handled within a materials supply contract. The partners would have had to procure or build cranes, hire a contractor or overseer to manage the quarries, a rope supplier, a tool supplier, a cart and carriage supplier, someone to manage the water transport and haulage.37 That the cathedral specified delivery to St Paul’s wharf, not the Cathedral would have saved them haulage and carting costs in London. Given these activities, and pressures, it is not surprising that contractors were sometimes paid to actually be on site. Clients had to offer them some sort of daily allowance to procure their time.38

Bargaining Costs For extraordinary work, the terms of each contract had to be individually stipulated and agreed. Contractors frequently attended meeting with commissioners and other contractors and paymasters, and these meetings required preparation and active bargaining, and often the presentation of much information. The Fabric Committee at Greenwich Hospital had at least three meetings with Edward Strong and Thomas Hill before issuing a contract in 1696. The minutes imply that Nicholas

35British

Library, MS 27,587, Andrews Jelfe Letter Book, pp. 30–147. Middlemen in English Business, Particularly Between 1660 and 1760. 37For a discussion of how this went wrong see Campbell, Building St Paul’s, Chap. 11. 38Westminster Abbey Muniments 34513, quoted p. 112 above. 36Westerfield,

120     J. Z. Stephenson

Hawksmoor, as surveyor and clerk-of-the-works had many more.39 At Westminster Abbey in October 1722 Edward Stanton was called to the Committee for the Repair of the Abbey to give an account of his works there. He had put up a bond of £2000 to take on the work and position of Edward Tuffnell, who had died in 1719. At St Martin’s in the Fields four contractors contested the bricklaying contract, and several for the masons. The business of appointing contractors took the summer of 1721; Cass, the mason, also gave a bond.40 It was not unusual for the process of agreeing a contract to take months. Once the contract was in hand, its operation imposed obligations on contractors. Measured work necessitated measurement, which was administered by an accountant and surveyors. Billing and accounts had to be managed to the calendar that the client institution adhered to, typically twice yearly. Contractors would also have to comply with the measurement or enforcement systems of clients. At Bridge House contractors not only had to make themselves available for inspection by the City surveyor, but possibly also had to pay a commission towards his costs.41 In the later part of the eighteenth century it was common for architects to charge workmen for assigning contracts and handling bills.42 The risk of a bad bargain, or one which did not fulfil the optimum (from the contractors’ point of view: the highest price) was obviously inherent in the process, but impossible to quantify here.

Enforcement Costs As discussed in Chap. 4, measurement fulfilled a monitoring role in construction, as it incentivised contractors to avoid non-payment or discounting through only delivering work that would meet the standard required. Measurement was enforcement. Because an unsatisfactory 39TNA

ADM 67/2, ADM 82/4. Building the Georgian City, Appendix II, pp. 234–235. 41Bills at Bridge House in the 1740s were signed by the surveyor, George Dance, and some have an amount, 4d., assigned them. 42Nisbet, A Proper Price, p. 34. 40Ayres,

5  What did Bosses (in London Construction) do?     121

measurement could wipe out as much as whole bill, the process of measurement was very important to the contractors’ profit, so how easy was it? What did it cost? Campbell has asserted that measurers were independent.43 Nisbet found evidence that measurement sessions always had both the clients and the contractor’s representatives present.44 Only the client’s measurer usually appears in the accounts however, (at Westminster Abbey at least). There, all bills were measured by William Dickinson, who was paid £50 per annum for his services, approximately 1–2% of the annual expenditure. The cost of measuring the work on Fleet Ditch carried out under John Fitch’s contract was reported at only 0.12% of the total cost of the project (£51,307), likely because the measurement was only carried out at the final account, or perhaps because there was only one contractor.45 At Westminster Abbey, by contrast, measurement was biannual, and covered several or all contractors, and at St Paul’s the process was quarterly. The contractors’ measuring costs were borne by themselves and can be presumed to have cost the same. Measurers or surveyor’s day rates were from 10s. to a guinea a day.46

The Costs of Operating: Managing Sites and Labour Having discussed the costs that are generally associated with contracting and transacting in markets and organisations we now need to consider the practical and financial costs of day-to-day production. Contractors were wholly responsible for hiring and paying all labour, skilled or unskilled, that they utilised on a project. As today, much building work, especially on large new build projects, required the

43Campbell,

‘The Finances of the Carpenter in England 1660–1710’, p. 336. A Proper Price, pp. 2, 26. 45Ibid., p. 2. 46LMA CLA/007/FN/04/001. 44Nisbet,

122     J. Z. Stephenson

supply of large numbers of workers for short periods of time.47 The construction of scaffolding, for instance, had to be performed prior to masonry work, and if scaffolding labour was not needed for the period that the masonry was ongoing, then the removal of the scaffolding meant calling in large numbers of carpenters again. Labourers were engaged in clearing and demolition one week and soldering and assisting masons the next. In order to meet the cost requirements for each job, time and resources had to be chosen carefully and deployed effectively. To recruit, organise, and let go the right casually hired labour at short notice, took experience, and trust. Yet, the search and information costs contractors faced in hiring labour are very difficult to ascertain. Despite the scale of the projects, there are virtually no records of contractors’ employment practices. As was mentioned earlier, Knoop and Jones’ survey of company searches indicates that the numbers employed in workshops were small (never more than eight or nine).48 Yet, St Paul’s call books show hundreds of workers on site. Task work and measured work, by their very nature, necessitated that labour was only deployed and paid for the time it took to complete the pieces or measures required. The labour requirement at sites varied substantially, as discussed in greater depth in the following chapters. By way of example, the number of labourers employed on site at St Paul’s in Januarys 1700–1710 varied from under 40 to 150.49 The fluidity of the workforce is clearly visible from Kempster’s day books for St Paul’s. These show a constantly evolving and adapting team—a team of approximately 17 men worked four to six days a week in October to December 1700, but then had no work for the whole of January. Eight of them were rehired in February.50 They may have been working for Kempster elsewhere or had agreed a

47Campbell, Building St Paul’s, p. 154; Knoop and Jones, The London Mason, Appendices A–C, pp. 66–72. 48Ibid., pp. 25–28. 49LMA St Paul’s MS 25574, 25473. 50TNA C 106/145, day books.

5  What did Bosses (in London Construction) do?     123

rest period with him. If so we cannot know. Of the 21 men who worked in the week ending October 12, 1700, 12 were also found working in the team a year later. In October 1702 only 9 of the 21 men he had employed in October 1700 were still working for him. At Bridge House although there are records of named men working which appear over two or three decades, very few men can be observed to work steadily throughout the year. The construction industry is always characterised by the periodic or short-term requirement for labour because specialist skills only need to be deployed for specific tasks. At St Paul’s some men’s employment seems to have been based on casual hiring, but importantly some long-term relationships can be observed. William Kempster seems to have not only taken over Jon Thompson’s contracts but many of his usual team too. Richard Richards, working for Thompson in 1694, was working for Kempster in 1702 and 1708.51 There were some very long-lasting employment relationships at Bridge House also. Bartholomew Sparruck is first observed on the books working for Jeremy Bower in 1707. His son or grandson finally relinquished the contract to Joseph Nixon in 1757. Nixon had been working for Sparruck since the early 1730s.52 Beyond hiring, labour on site needed monitoring and direction. The solution to the problem of search and monitoring presumably lay in employing a foreman. However, the costs of such a skilled agent or overseer are hard to evaluate. Jelfe’s letter book only records his correspondence with Roper not his payments to him. Roper’s pay would not, so far as we can tell, appear in day bills, but rather was one of the firm’s costs of operating—he may well have just taken a commission or percentage of the contracts he procured for Jelfe and Tuffnell. In Jelfe’s accounts for 1734–1735, there are a number of men regularly employed for most of that year at a number of locations. One, John Ogle, was paid a day rate of 4s. 2d. (50d.) per day, but is not listed as a mason or labourer in any of the bills. It seems that Ogle died

51Ibid.; 52See

Knoop and Jones, The London Mason, Appendix C. LMA CLA/007/FN/04/19–27.

124     J. Z. Stephenson

during the year. Jelfe paid Anne Ogle, his wife, a sum “owing on Ogle’s account”.53 Given the very high level of his pay, Ogle was possibly some kind of overseer, but without further information we cannot be sure of his role. Most bills submitted to clients to not detail the cost of overseers as a separate charge or expense. William Kempster’s foreman at St Paul’s, Joshua Fletcher appears in his day books paid 20s. a week in 1708, but none of Kempster’s bills charge any man at that rate.54 The detailed sawing records kept by Kempster and Strong from St Paul’s imply that there was a full time monitor present to record the amounts sawn and the distribution of the load and account for any wastage.55 Kempster’s day books refer to men who spent time ‘in the yard’, so there must have been someone overseeing that, and other men present whose work was not itemised on bills for work done. There were also other costs of labour and management—those of taking on apprentices and training new skill. It’s hard to know how much of a contractor’s time would have been given over to their apprentice or apprentices, given that there is evidence that many mason’s apprentices worked with others during their time, but literature on apprenticeship stresses the importance of the master–apprentice relationship.56

Administration, Accounting and Auditing Communication and administration between sites presented a particular challenge and added to costs. The letter book of Andrews Jelfe shows not just that letters were sent to his agent Roper, but to others also, and his fortnightly accounts show that he paid 3d. or 4d. for each letter to and from London when working away. There are frequent referrals to extra charges for transport, paper, pens, letters, and porters. In the six

53BL

MS 27587, March 1734. C 106/145 also see Wren Society, Volume XVI, Fletcher appears in the bottom of the list of pay in several weeks in 1708, he is referred to as a foreman in the Wren Society account, but he is not present throughout Kempster’s books. 55TNA C 106/145 and the marble sawing records at LMA CLC/313/I/B/003/MS25473. 56Wallis, ‘Apprenticeship and Training in Premodern England’, pp. 832–861. 54TNA

5  What did Bosses (in London Construction) do?     125

weeks prior to 30 September 1734 these sorts of costs were 2.5% of the amounts in his day book.57 These may have been exceptional costs because he was working away. Contractors working on projects, but not staying on site themselves, probably had similar high levels of correspondence to overseers or foremen, agents or subcontractors. The other skilled service that contracting firms needed was accountancy, and auditing. Accountants were required and undoubtedly used by contractors of all kinds. The letter books of Andrews Jelfe and the day books of William Kempster indicate that contractors made their own rough accounts of costs on a weekly or bi-weekly basis.58 However, as the bill book of Edward Strong shows, the costs had to be properly formed into bills to the client.59 Jelfe’s and Strong’s books have a number of hands in them, suggesting that Jelfe had assistants or scribes. Given the number of sites and contracts they were involved in, and the distance between sites, a contractor such as Strong, or Tufnell or any of those discussed in the previous chapter surely did not have time to sit down and account for all their measured rates for all sites for all workers. Foremen or other managers must have gathered data from all sites and collated and prepared it. Accountants, like surveyors, were well paid in this era. At Westminster Abbey the accounting and audit costs for the refurbishment of the Abbey in the years 1712–1720 were approximately £70 per annum, comprising the accountant’s fee of £60, the auditors’ assistant’s halfpenny fee in 1716 of £6 9s. 10½ d., and the clerk’s fee for writing up (£2 12s. and 14d). This represents 2.2 to 2.5% of the total building costs at Westminster. As a recipient of the coal tax the Abbey was required to submit abstracts to the Treasury, which would not have been needed by contractors. It is also not likely that contractors hired auditors and accountants who were as prestigious and pricey as the clients’. However, contractors had to employ accountants to write and submit bills throughout the years work—the client had only to receive and

57BL

MS 27587, pp. 4–5. Ms 27587, TNA C 106/145, day book. Jelfe’s accounts are fortnightly, Kempster’s weekly. 59LMA CLC/B/227-175. 58BL

126     J. Z. Stephenson

check them. It seems reasonable to put contractors’ accounting costs at the same level as those of the clients’.

Rent On top of the costs of operating on site, at least some large contractors maintained separate premises for off-site work and administration, so rent must also be considered. For example, Edward Stanton maintained premises at Holborn from the late 1690s (and possibly before) to the 1720s. Joshua Marshall’s yard was nearby in Shoe Lane, Edward Pearce was south of St Clement’s in Arundel St. We do not know the exact locations of other masons’ yards. Their yards would have stored goods, carving and other work for task or measured orders prior to delivery to sites. Kempster is believed to have had premises on Garlick Hill, but these may have been temporary due to work on the Church there. Rents are notoriously difficult to estimate for London in this period. A figure of 2% is an estimate based on Nicolas Barbon’s Apology, in which he quotes houses in Bishopsgate at £30 per annum and in Charing Cross at £60 per annum in 1685; thus, we might expect a median rent of £45 per annum, based on a conservative estimate of turnover at £2000 for a mason. However, it is also possible that commercial rents—where buildings were not maintained by the landlord were much lower, but maintenance of property should also be allowed for.

Carriage Carriage costs would have been considered and accounted for as part of materials costs, not labour bills, and obviously they would not have been part of ‘by the day’ labour rates. Since many bills charged for the final product by the piece, which involved a combination of materials, carriage, and labour the uncertainties and quantity of them is relevant to discussion of operating margin. For contractors who were providing materials or final products on contracts by the great or measurement, there were substantial costs of carriage to take into account. Peter Guillery’s recent case study of

5  What did Bosses (in London Construction) do?     127

construction at Cavendish Square in the 1750s and 1770s highlighted that as much as half the cost of stone used on such projects was the cost of carriage to site in London.60 Given this, the amounts that institutions accounted for carriage are surprisingly small. At Westminster Abbey 1713–1720 outwards carriage for the Abbey ranged from £37 to £48 per annum, on an approximate annual expenditure of £3000. Of the total works bill for that seven-year period of £15,789, £290 was billed by the carter, Richard Adams, at a rate of 1s. per load. At Westminster Abbey, Elizabeth Gregory provided two large mast trees in August 1714 at a total of £10 7s. 8d. and was allowed cash for the purchase. There is no mention of carriage for this in the accounts. There are no purveyor’s bills at Westminster Abbey, and no separate bills for stone during the period 1712–1719. This firmly implies Gregory and Tufnell were providing materials and providing and paying for inwards carriage for it. At Westminster Bridge “contracts entered into by the board” do not include carriage.61 The cost of carriage inwards would have been borne by the contractors’ supplying materials. Carriage was required for all goods inward: stone; mortar; timber and scaffolding; tools; ironmongery; lifting equipment and other items.62 The business of carriage was subject to significant risk for contractors due to weather and other problems. Thomas Gilbert and Thomas Wise’s contract to supply Portland to St Paul’s Cathedral discussed above, specifically stated including delivery to a London Wharf.63 When they attempted to charge cost for cranes and rope to assist the transport, the Cathedral would not allow the bills.64 At St Paul’s some carriage in was

60Guillery,

‘Cavendish Square and Spencer House: Neo-Classicism, Opportunity and Nostalgia’, pp. 75–96. Richard Wilson showed the percentage of the bills appearing in final accounts for country-house building in the same period was 4 to 4.5% in Wilson, ‘The Economic Impact of Building the English Country House 1660–1880’, p. 997. 61TNA Work 6/46. 62For lifting items see Campbell, ‘The Finances of the Carpenter in England 1660–1710’, p. 316. 63Above, p. 92. 64See Wren Society, Volume XVI, pp. 19, 51. There was a similar incident at St Martin in the Fields in 1721 referred to by Ayres, Building the Georgian City, p. 234.

128     J. Z. Stephenson

recorded, costing between 2s. and 2s. 6d. “per tun”. Whilst the cost of carriage was undoubtedly built into materials bills, the unpredictability of it necessitated management and a margin for error in financing works.

Tools The question of tools is complex. Smiths’ bills are a feature of site accounts throughout the eighteenth century, and so are mostly integrated into extraordinary work. Smiths on site charged for tools and tool repairs, but also made cramps and hinges and the other kinds of iron work that were part of construction. Many authors assert that journeymen provided their own tools, in which case their cost would have been met out of their wages directly.65 However, there are a number of pieces of evidence that suggest that what was supplied in site by either the client or the journeyman was not the full set of tool requirements. If so, contracting masons, and carpenters and others would have had tool costs as part of their overheads. When Thomas Wise took over the position of mason to Bridge House in 1685 he was paid a salary of £4 per annum, plus an extra £1 per annum in lieu of edge tools. The edge tools bill prior to this had been more than £1 per annum, so presumably it was expected that Wise’s margin on materials or wages would account for further costs. Wise was providing his own team’s tools. Smith’s bills for Bridge House were passed through the purveyor from this point onwards and are not always regular.66 Kempster’s books show that he made payments to smiths irregularly, mostly small sums of between 5s. and £2. Masons and carpenters with their own workshops faced tool costs. Smiths’ bills are a small proportion of total building charges, around 10% of that of masons at Somerset House.67

65Campbell,

Building St Paul’s, p. 79; Ayres, Building the Georgian City, p. 163.

66COL/CC/BHC/10/003. 67Colvin

et al., The History of the King’s Works, Vol. V, 1660–1782, p. 467.

5  What did Bosses (in London Construction) do?     129

They cannot be clearly distinguished from other costs at St Pauls, but, at Westminster Abbey, the accounts do include smiths’ bills: they were £78 in 1712–1713, £103 the following year, and £57 in 1716; they never exceeded 3% of the cost of construction, but the bills specify that the charges were for items used in building, not services such as mending tools, so it seems that the cost of tools and maintaining them was borne by contractors or their workers. Either way they added to the effective overhead in day bills. If a client such as St Paul’s or Westminster Abbey were willing to leave the employment and deployment of labour and work to the direction and autonomy of contractors, then it is highly unlikely that they could have effectively managed the contractors’ tool requirements—even had they been willing to.

The Costs of Finance As has been recognised by every author who has examined any of the records of the Office of King’s Works, the credit that had to be provided for clients (in this case the crown) was one of the major costs faced by building contractors.68 The Office of King’s Works owed money to approximately 85 contractors throughout the period 1709– 1724. At any one time, the monies outstanding ranged in total from £6000 to £11,000. Single contractors were owed as much as £1200.69 At Westminster and Greenwich payments were typically made 15–30 months after the work was carried out. For example, Edward Tuffnell’s work at Westminster Abbey between October 1712 and February 1713 was measured on the 29th of October 1713, the bill was allowed on December the 29th 1713, and payment signed for on May the 4th 1714.70 Tuffnell’s outlays of wages and materials costs were not

68Melody

Mobus, ‘Survining Late Payments: The Strategies of Christopher Wren’s Masons from Burford’; Colvin et al., The History of the King’s Works, Vol. V, 1660–1782, pp. 39–46. 69TNA WORKS 5 /56 Abstracts of accounts 1709–1726. 70Westminster Abbey Muniments, 34513.

130     J. Z. Stephenson

recouped for 19 months. This was not a bad credit period compared to the ­financial issues at St Paul’s in the 1680s and 1690s.71 At St Paul’s, payment initially was only running six months behind completion of work.72 By the mid-1690s however, payment was sufficiently delayed to such an extent that Wren was forced to convert the money owed to contractors into a bond.73 The rate St Paul’s paid for this line of working capital was 6% per annum from this point, after a significant delay in payment already.74 Remittance records do not show any compensation for delays before the 1690s, and this system was not found at other sites. Some contractors were owed thousands over many years.75 In 1670, to deal with the problem and maintain credit, officers of works were allowed 6% interest on works outstanding which were traded as a loan but only if they loaned a further sum. In 1692 the same deal was offered to contractors who would advance double the amount outstanding to him by way of a loan to the works.76 In other words late payment was so bad the Office of the King’s Works was forced into paying interest to maintain credit—but the interest rates were only available to those who would put more capital in. Whilst they waited for payment contractors were paid in tickets or vouchers, a sort of promissory-note arrangement. Those who could not afford the outlay any further could sell these at a discount and apparently a secondary market developed, with some unscrupulous officers taking advantage.77 The consequences for the works of perennial lack of cash and late payment were “bad credit and disadvantageous terms.”78 What was the cost to contractors of supplying this credit? Annual interest rates or discounts on trade bills of exchange were commonly

71Campbell,

Building St Paul’s, pp. 66–67, 73. p. 66. 73Ibid., p. 67. See remittance records at CLC/313/I/B/014/MS25483/001. 74Ibid., p. 69. 75Colvin et al., The History of the King’s Works Vol. V, 1660–1782. ‘Financial Stress’, pp. 39–44. 76Ibid. 77Ibid., p. 44. 78Ibid. 72Ibid.,

5  What did Bosses (in London Construction) do?     131

between 6 and 15% in the period, suggesting that a commonplace 18-month delay in payments cost a contractor about 10% of his bill.79 The loan system outlined above negated this cost for contractors with their own ready cash at disposal to lend. Those who did not have considerable sums to invest (double their sums outstanding), or to lend bore the 6% cost of credit. Suppliers further down the supply chain may have shared some of the pressure, but building contractors could not hire workers on credit. Experienced contractors would have known the risk and priced work accordingly. The ticketing system may have helped contractors’ liquidity, but given the loan arrangements the tickets must have been discounted by at least 6%, probably more. Aside from this, contractors received only token relief for the costs of credit they faced. Advances or imprests were commonly given to key contractors. However, at Westminster Abbey the maximum imprest in any quarter seems to have been £50, whereas contracts in progress in the same period were worth thousands.80 Campbell showed that carpenter contractors receiving imprest money at All Hallows’ Church had small regularly forthcoming payments.81 Yet by August 1691, Woodstock, the contractor, had billed just under £1000. That amount of funds had only been paid out to him by May 1693, a delay of almost two years. At Westminster Bridge the system seemed a bit more generous, with payment running only about six months behind bills throughout 1744. On a bill for £6185 Jelfe & Tuffnell had had imprests of £4000.82 The amount of time credit was out for likely reflects the differences between institutions and projects. We can see the impact of the need to provide working capital in the Kempster books for St Paul’s, which

79Temin

and Voth, ‘Private Borrowing During the Financial Revolution: Hoare’s Bank and Its Customers, 1702–241’, p. 545; Quinn, ‘The Glorious Revolution’s Effect on English Private Finance: A Micro History, 1680–1705’, pp. 593–614. 80Westminster Abbey Muniments, 34513, 34514, 34518. 81Campbell, ‘The Finances of the Carpenter in England 1660–1710’, p. 346, Appendix B. 82TNA WORK 6/46 throughout and p. 47.

132     J. Z. Stephenson

show the firm using borrowed capital from other family members, as well as from other sponsors, to pay weekly wage bills and keep their team working.83 At Greenwich due to delayed payment by the Hospital, Edward Strong seems to have laid off workers or stopped work at the site in 1702. Minutes refer to workers agitating and the closure of the gates at the east end.84 At Westminster Abbey in 1722 the Abbey was behind in payments to all contractors, but work seems to have carried on regardless, and contractors absorbed the cost of credit.85 The Bank of England rate was 4.50% from June 1699 to August 1716 and then 4% until late-April 1719 after which it was 5% for over a century until June 1822.86 Usury limits were 6%, falling to 5%, in this period, but a rate of return nearer 8% annually has been established for two contemporaneous private lenders.87 If credit was priced conservatively at 6% per annum, then a relatively rapid payment schedule of 18 months gives a figure of 9% (simple interest). Note, the relevant calculation for credit costs is over an 18 month-period—not annual—as I am seeking to evaluate the amount of margin in a bill that would have been dedicated to the cost of credit.

Deductions Contractors’ faced the risk that clients would make deduction from their bills for not meeting their expectations, terms or standards, or those of their measurers and surveyors. Measured contracts usually gave the client the right to deduct from bills on the grounds of quality, and wrote this right into construction agreements, as can be seen in Strong’s 83TNA PRO 106/145, also Mobus, ‘Survining Late Payments: The Strategies of Christopher Wren’s Masons from Burford’. 84ADM 67/2. 85Westminster Abbey Muniments 34517. 86Thomas and Dimsdale (2017), ‘A Millennium of UK Data’, Bank of England OBRA dataset, https://www.bankofengland.co.uk/-/media/boe/files/statistics/research-datasets/a-millenniumof-macroeconomic-data-for-the-uk.xlsx?la=en&hash=73ABBFB603A709FEEB1FD349B1C61F11527F1DE4. 87Quinn, ‘The Glorious Revolution’s Effect on English Private Finance: A Micro History, 1680–1705’; Temin and Voth, Prometheus Shackled: Goldsmith Banks and England’s Financial Revolution After 1700.

5  What did Bosses (in London Construction) do?     133 ϭϬϬй ϵϬй ϴϬй ϳϬй ϲϬй ϱϬй

ϰϬй ϯϬй ϮϬй ϭϬй ďŝůůŶƵŵďĞƌ ϰ ϴ ϭϮ ϭϲ ϮϬ Ϯϰ Ϯϴ ϯϮ ϯϲ ϰϬ ϰϰ ϰϴ ϱϮ ϱϲ ϲϬ ϲϰ ϲϴ ϳϮ ϳϲ ϴϬ ϴϰ ϴϴ ϵϮ ϵϲ ϭϬϬ ϭϬϰ ϭϬϴ ϭϭϮ ϭϭϲ ϭϮϬ

Ϭй

Fig. 5.3  Discount of Edward Strong's Bills Greenwich Hospital 1698–1708 showing actual percentage discount on each bill and 10 bill MAV (Source Bills listed in Edward Strong’s account book, LMA ref. CLC/B/227-175, formerly MS00233)

1706 mason’s contract for Greenwich Hospital.88 The contract did not give reciprocal rights to the contractor.89 Clients could also simply change the terms. At Middle Temple, Wren discounted the plasterer’s bill by 30% in 1682.90 Deductions occurred on day work as well as measured. The account book of Edward Strong for work building Greenwich Hospital 1698 to 1708–1709 shows that his bills were regularly deducted from. Strong’s day bills were cut by an average of 15%, his measured bills by 3%.91 The deduction or discount is shown in Fig. 5.3.

88LMA

ref. CLC/B/227-175, formerly MS00233. copy of the contract is in Strong’s account book LMA ref. CLC/B/227-175. 90Middle Temple Archive MT.6/RBW. 91LMA ref. CLC/B/227-175, formerly MS00233. 89A

134     J. Z. Stephenson

In the Strong bill book, 11 day bills out of 25 in total were cut. Two were slashed by more than 55%, the rest were cut by between 17 and 45%. The average day bill discount is 15%. Because his measured bills were deducted less heavily, the total nominal deduction by value he took on all his bills was only 3.84%. However, the risk of deduction was possibly perceived and priced by contractors as being greater than that. Since contractors did not have recourse if discount was applied, they must have had some kind of estimation rule of thumb based on experience for allowance. The 10-bill moving average at Greenwich, which is also shown in Fig. 5.3, ran substantially higher than 3.8% for substantial periods, implying that Strong would have been sensible to expect a higher discount in his pricing. I therefore allow 3.5–7% as the allowance for deductions.

Profit and Overall Margin The question of profit now arises. Whilst the various costs listed above offer some support for Campbell’s calculations by showing that contractor’s operating overheads fit with his estimate of a margin of between 21 and 41%, they do so before any actual profit is allowed for the ­contractor.92 Here we are largely in the dark. Campbell used the high accumulated wealth of contractors to infer that they took a substantial rate of profit from their work.93 While is it likely that there were some contracts where profits exceeded 20%, it is hard to believe that they were much higher than this. There were competitive tenders for many projects, and it was the stated aim of many a commission for a public project to take the lowest estimate possible. If profit margins had been that big they may not have survived the tendering process. There is not a great deal of available data on profit margins for seventeenth-century business people that would allow us to set a comparative benchmark. Grassby estimated that an acceptable 92Campbell,

‘The Finances of the Carpenter in England 1660–1710’, p. 334. p. 341. Campbell, ‘Building a Fortune: The Finances of the Stonemasons Working on the Rebuilding of St Paul’s Cathedral 1675–1720’; Colvin, A Biographical Dictionary of English Architects, 1660–1840, p. 3. 93Ibid.,

5  What did Bosses (in London Construction) do?     135

profit margin to seventeenth-century merchants was 13–20%.94 My working estimate of profit margin is thus set at 10%, which should be read with the proviso that it will probably have varied around this figure by a large margin.

Summary of Operating Costs and Profit, and Ordinary Work This examination of the accounts and organisation of building work on extraordinary public projects in London from 1660 onwards suggests that the operating margin that contractors worked with was substantial. The different elements of the operating costs are set out in Table 5.1. Allowing for profit, it must be concluded that contractors’ mark-up on goods and services on average or cumulatively was in the region of or approximately a third over prime cost. It is not likely that this margin was applied uniformly to all labour and services, however. The incentives of measured contracts will have called for high levels of skill at certain times, and lower at others. Changes in prices and conditions of the materials supplied would have allowed a large margin sometimes and a much thinner profit in others. As will be seen in Chaps. 5–7, this operating margin was also reflected in the difference between craftsmen’s and labourers’ pay, and the rates they were charged out at. Did the same set of conditions apply for the more mundane ­‘ordinary’ repair work that went on throughout the period on all building types? Maintenance work, like that conducted on the Bridge House, and in the ordinary accounts of the Office of the King’s Works, was financially distinctive in a few ways. Bridge House contractors were paid a retainer or salary of kinds (£4 per annum), King’s Works sites also had a permanent labourer-in-trust or clerk-of the-works to organise and procure works.95 Contractors at the Bridge billed weekly for materials 94Grassby, 95LMA

‘The Rate of Profit in Seventeenth-Century England’. CL/007/FN/05/7, 8.

136     J. Z. Stephenson Table 5.1  Calculation of operating margin for ‘extraordinary’ day work Type of cost

% share of bills

Credit Discounting Measurement, estimation, agency, overseeing Accountancy Rent Tools Total before profit With profit of 10%

6–9 3.5–7 5 2 2 1 19.5–26 29.5–36

Source See text

and labour, so they did not have to run long credit lines and associated costs for large bills, although this did not prevent profiteering.96 Bridge House work did not necessitate the same costs of estimation and measurement, nor risk of deductions. In short, the justification for a margin on wages might seem much weaker in ordinary work. On the other hand, Bridge House’s contractors still faced considerable financial or credit costs. They paid large sums to obtain their position, sums that only make sense if viewed as investments on which they expected a profit. The post of Land Carpenter ‘sold’ for £322 in 1706, (to Thomas Wilmor), that of Tide (water) Carpenter £450 in 1710, (to Bartholomew Sparruck), that of Mason in 1720 for £520, and £546 in 1724.97 Mark Latham has made a compelling case that there was corruption and false billing and showed that the price the Bridge paid for materials was higher than elsewhere.98 The Tide Carpenter justified such practices in 1742–1743, by claiming it was customary to seek profit through inflated bills.99 Taking Thomas Wilmore’s case as an example: he paid £322 for the position of Land Carpenter in 1706. The interest per annum alone on

96Mark Latham, ‘“The City Has Been Wronged and Abused!”: Institutional Corruption in the Eighteenth Century’. 97Ibid., Table 1, p. 1044. 98Ibid., pp. 1046, 1051; ‘The London Bridge Improvement Act of 1756: A Study of Early Modern Urban Finance and Administration’, pp. 166, 169, 171, 178. 99LMA COL/CC/BHC/01/08, 11 May 1743, 1 June 1743, and 5 October 1743.

5  What did Bosses (in London Construction) do?     137

that sum would have been nearly £20 if priced at a 6%, so the salary of £4 per annum would leave £16 per annum, or just over 6s. per week shortfall on interest. Wilmore’s labour bills were in the region of £5 to £10 per week in 1707.100 Assuming he worked to recoup his investment, he could, of course, also take a mark-up on his supply of materials to the Bridge. However, the remittance books show that at least some payments for materials and task work was not forthcoming every week, but rather evaluated twice annually. In later decades there are indications that deductions were made from bills for auditing costs by the Bridge.101 Furthermore in order to bill effectively a contractor to the Bridge would have required an accountant or auditor.102 While those Bridge House contractors who were supplying materials did so under separate contracts to the main labour work, their weekly bills do not itemise carriage.103 The uniformity of rates found in Bridge House bills after 1700 suggest that the amounts listed are charge-out rates.104 A contractor to the Bridge would have sought to recoup the costs of purchasing the position, the costs of financing that investment, the cost of meeting audits and administration, and some tools from labour the sale of materials. After 1731 Bridge House contractors could no longer buy or sell their positions; an Act of Common Council banned sale of place. In a new system, contractors—no longer position holders—had to bear audit costs, although they also gained the right to charge interest at 4% if their bills were outstanding for more than six months.105 All accounts were audited, and all materials and bills had to be passed by the Comptroller, and possibly the Surveyor. Notes in some accounts give the impression that contractors may have had to pay for this ‘service’.106 As Table 5.2.

100LMA

CLA/007/FN/06/01/014. CLA/007/FN/05 (series), CLA/007/FN/04/03. 102The final bill books for Bridge House are ‘audited’. Later books show that Wilmore was paid in lieu until his books were audited. CLA/007/FN/03 22–27, and CLA/007/FN/10/007. 103COL/CC/BHC/10/006, 003 Bills Accounts and sundry papers includes purveyors bill for years 1718 to 1730. 104See Fig.  7.1, p. 187 below. 105LMA COL/OF/02/173. 106Ibid., notes beside amounts are signed off by the surveyor. 101LMA

138     J. Z. Stephenson Table 5.2  Operating margin for contractors supplying weekly labour at Bridge House—ordinary work Type of cost

% share of bills

Cost of finance Accountancy and sundry Total before profit Profit margin Total

6 1.5 7.5 10 17.5

Source See text

shows, these costs might have been smaller relative to those of extraordinary contracts, with profit making up the largest component.

Early Modern Bosses Early modern building contractor’s business activities were complex and demanding. They managed supply chains which involved substantial logistics planning and correspondence, and which tied up and risked capital. The had significant human resource management activities arising from searching and retaining the skill appropriate to the construction stage at multiple sites (before the costs of managing any apprentices). Their administrative and accounting related tasks were vital to being able to navigate the complex bureaucracy and payment systems of their clients. Their costs and activities relating to finance: raising it, repaying it, and managing cash flow with funders or sponsors, were considerable. They spent time and resource on the activities of sales and relationship management with potential clients. All of these, even in the early modern pre-industrial era were specialised and time consuming. It is not likely that a trained stone cutter, carver, or fitter would have known how to manage such activities well, although it could be maintained that some sort of communal governance could have substituted. More work needs to be done on understanding how men and women trained for these management and administrative tasks. The likely answer lies in apprenticeship, although in much literature that institution is firmly linked with technological skill rather than administrative entrepreneurship.

5  What did Bosses (in London Construction) do?     139

A large part of what answers Marglin’s question for early modern building contractors is financial risk. The key input that these early modern bosses provided was capital. Contractors were risking their own familial capital or the capital of sponsors and partners. Without the capital they would not have been able to take on the contracts, and in the next chapter we will see evidence of many freemen, highly skilled, fully qualified, who were not able to trade on their own account.

6 Contracts and Pay at St Paul’s Cathedral, and at the Office of the King’s Works

By now, it should be apparent that ‘wage’ rates published by previous historians are not records of wages. Until now, there have not been any other types of ‘wage’ record with which to compare and test the thesis. This chapter presents the most significant proportion of the new evidence presented by this book, based on the day books of William Kempster, the mason contractor responsible for the south west tower, west front, geometric staircase, and library, at St Paul’s Cathedral in the decade after 1700. The new data these books provide are bound by some distinctive characteristics of the site, which affect its representativeness, and so which must be considered carefully. The most obvious characteristic is that of design and skill. St Paul’s represents the highest level of craft skill and architectural ambition in London at the time. It would not be credible to argue that the masons at St Paul’s were ‘average’ craftsmen. Rather, many of them will have been highly skilled specialists. The Cathedral was the most important building in the City, and Wren’s masterpiece. The process of its construction spanned more than

© The Author(s) 2020 J. Z. Stephenson, Contracts and Pay, Palgrave Studies in Economic History, https://doi.org/10.1007/978-3-319-57508-7_6

141

142     J. Z. Stephenson

thirty-five years and the site dominated not just the London skyline, but local and national supply chains and markets.1 The second, and potentially more confusing distinctive characteristic, is organizational. The rebuilding of the Cathedral was a unique project that involved the Crown and the City at the highest levels. The Commissioners formed the organizational features around their ideals of the importance of the project, and contracts were awarded to those who already had strong relationships with the City, the Crown, or Wren. Unusually, the clerk-of-works hired labourers directly until the mid-eighteenth century. We know this is because we have their names, and the number of days that they worked each month or quarter, from the late 1670s right through to the 1740s. The Commissioners knew that their practice was unusual and cited it as exemplary when they wished to defend themselves against accusations of malpractice.2 This seems to be the only site where such a payment system for labourers could be found at this time, as I have verified by reviewing records at Westminster Abbey, Middle Temple, Bridge House, the Office of the King’s Works, and Greenwich Hospital. At those other sites, it is only bills from contractors for supplying labourers by the day that can be found. Until now, it has not been possible to corroborate Richard Jenings’ assertion that his business practice, of paying men less than the rate he charged, was normal practice.3 The evidence from St Paul’s will show that craftsmen and labourers did not receive the day rates that contractors billed institutions, or the amounts that other institutions’ accounts show. Rather, major contractors charged out one rate to their clients, and paid their men another. It will show that labourers and craftsmen working on building the new St Paul’s were paid in the region of between 15 and 35% less than those prevailing charge-out rates, and

1See Campbell, Building St Paul’s, Chap. 1; Moffett, Fazio, and Wodehouse, A World History of Architecture, pp. 392–419; Beard, Craftsmen and Interior Decoration in England 1660–1820, pp. 117–120; Volumes XIII–XVI of the Wren Society. 2The original account books with the Labourers names can be seen at LMA CLC/313/I/B/003/ MS25473 nos. 11–41; Wren Society, Vol. XVI, pp. 139–180. 3It is fully discussed, however, in the work of Campbell, see ‘Seventeenth-Century Bricklayers’ contracts’; ‘The Finances of the Carpenter in England’; ‘Building a Fortune’.

6  Contracts and Pay at St Paul’s Cathedral …     143

that there was a large range or variation in day rates paid to craftsmen, but a more consistent, homogenous daily wage rate for labourers. The new evidence comes from William Kempster’s daybooks, which are call and pay records for his team of masons on-site at St Paul’s for 1700–1709, recording the weekly number of days and weekly pay given to his masons. They are to be found amongst some probate papers relating to chancery case with Kempster’s wife’s family. In a box at the National Archives are two daybooks in which are records of sawing done and day rates for masons and labourers; a small pocketbook; some bills for work at St Mary Somerset and St James’s Garlickhill; and some other family papers.4 The books have never been used by wage historians before, and, if the other contractors who carried out work at St Paul’s kept similar daybooks, we have not been lucky enough to find them. Given Kempster’s provenance, the strong relations between the masons at the Cathedral and all the contracting partners in the other trades, I assume that Kempster was pretty typical in the way he paid his men. The two daybooks cover the period from October 1700 to December 1702, and March 1706 to June 1709. They omit all of 1703–1705, and the years after 1709, when we know Kempster was working at the Cathedral. For the most part, each double page of the daybook refers to a working week, at the top of which Kempster has written ‘paid this week’, or, ‘paid by my sponsor this week at St Paul’s’ or ‘paid by my father this week’. Kempster was borrowing to keep skilled labour at the Cathedral and benefit from his contracts. One side lists the men and the number of days they worked, and the other side their pay. There are some small notes on the particulars of certain tasks peppered throughout, such as “Mr. Richards began to set the paving this week”, and “John Tuckey began to set the flight of steps at the west front May 26, 1709, ended the same August 24, 1709 it being Bartholomew’s day” from the summer of 1709. From November 1708 “John Fullwell and Will Brathen began to rub the black marble Paving, Filley Miles the same account”. We don’t have Kempster’s bill books, but we do have the Cathedral’s accounts of his bills, so we know what he was charging 4TNA

C 106/145.

144     J. Z. Stephenson

the men he was paying out for when working ‘by the day’. Consistently he charged 30d. for masons and 18d. for labourers. However, the vast majority of his bills were for measured work.5 As discussed in Chap. 2, there were a number of mason contractors working at St Paul’s, perhaps because no one mason contractor had the capital to undertake the risk, or because Wren resisted the very considerable bilateral monopoly risk inherent in choosing a single contractor. The mason contractors were Joshua Marshall, Thomas (and then Edward) Strong, Edward Pearce, Jasper Latham, Thomas Wise Senior, Christopher Kempster and Ephraim Beauchamp, Nathaniel Rawlins, Thomas Wise Junior and Thomas Hill, John Thompson, William Kempster, and Samuel Fulkes. From 1675 the Cathedral’s Commissioners set down rates that they would pay for day work, and measured units, and those rates were mostly set throughout the period of main construction (1675–1711). In fact, the day rates for men’s work were the rates still being paid at the Cathedral decades later. The contractors hired their own skilled men and were responsible for m ­ anaging and paying them. Most contracts with masons were by measure or task, but there was an unusually large amount of day work for carpenters. This was on Wren’s instruction due to the nature of the arches and vaulting, which relied on carpenter’s scaffolding for safety.6 The Cathedral only recorded men on site if they were working ‘on the call’ (on day work), so that the Commissioners would have a record of the correct number of days to pay for. If workers were attending to carry out task or measured work, however, they were not entered in the call books. This means that the Cathedral’s call books are not accurate counts of the number of men on-site, and so we cannot be entirely sure how many men worked on the project, either at any one time, or overall. At some points, particularly as the Dome was completed in 1708–1709 there were over 150 labourers listed per quarter by the Cathedral (not all whom worked every day, however). There were weeks where Kempster 5The Cathedral’s accounts at LMA CLC/313/B/I/003/25473 nos. 10–42, Kempster’s bills book nos. 36–42. 6Campbell, Building St Paul’s, pp. 72, 103–104 gives the most accessible visual schema of the mason’s contracts. See also John Schofield, ‘St Paul’s Cathedral’, pp. 48–51 also.

6  Contracts and Pay at St Paul’s Cathedral …     145

had over 60 men on-site. If we assume a second and subsequent mason’s team of the same and teams of bricklayers, and carpenters it is possible to guess that 350–400 men may have been on-site at any one time, but the stage-dependent nature of building makes this uncertain.7 Kempster had been working at the Cathedral before 1700 in partnership, or employed by John Thompson. It is his contract he assumed. Over the next decade he built the south west tower, the Dean’s staircase (or ‘geometric’ staircase) within it, and the library. He also shared in a contract with Samuel Fulkes in December 1708 to flute the pillars of the West End (what we think of today as the west front of the Cathedral). Kempster became master mason to the Cathedral in 1714 or 1715. In all there are records of payments to 179 men in the daybooks. Of those 120 worked for Kempster for a year or under, twenty-six for more than a year but less than two years, another twenty-three less than three years, and only ten men worked for him for more than four years.

Work and Pay in Practice at St Paul’s We would expect to find that craftsmen and others carrying out highly specialized work at St Paul’s would be paid more than the average wage predominant on other London jobs at the time. Surprisingly, the charge-out rates paid by the Cathedral were lower than those found in the Boulton series, which is derived from other, contemporaneous, more ‘representative’ kinds of building projects.8 Boulton found day rates for craftsmen of, on average, 36d. a day in the early 1670s declining to 30d. a day through the early 1680s, and ranging from 30 to 36d. a day throughout the 1690s and 1700s. The Cathedral paid contractors 30d. per day for all craftsmen on day work 1675–1720. Likewise, Boulton found labourers’ day rates were mostly 20d. a day until the late 1690s when they rose to 22d., 24d. and above. By contrast, the Cathedral never paid more than 18d. a day for a labourer in

7The

St Paul’s call books are at the LMA CLC/313/I/B/003/25471 and 25473. ‘Wage Labour’.

8Boulton,

146     J. Z. Stephenson Table 6.1  St Paul’s charge out rates for day work in d. (Wren Society, Vol. XV, p. xlvii) Trade

Foremen/masters

Journeymen

Labourers

Carpenters and joiners Masons Labourers Carvers

36 36 18–30 (mostly 24) 40

30 30 16–18 36–38

18 18–20 12

summer and 16d. in winter, from 1675 to 1748. The parsimony of the Cathedral’s Commissioners is a likely explanation for the fact that rates (shown in Table 6.1) offered to contractors at St Paul’s were lower than those elsewhere. But it is also possible that the Commissioners were able to negotiate lower rates, and tighter margins, because of the volume of work they were commissioning. The evidence from both the Cathedral and Kempster’s books suggests that labourers were hired on what could be considered a wage-posting model—where there was only one wage offered, which is common in lower-skilled jobs.9 Generally, the accounts describe labourers’ work as assisting carpenters and masons, digging, general laboring work, marking mortar, wheeling goods, watching the doors, taking deliveries, and assisting other crafts. The account books show that large teams of men paid 16d. in winter and 18d. in summer were overseen by a small number of foremen. Kempster’s books show he paid his labourers 18d. a day consistently. The Cathedral had a separate, standardized rate for a very small number of foremen, at about 20% premium to craftsmen or journeymen.10 As highlighted above, there were two types of labourers paid at St Paul’s: those employed by contractors who assisted craftsmen, and the Cathedral’s non-craft or general labourers, paid individually by the clerk-of-the-works. In the course of its dispute with Richard Jenings in 1710, the Commission reiterated that its labourers were all paid directly

9Kreuger and Hall, ‘Evidence on the Determinants of Choice Between Wage Posting and Wage Bargaining’; Hanna Brenzel, Hermann Gartner, and Claus Schnabel, ‘Wage Bargaining or Wage Posting? Evidence from the Employers’ Side’. 10LMA CLC/313/I/B/003/25473.

6  Contracts and Pay at St Paul’s Cathedral …     147

and without deduction.11 There is no evidence to suggest otherwise, although it is possible, and likely, that hierarchies and sub-contracts did develop over the long period of construction. Because St Paul’s was buying labourers’ work directly (at what we can only presume to be the market rate because it was so close to that of Kempster), this created a situation where contractors could not make margin on day labouring at St Paul’s. Yet we know from Boulton’s and Gilboy’s work that the rate that other institutions would pay was higher, so contractors would have made margin on labourers at other sites.12 Kempster paid his masons or craftsmen a wide range of rates. Given the high level of the output, we would expect the level of skill at St Paul’s to be of the very highest where required, but also of a considerable range. There would have been people on-site laying bricks or rough laying walls, but there would also have been people installing floating staircases; carving delicate faced cherubs, swags and decorative ­stonework; fitting complex engineered weight-bearing structures; smoothing ashlar etc. In such a situation, the flat craftsmen’s rate observed in traditional wage series would be puzzling, because given the contractual constraints we would expect varying rates for skill. As discussed in Chap.  5, in order to profit from measured work contractors had to allow for an operating margin and a profit margin in setting prices and wages. As they bore the risk of defective work, and their profit depended on the price of inputs and labour, contractors would have been incentivized to hire the best skill possible, and economize most on lower-skilled tasks, ensuring tasks of higher skill, with more valuable materials, were done by suitably qualified men. There is plenty of evidence from previous authors that this was the case. Knoop and Jones detailed a Mason’s Company search of the site in September 1694 (where the information about masons’ previous experience throughout this chapter comes from) which found 151 men in masons’ teams.13 In John Thompson’s team there were 13 men: William Kempster; 11Wren

Society, Vol. XVI, pp. 149–153. discussion of Boulton and Gilboy in Chap. 1 above. Also, it is possible that some of the men paid by Kempster at 18d. per day were charged out as masons. 13Knoop and Jones, The London Mason, Appendix C, p. 75. 12See

148     J. Z. Stephenson

John Magnus; Theo Allen; William Cooper; William Page; and Stephen Powell, who were all masons (Allen was not free). One ‘foreigner’, John Barker, a joiner’s apprentice; four other apprentices, a foreigner, and merchant-tailor. In Edward Strong’s team of 66 men there were 17 apprentices and 18 foreigners. The descriptions indicate that there was a wide range of skill on-site, some highly skilled craftsmen free of the city, some apprentices, some very experienced, some less so. Some men present in the 1694 search were still on-site working for Kempster in 1700: Tho. Magnus, Tho. Allen, William Cooper, and John Barker all appear in Kempster’s daybooks as masons in 1700–1702 and Barker and Magnus in 1708. A contractor like Kempster may have employed men nearby off-site too. At St Paul’s there were various types of stone used.14 It is thought that most of the rough stone or rubble for the inner walls was from the pre-existing structure. For the facing stones we know that the stone was rough cut at the quarry—either Portland or elsewhere, and transported to the Cathedral by water, arriving at the wharf south of the site. It may be that local stone yards worked the stone further; we know that Joshua Marshall’s yard was near St Andrew Holborn, and Kempster’s at Garlickhythe, both near enough to make transportation to St Paul’s feasible. Either there or at the Cathedral the stone was further scappelled to the shapes required, carving added and then finally fitted, and smoothed to ashlar appearance. We do know, however, that most of the carving at St Paul’s was carried out by the mason contractors.15 Given Kempster’s output, we can surmise that his team may have comprised predominantly fitters, carvers and the most specialist men who would have been capable of augmenting the design for the geometric staircase to make it workable, and who would have had experience of working with ironworkers as innovative and technically challenging as Jean Tijou. There are no other carving contracts in the St Paul’s records that could be for south west tower. Francis Bird was responsible for the panel on the West front adjacent for £329 18s. 4d., and the small models (statues) at the top of the Tower

14Wren 15Wren

Society, Vol. XV, ‘Introduction’, pp. xiii–xvi. Society, Vol. XV, pp. xiv.

6  Contracts and Pay at St Paul’s Cathedral …     149

and the Lantern tops at £8 4s. 6d. a few years later so we know that all the stonework and carving on the south west tower after 1700 was carried out by Kempster’s team.16 This presents the problem that some of the wage records are for carvers (who were charged out at a higher rate of 42–44d. in a similar period17) rather than masons, making it difficult to arrive at an all-important ‘average’ mason’s wage. Kempster paid his men varying rates from as little as 10d. a day for unskilled men, to 40d. a day for a highly skilled foreman (Joshua Fletcher). On a small number of bills Kempster could bill his own time at 36d. per day. His books show many more men than those in day bills submitted to the Cathedral, and many different rates, implying that he paid his men by the day, whether they were working on measured, task, or day contracts. The day books record work from October 1700, and a team of up to twenty-eight worked until the end of December 1700. Then no day rates or wages were recorded again until the beginning of February when a team of just 7 men restarted work. Over the next few weeks the team grew again, as before December, with one difference. Three men took a significant reduction in wage. Thomas Bayley and Thomas Loller had been earning 13s. a week in the autumn, now they earned 12s. a week. Roger Blissard who had been earning 12s. a week now received 10s. The new rates lower rates for these men were sustained until the at least the end of 1702. The day rates the men named in the books received in 1701–1702 are shown in Table 6.2. To interpret the rates shown in Table 6.2 properly, it helps to have some idea about the masons’ level of skill and experience. From the Mason’s Company search of 1694 we know John Magnus and William Cooper to have been freemen at that date. They were qualified enough to be masters, well-off enough to have taken their freedom, yet with at least 6 years’ further experience they were working for a large contractor. Tho. Allen had already been working at St Paul’s in 1694, although he was not free then; given that he was certainly more than

16Wren

Society, Vol. XV, pp. xxiii–xxiv, xxxix. Abbey Muniments 34513, however, craving was mostly paid for by the measure, not as day work.

17Westminster

150     J. Z. Stephenson Table 6.2  Day rates Kempster team St Paul’s 1700–1702 February 1701–December 1702 John Barker Edmund Gibson J. Magnus Peter Marrot Ed Healing Tho Knagg Will Cooper R. Parinoole Tho Allen Tho Minor Edmund Bibby J. Smith Tho Bayley Tho Loller John Slorkly M. Turrill John Renn Rog Blissard James Williams J. Garner John Parlour John Foster Jos Shepard G. Conaway W. Danifour Chs Thurland R. Florker R. Joanes

36 34 32 32 30 30 30 30 28 28 28 28 24–26 24–26 24 24 24 20–22 20 18 18 18 18 18 18 18 16 16

7 years into his working career in 1701, he must have been very experienced, and possibly free. John Barker only appears in the books after the beginning of July 1701, but stayed through the year after, in fact he was still working for Kempster in late 1708. We know he was not free in 1694, but we also know that he was then apprenticed to a joiner. That, and payments noted by Kempster to a ‘Barker’ in one of the books for ‘carving’ suggest he might have been a carver, and by this stage, 1701, an experienced one. This would justify his rate of 36d. a day, as would carrying out a supervisory role. Edmond Gibson only appears in November and December 1700 for 6 weeks of observations, and Peter Marriott only in September 1701, they may have been hired in for some

6  Contracts and Pay at St Paul’s Cathedral …     151

specialist task, or just did not work out within the team. The only ­person who was consistently within the team and paid over 30d. per day through 1700–1701 was Tho. Magnus. He may have been a foreman, but for many weeks he was only paid for five and a half or fewer days— fewer than others in the team. Kempster also noted paying his wife a pound on several occasions, and he also noted a number of days worked by “my man Day”. This turns out to be Richard Day, who was apprenticed to Kempster in 1700, and just as we would expect was not paid, nor appear in the payments list until late in 1706. From 1706 Kempster had a far bigger team on-site; furthermore, thanks to notes in the margins against entries in the second book we can know that James Williams, John Parlour, and Charles Thurland (and all listed below them), were labourers. Those marked as labourers were consistently paid 18d. per day, or 9s. for a 6-day week, 7s. 6d. for a 5-day week. Wages for those marked as masons, however, show highly varied day rates—between 10d. and 32d. and Joshua Fletcher, who acted as foreman commanded the rate of 20s. for a 6-day week or 40d. per day. There were also some other notes in the books that enable us to construct a plausible hierarchy. Kempster frequently had more than 40 men on-site most days in the years 1706–1709. He was also managing different sets of skills. In some weeks he listed men in his book as ‘Masons on call’, ‘Ruff layers’, and ‘labourers’ but mostly he just listed their names from highest to lowest paid. The rates for July 1708 are reported in Table 6.3, in the hierarchy or order he listed. Through 1706–1709 Richard Richards and Joseph Smith, both experienced masons—Smith was on Edward Strong’s call list in 1694, Richards on Christopher Kempster’s—were part of the team; Thos. Loller had regained the 26d. a day rate he had in 1700. A son of Edmund Bibby, Thomas Bibby was employed throughout 1707 and 1708 at 7s. for a 6-day week or just 14d. a day, but listed as a mason. Likewise, Thomas Sanders was paid 5s., or 10d. a day and listed as mason. (Sanders’ father (posssibly) had been apprenticed to Kempster in 1686.) Neither of these were apprenticed through the masons company, but it seems likely that they were training. Richard Day appears in the pay records as a fully trained, skilled mason at 28d. per day from late 1706.

152     J. Z. Stephenson Table 6.3  Day rates 1708, Kempster team, by hierarchy found in books July 1708, Kempster team

d. per day

John Barker Tho Knagg Peter Abraham Will Ash Charles Cawtharn Masons on Call Richard Richards Joseph Smith Joh Tukey Mark Abraham Ed. Maslin Ed. Bibby Th. Bibby Richard Duffell W. Kempster Richard Day Mick Growden Matt Wayne Thos. Loller Roger Healing Will Stonhouse Ruff layers and their labour Charles Dray R. Raymond James Bishop John Wren Thom. Sawbridge Sam Ledford John Smith John Fullwell Will Brathen Labourers Williams Parlour J. Foster Will Walbank Rob Onion E. Lintall

36 30 30 30 28 30 30 30 30 30 28 14 28 24 28 28 26 26 20 20 24 24 24 24 18 18 18 18 18 20 18 18 18 18 18

6  Contracts and Pay at St Paul’s Cathedral …     153 Table 6.4  Mason’s day rates on the columns (from “An account of masons time it was at the columns” (undated loose sheet) TNA C106/145) Name

Day rate in d.

John Harntnoll Thom Deene Thom Atkins Peter Abraham Edward Bibby Will Sutton Rich Duffell John Panton Matt Warner Nic Abraham C. Crowther Roger Healing Joseph Smith John Tuckey Andre Lewes Henry Turner Will Ash Mic Growder Will Stonouse Jam. Tilley Thom Bibby Gill Huett Thom Bradle Abr. White Thom Jones Richard Day

28 28 28 28 28 24 26 28 26 28 26 20 30 30 30 28 30 30 20 28 18 14 20 28 28 26

Men who were ‘on call’ were never paid more than 30 d. per day and some, such as Roger Healing (20d. per day) were paid much less. (William Showers, who worked for Kempster at 18d. a day appears in the Cathedral’s list of masons on call in 1708.18) Interspersed in the pages of Kemspter’s book is a single page list of the “masons who worked on the columns”. These refer to the fluting of the columns of the west front. This would have been not only skilled, but as the special singular page suggests, perhaps prestigious. The mean average day rate for those working on the columns was 26.5d. per day (Table 6.4). 18LMA

CLC/313/I/B/003/25473 no. 41, p. 12, July 1708.

154     J. Z. Stephenson Table 6.5  Levels of skill and ranges of day rates paid in Kempster team Skill level

Day rate range

# in data set

Unskilled men Labourers Skilled fitters, layers, hewers Master masons Master masons and carvers Specialists Foremen and carvers

10–17d. 18d. 19–27d. 28d. 30d. 32–34d. 36–40d.

13 63 27 37 23 13 2 179

Hierarchy, Skill and Pay How are we to arrive at a meaningful day wage figure that can give an indication about how much an average craftsman or labourer earned in this period? This is difficult, because, as can be inferred from the sources and tables, the average day rate for any construction team was a function of the level of skill on-site in any particular period, and that was a function of the stage-dependent nature of construction. Generally, Kempster’s records indicate a wide distribution of masons pay, whilst the rate for labourers was fixed. Table 6.5 gives a general summary and banding of skills and rates. To address the question of labourers first, after 1700-1 when some were paid 16d. a day, Kempster pretty much consistently paid labourers a rate of 18d. per day year-round, but the Cathedral consistently paid 16d. in winter, and 18d. in summer, which would equate to a mean day rate of 17d. Despite this lower rate some of Kempster’s labourers agreed to be paid ‘on the call’ or join the Cathedral labourers team occasionally. Charles Thurland who was related to Kempster, and worked for him throughout this period, was employed by the Cathedral in 1709.19 There may have been other perceived benefits to being paid by the Cathedral—Cathedral labourers had the possibility of the opportunity to work as watchmen during the night. This night shift paid a full shilling

19LMA

MS 9172/113; Will Number: 100.

6  Contracts and Pay at St Paul’s Cathedral …     155

after 1700 (12d.), but there were only a very few watchmen hired, out of a large labourers’ team, and those small number of men were limited to a maximum of eight nights’ work a month. The men who did get watchmen’s work tended to be those who had worked at the Cathedral for a very long time. For instance, in 1702–1704, out of 77 labourers who were entered on the books as labourers throughout the 2 years, watchmen’s work went to just 12 men; Simon Satchel (observed on the books since the 1670s) nine other men most of whom who had a history at the Cathedral stretching a decade or more, and two foremen. This suggests that the Cathedral’s labourers’ team may have operated some sort of internal labour market where length of tenure and seniority returned an excess wage. The number of labourers on-site varied hugely throughout the term of building, being as few as eight in one month, to over 150 in a quarter. The number of watchmen did not expand in line with the numbers of labourers and sometimes there was no watchmen’s work (perhaps if there were no valuable materials on-site). The watchmen’s work records also throw up some highly unlikely working patterns, with some men listed as having worked 30 days and 8 nights in a month. Perhaps such experienced men subcontracted the nights—or, indeed, the days. The effect of the night shifts was that it raised the annual or monthly pay of some of the most long-standing labourers by about 20%, if, and only if they worked the full number of days open to them also. The effect on average pay, however, was negligible as such a small number of men had access to the excess wage of the nights. The exact difference it would make in the average wage would be a function of the total of number of men who worked on-site, and this was highly variable. Since the men who received the excess wage of nights were more ‘senior’ than the other labourers, the representative average day rate for a labourer at St Paul’s is between 17d. and 18d. One should bear in mind that the Cathedral hired boys, lads, and disabled men through the 1670s–1690s at 8–12d. a day. For masons the question of an average wage is far more complex because the nature of skill employed at St Paul’s was probably more specialist than at other sites. A modal wage taken from these records would give too much weight to carvers and foremen, whilst an average wage would not have been received by very many men. For Kempster’s team a straight non-weighted average of all craft day rates found in Kempster’s

156     J. Z. Stephenson

books gives a figure of approximately 23.5d., but a weighted average of all paid is 27d. Averages for given period or task like the columns are about 26d. Masters—that is freemen—earned mostly 28d. a day. Lesser skilled journeymen earned between 20d. and 26d. Training or unskilled men 10–18d. Skilled men’s day pay was variable for both skill, seniority and productivity. Work that did not include carving would perhaps have had a different profile of pay. The likelihood is that ‘ruff’ layers and journeymen on other projects which did not result in the west front of St Paul’s and the geometric staircase in the south west tower would have been paid the lower rates found here. We can only speculate whether highly specialist, skilled men would have also worked at lower levels of skill and pay when such specialism was not required. Margin is only partly a guide to mason’s pay, because the bargaining power of the Cathedral Commissioners was so substantial. If we only take into account mason’s pay on day rates, Kempster’s margin on labour at St Paul’s was 13%, with a range from zero to 30%. Since we cannot observe the costs of his materials because we do not have access to his other books and journals we have no idea what his overall margin was. Because of the unusual organizational arrangements at St Paul’s, he was making no margin on labourers’ day work, but had he been employing them at other sites where the charge-out rates were higher he would have been making a substantial margin. For these years, Boulton’s data has no craftsmen’s observations lower than 30d. per day, and only one less than 22d. per day for labourers.20 This implies that the St Paul’s rates had not kept up with a rise in charged rates elsewhere from the 1660s. Table 6.6 calculates the differences found between the rates published by Boulton, as representative of charge-out rates in a broader sample of locations and the Kempster pay, expressed as a percentage difference. I have taken 17d. as the average rate for labourers, and 26d. for masons. On average, the difference between Boulton’s rates and Kempster’s for craft is 20%, and for labour 31%. However, the day rate that men received was not the main determinant of a man’s pay or income, because the amount of days that men worked for Kempster varied enormously. 20Boulton,

‘Wage Labour’.

6  Contracts and Pay at St Paul’s Cathedral …     157 Table 6.6  Percentage difference between Boulton series and Kempster pay per day for craftsmen Year

Boulton Kempster % difference Boulton craft d. craft d. labour d.

St Paul’s labour d.

% difference

1701 1702 1705 1706 1707 1708

32 36 34 32 32 31

17 17 17 17 17 17

−29.1 −29.1 −32 −34.6 −34.6 −32

26 26 26 26 26 26

−18.75 −27.7 −23.5 −18.75 −18.75 −16.1

24 24 25 26 26 25

The Number of Days Worked and Income In addition to giving us new information on the level of day pay for labourers and craftsmen, Kempster’s books give us a unique insight into the hours and days of work that London’s building workforce might have experienced. A building worker’s income was dependent on two things—his day rate, and how much work he got. The number of days worked in a year is essential information with which to turn ‘day wages’ into the income with which we estimate early modern standards of living.21 Surprisingly, we have very little information about the number of days a man might have worked from other sources. Any evidence from this period is relevant to theories about an ‘industrious revolution’, or how an increase in labour inputs per head in the long eighteenth century contributed to economic growth.22 The most substantive contribution to this subject, and virtually the only data for London (although not for construction workers) has been made by H.J. Voth, who analyzed the depositions of those giving evidence at the Old Bailey to show that working days in London increased between 1760 and 1800.23 Voth’s calculations for number of working days per year

21See Allen, ‘Prices and Wages’. Current average income estimates assume 250 days’ work per year for urban workers. Clark estimates 300 in The Condition of the Working Class. 22See de Vries, The Industrious Revolution; Allen and Weisdorf, ‘Was There an Industrious Revolution Before the Industrial Revolution’; for a summary of estimates see Broadberry et al., British Economic Growth, pp. 263–265. 23Voth, ‘Time Use in Eighteenth Century London’, p. 38.

158     J. Z. Stephenson

turned on the evidence that Mondays were not worked by many until the later part of the eighteenth century. The view, simply stated, is that before 1750 there was a twofold limitation on Monday work. The first, a labour demand constraint, where ‘Saint’s’ days were stipulated by employers or communities as non-working days. The second limitation can be thought of as more related to labour supply, where casual or informal Saint’s Mondays were taken to recover from the weekend. Voth postulates that both these had been mostly eradicated by 1800. According to Voth’s calculations before the 1750s there were 208 working days a year of (on average) 11 hours, which rose in 1800 to 306 days of 11 hours.24 Voth included a construction case study, that of the Burnton and Western canal in Cheshire in 1801, as an example of this later industriousness where he found a high number of days worked, and Mondays regularly worked. The canal site was operational 6 days a week. 5924 days were worked on the canal in one year, but an average number of days per man was not given.25 Essentially, the site in 1801 has some strong similarities to St Paul’s and sites 100 years before. Voth argued that the case indicated increased industriousness generally by 1800. In fact, the case shows how difficult it can be to find evidence of working hours for the period before the nineteenth century (something that should alert us to whether it was an important part of employment, wage formation and working practice generally). In order to understand what a day’s wage means, and because Voth calculated working hours per year rather than days, we need to know what the length of the working day was. The call bell at St Paul’s reputedly was rung at 6 a.m., 1 p.m., and 6 p.m.26 Winter times are not differentiated in any of the records. With a break, or breaks, for men to eat in the middle of the day, this implies a working day of about 11 hours. By comparison, the Middle Temple’s records from June 1722 have a bill from Edward Stanton, mason, which details half a day’s work for 24Voth,

‘Time and Work in Eighteenth-Century England’, pp. 37–39. pp. 47–49. 26Lang, Rebuilding St Paul’s, p. 81. This sort of day differs from the system in the Dutch Republic where ‘shoften’ were punctuated by three breaks throughout the day: de Vries, ‘An Employer’s Guide to Wages and Working Conditions in the Netherlands’, pp. 47–64. 25Ibid.,

6  Contracts and Pay at St Paul’s Cathedral …     159

a mason and 3 hours for a labourer costing a total of 2s.27 If masons were charged out at 3s. a day and labourers at 2s. a day (as was the rate in the other accompanying lines of the bill28), then this would imply the amount for the labourers’ 3 hours was just 6d., and it follows that 3 hours was a quarter of a day and 12 hours was a full day. Given the St Paul’s evidence of bells, and by implication breaks, it seems a plausible working assumption that of those 12 hours, 11 were worked.29 If shorter hours were worked in winter there is no evidence of what these might have been. It should be highlighted however, that not everyone was paid by the day (as Chap. 7 will show the unit of working time for most workers at London Bridge was the tide). Although tides and nights were also a feature of Dockyard workers’ pay schemes, frequently as overtime, there is not generally a provision found for overtime (with some small exceptions discussed in the section below) found in building accounts, and no evidence for it in Kempster’s books. All large sites reviewed for this study show 6 days a week worked throughout the seventeenth and eighteenth century with a conspicuous absence of St Mondays or other holy days. Annual account books at St Paul’s, the Office of King’s Works and London Bridge all indicate that it was usual for there to be one week of four days’ work only in December, (Christmas) and two weeks with only five days worked in the Spring, usually after Lady Day (Easter). There is no evidence of the fifty-three saint’s days that Voth and others assumed were not worked in the seventeenth or early eighteenth century.30 As a result the total number of possible working days in the year was a considerable 308–309. A normal working week was up to 66 hours on normal time, very long by twenty-first century standards.31 27In

folder ACCVOUBI MT.2/TUT, Middle Temple Archives. Paul’s labourers were still paid 18d. in summer and 16d. in winter at this time. 29Voth actually notes eating or break times of 1.5 hours, in ‘Time and Work in EighteenthCentury London’, n. 31, but since 3 hours is ¼ of a day in the Middle Temple accounts described here, more than 1 hours break seems unlikely; therefore, I accept Voth’s figure of 11 hours per day as generally applicable. 30Freudenberger and Cummins, ‘Health, Work, and Leisure’, p. 9. 31The literature indicates that the trend from the beginning of the nineteenth century was a reduction in hours from eleven, through ten, toward eight with a half day on Saturdays; Postgate, Builder’s History, pp. 97, 135–136. 28St

160     J. Z. Stephenson Table 6.7  Carpenters on day work, quarterly accounts, St Paul’s, 1696–1700 (quarterly averages of total numbers of men observed and modal days worked) Quarter

Men

Days

Michaelmas Christmas Lady day Midsummer Total annual

28.4 19.8 26.8 28.25

57.6 31 45.8 56.75 191.15

Table 6.8  Average number of days worked for all men 1700–1701 1701–1702 1706–1707 1707–1708 1708–1709 Average Coefficient of variation (%)

152.89 62

167.44 55

122.36 73

142.42 68

113.05 88

Although sites were operational on this number of days, in all accounts there is a distinct seasonality, with less work, bills, and men recorded in work in January and February than in other months. At St Paul’s, Christmas quarter or the Candlemas period (December 25– March 25) shows up to 30% fewer bills and carpenters’ and labourers’ days than Michaelmas—which was usually the busiest period (Table 6.7). Work continued through the winter for a small band of men. Kempster’s team was present on site 47–49 weeks in the years given in the daybooks. The weeks not worked were all in January/early February. Overall this means that if work intensity was highest in July. This level of seasonality is significant, but it is less than that suggested by some previous historians of construction.32 Throughout the records there is a very low number of average days worked per man in any 52-week period (Table 6.8). As in construction everywhere there was a high level of churn or turnover at every level of skill (Table 6.9). Men worked for other contractors and employers, of course, but the high churn and level of weeks worked imply high search and matching costs, something still found in construction today.33 32See

Woodward, Men at Work, pp. 132–141. ‘Duration of Unemployment in the Construction Industry’, pp. 113–118; Anagboso, ‘Understanding the Divergence Between Output and Employment in the UK Construction Industry’, pp. 42–51.

33Perloff,

6  Contracts and Pay at St Paul’s Cathedral …     161 Table 6.9  Average number of days worked per annum by day rate paid Day rate

Number of days 1700–1709

18d.