A research agenda for political demography 2021938690, 9781788975742, 9781788975735, 178897574X

Table of contents :
Front Matter
Copyright
Contents
Figures
Contributors
1 Introduction to A Research Agenda for Political Demography
PART I Foundations of demographic change
2 Research in population aging: uncharted territory
3 Progressing research on forced migration
4 Urbanization: poverty,conflict, and climate change as causes and consequences
5 Sex, demographics and national security
PART II Implications of demographic change: context and connections
6 Whiteshift: demographic change, populism and polarization in the West
7 Wombfare: the weaponization of fertility
8 Health and demography
9 Population, rebellion and revolution
10 A research agenda for youth policies and investments
11 Demography and democracy
12 Demographic engineering and strategic demography
13 The demographic dividend: positive prospects, unclear path
14 Forecasting in age-structural time
15 A twenty-first century agenda for policy-relevant demographic research
Index

Citation preview

A Research Agenda for Political Demography

Elgar Research Agendas outline the future of research in a given area. Leading scholars are given the space to explore their subject in provocative ways, and map out the potential directions of travel. They are relevant but also visionary. Forward-looking and innovative, Elgar Research Agendas are an essential resource for PhD students, scholars and anybody who wants to be at the forefront of research. Titles in the series include: A Research Agenda for Social Finance Edited by Othmar M. Lehner

A Research Agenda for Animal Geographies Edited by Alice Hovorka, Sandra McCubbin and Lauren Van Patter

A Research Agenda for Regional and Local Government Edited by Mark Callanan and John Loughlin

A Research Agenda for Experimental Economics Edited by Ananish Chaudhuri

A Research Agenda for Geographies of Slow Violence Making Social and Environmental Injustice Visible Edited by Shannon O’Lear A Research Agenda for Evaluation Edited by Peter Dahler-Larsen

A Research Agenda for International Business and Management Edited by Ödül Bozkurt and Mike Geppert A Research Agenda for Political Demography Edited by Jennifer D. Sciubba

A Research Agenda for Political Demography Edited by

JENNIFER D. SCIUBBA

Department of International Studies, Rhodes College, USA

Elgar Research Agendas

Cheltenham, UK • Northampton, MA, USA

© Jennifer D. Sciubba 2021

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA

A catalogue record for this book is available from the British Library Library of Congress Control Number: 2021938690 This book is available electronically in the Political Science and Public Policy subject collection http://dx.doi.org/10.4337/9781788975742

02

ISBN 978 1 78897 573 5 (cased) ISBN 978 1 78897 574 2 (eBook)

Contents

List of figuresvii List of contributorsix 1

Introduction to A Research Agenda for Political Demography1 Jennifer D. Sciubba

PART I

FOUNDATIONS OF DEMOGRAPHIC CHANGE

2

Research in population aging: uncharted territory Jennifer D. Sciubba

17

3

Progressing research on forced migration Tiffany S. Chu

29

4

Urbanization: poverty, conflict, and climate change as causes and consequences Matthew Cobb and Alex Braithwaite

5

Sex, demographics and national security Valerie M. Hudson

PART II

45 61

IMPLICATIONS OF DEMOGRAPHIC CHANGE: CONTEXT AND CONNECTIONS

6

Whiteshift: demographic change, populism and polarization in the West Eric Kaufmann

7

Wombfare: the weaponization of fertility Monica Duffy Toft

101

8

Health and demography Jeremy Youde

115

81

v

vi

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

9

Population, rebellion and revolution Jack A. Goldstone

131

10

A research agenda for youth policies and investments John F. May

147

11

Demography and democracy Hannes Weber

161

12

Demographic engineering and strategic demography Michael S. Teitelbaum

179

13

The demographic dividend: positive prospects, unclear path 199 Kaitlyn Patierno, Elizabeth Leahy Madsen and Smita Gaith

14

Forecasting in age-structural time Richard Cincotta

15

A twenty-first century agenda for policy-relevant demographic research Suzanne E. Fry

215

235

Index245

Figures

3.1

Global count of refugees since 1960

30

3.2

Region of origin of refugees

31

6.1

Highest populist right poll vs medium Muslim share projection

91

6.2

News stories about immigration, those mentioning immigration as an issue and net migration over time

93

11.1

Population pyramids for Germany, 1925 and 2018

162

11.2

Youth ratio in 182 countries of the world, 2015

169

11.3

Youth ratio vs Freedom House Index in 182 Countries, 2015

170

13.1

Percent likelihood of achieving four dividends by median age in Nepal

202

14.1

Logistic curve fitting across the age-structural transition 218

14.2

Categorical age-structural functions

222

14.3

Probability of conflict

226

vii

Contributors

Alex Braithwaite is a professor in the School of Government & Public Policy at the University of Arizona. His teaching and research address the causes and consequences of political violence, strategic nonviolence and armed conflict between states, with a special focus on patterns of forced migration. Tiffany S. Chu is a researcher using a multi-method approach to answer policy-driven questions. Her research focuses on decisions taken by forced migrants and conflict processes and has been showcased in multiple publications. She received her PhD and MA from the University of Arizona and completed her BA at UC Berkeley. Richard Cincotta is a demographer and population biologist whose recent research has focused on the demographic transition’s influences on political, social and environmental conditions. He is active as a Wilson Center Global Fellow and with the Atlantic Council and Population Reference Bureau. He formerly served as Director of Demographic and Social Science Programs in the (US) National Intelligence Council’s Long-Range Analysis Unit, as a senior demographer at PAI, and in USAID’s Office of Population and Reproductive Health. Matthew Cobb is an assistant professor at Coastal Carolina University, teaching in the Intelligence and Security Studies department. His research focuses on political violence, security policy and political developments in developing countries. He received his PhD from the University of Arizona in 2020, where he researched the impact of urbanization on patterns of armed conflict in developing countries. Suzanne E. Fry is a member of the Central Intelligence Agency’s Senior Analytic Service and former Director of the Strategic Futures Group at the National Intelligence Council (NIC), where she was responsible for long-range and global issues analysis, including NIC’s flagship unclassified assessment of the future strategic context, Global Trends: Paradox of Progress (2017). Dr. Fry received her PhD in Politics from New York University and BA from ix

x

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

the University of Notre Dame. She is a member of the Council on Foreign Relations. Smita Gaith is a former Senior Policy Advisor at Population Reference Bureau (PRB), where she focused on translating and communicating research on population and development. Prior to joining PRB, Smita contributed to research and advocacy activities at Women Fighting AIDS in Kenya, Women Deliver and Population Action International. Smita holds a Master of Public Health from New York University and a Bachelor of Science from The College of New Jersey. Jack A. Goldstone (PhD Harvard) holds the Virginia E. and John T. Hazel, Jr. Chair in Public Policy at George Mason University. He has received the Distinguished Contribution to Scholarship Award from the American Sociological Association, and fellowships from the MacArthur Foundation, the JS Guggenheim Foundation, the Carnegie Foundation and the Mellon Foundation. His recent books include Revolutions: A Very Short Introduction (Oxford University Press, 2014) and Revolutionary Waves of the 21st Century, coedited with Leonid Grinin and Andrey Korotayev (Springer, forthcoming). Valerie M. Hudson is a University Distinguished Professor in the Department of International Affairs at The Bush School of Government and Public Service at Texas A&M University, where she directs the Program on Women, Peace and Security. Eric Kaufmann is professor of politics at Birkbeck College, University of London. His books include Whiteshift: Populism, Immigration, and the Future of White Majorities, Shall the Religious Inherit the Earth? (Abrams Press, 2019) and The Rise and Fall of Anglo-America (Harvard University Press, 2004). Elizabeth Leahy Madsen is an Associate Vice President at Population Reference Bureau (PRB), has over 15 years of experience leading research and advocacy initiatives that inform policy change in the global health sector. She has implemented programs and partnerships across sub-Saharan Africa and South Asia and has published widely on political demography and the intersections between population and multisectoral development. Elizabeth holds an MA in international affairs from The George Washington University and a BA from Knox College. John F. May is a specialist on population policies. He was a lead demographer at the World Bank and has worked for the United Nations and the Futures Group International, a US consulting firm. He currently teaches demography at the Schar School of Policy and Government, George Mason University. He has published World Population Policies (Springer, 2012) and co-edited

CONTRIBUTORS

xi

Africa’s Population: In Search of a Demographic Dividend (Springer, 2017). He received his PhD in Demography Summa cum laude from the University of Paris-V (Sorbonne) in 1996. Kaitlyn Patierno is a Program Director at Population Reference Bureau (PRB), specializing in global family planning policy, program and advocacy initiatives. Prior to joining PRB, Kaitlyn supported reproductive, maternal and child health programs at the US Agency for International Development in the Bureau for Africa. Kaitlyn holds a Master of Public Health from the University of California at Berkeley and a BA from The George Washington University. Jennifer D. Sciubba (PhD University of Maryland) is affiliated with the Center for Strategic and International Studies, the Environmental Change and Security Program at the Wilson Center and the Rising Powers Initiative at Boston University. She has authored The Future Faces of War: Population and National Security (ABC-Clio, 2011) and 8 Billion and Counting: How Sex, Death, and Migration Shape Our World (W.W. Norton, 2022). Dr. Sciubba teaches as an Associate Professor at Rhodes College and is a member of the Council on Foreign Relations and Phi Beta Kappa. Michael S. Teitelbaum is a demographer at Harvard Law School’s Labor and Worklife Program. His current research interests include political demography; low fertility; international migration; and science and engineering labor markets. Previously, he was a faculty member at Princeton and Oxford; Vice President of the Alfred P. Sloan Foundation; and Vice Chair and Acting Chair of the US Commission on Immigration Reform.  He was educated at Reed College and at Oxford University, where he was a Rhodes Scholar. Monica Duffy Toft is Professor of International Politics and Director of the Center for Strategic Studies at Tufts University’s Fletcher School. Toft is a Global Scholar of the Peace Research Institute in Oslo, a member of the Council on Foreign Relations and a non-resident fellow of the Quincy Institute. She is the author of seven books and has published widely on international relations. Before college, she spent four years in the US Army as a Russian linguist (honorably discharged). Hannes Weber holds a PhD in Social Sciences and Economics from the University of Stuttgart, Germany. He was a post-doctoral fellow at the Mannheim Centre for European Social Research, University of Mannheim, before leaving academia to join the industrial sector in 2019. His research mainly focuses on the social and political consequences of demographic change and migration. He recently published a book on Germany’s demographic future, as well as scholarly articles in journals such as European Sociological Review, European Journal of Population, and Demographic Research.

xii

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Jeremy Youde is the Dean of the College of Arts, Humanities, and Social Sciences at the University of Minnesota Duluth. A political scientist by training, he previously held appointments at the Australian National University, Grinnell College, and San Diego State University. He is the author of five books on global health politics, with the most recent being Globalization and Health (Rowman and Littlefield, 2019).

1

Introduction to A Research Agenda for Political Demography

Jennifer D. Sciubba

Introduction The human population has fundamentally transformed over the last several centuries, with radical changes in our birth and death patterns and our distribution across the continents. In the wake of these demographic shifts, how we organize ourselves politically, the design of our economic systems, and even our social relationships have changed in response. As the chapters in this book show, demographic dynamism continues to shape the character of societies. In high- and middle-income countries, the most recent transition is to extremely low fertility and low mortality, leading to a shift in the composition of various age groups – far more elderly than youth and declining proportions of those in the middle ages. For the world’s most developed countries, national goals of economic growth of 2 percent or more are mismatched with shrinking populations – the idea of infinitely expanding economies is rubbing up against demographic reality. In some states with low fertility, immigration is eroding the advantages of long-time ethnic majorities and political tensions are high. Rising support for anti-immigrant far right parties and populists, particularly in the USA and Europe, are demonstrations of the connection between demographics and politics. In lower-income countries, fertility remains high, but declining mortality means that these populations are growing exponentially – a different transformation. Population density is increasing as the amount of available land stays constant and the number of people who inhabit it grows two- or three-fold. Climate change is accelerating strains on the land itself, and economic forces like globalization are restructuring economies, often toward production for 1

2

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

export, rather than for subsistence. Economic crises too often turn into civil conflict, which then pushes populations into new communities and across borders, and creates a new set of problems for both senders and receivers. The scholars contributing to this volume devote their research to understanding what such demographic transformations mean for how the world works. As such, we are part of a long tradition. Scholars have been conducting research on population and politics for centuries. The Reverend Thomas Malthus warned in 1798 that human population growth was outpacing food production; French demographers and military leaders blamed France’s slow population growth for the country’s loss to Germany in the war of 1870. In his chronicle of world population changes, Massimo Livi-Bacci observed that Europe’s demographic changes were “an integral part of the transformation of European society” (Livi-Bacci, 2007, p. 116). Today, we describe such research under the term “political demography”, the study of the politics of population change. As scholar Myron Weiner defined it in a 1971 report for the National Academy of Sciences: “Political demography is the study of the size, composition, and distribution of population in relation to both government and politics” (Weiner, 1971, p. 567). Despite the tradition of studying the effects of population changes, only a small community of scholars knows what “political demography” is. The purpose of this volume is to capture the state of the field, demonstrate its relevance for other fields, and chart a course for advancing research. To that end, this volume brings together a vibrant collection of political demography scholars. Their chapters discuss the causes and consequences of demographic change, and connections between demographic, political, economic, and social dynamics. These authors not only describe accumulated knowledge for their issue area, but also ask: What don’t we know? What are the key puzzles, data, and theories that need to be developed? Overall, we find more questions than answers, making this a field ripe for further study. With nearly 8 billion diverse people on the planet divided into some 200 countries, exploring the impact of demographics has never been more relevant. Headline after headline reports tragic refugee flows, public health crises, and scarce resources amidst growing populations. Yet, humans are living longer than ever and the achievements of aging societies illustrate the huge leaps humankind has made in controlling its own destiny. There is a growing acceptance that at a minimum demographics form an important part of the background against which larger political, economic, and social processes take place. Arguably, there is an even greater role for population as a causal factor in

INTRODUCTION

3

some of the most pertinent debates of today, such as causes of war and peace, or poverty and abundance. Population dynamics are also recognized for their role in shaping the domestic distribution of power. This brief introduction first describes the basics of population dynamics. Then, through exploring the relationship between population and politics, it summarizes the types of dynamics population helps determine. Political demography research is not without controversy and has many limitations, as the third section describes. The final section offers an outline for the chapters that follow.

The relationship between population and politics Population change results from fluctuations in fertility, mortality, and migration. Imagine these three components as dials, like the volume dials on a radio, that turn up or down to produce different outcomes in combination. Some societies have high fertility, high mortality, and low migration – their overall population size, composition, and distribution will look quite different from a society where the dials are turned in different directions. With a total population of 24 million in 2020, Niger’s fertility was the highest in the world at 7.1 children per woman, life expectancy was only 59 years, and just 16 percent of the population lived in urban regions. There were more people leaving Niger than coming in. In contrast, Singapore’s fertility was one of the world’s lowest, barely above one child per woman. With a smaller total population of 5.8 million, immigration was strong, life expectancy at birth was 83 years, and the population was 100 percent urbanized (Population Reference Bureau, 2020). These are just two examples, but infinite combinations of fertility, mortality, and migration mean there is no shortage of population dynamics to study. What causes such differences in demographic dynamics? It should not surprise anyone – especially anyone who has made the decision to move away from home or weighed whether to have a child – that the reasons for demographic change are many and complex and we can draw on the wider discipline of demography to better understand those factors. But one determinant is political, and among political demographers there is some discussion of these determinants of population change. Political demography scholars may study how the infrastructure put in place by governments, including physical infrastructure like sanitation systems and networks of health clinics, can shape mortality outcomes. For example, Nigeria’s response to the West African Ebola outbreak in 2014 was swift and effective, limiting the country’s outbreak to 19 cases and

4

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

7 deaths, while Liberia faced 300 to 400 new cases per week during the peak (World Health Organization, 2014; World Health Organization Africa, 2015). The impact of political response holds true in wealthier countries as well. Both Russia and the United States have lagged far behind peers in improving societal life expectancy as health systems fail to address unhealthy lifestyles and the rampant heart disease and stroke that result. The gap between life expectancy for men and women is wider in Russia than anywhere else in the world (Starodubov et al., 2018). Even though the USA spends more on healthcare than any of its peers, access to and quality of healthcare is uneven (Tikkanen & Abrams, 2020). As a result, with a life expectancy at birth of 78.7 years, in 2020 the USA ranked 27th out of 37 OECD countries, just behind Chile and tied with the Czech Republic (Life expectancy at birth (indicator), 2020). Differences in fertility, mortality, and migration produce changes in population size, composition, and distribution. Between 1950 and 2050 there will have been tremendous shifts in the overall size of the global population, from 2.5 billion to somewhere between 8.5 and 9.2 billion (arguably more if fertility declines slow), and in the list of the world’s largest countries. Table 1.1 shows a clear geographic shift away from Europe and East Asia to Africa and South Asia among the world’s most populous countries between 1950 and 2050. A key driver of that geographic shift is the pace at which countries (and regions) are going through or completing the demographic transition from high to low fertility and mortality. While Europe, North America, East Asia, and most of Latin America have already completed the transition and are experiencing slow or no population growth, some scholars argue that Africa’s fertility transition is stalled (Bongaarts, 2017). These global patterns mean that there is a demographic divide between an aging global North and a youthful and growing global South. The benefit of a demographic lens is that through it we can see the future – not clearly, of course, but the general contours and trends. Today’s workers are tomorrow’s retirees; the kindergarteners filling desks today will fill the ranks of soldiers in the future. A demographic lens is particularly useful for long-range economic and foreign policy forecasting. The world is complex, but demographics are relatively predictable. One certainty is that development is accompanied by demographic aging. As Massimo Livi-Bacci observed in his Concise History of World Population, “no population has maintained high levels of fertility for long in the face of increasing well-being and declining mortality” (Livi-Bacci, 2007).

China

India

United States

Russia

Japan

Germany

Indonesia

Brazil

United Kingdom

Italy

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

46

50

53

69

69

82

102

158

376

554

Population (millions)

Nigeria

Japan

Bangladesh

Pakistan

Russia

Brazil

Indonesia

United States

India

China

Country

122

128

132

139

146

175

212

282

1,053

1,283

Population (millions)

2000

Ethiopia

Democratic Republic of Congo

Bangladesh

Brazil

Pakistan

Indonesia

United States

Nigeria

China

India

Country

191

198

202

233

307

322

390

411

1,364

1,659

Population (millions)

2050

Source: UN Department of Economic and Social Affairs, Population Division (2019); UN World Population Prospects, the 2019 Revision, United Nations DESA.

Country

1950

Top 10 most populous countries, 1950, 2000, and 2050

Rank

Table 1.1

INTRODUCTION 5

6

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Although the causal arrow runs both ways between demographics and politics, the scholars in this book mostly look at the effect of the former on the latter. With this brief review of population dynamics, we now turn to the themes that emerge from this collection of chapters.

Research in political demography Relationship to development Countries with high fertility and extremely youthful populations are some of the poorest on the planet, and some scholars have gone so far as to argue that lowering fertility is a precondition for economic development. It is easy to see why these arguments have traction: when half of a population is children and adolescents, per child investment at both the family level and the state level is often low and human capital accumulation is slow. Lower fertility eventually produces a higher ratio of working-age population to dependents, an age structure seen as favorable for accelerated economic growth, known as the demographic dividend. But lower fertility and the age structure that results does not automatically produce development – states have to capture their demographic dividend through targeted policies in sectors like education, health, and finance. While the promise of a demographic dividend is seductive, as Patierno, Madsen, and Gaith describe in Chapter 11, only a handful of states have truly made the most of the window after initial fertility declines in which the majority of the population is of working ages. Demographic engineering has taken place to speed up the demographic dividend in many places around the world, most notably India and China. However, such policies (like China’s one-child policy) have led to unforeseen consequences, like gender imbalance and rapid population aging. As scholars, we know little about economic development under conditions of population aging because aging is so new. There are many countries that are growing old before becoming rich. Have they missed their window of opportunity forever? Is it possible for wealthy countries reaching median ages over 40 years to maintain economic strength, even as the proportion of older age population grows relative to those in the prime of life? Many such questions about the economics of aging remain. Finally, we know that the countries that industrialized first urbanized first, and up to two-thirds of the global population will be urbanized by 2050 (Chapter 4). If past were precedent, we would expect this distribution to confer eco-

INTRODUCTION

7

nomic benefits, but like with lower fertility, this process is not automatic, and scholars are trying to understand the growing phenomenon of urbanization without growth. Relationship to domestic politics Those countries with high fertility are not only more likely to be poor; they have a greater likelihood of civil conflict, instability, and violence. States with youthful age structures are more likely to be nondemocratic, and some political demography scholars argue that “premature” transitions to democracy – before populations attain a demographic transition to older median age – are at greater risk of “backsliding” to authoritarianism. While the converse has been true with rare exception, as median ages rise across the world the roster of states with older age structures is growing more diverse and less democratic. Cuba, Iran, Singapore, and Russia are just a few of the states with less than full democratic institutions but rising median ages. The correlations between youthful age structure and civil conflict are well-documented but the causal mechanisms are less understood. We know that rapidly growing populations usually face intense pressures on job markets, and lack stable or effective institutions. Scholars have focused on experiences of youth, from around ages 15 to 25, to try and understand why risk of violence is higher in youthful countries than demographically older ones. It seems that increasingly large cohorts of youth are closed out of opportunities to exercise political voice, establish social ties, and make a living. For example, as described by Valerie Hudson in Chapter 5, an inability to get married in societies where marriage is a key marker of adulthood can lead to increased grievances among non-elite men closed out of the marriage market, and thus a higher likelihood of violence. But not all youthful societies experience violence, and there is much left to discover about the conditions under which a youthful age structure is a bane or a blessing. Political demographers also study the relative size of groups and competition for political dominance. In societies without firmly established democratic institutions, such competition can take the form of violence among ethno-national groups. In democracies, competition is more likely to take place at the polls, but is still contentious. Demography is not necessarily about “natural processes”, and in fact states and sub-state groups have tried to manipulate their demographics by encouraging higher fertility, suppressing minority groups, and even advocating for genocide. Monica Duffy Toft in Chapter 7 describes several cases, including Lebanon, Israel, and Palestine,

8

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

that demonstrate the strong link between demographic manipulation and instability or conflict. Relationship to global power There is some connection between the population of countries and the distribution of political and economic power in the international system. The most populous countries tend to be some of the most powerful. But military prowess is about more than the number of boots on the ground – technology and quality of soldiers are important as well. Alliances can boost a state’s numbers, if their interdependent economies and cooperative populations are considered together (Sciubba, 2012). And, we know that a state need not be one of the world’s most populous to capture the world’s attention, as Cuba and North Korea have shown. Population plays a role in economic strength, but so do factors such as productivity. In fact, a rapidly increasing population can inhibit economic growth. The Democratic Republic of Congo and Ethiopia will be on the list of the top ten most populous states in the world by 2050, but are highly unlikely to see concomitant growth in economic and political power. Over the last two decades, Mexico replaced Japan on the top ten list, but hardly replaced them in the global order. Importance of institutions Political demography insists that population cannot be understood in a vacuum. One important aspect is institutions, which determine the impact of population trends. Does the state have primary responsibility for caring for the elderly population, or does the family? How open is a state’s immigration policy and, once immigrants arrive, how liberal are citizenship laws? The international refugee regime is stressed and inadequate – it creates temporary solutions, if it creates solutions at all. It does not include the environment, even though climate change displaces more persons annually than does conflict. Likewise, although there are many global health institutions, their limited influence and challenges coordinating among them have meant slow responses to new issues and to crises. Institutions are often sclerotic and not responsive to population changes. They are reactive, not proactive, even if we can see the demographic changes on the horizon. Evolving challenges As we see, political demographers have used the lens of population to understand some of the most important questions in political science, including those about power, about war and peace, and about poverty and prosperity.

INTRODUCTION

9

A host of evolving issues, from climate change to disease to technology, continues to expand the political demography research agenda and challenge scholars to consider how existing theories hold up under changing conditions. As the human population increases in number and the human footprint increases in extensity through urbanization, suburbanization, and landscape changes for agricultural practices, humans come into contact with more animals and diseases but without immune resistance to them. The appearance of Covid-19 in 2020 showed just how intertwined populations were and how quickly disease could spread across political borders. Climate-induced migration is also growing in importance. In sub-Saharan Africa, Asia, and Latin America, millions have already been displaced with more expected, but the intersection between refugees, climate change and health is understudied. Finally, what is the role of technology vis-à-vis demography? As technology makes some jobs obsolete, certain groups inevitably get left behind. To the extent that this correlates with certain demographic subgroups – i.e., youth, certain ethno-national groups – this becomes a demographic issue.

What is problematic about political demography? While a main goal of this book is to demonstrate the strengths and utility of a political demography approach, we recognize that the study of population can be highly controversial. Part of the reason is that population counts have been used for nefarious purposes at many times and in many places. Political demography has been problematic in practice, as the loudest voices have often been the least informed, including non-academics who understand a bit about demographics but do not have a full picture of the institutional context within which population change takes place. Michael Teitelbaum has noted that pundits often make exaggerated claims for ideological or political ends, which undermines the whole field’s credibility (Teitelbaum, 2015). Some criticisms are of the field itself. Political demography involves long causal chains, and as Monica Duffy Toft mentions in Chapter 7, the field of political science prefers to study questions that capture direct causality, not longer-term, indirect factors such as population changes. There are also issues within the field itself, particularly definitions and measurements. While categories like “youth”, “urban”, or “old” are the cornerstones of research on composition, there are no firm boundaries or universal definitions for any of those categories. Some groups or countries engineer collection, classification, and reporting of data. As Cobb and Braithwaite point out in

10

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Chapter 4, “urban” is variously defined by land area, number of residents, administrative status, population size, or density. Assumptions about economic status and political participation of various age groups undergird political demography theories, yet “youth” and “elderly” are socially constructed categories that differ cross-culturally. Thus, cross-national comparison of such categories is highly problematic. I discuss these challenges in Chapter 2 on aging, pointing out that average age of exit from the workforce – which would qualify someone as an old-age dependent – differs widely even among the few countries of the OECD. When we look globally, comparisons of dependency are even more problematic. Likewise, as John F. May discusses at length in Chapter 10, the transition to adulthood is a murky one, yet research about likelihood of civil conflict in countries with high youth populations requires definitive boundaries. Is the experience of a 15-year-old in Nigeria comparable to that of a 15-year-old in Thailand? Should the upper limit for youth be 25 years, or 29? If our theories rely on assumptions about educational and marital status, then our measurements need to align. Finally, while demographics may provide the closest thing to a crystal ball we have for projecting future trends and scenarios, it is far from an exact science. Africa’s fertility transition has been slower than many demographers had projected (although in line with what other demographers expected); scholarly disagreements on the direction low-fertility societies will take abound as well. For example, many demographers and users of demographic data (like myself) took issue with widely publicized projections by the Institute for Health Metrics and Evaluation (IMHE) in the summer of 2020 and wrote an open letter to The Lancet journal, where the IMHE projections had been published, calling into question the data, models, scenarios, and interpretation of the projections (Gietel-Basten, Sobotka et al., 2020). And migration is even harder to project. As Michael Teitelbaum has said, you are born once and you die once, but you can migrate many times. Models that try to account for migration are necessary in a globalized world like ours, but they are messy. Migration involves many moving parts – pardon the pun – and understanding causality is difficult. Even in the realm of health, structural devaluation of women has embedded in data collection practices such that it is difficult to impossible to get health data disaggregated by gender, even though gender is a key health determinant (for both women and men), as Jeremy Youde points out in Chapter 8. Not only do women live longer, but many diseases manifest differently according to gender.

INTRODUCTION

11

Outline of the book Part I of the book introduces readers to the “Foundations of Demographic Change”: fertility, mortality, and migration. In Chapter 2, I look at how low fertility and lengthening life expectancy have led to the most aged societies the planet has ever seen. Because population aging is a relatively new phenomenon, there is much we do not know about the political consequences, and this chapter suggests deeper exploration of institutions as forces that magnify or dilute the impact of population aging. In Chapter 3, Tiffany S. Chu takes up the salient topic of forced migration, looking at how global migration governance has fallen woefully behind refugee flows. She describes the state of research on projecting migration crises, understanding the causes of migration, and governing migration. In Chapter 4 on urbanization, Matthew Cobb and Alex Braithwaite describe how the wide variety of urban experiences, from gleaming skyscrapers to crowded slums, carry both opportunities and challenges for peace and prosperity. With the world’s human population increasingly urbanizing, there is an immediate need for more research on the consequences for equality, governance, and security. In Chapter 5, Valerie M. Hudson explains how gender is central to political demography, not only because women are half of the population or because their numbers drive fertility statistics, but also because their place in society can determine outcomes for the entire state. Her research has shown that violence against women is connected to state-level violence, and skewed sex ratios at birth in multiple countries, including China and India, necessitate further research. Political demography draws liberally from other fields. Whereas a scholar of religious politics might put identity at the center of analysis and only tangentially consider demographics, a political demography perspective on religion sees the two as inextricably linked. Likewise, ethno-national issues are as much about making meaning and forming identity as they are about demographic change or perception of that change. Part II, “Implications of Demographic Change”, explores the connections between demographics and closely related issues, such as identity, religion, environment, and health. Eric Kaufmann, in Chapter 6, describes the intersection of race, identity, and electoral politics. In what he terms “Whiteshift”, majority white populations respond to rapid ethnic change by shifting allegiances to the far right. These dynamics are closely related to religious ones, as Kaufmann has found in prior research, and in Chapter 7, Monica Duffy Toft argues that religious and ethnic groups

12

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

(which often overlap) sometimes use demographics to increase their political position in a society – a phenomenon she terms “wombfare”. Closely connected is Chapter 8 on health, by Jeremy Youde. As he shows, disease can affect age structure just as fertility can. There is much work to be done in connecting health outcomes and political demography, and his chapter raises interesting questions about the links between maternal and child mortality and conflict, and between infectious disease and governance. The next two chapters of Part II look at both threats and opportunities from demographic change. In Chapter 9, Jack Goldstone examines how demographic trends influence a range of contentious politics, from protest to civil war. Similarly, in Chapter 10, John F. May discusses how youthful age structures are more likely to experience civil conflict, be mired in poverty, and have non-democratic governments. Hannes Weber in Chapter 11 explores the role of demography in why democracies fail. As he explains, when large cohorts of youth are excluded from social institutions such as marriage and jobs, they are more likely to take part in political violence. Michael S. Teitelbaum in Chapter 12 looks at how political actors have intentionally manipulated population trends to give themselves or their group advantages. Far from an anomaly, such “demographic engineering” is pervasive from ancient times to now and across cultures, regions, and political systems. In the book’s final section, Part III, “Putting Political Demography into Practice”, Kaitlyn Patierno, Elizabeth Madsen, and Smita Gaith describe the demographic dividend in Chapter 13. As they argue, while youthful populations are often associated with conflict, certain other age structures bring hope for improved peace and prosperity. Countries with “middle-aged” age structures, in those middle stages of the demographic transition from high to low fertility and mortality, are poised to reap the benefits of a large working-age population and relatively few dependents – but only with the right policies in place. The demographic dividend is a seductive idea for a developing country with a large and growing population. It is no surprise that countries in Africa have wanted to emulate the success of the “Asian tigers”. Richard Cincotta and Suzanne E. Fry show how useful political demography is for forecasting and how research in this field has and can continue to shape public policy in areas from economics to national security. In Chapter 14, Richard Cincotta describes how what he terms an age-structural maturity approach can provide a window to the future and help researchers and poli-

INTRODUCTION

13

cymakers understand where a population is headed in terms of regime type or level of economic development. In Chapter 15, Suzanne E. Fry rounds out the volume by offering a practical take on how political demography can directly influence public policy through the intelligence communities and beyond. As this collection of chapters shows, there are many fascinating and important questions about the relationship between population and politics yet to study. What scholars should find especially appealing about the field of political demography is that even as we find those answers, population continues to change, an infinite font of dynamics to study.

References Bongaarts, J. (2017). Africa’s unique fertility transition. Population and Development Review, 43, 39–58. Retrieved from https://​www​.jstor​.org/​stable/​pdf/​26622873​.pdf. Gietel-Basten, S., Sobotka, T. et al. (2020). Letter on ‘Fertility, Mortality, Migration, and Population Scenarios for 195 Countries and Territories from 2017 to 2100: A Forecasting Analysis for the Global Burden of Disease Study’ by S.E. Vollset et al. Retrieved from https://​osf​.io/​ytf6m/​?view​_only​=​53​89a004f7e9​4917b30b5a​ 4d4ea7a154. Life expectancy at birth (indicator) (2020). Retrieved from https://​ data​ .oecd​ .org/​ healthstat/​life​-expectancy​-at​-birth​.htm. Livi-Bacci, M. (2007). A Concise History of World Population (4th edn). Malden, MA: Blackwell. Population Reference Bureau (2020). 2020 World Population Data Sheet, Washington, DC. Retrieved from https://​www​.prb​.org/​wp​-content/​uploads/​2020/​07/​letter​ -booklet​-2020​-world​-population​.pdf. Sciubba, J. D. (2012). A new framework for aging and security. In J. A. Goldstone, E. P. Kaufmann & M. D. Toft (eds), Political Demography: How Population Changes are Reshaping International Security and National Politics (pp. 63–80). Boulder, CO and London: Paradigm Publishers. Starodubov, V., Marczak, L. B., Varavikova, E., Bikbov, B., Ermakov, S. P., Gall, J., … Naghavi, M. (2018). The burden of disease in Russia from 1980 to 2016: a systematic analysis for the Global Burden of Disease Study 2016. The Lancet, 392, 1138–46. Teitelbaum, M. S. (2015). Political demography: powerful trends under-attended by demographic science. Population Studies: A Journal of Demography, 69 (Population: the long view), S87–S95. Tikkanen, R. & Abrams, M. K. (2020). U.S. health care from a global perspective, 2019: higher spending, worse outcomes?, January 20. Retrieved from https://​www​ .commonwealthfund​.org/​publications/​issue​-briefs/​2020/​jan/​us​-health​-care​-global​ -perspective​-2019. Weiner, M. (1971). Political demography: an inquiry into the political consequences of population change. In N. A. O. Sciences (ed.), Rapid Population Growth: Consequences and Policy Implications (Vol. II, pp. 567–617). Baltimore, MD: Johns Hopkins University Press.

14

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

World Health Organization (2014). WHO declares end of Ebola outbreak in Nigeria [press release]. Retrieved from https://​www​.who​.int/​mediacentre/​news/​statements/​ 2014/​nigeria​-ends​-ebola/​en/​. World Health Organization Africa (2015). The Ebola outbreak in Liberia, May 9. Retrieved from https://​www​.afro​.who​.int/​news/​ebola​-outbreak​-liberia​-over.

PART I Foundations of demographic change

2

Research in population aging: uncharted territory

Jennifer D. Sciubba

Introduction In 1950, the cover of the popular American periodical, Newsweek magazine, drew readers’ attention to a worrisome population explosion in postwar Japan. The magazine observed that of all the pressing policy issues facing the Asian nation, including a peace treaty with Western powers, US presence on Japanese territory, and relations with Russia and “Red China”, “hardly anyone discussed the great long-range problem Japan will face after a peace treaty: How to support its rapidly growing population or control it if it can’t be supported” (“Too Many Babies? Japan Tries All-Out Birth Control”, 1950, p.  42). With direct echoes of the Malthusian ideas resurging at the time, Newsweek pointed out that Japan only produced enough food for a population of 65 million, but was on track to reach 100 million people by 1975. That projection was wrong: Japan reached 100 million before even 1970. Today, Japan still contends with tense regional relations and the presence of US soldiers on Japanese soil, and Newsweek still covers Japan’s population issues. Instead of population explosion, though, its stories are about the rapid shrinking and aging of Japan’s population, which peaked around 2010 at 128.5 million. Japan’s median age at the time of the 1950 cover story was 22.3 years. Today it is the world’s highest, at 48.4 years (UN Department of Economic and Social Affairs, 2019). The initial demographic transition described in this volume’s introduction has given way to a second demographic transition, as we see in Japan, with a host of novel implications. Demographers conceived the concept of a second demographic transition to understand observed patterns of postponement of marriage and childbearing and the shift to super-low fertility, far below 17

18

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

replacement levels of about two children per woman. As demographer Ron Lesthaeghe (2014) describes it, the first demographic transition focused on the child, including benefits of greater investment per child once smaller families were achieved; the second demographic transition is about “adult self-realization” and choices adults make about what type of lifestyle they want. What today’s adults want, so it seems, is much smaller families than their parents or grandparents desired. While high-income countries (using the World Bank’s classification) had an average total fertility rate (TFR) of 2.99 in 1950, that number is only 1.67 today (UN Department of Economic and Social Affairs, 2019). Yet, it is not only high-income countries that are seeing the shift to very low fertility and population aging. We can describe the evolution of population aging in terms of a first wave and a second wave. The first wave of countries to achieve a median age over 35 years comprised countries in North America and Europe, plus Japan. These were almost exclusively stable liberal democracies. The second wave of states to achieve the median age milestone is more geographically, culturally, economically, and institutionally diverse, meaning that our initial findings about the political implications of population aging are likely to be repeatedly challenged. This is good news for researchers new to the field, who will find many fruitful avenues to explore. Although ubiquitous among developed countries today and spreading rapidly, population aging is a relatively new demographic phenomenon and, as a result, research on the issue is somewhat underdeveloped compared to research on other demographic issues, like migration. However, the political demography field and other social sciences are catching up to this new reality, as population aging increasingly affects the world’s most powerful and developed countries – a great motivator to understand the implications of this novel demographic shift. This chapter describes the two main research areas of the politics of population aging. The first area explores intergenerational issues and the degree to which gerontocracy – rule by the old – exists in societies with higher proportions of older people. In short, does the size of population groups translate to political power? This is the most developed research line in the politics of aging, since the first wave of aging states was mostly democratic. But now that the group of aging states is becoming more diverse as subsequent waves appear, research on the politics of aging has expanded to other questions typically of interest to political scientists. A subset of this research area focuses more on political economy, namely whether there is a reliable relationship between regime type, economic development, and population aging. The second, and least developed, research area is national security, which to date mostly debates whether an aging country is more or less inclined to peace.

RESEARCH IN POPULATION AGING: UNCHARTED TERRITORY

19

Intergenerational relations, institutions, and gerontocracy One of the most developed research lines in the political demography of population aging concerns the degree to which the size of demographic groups translates into political power: the extent to which gerontocracy exists in aging states. One line of questioning looks at the voting rates of older people versus younger people. Voter turnout among older age groups is consistently higher than among younger age groups, a remarkably stable finding across political system, level of economic development, experience with democracy, generational changes, and so on (Goerres, 2007b). Although counteracted by factors like health conditions at the eldest ages or lack of a mobilizing partner, older people are more likely to vote compared to younger people because voting becomes habituated over one’s lifetime and because older people have generally been settled in an area longer than younger people (Goerres, 2007b). In the USA, voters aged 65 and older consistently have the highest turnout rates, with a significant gap between their turnout and those of the youngest voters. In the US midterm elections of 2018, for example, 66 percent of the oldest voters showed at the polls, while only 36 percent of voters ages 18 to 29 voted (Frey, 2019). Voter turnout is one thing, but voter preferences are another. Older voters are often perceived to be more conservative than younger voters, and indeed some studies find evidence that age itself leads to conservative preferences. Other studies, however, argue that any observed correlation between conservative preference and older ages is driven by generational effects, not age effects. Goerres (2007a) argues that age is not a salient political cleavage because it is transitional, meaning everyone passes through various ages while not everyone has a stint as a white person, or as a Social Democrat. Second, older people are interested in the welfare of other age groups besides their own (Goerres, 2007a). More research is needed on whether or not older people vote only in their own interests or whether their voting patterns reflect intergenerational concerns. Repeating such studies as new generations become the “oldest voters” will help researchers disentangle generational effects from age effects and give us a greater sense of what the future portends in polities that give citizens – including older persons – a say in governance. A second question in this line of research asks whether societies with higher proportions of elderly have more generous social security systems. The elderly power view holds that the size of demographic groups does translate to political power, namely that societies with larger proportions of older persons will have more generous social welfare benefits (Cutler, 1977; Mulligan &

20

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Sala-i-Martin, 1999; Persson & Tabellini, 2000). In contrast, the “fiscal leakage” hypothesis predicts the opposite, that working voters in pay-as-you-go systems will favor more austere social welfare for the aged because they have to foot the bill (Breyer & Stolte, 2001; Razin & Sadka, 2007). Empirical evidence is mixed, but an institutional view can offer insight. There is a robust research agenda on family policies, including intergenerational issues like care for children and for the aged, in social policy research, particularly in the nexus between sociology and political science. Much of the research on this politics of aging has been comparative within the Organization of Economic Cooperation and Development (OECD) states, particularly within Europe. Broadening research beyond OECD states, now that other regions are aging, is likely to result in different conclusions about the institutional politics of aging. While research on most of the OECD states has tried to disaggregate the various levels of entitlement generosity towards the elderly, the bias among researchers that the state holds significant responsibility for elder welfare is clear. In Asia, however, there is a common tradition of filial duty and familialism, whereby the state is the last resort for care duties, both for children and for elderly. Scholars have observed changes in some Asian states, particularly those that have democratized, as various forms of a welfare state have emerged and begun to take over some functions of the family, such as long-term care for the elderly. For example, Japan introduced publicly funded long-term care insurance in 2000, South Korea in 2008, and Taiwan in 2015 (Kodate & Timonen, 2017). Both Japan and Korea modeled their programs on Germany’s. Early research on the second wave of aging states underscores findings from the first wave: that institutions are a greater determinant of the power of demographic groups than is their size alone. In a study of Singapore and Taiwan, Chien-Kai Chen and I found that political institutions determined the negotiation of social welfare policies toward the aged. In liberal democracies, leaders must consider the demands and influence of various groups, including pressure groups and voters, while less liberal regimes with restricted citizen roles allow leaders to ignore some of these demands and enact less generous social welfare policies for older persons. The burden on the state is less in illiberal regimes like Singapore’s, while in Taiwan the burden has shifted from the family to the state as Taiwan moved from an authoritarian to a democratic government (Sciubba & Chen, 2017). Now, democratized Taiwan must respond to interest groups. Research on the politics of aging in the second-wave states is nascent, and more comparative research, especially, is needed. Early empirical findings

RESEARCH IN POPULATION AGING: UNCHARTED TERRITORY

21

indicate that the politics of aging in the USA and China are likely to be radically different – even as the size of elderly groups in both increases – because the institutions of each are so different. Political economy of aging One subset of this research area concerns political economy, particularly questions about whether aging states can maintain economic growth even as their proportion of working ages declines and elderly dependents expands. We know that as states get richer they are more likely to grow older – fertility declines and life expectancy increases. The converse is not always true, however. States that are not yet rich are growing older, and we aren’t sure what that means. Is there a relationship between regime type, economic development, and population aging? And is democracy more likely as a country ages? None of our prevailing economic theories were developed under conditions of widespread population aging, and that most theories see infinite economic growth as the goal is unsurprising given that they were formulated when population seemed to be similarly infinitely expanding. It is time to ask: Is unending economic growth possible in a shrinking population and, given the planetary environmental crises, is it even desirable? Institutions are again relevant here. Is the aging burden the same on non-democracies as it is for democracies? One area for research is looking at the burden on the state, on women, and on elderly themselves in a context of population aging across regime types. Economic system and regime type are tied, and perhaps the question scholars should be asking is how a poor country is likely to allocate its economic resources – among various segments of the population and among various budgetary areas – when it is non-democratic. Are countries where decision-making is concentrated in the hands of a few powerful figures freer to put policies in place that require tough choices, such as higher retirement ages or more austere social benefits for the elderly? Measurements of population aging can be problematic in political economy studies. While any country where fertility is falling and median age is rising is technically aging, the countries of interest for population aging research are those where such dynamics have been taking place for a while and median age is advanced – this is a small set of cases. Such cases have shifting proportions of dependents, relatively more elderly persons, and fewer working age and younger persons. The challenge comes in defining the boundaries of “dependency”. Typically, data is available to define dependency at ages 60 or 65. Using that data means applying those cut offs cross-nationally, no matter what the actual effective retirement age, which would mark someone as a “dependent”,

22

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

actually is. Those cutoffs, then, are often inaccurate. Even within Europe, the average age of exit from the workforce varies widely – effective retirement age for men in 2018 was 68 in Iceland, but 60.8 in France (OECD, 2019). Thus, the very definition and measures of dependency should be critiqued in political economy research. Gray social movements A second subset of institutional research examines nascent gray social movements and the tie between issues of old age and human rights. One reason there is so little research into issues of old age and protests is that studies of demographics and protest are biased toward youth populations. This has been mostly for good reasons, as innumerable student movements throughout history have driven large-scale political change, and states with youthful or youth bulge age structures today are more inclined to see expressions of dissatisfaction with the status quo. But the assumption that older people don’t protest and thus are not worth studying has been to the detriment of interesting political dynamics reaching all the way to the United Nations. A Transnational Advocacy Network (TAN) around aging seems to be emerging, particularly around the proposed UN Convention on the Rights of Older Persons (CROP) (Sciubba, 2014, 2015; Herro, 2019). In previous research, I have found that advocates gained some traction in getting the UN community to consider a human rights instrument on older persons because of successful norm diffusion from Latin America to the international level. The debate and passage of the UN Convention on the Rights of Persons with Disabilities in 2006 helped spark progress on the contemporary CROP campaign. States are divided in their support for, or resistance to, a potential CROP, in large part because of their divergent interests and incentives, which have shifted as population aging has intensified. Whether the campaign for a UN Convention on the Rights of Older Persons will ultimately be successful remains to be seen but the aging advocacy network has been significantly weaker than the successful disability network and faces a relatively more closed political opportunity structure. The neglect of old age issues and older age structures in contentious politics research is an avenue for potential exploration, particularly as population aging strains generous social welfare systems and protests have erupted across the globe when pensions have come under threat of reform, as happened in Russia in 2018.

RESEARCH IN POPULATION AGING: UNCHARTED TERRITORY

23

Population aging and national security The second major area of research on the politics of population aging is linked to national security. Discussions about aging and national security tend to focus on the ability and willingness of an aging state to engage in conflict. Ability is most often operationalized as manpower and budget. Personnel costs are thought to be relatively higher in states with older age structures, as lower numbers of military-age populations mean that the costs of recruiting and retaining each solider are high. This discussion is similar to general concerns about the pitfalls of recruiting an all-volunteer army, namely that the most capable potential recruits are often lost to the private sector, where they can command higher wages and avoid the strictures of military life. As the number of potential recruits shrinks, states could substitute technology for manpower, but some military functions, such as humanitarian relief missions and counterinsurgency operations, remain labor-intensive. Technology is not a perfect substitute for manpower. There are two layers to the budget aspect of “ability” The first is an assumption that rising personnel costs will crowd out other military spending, taking up a larger portion of the pie. The second is that military spending in general will suffer in light of increased spending on elderly entitlements – non-military spending crowds out military spending. Furthermore, many worry that the overall budget – the pie itself – is likely to shrink as economic productivity declines in the face of population aging. Casualty aversion in the face of smaller families is the main focus of discussions of an aging society’s willingness to engage in conflict. Brooks et al. (2019) find that states with older societies are least likely to engage in interstate conflict, citing evidence of casualty aversion among some aged states. Specifically, they find that “By the time median age reaches 30 years, the probability of conflict falls to about one-quarter of its peak value” (Brooks et al., 2019, p. 77). Employing a political demographic perspective yields very different theoretical insights than one absent the political element. The assumption that aging states will spend more on elderly than on younger generations, for example, is apolitical. To validate that assumption we need to examine the policymaking processes in aging states and refer to research in the preceding section on elderly power. We must recognize that the size of demographic groups does not automatically equate with the size of their political influence. Spending on entitlements is as much a political decision as level of immigration, spending on education, or spending on the military.

24

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Another issue is that many of the conclusions in this area of research are still prospective, arguing that older people may be less supportive of war, and their increasing numbers are likely to amplify their voices. But is that true? Are older people actually less supportive of war and, if so, is this opposition a salient issue for them, meaning a high priority when it comes to voting? If it is a low priority then knowing that older people in the world’s oldest societies are averse to war does not help us predict their state’s propensity for conflict. As Brooks et al. (2019, p. 82) note, although there is some evidence that older persons in the US, UK, and a few other countries are more “dovish” than other age groups, every war is unique and “it can be hard to assess whether the extent of the age gap might be dependent on key features of the wars in question”. And, combined with research on political institutions, if they are less supportive does the political system amplify their voices relative to other age groups? These are all questions that remain unanswered in the literature. There are theoretical reasons to expect older societies to be more peaceful, but the empirical evidence is still weak and inconclusive. Whether an aging state is willing to spend money on entitlements, the military, or any other policy area is to a large extent dependent on the political institutions and dynamics in that country, not solely on its demographics.

Research limitations In general, the most reliable studies of the political implications of population aging are qualitative ones, since there are so few states with high median ages. Research on gerontocracy can be and has been comparative and case study based, and thus is the most well developed. Political economy and national security studies are often quantitative and, because they require a larger set for robust results, tend to take liberties with defining cases as “old”. National security studies, especially, suffer from this dilemma. If the research question is about the relationship between youthful age structures and interstate or civil war, the n can be large – there are plenty of cases over plenty of years. If researchers attempt the same question but with mature age structures, the n is very small – few cases over few years. In 2000, only 32 states with populations over 1 million had median ages of 35 years, with only three states achieving a median age of 40. The number of cases is one significant methodological limitation; the nature of cases is another. The set of cases with advanced median age includes the world’s great powers, and arguably requires a finer analytical hand than large-n studies are capable of. The conflict–peace calculus of such states is complicated and, while war is rare, lower-intensity acts of aggression

RESEARCH IN POPULATION AGING: UNCHARTED TERRITORY

25

are more common, posturing and preparing for conflict even more so. Thus, until the case set of states with advanced median ages – at least over 35, if not over 40 – is much larger to increase the number of observations, a mixed methods approach or case study research will be the most reliable for understanding the implications of aging on economic development and national security issues. Additional areas for future research There are two areas, in particular, where research on the intersection of aging and with other issues is underdeveloped but with potentially fascinating questions. The first is gender and aging, as gender implicitly and explicitly shapes fertility trends and economic implications of low fertility and population aging. Women live longer than men but often have lower official retirement ages when they do work, or are out of the workforce completely. In short, understanding the changing situation of women across the life course and across cultural contexts is essential to a full picture of the politics of aging, yet, as Gemma M. Carney says, “These women were never a political priority, from their education as little girls, through the free labour they provided as mothers, through the years they spent working for lesser pay and now as widowed old ladies who need care” (2017, p.  9). Policies toward the elderly are often gender-blind, which flies in the face of reality. Old age itself is feminized, yet the political context within which old-age policies are made can put women at a disadvantage. As Carney says, “By the time women reach the stage whereby they need eldercare, their accumulated disadvantage is seen as a problem of old age – it is rarely linked to the gendered institutions that have shaped their life trajectories” (2017, pp. 9–10). To the extent that policymakers try to increase women’s participation in the labor force to make up for shrinking working-age populations in low-fertility societies, they work at cross-purposes if they do not concomitantly address women’s needs for caregiving support. If women are being asked to care for children and aging parents and work outside the home, the stress is often too much to bear and the effect could be even lower fertility. Some policymakers realize this. Shinzo Abe of Japan, for example, increased the availability of childcare facilities at the same time that he tried to encourage women to join the formal labor force. There is also significant room for comprehensive political demography research that looks not at aging as a single trend, but at the interaction of aging with other population trends, particularly immigration. Teitelbaum and Winter’s early work on high migration/low fertility societies was a useful

26

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

initial chronicling of an early trend we still observe today in Europe and North America, and the 2010s European migrant crisis has renewed interest in those questions (Teitelbaum & Winter, 1998). David Coleman has offered the Third Demographic Transition as a framework for considering these simultaneous changes (Coleman, 2006). Coleman argues that sub-replacement fertility and high immigration from ethnically different areas are together changing the ancestry of national populations, transforming the ethnic character of such countries. There is much to explore about the political consequences of such a transition, as Brexit shows. With regard to Britain’s historic vote to leave the European Union, Coleman (2016) found that Leave voters cared most about Britain regaining autonomy over immigration. Economic arguments, which the Remain campaign focused on, did not resonate. As some of the newer aging societies seem to face very different immigration situations, an expansion of this research beyond Europe and North America would be useful. In particular, Japan and South Korea, two rapidly aging Asian countries, are thus far choosing a path of aging that preserves ethnic homogeneity rather than choosing to supplement a declining working-age population with substantially increased immigration. Whether they stick to this path or chart a new one is an entirely political question, and thus exactly the type of research political demography was designed to pursue.

Conclusion As this review has shown, research on population aging is nascent in political demography because the trend itself is so new. But, given the speed with which past fertility declines are showing up in shifting proportions of young and old in most regions of the world, the issue is only growing in importance. Researchers can build on political science work in voting behavior and political parties to understand how aging is likely to impact politics – those studies are always about the relative power of groups, age or otherwise. But the theoretical base for understanding economic development and military implications in the context of aging is far weaker, as the very theories that undergird research in those areas were formed in a time of expanding population. Thus, not only is there room for empirical work on population aging, there is also significant need for theoretical development. This need is made more pressing by the intensification and spread of population aging beyond East Asia, North America, and Europe, to more places where the level of economic development is lower and democracy is scarce.

RESEARCH IN POPULATION AGING: UNCHARTED TERRITORY

27

References Breyer, F. & Stolte, K. (2001). Demographic change, endogenous labor supply and the political feasibility of pension reform. Journal of Population Economics, 14, 409–24. Brooks, D. J., Brooks, S. G., Greenhill, B. D. & Haas, M. L. (2019). The demographic transition theory of war: why young societies are conflict prone and old societies are the most peaceful. International Security, 43(3), 53–95. Carney, G. M. (2017). Towards a gender politics of aging. Journal of Women & Aging, 1–17. doi: https://​doi​.org/​10​.1080/​08952841​.2017​.1301163. Coleman, D. (2006). Immigration and ethnic change in low-fertility countries: a third demographic transition. Population & Development Review, 32(3), 401–46. Coleman, D. (2016). A demographic rationale for Brexit. Population & Development Review, 42(4), 681–92. Cutler, N. E. (1977). Demographic, social-psychological, and political factors in the politics of aging: a foundation for research in ‘political gerontology’. The American Political Science Review, 71(3), 1011–25. Frey, W. H. (2019). 2018 voter turnout rose dramatically for groups favoring Democrats, census confirms. Retrieved from https://​www​.brookings​.edu/​research/​2018​-voter​ -turnout​-rose​-dramatically​-for​-groups​-favoring​-democrats​-census​-confirms/​. Goerres, A. (2007a). Can We Reform the Welfare State in Times of ‘Grey’ Majorities? MPIfG Working Paper. Cologne, Max Planck Institute for the Study of Societies. Goerres, A. (2007b). Why are older people more likely to vote? The impact of ageing on electoral turnout in Europe. British Journal of Politics and International Relations, 9, 90–121. Herro, A. (2019). Advocating a UN Convention on the Rights of Older Persons in the United Kingdom: the case for a radical flank. Journal of Human Rights Practice, 11(1), 132–50. Kodate, N. & Timonen, V. (2017). Bringing the family in through the back door: the stealthy expansion of family care in Asian and European long-term care policy. Journal of Cross-Cultural Gerontology, 32, 291–301. Lesthaeghe, R. (2014). The second demographic transition: a concise overview of its development. PNAS, 111(51), 18112–15. Mulligan, C. B. & Sala-i-Martin, X. (1999). Gerontocracy, Retirement, and Social Security. NBER Working Paper Series. Cambridge, MA, National Bureau of Economic Research. OECD (2019). Average effective age of retirement versus the normal age in 2018 in OECD countries. Paris: OECD. Persson, T. & Tabellini, G. (2000). Political Economics. Cambridge, MA: MIT Press. Razin, A. & Sadka, E. (2007). Aging population: the complex effect of fiscal leakages on the politico-economic equilibrium. European Journal of Political Economy, 23(2), 564–75. Sciubba, J. D. (2014). Explaining campaign timing and support for a UN Convention on the Rights of Older People. The International Journal of Human Rights, 18(4–5), 462–78. Sciubba, J. D. (2015). Securing rights in the 21st century: a comparison of the Disability and Older Persons’ Rights Conventions. Journal of Human Rights, 15(4), 533–49. Sciubba, J. D. & Chen, C.-K. (2017). The politics of aging in Singapore and Taiwan. Asian Survey, 57(4), 642–64.

28

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Teitelbaum, M. S. & Winter, J. (1998). A Question of Numbers: High Migration, Low Fertility, and the Politics of National Identity. New York: Hill and Wang. “Too Many Babies? Japan Tries All-Out Birth Control” (1950). Newsweek, May 8, 35(19), 42–3. UN Department of Economic and Social Affairs, Population Division (2019). UN World Population Prospects, the 2019 Revision. https://​population​.un​.org/​wpp/​.

3

Progressing research on forced migration

Tiffany S. Chu

Introduction Imagine armed nonstate groups are clashing with your government. The fighting started in the capital city, which is somewhat far from your hometown, and you hope tensions will de-escalate before they reach you. However, the fighting continues to get closer. You start serious discussions with your family about ensuring their safety, and try to get advice from neighbors. You find out some have already fled, realizing they would rather get away early than wait until the fighting gets closer. You reach out to family in other countries to ask about staying with them, although you hear aid organizations are starting to assist those that flee. The next day, you get news that bombs hit one town over. The time has come. You decide to take the offer to stay in your cousin’s house across the border. Surely, you’ll only be there for a couple of months. But months turn into years, with no end to the war in sight in your home country. This scenario outlines a typical narrative of displacement: there are many moving parts, few choices, and multiple factors weighing on the decision to flee or stay. The international refugee protection regime is supposed to allow those who can make it to another country to apply for refugee status and expects host states to offer sanctuary as well as receive aid and assistance. These safeguards arguably can make fleeing more appealing; however, international and nonprofit relief organizations tend to be under-resourced and require approval by host states. While the refugee protection regime is ideal in theory, increasing trends in the numbers of people forcibly displaced and the association of migration flows with national security risks has put the regime in peril. The United Nations High Commissioner for Refugees (UNHCR) is the primary organization assisting refugees globally. Created in the aftermath of 29

30

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

World War II, the 1951 Convention Relating to the Status of Refugees and the 1967 Protocol affirming and extending the Convention are the main treaties documenting international refugee law. Beginning in the 1980s, there was a stark increase in the number of individuals fleeing conflict and violence who needed protection under the international refugee regime (Barnett and Finnemore 2004). Figure 3.1 displays the global count of refugees since 1960, demonstrating this upward trend.

Source: UNHCR 2019.

Figure 3.1

Global count of refugees since 1960

As the reason individuals fled changed from political persecution to armed conflict, international policies and attitudes toward accommodating refugees changed as well. During the Cold War, especially, Western states were willing to accommodate political refugees because it was in the host state’s strategic interest to grant asylum to individuals fleeing communist regimes. The smaller numbers of forcibly displaced also meant it was easier to resettle refugees into Western developed states, as long as the refugees’ country was of strategic value to the potential host (Chu 2020). With the increase in civil wars across Africa, the Middle East, and Southeast Asia, more individuals were fleeing generalized violenced and applying for asylum in other states. The increase in numbers overwhelmed the bureaucratic pipeline to resettle refugees, slowing

PROGRESSING RESEARCH ON FORCED MIGRATION

31

down the process significantly, and consequently leading to fewer states willing to take on a large share of non-citizens. With the Cold War’s end, there was less of a strategic foreign policy incentive for developed states to host refugees (Toft 2007). The 2010s continued to put stress on the existing international refugee regime, as individuals fled intrastate violence in countries such as Sudan and Iraq and the outbreak of civil war in Syria. The majority of refugees originate from the Middle East and Africa, as displayed in Figure 3.2, with the same regions hosting the most refugees but lacking the political or economic resources to provide for the displaced. Simultaneously, developed countries implemented policies deterring the inflow of refugees, such as the European migrant crisis in 2015 leading to the construction of border walls and patrols (Benedicto and Brunet 2018) and the 2017 travel ban adopted by the Trump administration that denied entry of asylum seekers from predominantly Muslim countries.

Source: UNHCR 2019.

Figure 3.2

Region of origin of refugees

Ultimately, the existing refugee regime is failing in terms of protecting individuals under its purview. The international community endorses a “three durable solution” framework for displacement: integration into host communities,

32

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

repatriation to the home state, or resettlement into a third country, with repatriation as the preferred solution (Long 2014). However, the framework and promotion of repatriation provides incentives for temporary policies that do not offer a robust solution to refugees’ situations. The lack of burden-sharing among states puts pressure on under-resourced countries that lack the capacity to accommodate both refugee and citizen populations. Moreover, international protection for refugees does not cover those fleeing environmentally induced disasters or violence perpetrated by criminal organizations. This gap allows states to refuse relief for such individuals, even though these problems continue to grow. As scholars, continuing to study forced displacement with an eye toward connecting research with existing policy can help evaluate and improve the policies in place to reduce the number of forced migrants and offer better solutions to their plight. The following sections outline integral scholarship on forced migration. Mainly, studies on displacement began with a macro-level approach at the country level and have, over time, shifted to the individual and family levels. New directions in research have also switched from examining the determinants of flight and host country impacts to evaluating existing solutions to forced displacement. Several topics have received minimal attention: environmental- and criminal organization-driven displacement, and internal displacement. The final section offers some suggestions for how we can conduct ethical research on these vulnerable populations in order to capture their context and accurately propose better policy solutions.

Theories and findings on forced displacement Scholars have categorized four phases of involuntary migration: initial flight, resettlement in asylum, repatriation, and reintegration after return (Bascom 2002). While these phases help conceptualize stages of involuntary migration, not every displaced person goes through all four steps. The literature has mostly focused on initial flight and resettlement in asylum states. Studies focusing on forced displacement utilize theories of migration that take a rational actor approach in the decision to flee. They assume that individuals have agency, or the capacity to reflect on their current position, devise strategies, and take action to achieve their desires (Bakewell 2010). The literature on factors influencing the decision to flee is well developed. In general, scholars have identified violence and political instability as important factors driving displacement. Some people flee their homes and become

PROGRESSING RESEARCH ON FORCED MIGRATION

33

displaced when there are threats to their physical integrity (Davenport et al. 2003). Schmeidl (1997) finds civil wars with foreign military intervention are more likely to produce massive and prolonged refugee migrations and ethnic rebellions cause smaller flows, not mass exoduses. Moore and Shellman (2006) argue that the ratio of refugees to internally displaced people (IDPs) increases when there is state-sponsored genocide/politicide compared to other forms of armed violence. Other factors influencing whether people are internally or externally displaced are the geographic distance of another country (Moore and Shellman 2007) and the existence and size of transnational ethnic kin in another state (Cederman et al. 2009; Ruëgger and Bohnet 2018). Beyond general violence, governments and nonstate actors sometimes strategically displace populations to further their agendas. Greenhill (2010) demonstrates how governments coercively engineer population outflows to create a homogeneous nation state while also inundating a rival state with people that can cause a strain on resources and spark crises domestically. Turkey allowing Syrians to continue to Europe, instead of hosting them, is an example of this dynamic (Greenhill 2016). Turkey employed this strategy because they lacked the resources to provide for refugees, desired financial assistance from Europe, and wanted to reignite talks about joining the European Union. This strategy was partially responsible for the so-called migrant “crisis” in Europe that culminated in a $7 billion deal to keep Syrian refugees in Turkey. Forced displacement as a strategy is also a pattern in civil wars, mainly to gain territorial control of a region (Balcells and Steele 2016) and to achieve civilian support (Zhukov 2013). Noncombatants are more likely to be targeted by armed groups if territory is contested (Kalyvas 2006). Armed groups can gain information on civilian loyalties based on election results and subsequently change the demographics of the electorate by uprooting populations from these regions (Steele 2011). If combatants cannot identify opponents, indiscriminate forced resettlement is a more efficient alternative than trying to win hearts and minds (Zhukov 2015). Civilians who believe they will not be targeted by either group, or feel protected by the strongest group, are more likely to decide to stay (Steele 2009). While conflict and violence are strong predictors of displacement, economic opportunity also factors into the decision-making process. The outbreak of war destroys infrastructure and limits job opportunities, especially if fighting is taking place in that area. In other words, while people may not be explicit targets of violence, they are more likely to leave if the fighting reduces the availability of the economic resources necessary to secure a livelihood (Adhikari 2012).

34

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

In sum, refugees, especially since the 1980s, are fleeing the violence and infrastructure damage that disrupt everyday life and lead to hunger and loss of economic opportunities (Adhikari 2013; Hammond 2004). Given the increasing number of individuals fleeing, states receiving the bulk of the population become quickly inundated and lack the capacity to support the refugees on their own. Although some states are averse to shouldering the burden of hosting refugees, host states can enjoy certain advantages. The arrival of refugees compels states to bolster their strength along border regions, which paves the way for increasing state legitimacy through territorial control (Jacobsen 2002). Refugees that settle are associated with a boost in local economies, especially if refugee families receive cash assistance (Taylor et al. 2016). While refugees can benefit host state economies and state capacity, migrants, in general, have become inextricably linked with issues of national security (Adamson 2006). Conflict scholars demonstrate how civil war violence can spill over into hosting states (Lischer 2005; Salehyan and Gleditsch 2006). Although refugees are rarely the root cause or the perpetrators of the violence, states become less willing to accommodate refugees given this link (Wright and Moorthy 2018). The association between incoming refugees and threats to national security propagates negative public opinion toward hosting refugees, who are seen as sources of economic competition and crime (Adida et al. 2017; Ghosn et al. 2019; Martin 2005; Wright and Moorthy 2018). Direct exposure to incoming refugees leads to a general shift toward right-wing parties, which are traditionally anti-immigrant (Dinas et al. 2019). Some citizens blame refugees for domestic issues, resulting in violence against refugees (Benček and Strasheim 2016; Savun and Gineste 2019). In order to contain the flow of refugees, host states can construct borders (Carter and Poast 2017), promote policies advising non-assistance toward refugees (Gerver 2014; Heller and Pezzani 2016), and intervene directly in the conflict, producing even more refugees (Salehyan 2008; Uzonyi 2015). Finally, an emerging literature evaluates the three “durable” solutions. While refugee repatriation is the preferred solution, the ratio of returnees to refugees remains low. Since the end of the Cold War, the highest ratio is 25 percent and hovers at an annual average of approximately 15 percent (UNHCR 2019). Bradley (2013) documents how repatriation is an imperfect process that can lead host states to pressure refugees to return through monetary incentives. These dynamics put into question whether return flows are in the best interest of refugees (Chimni 2004). Other studies evaluate how experiencing displacement and subsequent return affects the origin state. Schwartz (2019) finds return migration generates

PROGRESSING RESEARCH ON FORCED MIGRATION

35

new social divisions between those that were displaced and populations that remained in the country. These cleavages lead to the onset of new sources of disputes in the country of conflict unless local institutions can properly mitigate these disputes. For example, returnees in Burundi are less likely to own land, leading to worse economic outcomes compared to individuals who stayed (Fransen 2017; Fransen et al. 2017). These studies are just the beginning of a compelling research agenda investigating determinants, preferences, and consequences of return migration.

Gaps in research There are several under-developed areas of forced migration scholarship that require more significant attention. The first is understanding policy responses toward individuals displaced by environmental factors and natural disasters, especially since these events are predicted to increase in frequency and severity. Existing asylum law does not protect individuals who flee because of environmental conditions. Instead, individual countries have discretion to determine if they will protect fleeing civilians. This leads to an erratic array of policies that are event-specific, such as the United States allowing some, but not all, Bahamians refuge in the aftermath of Hurricane Dorian. Moreover, fast-moving disasters, such as hurricanes and earthquakes, are more likely to trigger international relief efforts than slow-moving environmental conditions, such as drought or changes in arable lands (Reuveny and Moore 2009). Another topic that demonstrates the failure of existing asylum law is crime-induced migration. Individuals fleeing political persecution are granted refugee status. During the Cold War, this was meant only to include government actors, thereby not covering violence from nonstate actors as grounds for refugee protection. Over time, this stipulation was relaxed to include rebel groups and terrorist organizations (Orchard 2014). However, many asylum seekers today, particularly those from Latin America, are fleeing violence perpetrated by drug cartels (Cantor 2014). Since cartels and gangs are not considered political organizations, their victims are denied refugee status even though they are fleeing extremely violent groups. This topic continues to be understudied even as forced displacement generated by organized criminal groups has grown and contributed to increased migrant flows to the United States since 2014. As many states are increasingly reticent to expand the number of refugees within their borders, or are actively working to reduce the size of the existing

36

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

refugee population, the number of IDPs are growing. IDPs do not fit into any of the four categories or stages of forced migration referenced previously (Bascom 2002), yet are the highest percentage of forcibly displaced individuals. Many refugees who return to their country of origin actually become IDPs, yet researchers do not pay enough attention to their varied struggles since return is considered the “end” of forced migration (IDMC 2017). At this time, there are comparatively few studies on the causes and consequences of internal displacement compared to refugees. Existing scholarship on the determinants of internal displacement mirrors the same mechanisms as refugee flight (Adhikari 2012, 2013). IDPs do not have a legally binding instrument upholding their rights; and lack a “durable solution” to their plight. Moreover, the responsibility for protecting and assisting IDPs lies with national authorities (Ferris 2014) – a problem, since many countries with large IDP populations are involved in civil conflict that could have been the cause of their displacement. IDPs complicate issues for returning refugees, since many of the internally displaced settled in areas from which those refugees had fled (Bradley 2013). Unless there are certain dispute resolution mechanisms in place, conflicts can easily arise between these populations. While international visibility and responses to IDP issues have improved, a lack of a systematized policy and fewer protections provide challenges and opportunities to expand knowledge on the issue.

Methodological advances in displacement research Beginning in the 1990s, studies on forced displacement tend to take a macro-level approach to explain refugee flows and responses by host states. Most of these studies use statistics from the UNHCR, which collects and reports yearly data on their seven populations of concerns: refugees, asylum-seekers, internally displaced people (IDPs), returned refugees, returned IDPs, stateless persons, and others. However, these studies are unable to capture individual-level mechanisms in the decision-making of flight. Importantly, these studies miss micro-level variation in when and why individuals would flee to another country. Recently, a growing number of studies are beginning to shift the level of analysis toward the individual or family unit, where most of the decisions to flee take place. Researchers are increasingly using surveys to ask a battery of questions to a broader number of respondents. While questions can lack depth, since most require a yes/no answer, these surveys are still better able to capture the micro-level dynamics of flight. For example, Schon (2019) shows how Syrian civilians with opportunity (money and connections) and motiva-

PROGRESSING RESEARCH ON FORCED MIGRATION

37

tion to flee left their homes approximately one-and-a-half years earlier than those without opportunity or motivation to leave. Researchers are also using survey designs to embed experiments where the survey population is randomly assigned to a control or treatment group. The treatment group is asked a version of a question in order to elicit a potentially different response compared to the control group (Sniderman 2011). For instance, Fabbe et al. (2019) show how Syrian refugees are much more likely to agree to a ceasefire proposed by a civilian than one proposed by any of the involved armed actors in the war. Others take a “natural” experiment approach by exploiting exogenous variation unrelated to the design of the study. Braithwaite et al. (2019) used conflict events that occurred while their survey was deployed to compare responses from individuals surveyed before or after the events. Dinas et al. (2019) used geography to their advantage to test for increasing support for right-wing parties when some islands in the Aegean Sea close to the Turkish border received a notable influx of Syrian refugees while other islands did not. Ethnographic approaches, semi-structured interviews, and focus groups have been more frequently employed to capture the depth and nuance of decisions made at various stages of displacement (Cyr 2019). Moreover, interacting with participants reveals useful information about the context of the answers as well as how interviewees portray themselves to the interviewer (Fujii 2010). These studies have helped further our understanding of how individuals decide to flee, and allow us to speculate about future movement by underscoring the fact that refugees are not homogeneous in their preferences or in how they resolve to flee (Schwartz 2019; Stefanovic and Loizides 2011). Researchers are also implementing surveys among residents of host states to gauge attitudes toward hosting, which vary. Host populations that share ethnic ties with incoming refugees (Getmansky et al. 2018), have close contact (Ghosn et al. 2019), or have previously experienced conflict and displacement can, therefore, empathize with their plight and are more willing to support hosting refugees (Hartman and Morse 2020). However, these effects do not hold over time. The longer the refugee presence, the less supportive host populations become (Feldman et al. 2017; Ghosn and Braithwaite 2018).

38

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Best practices for research Several factors are crucial in developing a research agenda that builds upon existing work. The first is the population of concern included in the analysis. For example, a project seeking to explain decisions to return to the country of origin should have a sample including those that have returned and those that have not. Without engaging with both of these populations, one part of the process will be biased in the results (Schewel 2019). One way we can critically and more successfully collect and analyze data on forced migration is to adopt a mixed-methods approach (Seawright 2016). For example, survey data can broadly inform on how a sample population feels about specific topics, whereas focus groups, including some of those surveyed, can provide the reasoning and context behind certain answers. While surveys and focus groups require more resources than openly sourced data, the former would ensure a more precise depiction of return migration processes and capture information from a variety of angles. Finally, we should be mindful of potentially damaging effects on the participants of our studies, especially when researchers are foreigners to the contexts of the areas and locations they study (Getmansky 2019). A lack of contextual knowledge of an area or harboring an uncritical point of view can severely undermine the validity of the research (Ghosn and Parkinson 2019). Additionally, asking displaced populations to relive traumatic experiences for the sake of “quality” data should not be the goal of our work (see Adhikari and Bryant 2018 for more information about surveying hard-to-reach and vulnerable populations). Since the locations we study can be fragile or continue to be in a state of protracted conflict (Cronin-Furman and Lake 2018), we must take into consideration the ethical implications of the work we do and how the communities we interact with can benefit from, or be endangered by, our research.

Conclusion The large and growing number of individuals forcibly displaced from their homes and lack of sustainable policy solutions means the issue will require global attention for the foreseeable future. This chapter has outlined existing research on the determinants of refugee flows, where they tend to flee to, and the influence of their arrival on host states. Scholarship on forced migra-

PROGRESSING RESEARCH ON FORCED MIGRATION

39

tion has moved from country-level studies that explain aggregate trends to micro-level research designs that capture the heterogeneity of refugee and host resident populations. There is a growing literature on evaluating the feasibility and implications of the durable solutions to forced displacement. In particular, research on refugee return has expanded and provided insights into when individuals are returning, refugee attitudes toward repatriation, and how returnees shape domestic society in their country of origin. These studies provide critical insight into the existing practice of promoting repatriation as the preferred solution for refugee issues. Current gaps in our knowledge concern areas where asylum laws do not technically apply but continue to be a pressing issue. Individuals displaced by environmental conditions and natural disasters, as well as targets of violence by criminal organizations, pose a problem to the existing international refugee regime since most states do not want to take in any refugees, let alone expand the definition to include these individuals. Some states will pass temporary laws to protect individuals displaced by natural disasters, yet there are no attempts to generate a standard of practice toward these individuals. Additionally, as states continue to limit the number of refugees they are willing to host, the number of internally displaced increases. As these issues persist in the coming years, research on these areas can help inform the implications brought about by gaps in the international refugee regime. Solutions to forced migration require international and regional cooperation. The profile of involuntary migration is not static – over time, characteristics and heterogeneity of the displaced have morphed. Accordingly, solutions and research should reflect these changes in order to best serve these populations. Researchers can play a critical role in offering recommendations by engaging in ethical scholarship that will benefit the communities of displaced individuals. Therefore, as the study of forced displacement continues to expand, researchers should be cognizant of such goals and of the effects their work can have.

References Adamson, F. B. (2006), ‘Crossing borders: international migration and national security’, International Security, 31 (1), 165–99.

40

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Adhikari, P. (2012), ‘The plight of the forgotten ones: civil war and forced migration’, International Studies Quarterly, 56, 590–606. Adhikari, P. (2013), ‘Conflict-induced displacement, understanding the causes of flight’, American Journal of Political Science, 57 (1), 82–9. Adhikari, P. and L. A. Bryant. (2018), ‘Sampling hard-to-locate populations: lessons from sampling internally displaced persons (IDPs)’, in L. R. Atkeson and R. M. Alvarez (eds), The Oxford Handbook of Polling and Survey Methods, pp.  155–80, Oxford: Oxford University Press. Adida, C. L., A. Lo and M. Platas (2017), ‘Engendering empathy, begetting backlash: American attitudes toward Syrian refugees’, Immigration Policy Lab Working Paper, 1701, 1–50. Bakewell, O. (2010), ‘Some reflections on structure and agency in migration theory’, Journal of Ethnic and Migration Studies, 36 (10), 1689–708. Balcells, L. and A. Steele (2016), ‘Warfare, political identities, and displacement in Spain and Colombia’, Political Geography, 51, 15–29. Barnett, M. and M. Finnemore (2004), Rules for the World: International Organizations in Global Politics, Ithaca, NY: Cornell University Press. Bascom, J. (2002), ‘Refugees: geographical aspects’, in N. J. Smelser and P. B. Baltes (eds), International Encyclopedia of the Social and Behavioral Sciences, Oxford: Pergamon. Benček, D. and J. Strasheim (2016), ‘Refugees welcome? Introducing a new dataset on anti– refugee violence in Germany, 2014–2015’, Research and Politics, October– December, 1–11. Benedicto, A. R. and P. Brunet (2018), ‘Building walls: fear and securitization in the European Union’, Transnational Institute Executive Summary, September, 1–56. Bradley, M. (2013), Refugee Repatriation: Justice, Responsibility and Redress, New York: Cambridge University Press. Braithwaite, A., T. S. Chu, J. Curtis and F. Ghosn (2019), ‘Violence and the perception of risk associated with hosting refugees’, Public Choice, 178 (3–4), 473–92. Cantor, D. J. (2014), ‘The new wave: forced displacement caused by organized crime in Central America and Mexico’, Refugee Survey Quarterly, 33 (3), 34–68. Carter, D. B. and P. Poast (2017), ‘Why do states build walls? Political economy, security, and border stability’, Journal of Conflict Resolution, 61 (2), 239–70. Cederman, L.-E., L. Girardin and K. S. Gleditsch (2009), ‘Ethnonationalist triads: assessing the influence of kin groups on civil wars’, World Politics, 61 (3), 403–37. Chimni, B. S. (2004), ‘From resettlement to involuntary repatriation: towards a critical history of durable solutions to refugee problems’, Refugee Survey Quarterly, October, 55–73. Chu, T. S. (2020), Hosting your enemy: accepting refugees from a rival state and respect for human rights, Journal of Global Security Studies, 5 (1), 4–24. Cronin-Furman, K. and M. Lake (2018), ‘Ethics abroad: fieldwork in fragile and violent contexts’, PS: Political Science, 51 (1), 607–14. Cyr, J. (2019), Focus Groups for the Social Science Researcher, Cambridge: Cambridge University Press. Davenport, C., W. H. Moore and S. Poe (2003), ‘Sometimes you just have to leave: domestic threats and forced migration, 1964–1989’, International Interactions, 29 (1), 27–55. Dinas, E., K. Matakos, D. Xefteris and D. Hangartner (2019), ‘Waking up the Golden Dawn: does exposure to the refugee crisis increase support for extreme-right parties?’, Political Analysis, 27 (2), 244–54.

PROGRESSING RESEARCH ON FORCED MIGRATION

41

Fabbe, K., C. Hazlett and T. Sınmazdemir (2019), ‘A persuasive peace: Syrian refugees’ attitudes towards compromise and civil war termination’, Journal of Peace Research, 56 (1), 103–17. Feldman, S., L. Huddy, J. Wronski and P. Lown (2017), ‘Compassionate policy support: the interplay of empathy and ideology’, Working Paper: Stony Brook University. Ferris, E. (2014), Ten Years After Humanitarian Reform: How Have IDPs Fared? Washington, DC: Brookings Institution. Fransen, S. (2017), ‘The socio-economic sustainability of refugee return: insights from Burundi’, Population, Space and Place, 23 (1), 1–16. Fransen, S., I. Ruiz and C. Vargas-Silva (2017), ‘Return migration and economic outcomes in the conflict context’, World Development, 95, 196–210. Fujii, L. A. (2010), ‘Shades of truth and lies: interpreting testimonies of war and violence’, Journal of Peace Research, 47 (2), 231–41. Gerver, M. (2014), ‘The role of non-governmental organizations in the repatriation of refugees’, Philosophy and Public Policy Quarterly, 32 (1), 2–13. Getmansky, A. (2019), ‘“Englishman in New York”: conducting research in the Middle East as a foreign scholar’, PS: Political Science & Politics, 52 (3), 490–93. Getmansky, A., T. Sınmazdemir and T. Zeitzoff (2018), ‘Refugees, xenophobia, and domestic conflict: evidence from a survey experiment in Turkey’, Journal of Peace Research, 55 (4), 491–507. Ghosn, F. and A. Braithwaite (2018), ‘Could contact stem the rising tide of negative attitudes towards hosting Syrian refugees in Lebanon?’, Oxford Monitor of Forced Migration: Field Monitor, 7 (2), 69–74. Ghosn, F., A. Braithwaite and T. S. Chu (2019), ‘Violence, displacement, contact, and attitudes toward hosting refugees’, Journal of Peace Research, 56 (1), 118–33. Ghosn, F. and S. E. Parkinson (2019), ‘“Finding” sectarianism and strife in Lebanon’, PS: Political Science & Politics, 52 (3), 494–7. Greenhill, K. M. (2010), Weapons of Mass Migration: Forced Displacement, Coercion, and Foreign Policy, Ithaca, NY: Cornell University Press. Greenhill, K. M. (2016), ‘Migration as a weapon in theory and in practice’, Military Review, November–December, 23–36. Hammond, L. (2004), ‘The making of a good citizen in an Ethiopian returnee settlement’, in L. D. Long and E. Oxfeld (eds), Coming Home? Refugees, Migrants, and Those Who Stayed Behind, pp.  187–205, Philadelphia, PA: University of Pennsylvania Press. Hartman, A. C. and B. S. Morse (2020), ‘Violence, empathy and altruism: evidence from the Ivorian refugee crisis in Liberia’, British Journal of Political Science, 50 (2), 731–55. Heller, C. and L. Pezzani (2016), ‘Ebbing and flowing: the EU’s shifting practices of (non)assistance and bordering in a time of crisis’, Near Futures Online: Europe at a Crossroads, 1–33, http://​nearfuturesonline​.org/​ebbing​-and​-flowing​-the​-eus​ -shifting​-practices​-of​-non​-assistance​-and​-bordering​-in​-a​-time​-of​-crisis/​. IDMC (2017), ‘Today’s returning refugees, tomorrow’s IDPs’, http://​www​.internal​ -displacement​.org/​expert​-opinion/​todays​-returning​-refugees​-tomorrows​-idps. Jacobsen, K. (2002), ‘Can refugees benefit the state? Refugee resources and African statebuilding’, The Journal of Modern African Studies, 40 (4), 577–96. Kalyvas, S. N. (2006), The Logic of Violence in Civil War, Cambridge: Cambridge University Press. Lischer, S. K. (2005), Dangerous Sanctuaries: Refugee Camps, Civil War, and the Dilemmas of Humanitarian Aid, Ithaca, NY: Cornell University Press.

42

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Long, K. (2014), ‘Rethinking “durable” solutions’, in E. Fiddian-Qasmiyeh, G. Loescher, K. Long and N. Sigona (eds), The Oxford Handbook of Refugee and Forced Migration Studies, pp. 475–88, Oxford: Oxford University Press. Martin, A. (2005), ‘Environmental conflict between refugee and host communities’, Journal of Peace Research, 42 (3), 329–46. Moore, W. H. and S. M. Shellman (2006), ‘Refugee or internally displaced person? To where should one flee?’, Comparative Political Studies, 39 (5), 599–622. Moore, W. H. and S. M. Shellman. (2007), ‘Whither will they go? A global study of refugees’ destinations, 1965–1995’, International Studies Quarterly, 51 (4), 811–34. Orchard, P. (2014), A Right to Flee: Refugees, States, and the Construction of International Cooperation, Cambridge: Cambridge University Press. Reuveny, R. and W. H. Moore. (2009), ‘Does environmental degradation influence migration? Emigration to developed countries in the late 1980s and 1990s’, Social Science Quarterly, 90 (3), 461–79. Ruëgger, S. and H. Bohnet (2018), ‘The ethnicity of refugees (ER): a new dataset for understanding flight patterns’, Conflict Management and Peace Science, 35 (1), 65–88. Salehyan, I. (2008), ‘The externalities of civil strife: refugees as a course of international conflict’, American Journal of Political Science, 52 (4), 787–801. Salehyan, I. and K. S. Gleditsch (2006), ‘Refugees and the spread of civil war’, International Organization, 60 (2), 335–66. Savun, B. and C. Gineste (2019), ‘From protection to persecution: threat environment and refugee scapegoating’, Journal of Peace Research, 56 (1), 88–102. Schewel, K. (2019), ‘Understanding immobility: moving beyond the mobility bias in migration studies’, International Migration Review, 1–28, https://​doi​.org/​10​.1177/​ 0197918319831952. Schmeidl, S. (1997), ‘Exploring the causes of forced migration: a pooled time-series analysis, 1971–1990’, Social Science Quarterly, 78 (2), 284–308. Schon, J. (2019), ‘Motivation and opportunity for conflict-induced migration: an analysis of Syrian migration timing’, Journal of Peace Research, 56 (1), 12–27. Schwartz, S. (2019), ‘Home, again: refugee return and post-conflict violence in Burundi’, International Security, 44 (2), 110–45. Seawright, J. (2016), Multi-Method Social Science: Combining Qualitative and Quantitative Tools, Cambridge: Cambridge University Press. Sniderman, P. M. (2011), ‘The logic and design of the survey experiment’, in J. N. Druckman, D. P. Green, J. H. Kuklinski and A. Lupia (eds), Cambridge Handbook of Experimental Political Science, pp. 102–14, Cambridge: Cambridge University Press. Steele, A. (2009), ‘Seeking safety: avoiding displacement and choosing destinations in civil wars’, Journal of Peace Research, 46 (3), 419–30. Steele, A. (2011), ‘Electing displacement: political cleansing in Apartadó, Colombia’, Journal of Conflict Resolution, 55 (3), 423–45. Stefanovic, D. and N. Loizides (2011), ‘The way home: peaceful return of victims of ethnic cleansing’, Human Rights Quarterly, 33 (2), 408–30. Taylor, J. E., H. Zhu, A. Gupta, M. Filipski, J. Valli and E. Gonzalez (2016), ‘Economic impact of refugee settlements in Uganda’, World Food Programme Kampala, 1–10, https://​reliefweb​.int/​sites/​reliefweb​.int/​files/​resources/​wfp288256​.pdf. Toft, M. D. (2007), ‘The myth of the borderless world: refugees and repatriation policy’, Conflict Management and Peace Science, 24 (2), 139–57. UNHCR (2019), ‘United Nations High Commissioner on Refugees population statistics database’, http://​popstats​.unhcr​.org/​en/​overview/​.

PROGRESSING RESEARCH ON FORCED MIGRATION

43

Uzonyi, G. (2015), ‘Refugee flows and state contributions to post-Cold War UN peacekeeping missions’, Journal of Peace Research, 52 (6), 743–57. Wright, T. M. and S. Moorthy (2018), ‘Refugees, economic capacity, and host state repression’, International Interactions, 44 (1), 132–55. Zhukov, Y. M. (2013), ‘An epidemic model of violence and public support in civil war’, Conflict Management and Peace Science, 31 (1), 24–52. Zhukov, Y. M. (2015), ‘Population resettlement in war: theory and evidence from Soviet archives’, Journal of Conflict Resolution, 59 (7), 1155–85.

4

Urbanization: poverty, conflict, and climate change as causes and consequences

Matthew Cobb and Alex Braithwaite

Introduction While large agglomerations accounted for a small fraction of the world’s population several millennia ago, technological advances and economic growth, especially over the last two centuries, have fueled the growth of cities around the world (Davis 1955; Frey & Zimmer 2001). The countries that industrialized first, urbanized first; but the rest of the world is catching up. Following decolonization in the first three decades after World War II, many countries from the developing world also began urbanizing. These countries increased their industrial outputs at more rapid rates than had Western countries that urbanized in the nineteenth and early twentieth centuries (Frey & Zimmer 2001; Henderson 2002). The meteoric rise of China’s industrial capacity is emblematic of this trend. Throughout the latter half of the twentieth century, the Chinese government encouraged rapid population urbanization to meet the country’s fast-growing need for factories and construction (Friedman 2006). In 2018, the United Nations declared that 55 percent of the world’s population lived in urban areas (United Nations 2019), a sharp increase from the level of 30 percent as recently as 1950. While urbanization is reshaping the global population, it is unevenly distributed across the regions of the world. North America (82 percent in urban areas), Latin America and the Caribbean (81 percent), and Europe (74 percent) are highly urbanized, but Asia (50 percent) and Africa (43 percent) have at least as many individuals residing in rural areas.

45

46

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Nonetheless, urbanization throughout the developing world will continue to expand at high levels for decades. By some estimates, as many as 2.5 billion new urban dwellers will bring the urban total to two-thirds of global population by 2050 (United Nations 2019). As this trend continues, policymakers and scholars will need to develop a better understanding not just of the causes and consequences of urbanization, but the policies most effective for managing it sustainably. Yet, the countries most affected by rapid urbanization have the fewest resources with which to address the demographic transition, complicating policy implementation. In this chapter, we first discuss various conceptual and operational definitions of urbanization. Second, we dedicate the heart of the chapter to discussing various sources and consequences of urbanization. Specifically, we focus on the relationship between urbanization and economic development, conflict, and climate change, respectively. Throughout, we look to identify key findings and theoretical insights from the existing empirical literature and plausible areas for future research to address important research gaps and emerging trends.

What is urbanization? Urbanization is a complex demographic and socio-economic phenomenon with critical implications for a country's economic, political, and social situations. At its core, urbanization refers to a change in the number, land areas, and population size of urban settlements, as well as the number and share of residents in urban versus rural areas (United Nations 2019). Of course, what qualifies as an urban settlement varies by country and region. Of 233 countries or areas monitored by the UN in its World Urbanization Prospects publication, only 108 explicitly include population size or density as a criterion for designating what is urban. In many instances, they place a stronger focus on specific, local administrative designations (United Nations 2019). For example, according to the UN Demographic Yearbook (2017), Iceland requires that urban localities have only 200 residents, whereas the threshold is 2,000 in Bolivia, 10,000 in Malaysia, 20,000 in Turkey, and 50,000 in Japan. Other countries focus on divisions of public administration in determining urban status. In South Africa and Sri Lanka, any place with a local government is considered urban. Egypt, Niger, and Ecuador only consider areas urban if they are designated as national, provincial, or district capitals. Definitional disparities greatly impede cross-country comparison or urban and rural population data.

URBANIZATION: POVERTY, CONFLICT, AND CLIMATE CHANGE

47

The term “urbanization” refers to several related demographic concepts. First, it can refer to static urbanization levels, the proportion of a country’s population living in urban areas. In high-income countries, an average 81 percent of the population lives in urban areas. By contrast, in lower-middle-income countries, such as Indonesia or Kenya, the average is almost half this at just 41 percent. The most urban countries are those whose citizens live almost exclusively in cities, such as Singapore, with effectively 100 percent of its population urbanized (McGranahan & Satterthwaite 2014; United Nations 2014, 14). Second, urbanization is often used dynamically in reference to the annual growth in the percentage of a country’s population living in urban areas. A country is said to be urbanizing if the proportion of a country’s population living in urban areas increases relative to those living in rural areas. Between 1950 and 2018, the average level of urbanization across Asian countries rose considerably from 27 percent to 59 percent. In China, which now has a population in excess of 1.4 billion, 59 percent reside in urban areas, massive growth from levels lower than 20 percent in the middle of the twentieth century (United Nations 2019). Third, scholars and practitioners increasingly use the term urbanization as a synonym for urban growth, the absolute annual increase in a country’s population living in urban areas, irrespective of the size of a country’s rural population (McGranahan 2015). Mexico’s absolute urban population has grown considerably over the past half century. In 1950, 12 million Mexicans resided in urban areas; by 2018, almost 105 million did because of both general trends in population growth and increased urbanization (United Nations 2019). Urbanization is often used in reference to country-wide demographic trends. However, at lower levels of analysis it is possible to see sub-national variation in urbanization patterns. Some of a country’s provinces may urbanize quickly, while others might experience no urbanization. In many countries, urbanization rates are especially high in the largest cities, which tend to have relatively high levels of infrastructure and markets, conducive to higher rates of growth. This is often the case for national capitals and, certainly, among “megacities” with populations of ten million or more. This growth is often encouraged by governments, which focus on service provision to key locations, but it can be detrimental to a country’s long-term development prospects. Essentially, primate cities – cities that are disproportionately larger than other cities within their respective countries – may struggle to accommodate high rates of growth and smaller, less influential cities may be left behind (Duranton 2008; Gotham & King 2020; Henderson 2002). The world’s most populous countries, China and India, have seen especially large growth in megacities, often influenced by

48

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

public policies aimed at harnessing urban growth to achieve economic growth (Sankhe et al. 2011). Other parts of the developing world have also experienced urbanization in ways that showcase the important role of megacities or primate cities. Known as metropolization, the growth of primate cities is associated with rising agricultural prices in Ethiopia (Vandercasteelen et al. 2018), economic inequality and declining living standards in Brazil (Campolina Diniz & Vieira 2016), and conflict over governmental incompatibilities in much of Africa (Schulz 2015). This sub-national variation within and across countries poses problems for urbanization researchers. After all, even at the country-level, official statistics depend upon conceptual and operational definitions that vary by government. An “urban dwelling” in one location might not correspond to that of another. Accordingly, we should anticipate that while interesting research questions lurk in the study of sub-national urbanization, so, too, do problems of measurement and comparison across cases.

Sources of urbanization Regardless of definition, urbanization tends to result from either the migration of people from rural to urban areas, or as a consequence of birth rates in urban areas exceeding those of rural areas (e.g. through relatively higher birth rates in cities) (McGranahan & Satterthwaite 2014; United Nations 2014, 14). A country’s urbanization rate – the annual percentage growth in urban populations as a percentage of the country’s overall population – is usually driven by the former, whereas absolute growth in urban population stems from the latter (Jedwab et al. 2014; Tacoli et al. 2015). Populations are becoming increasingly mobile, with internal migration more common than international. Worldwide, cities are the most common destinations for internal migrants, most of whom move in rural–urban or urban– urban patterns (Skeldon 2017). It is important to note, however, that in many developing countries, rural–urban migration patterns may be circular. In such cases, rural people may move to cities only for a short time, later returning to their homes in rural areas. Such trends are especially prominent in parts of Asia and Africa and may substitute for permanent relocation to cities, thereby slowing long-term demographic shift toward urban areas (Potts 2012; Wickramasekara 2014).

URBANIZATION: POVERTY, CONFLICT, AND CLIMATE CHANGE

49

This section focuses on rural–urban migration. An individual’s decision to migrate is complex, but some factors stand out as important, especially given the situations of developing countries where contemporary urbanization is most prevalent. Poverty, conflict, and climate change push people out of one place and into another, primarily from rural to urban areas. Our understanding of each of these issues would benefit greatly from more sustained research. In particular, we would benefit from improved understanding of the numerous and complex effects of urbanization upon economic development, political conflict, and climate change. What follows is a discussion of how each of these processes may be affected by the demographic shifts associated with urbanization. Poverty and economic development Urbanization is often associated with economic development. Most of the world’s population who live in extreme poverty (surviving on less than US$1.25 per day) are concentrated in rural areas of developing countries (Ahmed et al. 2014). Scholars and policymakers often contrast rural poverty with the seemingly vast opportunities of an urban society and discuss urbanization as a pathway to prosperity. Scholars linking urbanization to modernization (e.g. Harris & Todaro 1970; Huntington 1971; Rostow 1959; Todaro 1969) anticipate urbanization processes triggering economic growth and, in turn, bolstering state capacity (Hendrix 2010; McBride et al. 2011; Sobek 2010) and lowering the risk of intrastate conflict (Fearon & Laitin 2003; Hegre & Sambanis 2006). However, it seems unreasonable to assume that the achievement of growth, capacity, and peace is automatic or rapid for urbanizing countries. According to Turok and McGranahan (2013), in the developing world, the link between economic and urban growth varies greatly among countries. In Singapore and Hong Kong, urbanization comes with significant economic growth, while India and Brazil continue to experience economic difficulties. Urbanization provides a relief valve for those unhappy with rural living conditions or career prospects, especially in the developing world. Moving to urban areas does not always pan out as expected for migrants, however. The movement of migrants into cities is associated with the growth of informal housing and labor markets as low-income people arriving in cities work outside of officially registered businesses or live-in slum communities (Auerbach et al. 2018). In some countries, rural–urban migrants may outnumber available jobs, contributing to “urbanization without growth”. Moreover, higher rates of urbanization in developing countries often do not translate into higher rates of economic growth and may concentrate poverty (Fay & Opal 2000; Gollin et al. 2016). As a result, growing cities in the developing world often see great

50

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

socio-economic inequality and low living standards, particularly for recent rural migrants. This is exemplified by the growth of slum communities in cities such as Mumbai, Johannesburg, and Rio de Janeiro. Slums have informal housing and low access to public services but many rural people are still willing to move to them in hopes of a better life. Some scholars view slum formation optimistically as a temporary phenomenon, a stepping stone to future economic development (Glaeser 2011). Others see it as a “poverty trap”, creating inefficiencies that reduce governments’ abilities to improve living standards through investments in infrastructure and public welfare programs (Marx et al. 2013). If individuals who move to cities are unable to make a decent living, they will be less able to send remittances back home to their families in rural areas. In this way, urbanization in some developing countries may fall short of people’s expectations that migration into cities should benefit migrants who move to cities as well as their original rural communities. This phenomenon is exemplified by findings from Arouri et al. (2017), who reveal that rural Vietnamese households tend to become poorer as people move to the cities. Nevertheless, perceived opportunities may be strong enough to attract new migration even if actual opportunities are less plentiful. For those who find employment, migration into cities provides an opportunity to send money back to their rural family members. For example, in their respective studies of India and China, Cali and Menon (2012) and Du et al. (2005) argue that urbanization leads to improvements in rural household incomes. Du et al. (2005) note that although rural households benefit significantly from remittances sent from family members in urban areas, people from the poorest rural households are unlikely to migrate; as with international migration, the poorest are the least likely to move across borders (see Chapter 3, this volume). The pace of urbanization and levels of urban resources also impact a country’s chance of economically benefiting from urbanization. If urban areas severely lack infrastructure – as in much of the developing world – a country becomes less likely to experience growth or poverty reduction through rural–urban migration (Turok & McGranahan 2013). Thus, economic development is somewhat dependent on whether improvements to urban infrastructure can keep pace with influxes of new migrants. Researchers commonly refer to contexts in which urban growth outpaces infrastructure development as “over-urbanization” (see e.g. Bienen 1984), as in China where dozens of cities were built over recent decades to accommodate up to 300 million rural migrants but now lie empty (Shepard 2015).

URBANIZATION: POVERTY, CONFLICT, AND CLIMATE CHANGE

51

Even if urbanization benefits economic growth in developing countries it may contribute to their political instability. Olson (1963), for example, argues that economic growth is likely to increase economic inequalities between groups or otherwise disrupt existing social orders within a country. As shifting economic conditions either threaten a dominant group’s hold on political power or increase domestic grievances through economic disparities, armed conflict may become more likely. This argument presents the idea that, while development may contribute to a country’s stability in the long term, it is not a sufficient condition for stability and may, in fact, be subject to short-term setbacks. Clearly, we need to improve our measurement and assessment of the role of urbanization in undergirding or undermining economic growth. Other resource-based incentives may also pressure people to relocate. Based on research on migration within Nepal, Dudwick et al. (2011, 186) demonstrate that workers from impoverished rural areas prefer to migrate to proximate, wealthier, urban areas. These economic preferences may explain why development aid targeted at rural areas of developing countries significantly reduces the likelihood that people from those areas will emigrate (Gamso and Yuldashev 2018). The effects of these preferences raise an interesting puzzle. On the one hand, the very poorest rural residents have an incentive to migrate toward urban areas; on the other hand, dire circumstances limit the ability to move. According to Lipton (1977, 217–18; 1980), economists should not be overly optimistic about urbanization as a solution to rural poverty. Even if people from poor rural areas are able to relocate to cities – an especially difficult task for those whose poverty makes it hard to afford the cost of moving – they often have less education and job-related skills than their urban counterparts and fewer connections or knowledge about the urban job market. In the cities, they may have difficulty finding employment or be forced into poor living conditions (Cohen 2006; Hove et al. 2013). Rural–urban migration can indeed come with economic growth. However, cities with better infrastructure are often better able to benefit from rural–urban in migration (Castells-Quintana 2017). Although empirical studies often uncover a statistical link between urbanization and development (e.g. Njoh 2003; Ravallion et al. 2007), urbanization alone is not a sufficient condition of growth. While some research shows that economic development in rural areas plays a larger role in reducing a country’s overall levels of poverty than does urbanization, urbanization is still a correlate of national economic development and can help countries develop and to reduce income disparities between urban and rural areas (Imai et al. 2017; Liddle 2017). At the same time, increasing urbanization rates do not translate

52

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

into higher economic growth rates, especially when urbanization results from the actions of government-based programs (Chen et al. 2014). Sekkat (2013) identifies a U-shaped relationship between urbanization and poverty, such that one can expect the highest levels of poverty either in countries whose populations are most spread out across rural lands, or in those whose populations are clustered together in crowded cities. Conflict The civil war literature traditionally discusses civil conflict as a rural phenomenon in which the state’s difficulty in governing remote, rural locations enables rebels to hide and evade detection (Fearon and Laitin 2003; Hegre and Sambanis 2006). Rural insurgencies also emerge in response to locally concentrated grievances stemming from the inherent poverty of these regions. This is seen in the rural-focused operations of the Shining Path in Peru (de Soto 1989) and the Maoist insurgents of India (Hoelscher et al. 2012). The corollary of these logics is that urban areas were traditionally viewed as being somewhat immune from experiencing large-scale guerilla warfare, and much more likely to play host to social disturbances, such as riots and criminal violence, as well as opportunistic terrorist violence. However, a series of campaigns fought in urban areas in Iraq, Syria, and elsewhere over recent years might prompt scholars to revise this traditional perspective. In remote locations, it is much more difficult for government forces to track and fight armed non-state actors. Rebel groups can take advantage of the situation by promoting anti-government messages and recruiting rural citizens, especially in places where the population may have preexisting political or social grievances (Bienen 1984, 663–4; Collier 2003, 71; Kalyvas 2006). Rural young men are prime candidates for recruitment into rebel groups and their participation is often critical to rebel campaigns (Kunkeler and Peters 2011). Accordingly, migration to cities provides rural people with a non-violent means of improving their lives and avoiding recruitment into ongoing insurgency. Furthermore, internal migration of this kind can provide opportunities for individuals to send remittances back to their family in rural areas suffering from conflict-related problems. Traditional perspectives highlight that rural areas are more prone to large-scale anti-government violence than are urban areas. Newer claims have emerged, however, to suggest that urbanization is a factor that fuels violence. Some research on developing countries shows that the concentration of people in cities may sometimes worsen a country’s risk of armed violence. Several studies of conflict in Africa indicate that civil conflict, particularly that involv-

URBANIZATION: POVERTY, CONFLICT, AND CLIMATE CHANGE

53

ing competition for control of the government, is most common in and around major cities, particularly poor urban areas with numerous potential recruits (Buhaug & Rød 2006; Raleigh 2015; Raleigh & Hegre 2009). More consistently demonstrated is a relationship between urbanization and criminal violence. Countries with more urban populations tend to experience higher rates of homicide, robbery, gang activity, and rioting (Beall et al. 2013; Kunkeler & Peters 2011). Rebel militias typically focus on enlisting young men from low-income rural communities (Kunkeler & Peters 2011). This strategy, common in many developing world conflicts, is a challenge in urbanizing societies because there is a narrower pool of recruits to flesh out the ranks of rebel forces. Urbanization involves the relocation of many poor individuals whose economic deprivation may elevate their willingness to engage in violence. Increasing incomes in urban environments may reduce incentives to rebel. In the developing world, many of the people who move to the cities for work often send money home to their families in rural areas, thus reducing the bite of poverty there (Arouri et al. 2017; Cali & Menon 2012; Muggah 2012, 17), and lessening the effect of grievances that drive conflict. Cities may be ill-prepared to accept an influx of workers or to accommodate the added strain these new individuals place on existing public infrastructure. This may disadvantage internally displaced people (IDPs) in urban environments relative to pre-existing urban residents. For example, IDPs cluster in the Soacha municipality, near Colombia’s capital, but are often less educated than non-migrant residents and more likely to be unemployed or relegated to informal sector employment and thus often poorer than other urban laborers (Aysa-Lastra 2011). When migration into urban areas brings rural people into a city with a poor economy and little infrastructure, urban residents may see themselves as losing out in a zero-sum competition with new arrivals for jobs, housing, and access to public services. If this happens, normally peaceful urban residents – those not already involved in an armed group – suddenly find themselves under increased economic pressure. This socio-economic strain can generate grievances amongst urbanites displeased with the new competition. More general conflict research shows that the risk of conflict increases when people believe their identity group is threatened. New research is required to explore this specific dynamic in the context of processes of urbanization. If incoming migrants are of different ethnic or religious backgrounds from those already living in the cities, the mixing of people may heighten the perception of out-group threat toward separate or competing groups (Riek et al. 2006), increasing the role of

54

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

identity in urban violence (Gubler & Selway 2012). Under what conditions is this likely? And what policies and government interventions might be able to balance against this and ensure the sustainability of urbanization? Climate change Given current warming and emissions trends, the World Bank estimates that we will observe 86 million new forced migrants in sub-Saharan Africa over the next three decades, 40 million in South Asia and 17 million in Latin America (World Bank 2018). The United Nations’ projections are even bleaker, with more than 200 million expected to be forced from their homes by direct and indirect effects of climate change by 2050 (IOM 2009). Climate-related phenomena can make life harder for rural people, incentivizing them to move to cities, a trend highlighted in studies analyzing the effects of rainfall on people’s decisions to migrate. In rural areas, rainfall provides the water essential for growing crops and maintaining grazing lands needed for cattle raising. Empirical research shows that in Africa, decreasing rainfall undermines rural livelihoods, prompting internal migration to urban areas (Hope 2011; Marchiori et al. 2012). Alternatively, rural–urban migration is also spurred by severe flooding in rural places, especially in low-lying coastal areas and flood plains (Douglas et al. 2008; Neumann et al. 2015, 27). One example is flood-prone Bangladesh, where climate change has both short- and long-term impacts on urbanization. Rural people in Bangladesh sometimes temporarily migrate to cities, seeking shelter from severe weather or aiming to earn money to replace disaster losses (Poncelet et al. 2010, 215–16). When salty flood waters diminish soil quality, causing long-term reductions in crop yields, many farmers feel compelled to permanently migrate to cities (Gray & Mueller 2012). Although climate change is not the source of every natural disaster, increases in severe weather events associated with climate change come with great consequences for urbanized populations. This is especially disturbing given the number of major coastal cities threatened by rising sea levels. Jakarta, one of the world’s fastest growing cities, could be entirely under water by 2050. Shanghai, Hong Kong, Mumbai, and Kolkata are also under threat, and two-thirds of all major cities are in coastal regions (Wallace-Wells 2019). These problems are especially severe in developing countries where the risk of death from natural disasters is much higher in cities than in the countryside (Kahn 2017, 235; Neumann et al. 2015). In such countries, housing structures may be poorly constructed or citizens and governments may be ill-equipped to respond to emergency situations. Infrastructure problems, particularly

URBANIZATION: POVERTY, CONFLICT, AND CLIMATE CHANGE

55

sanitation and water systems, worsen the effects of natural disasters in the developing world, where flood waters often lead to water contamination and the spread of disease (Kahn 2017, 235; Zorn 2018, 60–61). Although natural disasters may cause high fatality levels in urban areas, they are unlikely to deter rural–urban migration in places where urban labor markets remain a strong draw for rural citizens. Research points to urbanization as a process that negatively impacts climate change for several reasons. First, dense infrastructure in cities absorbs heat (Poon 2018), which can exacerbate direct effects during heat waves, such as Chicago’s deadly 1995 wave (Klinenberg 2002). The scale of these problems looks destined to increase. There are 354 major cities with average maximum temperatures above 95 degrees Fahrenheit. By 2050, this number could be 970 (Wallace-Wells 2019). Relatedly, 14 out of the 20 biggest cities globally now experience water scarcity or drought conditions on a frequent basis. Accordingly, scenarios such as the taps running dry in Cape Town in the springs of 2018 and 2019 will only become more familiar. Second, cities are home to factories and power plants that emit greenhouse gases (Sharma 2011). Compared to rural areas, cities’ economies tend to be much more focused on industrial production and as a country becomes more urban, its economic activities usually shift away from agriculture and toward heavily polluting industries. Urban populations also tend to use more electricity than rural populations, so there is also a potential for severe environmental impacts of generating electricity, running transportation systems, and meeting other consumer and residential needs for these growing cities (Dhakal 2009; Jones 1991). This point is especially clear in large cities like Beijing and New Delhi, where coal-burning power plants – in addition to their many manufacturing plants – come with enormous carbon footprints and degrade the air quality (Aunan et al. 2018; Guttikunda et al. 2014; Krzyzanowski et al. 2014, 188–9). Indeed, New Delhi’s population of more than 26 million were breathing such poor-quality air in 2017 that it was the equivalent of smoking more than two packets of cigarettes each day (Wallace-Wells 2019). It is in this area – the relationship between urbanization and climate change – that arguably the largest gaps in research remain. For instance, climate change exacerbates natural disasters of various kinds, which, in turn, provoke migration patterns with rather distinct signatures. More research could help us better understand the triggers of sudden versus slow onset migrations, as well as those that represent permanent resettlements as compared to temporary or seasonal responses to events. Moreover, while much attention is paid to questions regarding when we might observe climate migration and what effect

56

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

this might have on receiving (urban) areas, less is asked about whether climate migration helps individuals to avoid the catastrophic effects of climate change and natural disasters. In other words, is climate migration a sustainable form of adaptation to the increasingly severe threats that our global community faces?

Conclusion In this chapter, we have attempted to convey some of the central debates and findings regarding urbanization globally. We have focused on three substantive domains in the form of the relationship between urbanization and economic development, conflict, and climate change. This review has demonstrated that processes of urbanization are associated with both positive and negative externalities. Urbanization, when sustainable, is associated with an escape from conditions of poverty, warfare, and, potentially, a means toward climate adaptation. However, given gross demographic challenges posed by population growth across developing regions, it is easy to anticipate that we might more frequently observe the negatives of resource scarcity, social disturbances and rising temperatures. We have identified a variety of gaps in the empirical literature, areas where research could provide a boon to our existing knowledge base. We leave to the end perhaps the greatest research challenge of them all: How can we best understand the nexus between these three substantive areas? How do development, conflict, and climate change intersect in urban areas and during processes of urbanization? The preceding pages appear to highlight two important considerations that surely warrant investigation in future studies. First, internal migration lies at the heart of each of these relationships and their intersection. Second, government interventions will be central to liberating the positive and constraining the negative externalities of urbanization. To understand what the future of urbanization promises, scholars must understand both patterns of migration and government interventions in processes of urbanization.

References Ahmed, A.U., R.V. Hill and F. Naeem (2014), ‘The Poorest: Who and Where They Are?’, in J. von Braun and F.W. Gatzweiler (eds), Marginality: Addressing the Nexus of Poverty, Exclusion and Ecology, Dordrecht: Springer, 85–100.

URBANIZATION: POVERTY, CONFLICT, AND CLIMATE CHANGE

57

Arouri, M., A.B. Youssef and C. Nguyen (2017), ‘Does Urbanization Reduce Rural Poverty?: Evidence from Vietnam’, Economic Modelling, 60, 253–70. Auerbach, A.M., A. LeBas, A.E. Post and R. Weitz-Shapiro (2018), ‘State, Society, and Informality in Cities of the Global South’, Studies in Comparative International Development, 53 (3), 261–80. Aunan, K., M.H. Hansen and S. Wang (2018), ‘Introduction: Air Pollution in China’, The China Quarterly, 234, 279–98. Aysa-Lastra, M. (2011), ‘Integration of Internally Displaced Persons in Urban Labour Markets: A Case Study of the IDP Population in Soacha, Colombia’, Journal of Refugee Studies, 24 (2), 277–303. Beall, J., T. Goodfellow and D. Rodgers (2013), ‘Cities and Conflict in Fragile States in the Developing World’, Urban Studies, 50 (15), 3065–83. Bienen, H. (1984), ‘Urbanization and Third World Stability’, World Development, 12 (7), 661–91. Buhaug, H. and J.K. Rød (2006), ‘Local Determinants of African Civil Wars, 1970–2001’, Political Geography, 25 (3), 315–35. Cali, M. and C. Menon (2012), ‘Does Urbanization Affect Rural Poverty? Evidence from Indian Districts’, World Bank Economic Review, 27 (2), 171–201. Campolina Diniz, C. and D.J. Vieira (2016), ‘Brazil: Accelerated Metropolization and Urban Crisis’, Area Development and Policy, 1 (2), 155–77. Castells-Quintana, D. (2017), ‘Malthus Living in a Slum: Urban Concentration, Infrastructure, and Economic Growth’, Journal of Urban Economics, 98, 158–73. Chen, M. et al. (2014), ‘The Global Pattern of Urbanization and Economic Growth: Evidence from the Last Three Decades’, PloS One, 9 (8), e103799. Cohen, B. (2006), ‘Urbanization in Developing Countries: Current Trends, Future Projections, and Key Challenges for Sustainability’, Technology in Society, 28 (1–2), 63–80. Collier, P. (2003), Breaking the Conflict Trap: Civil War and Development Policy, Washington, DC: World Bank Publications. Davis, K. (1955), ‘The Origin and Growth of Urbanization in the World’, American Journal of Sociology, 60 (5), 429–37. De Soto, H. (1989), The Other Path: The Economic Answer to Terrorism, New York: Basic Books. Dhakal, S. (2009), ‘Urban Energy Use and Carbon Emissions from Cities in China and Policy Implications’, Energy Policy, 37 (11), 4208–19. Douglas, I., K. Alam, M. Maghenda, Y. Mcdonnell, L. McLean and J. Campbell (2008), ‘Unjust Waters: Climate Change, Flooding and the Urban Poor in Africa’, Environment and Urbanization, 20 (1), 187–205. Du, Y., A. Park and S. Wang (2005), ‘Migration and Rural Poverty in China’, Journal of Comparative Economics, 33 (4), 688–709. Dudwick, N. et al. (eds) (2011), From Farm to Firm: Rural-Urban Transition in Developing Countries, Washington, DC: World Bank. Duranton, G. (2008), ‘From Cities to Productivity and Growth in Developing Countries’, Canadian Journal of Economics/Revue Canadienne d’Economique, 41 (3), 689–736. Fay, M. and C. Opal (2000), ‘Urbanization without Growth: A Not so Uncommon Phenomenon’, World Bank, Policy Research Paper 2412. Fearon, J.D. and D.D. Laitin (2003), ‘Ethnicity, Insurgency, and Civil War’, American Political Science Review, 97 (1), 75–90. Frey, W.H. and Z. Zimmer (2001), ‘Defining the City’, in  Ronan Paddison (ed.), Handbook of Urban Studies, London: Sage Publications, 14–35.

58

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Friedmann, J. (2006), ‘Four Theses in the Study of China’s Urbanization’, International Journal of Urban and Regional Research, 30 (2), 440–51. Gamso, J. and F. Yuldashev (2018), ‘Does Rural Development Aid Reduce International Migration?’, World Development, 110, 268–82. Glaeser, E. (2011), Triumph of the City, London: Penguin Books. Gollin, D., R. Jedwab and D. Vollrath (2016), ‘Urbanization with and without Industrialization’, Journal of Economic Growth, 21 (1), 35–70. Gotham, K.F. and A.J. King (2020), ‘Urbanization’, in George Ritzer and Wendy Wiedenhoft Murphy (eds), Wiley Blackwell Companion to Sociology, 2nd edn, Chichester: John Wiley & Sons, 267–82. Gray, C.L. and V. Mueller (2012), ‘Natural Disasters and Population Mobility in Bangladesh’, Proceedings of the National Academy of Sciences, 109 (16), 6000–6005. Gubler, J.R. and J.S. Selway (2012), ‘Horizontal Inequality, Crosscutting Cleavages, and Civil War’, Journal of Conflict Resolution, 56 (2), 206–32. Guttikunda, S.K., R. Goel and P. Pant (2014), ‘Nature of Air Pollution, Emission Sources, and Management in the Indian Cities’, Atmospheric Environment,  95, 501–10. Harris, J.R. and M.P. Todaro (1970), ‘Migration, Unemployment and Development: A Two-Sector Analysis’, American Economic Review, 60 (1), 126–42. Hegre, H. and N. Sambanis (2006), ‘Sensitivity Analysis of Civil War Onset’, Journal of Conflict Resolution, 50 (4), 508–35. Henderson, V. (2002), ‘Urbanization in Developing Countries’, World Bank Research Observer, 17 (1), 89–112. Hendrix, C.S. (2010), ‘Measuring State Capacity: Theoretical and Empirical Implications for the Study of Civil Conflict’, Journal of Peace Research, 47 (3), 273–85. Hoelscher, K., J. Miklian and K.C. Vadlamannati (2012), ‘Hearts and Mines: A District-Level Analysis of the Maoist Conflict in India’, International Area Studies Review, 15 (2), 141–60. Hope, K.R. (2011), ‘Climate Change in the Context of Urban Development in Africa’, in B. Yuen and A. Kumssa (eds), Climate Change and Sustainable Urban Development in Africa and Asia, Dordrecht: Springer, 37–55. Hove, M., E.T. Ngwerume and C. Muchemwa (2013), ‘The Urban Crisis in Sub-Saharan Africa: A Threat to Human Security and Sustainable Development’, Stability, 1 (2), 1–14. Huntington, S.P. (1971), ‘The Change to Change: Modernization, Development, and Politics’, Comparative Politics, 3 (3), 283–322. Imai, K.S., R. Gaiha and A. Garbero (2017), ‘Poverty Reduction during the Rural–Urban Transformation:  Rural  Development is Still More Important than Urbanisation’, Journal of Policy Modeling, 39 (6), 963–82. IOM (2009), ‘Migration, Environment, and Climate Change: Assessing the Evidence’, Geneva: United Nations International Organization for Migration. Jedwab, R., L. Christiaensen and M. Gindelsky (2014), ‘Rural Push, Urban Pull and … Urban Push? New Historical Evidence from Developing Countries’, Washington, DC: Institute for International Economic Policy. Jones, D.W. (1991), ‘How Urbanization Affects Energy-Use in Developing Countries’, Energy Policy, 19 (7), 621–30. Kahn, M.E. (2017), ‘Will Climate Change Cause Enormous Social Costs for Poor Asian Cities?’, Asian Development Review, 34 (2), 229–48. Kalyvas, S.N. (2006), The Logic of Violence in Civil War, Cambridge: Cambridge University Press.

URBANIZATION: POVERTY, CONFLICT, AND CLIMATE CHANGE

59

Klinenberg, E. (2002), Heat Wave: A Social Autopsy of Disaster in Chicago, Chicago, IL: Chicago University Press. Krzyzanowski, M., J.S. Apte, S.P. Bonjour, M. Brauer, A.J. Cohen and A.M. Prüss-Ustun (2014), ‘Air Pollution in the Mega-Cities’, Current Environmental Health Reports, 1 (3), 185–91. Kunkeler, J. and K. Peters (2011), ‘“The Boys Are Coming to Town”: Youth, Armed Conflict and Urban Violence in Developing Countries’, International Journal of Conflict and Violence, 5 (2), 277–91. Liddle, B. (2017), ‘Urbanization and Inequality/Poverty’, Urban Science, 1 (4), 35. Lipton, M. (1977), Why Poor People Stay Poor: A Study of Urban Bias in World Development, London and Canberra: Temple Smith; Australian National University Press. Lipton, M. (1980), ‘Migration from Rural Areas of Poor Countries: The Impact on Rural Productivity and Income Distribution’, World Development, 8 (1), 1–24. Marchiori, L., J. Maystadt and I. Schumacher (2012), ‘The Impact of Weather Anomalies on Migration in Sub-Saharan Africa’, Journal of Environmental Economics and Management, 63 (3), 355–74. Marx, B., T. Stoker and T. Suri (2013), ‘The Economics of Slums in the Developing World’, Journal of Economic Perspectives, 27 (4), 187–210. McBride, M., G. Milante and S. Skaperdas (2011), ‘Peace and War with Endogenous State Capacity’, Journal of Conflict Resolution, 55 (3), 446–68. McGranahan, G. (2015), ‘Urbanization’, in N. Smelser and P. Baltes (eds), International Encyclopedia of the Social & Behavioral Sciences, vol. 2 (24), Amsterdam: Elsevier, 958–64. McGranahan, G. and D. Satterthwaite (2014), Urbanisation: Concepts and Trends, London: International Institute for Environment and Development (IIED). Muggah, R. (2012), ‘Researching the Urban Dilemma: Urbanization, Poverty and Violence’, Ottawa: International Development Research Centre (IDRC). Neumann, B., A.T. Vafeidis, J. Zimmermann and R.J. Nicholls (2015), ‘Future Coastal Population Growth and Exposure to Sea-Level Rise and Coastal Flooding – A Global Assessment’, PloS One, 10 (3), e0118571. Njoh, A.J. (2003), ‘Urbanization and Development in Sub-Saharan Africa’, Cities, 20 (3), 167–74. Olson, M. (1963), ‘Rapid Growth as a Destabilizing Force’, Journal of Economic History, 23 (4), 529–52. Poncelet, A., F. Gemenne, M. Martiniello and H. Bousetta (2010), ‘A Country Made for Disasters: Environmental Vulnerability and Forced Migration in Bangladesh’, in A. Tamer and J. Jäger (eds), Environment, Forced Migration and Social Vulnerability, Berlin, Heidelberg: Springer, 211–22. Poon, L. (2018), ‘Street Grids May Make Cities Hotter’, City Lab, April 27, https://​ www​.bloomberg​.com/​news/​articles/​2018​-04​-27/​city​-grids​-intensify​-the​-urban​-heat​ -island​-effect. Potts, D. (2012), ‘What Do We Know about Urbanisation in Sub-Saharan Africa and Does It Matter?’, International Development Planning Review, 34 (1), v–xxii. Raleigh, C. (2015), ‘Urban Violence Patterns across African States’, International Studies Review, 17 (1), 90–106. Raleigh, C. and H. Hegre (2009), ‘Population Size, Concentration, and Civil War: A Geographically Disaggregated Analysis’, Political Geography, 28 (4), 224–38. Ravallion, M., S. Chen and P. Sangraula (2007), ‘New Evidence on the Urbanization of Global Poverty’, Population and Development Review, 33 (4), 667–701.

60

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Riek, B.M., E.W. Mania and S.L. Gaertner (2006), ‘Intergroup Threat and Outgroup Attitudes: A Meta-Analytic Review’, Personality and Social Psychology Review, 10 (4), 336–53. Rostow, W.W. (1959), ‘The Stages of Economic Growth’, Economic History Review, 12 (1), 1–16. Sankhe, S., I. Vittal and A. Mohan (2011), ‘Urban Giants: India and China, and Their Urbanization Paths’, Environment and Urbanization Asia, 2 (1), 1–12. Sekkat, K. (2013), ‘How Does Urban Concentration Affect Poverty in Developing Countries?’, Economic Research Forum Working Papers, No. 809. Sharma, S.S. (2011), ‘Determinants of Carbon Dioxide Emissions: Empirical Evidence from 69 Countries’, Applied Energy, 88 (1), 376–82. Shepard, W. (2015),  Ghost Cities of China: The Story of Cities Without People in the World’s Most Populated Country, London: Zed Books. Shulz, N. (2015), ‘Dangerous Demographics?: The Effect of Urbanization and Metropolisation on African Civil Wars, 1961–2010’, Civil Wars, 17 (3), 291–317. Skeldon, R. (2017), ‘International Migration, Internal Migration, Mobility, and Urbanization: Towards More Integrated Approaches’, New York: UNDESA. Sobek, D. (2010), ‘Masters of their Domains: The Role of State Capacity in Civil Wars’, Journal of Peace Research, 47 (3), 267–71. Tacoli, C., G. McGranahan and D. Satterthwaite (2015),  Urbanisation, Rural-Urban Migration, and Urban Poverty, London: Human Settlements Group, International Institute for Environment and Development. Todaro, M.P. (1969), ‘A Model of Labor Migration and Urban Unemployment in Less Developed Countries’, American Economic Review, 59 (1), 138–48. Turok, I. and G. McGranahan (2013), ‘Urbanization and Economic Growth: The Arguments and Evidence for Africa and Asia’, Environment and Urbanization, 25 (2), 465–82. United Nations (2014), ‘World Urbanization Prospects 2014’, New York: UN Department of Economic and Social Affairs. United Nations (2017), United Nations Demographic Yearbook 2017, New York: UN Department of Economic and Social Affairs. United Nations (2019), World Urbanization Prospects: The 2018 Revision (ST/ESA/ SER.A/420), New York: UN Department of Economic and Social Affairs. Vandercasteelen, J., S.T. Beyene, B. Minten and J. Swinnen (2018), ‘Cities and Agricultural Transformation in Africa: Evidence from Ethiopia’, World Development, 105, 383–99. Wallace-Wells, D. (2019), The Uninhabitable Earth: Life after Warming, London: Tim Duggan Books. Wickramasekara, P. (2014), ‘Circular Migration in Asia: Approaches and Practices’, in Graziano Battistella (ed.), Global and Asian Perspectives on International Migration, Cham: Springer, 51–76. World Bank (2018), Groundswell: Preparing for Internal Climate Migration, Washington, DC: The World Bank. Zorn, M. (2018), ‘Natural Disasters and Less Developed Countries’, in S. Pelc and M. Koderman (eds), Nature, Tourism and Ethnicity as Drivers of (De) Marginalization, Cham: Springer, 59–78.

5

Sex, demographics and national security

Valerie M. Hudson

Introduction One of the most important, if not the most important, determinant of demographics is the character of male/female relations within a given society. Until and unless artificial reproductive technologies become more pervasive, males and females will still need to interact for children to be born. The conditions under which that interaction occurs are thus critical to understanding demographic trajectories. Since demography has significant effects on the security, stability and resilience of nation states, the character of male/female relations thus has ramifications for those outcomes as well. In this chapter, we will explore these assertions through an examination of how the character of male/female relations affects the fate of a nation state. There is a small, but growing empirical literature linking the structuration of those relations with larger outcomes of national security, stability and resilience. In this particular chapter, and as an example of these linkages, we focus on how that structuration affects two specific phenomena of great interest to demographers: (a) birth and death rates, and (b) degree of marriage market obstruction resulting from sex ratio alteration, polygyny and brideprice unaffordability. Last, we raise research questions and data gaps at this intersection of gender, demography and security. First, however, we outline a conceptualization of how the structure of male/ female relations affects collective outcomes for the nation state.

61

62

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

The societal structuration of male/female relations Humankind is roughly divided into those who produce male gametes (hereafter, men or males) and those who produce female gametes (hereafter, women or females), with about half of the population male and half female. The union of a male gamete and a female gamete has been to this point the only manner in which new human beings are produced. Thus, any discussion of demography must of necessity examine the relationship between males and females. However, reproduction and sex are not identical between the two sexes. A male’s reproductive labor is limited to conception, whereas a lengthy gestation and potentially dangerous delivery of the child, followed by possibly years of lactation, is the sole province of the female. In addition, humankind exhibits a significant sexual dimorphism that sees men, on average, larger, heavier and stronger than women. With just these biological facts in hand, it is clear how an amazingly complex social structuration between these two halves of any human collective emerges and permeates the character of a society. Consider fundamental choices every human society confronts: Will these two groups, male and female, be considered equals, or will one group be considered superior and the other group inferior? Will decision-making for the collective be made by one or both groups? If males and females disagree, how is that conflict resolved? Are resources distributed and controlled equally by men and women? At heart, these are all political questions, and thus the answers a given society offers to these foundational questions will establish a first political order within that collective (Hudson et al., 2020). We assert that when the first political order – the sexual political order – is configured in such a way that it subordinates women through women’s exploitation, violence against women and autocratic rule over women, this in turn undermines the stability, security and resilience of the overall collective. It does so by making economic exploitation, corruption, violence, oppression and autocracy seem normal and unremarkable at the level of every household in society, laying the foundation for these practices to seem equally normal and unremarkable at the nation state level where they cause collective insecurity, collective instability and collective fragility. In other work, we have expounded a conceptual framework concerning the specific mechanisms of coercion and control that are put into place by societies choosing to establish a first political order where women are subordinated to

SEX, DEMOGRAPHICS AND NATIONAL SECURITY

63

men (Hudson et al., 2020). That conceptualization allows us to compare the level of women’s subordination across nation states, enabling a statistical analysis of trends, in addition to conventional qualitative process-tracing research.1 In the following section, we give two examples of how the structure of male/ female relations profoundly affects demographic factors that, in turn, affect national security, stability and resilience. These two linkages were empirically confirmed using our conceptualization of the subordinative first political order, using multivariate statistical modeling. First, is the treatment of women associated with trends in birth and death rates, and do these rates in turn affect national stability? Second, is the treatment of women associated with marriage market obstruction, and does such obstruction in turn affect national stability?

How the treatment of women affects a nation’s birth and death rates, in turn affecting national stability The subordination of women almost always entails women’s diminished control over their own reproductive capabilities. In subordinative societies, women have very little say over when and under what circumstances they engage in sex that may lead to procreation. Furthermore, they may have few legal or practical rights to access family planning methods that might serve to offset their lack of control over male sexual access to their bodies. Indeed, the value of women in highly subordinative societies may be reduced to their childbearing capability. Even more specifically, it may be reduced to their capability for bearing sons, who will belong to the extended male kin groups of their husbands. Consider how the context of normalized violence against women, plus patrilocal marriage, plus son preference, plus early age of marriage for girls has led to the plight of this Afghan woman who was interviewed by the Afghan Women’s Writing Project in 2014: Wounds on her face show she has recently been beaten. She cannot tell me much, she is afraid and cannot say her name to me, but she talks about her life. She does not tell her age; she says only that she was the youngest in her family. She is pregnant with her ninth child. She has eight children who are all girls, and in three months her next child will be born. She says her husband’s family forced her to continue to become pregnant to give birth to a boy and she says this time she knows she will have a son because the pregnancy feels different from the other times. All of her children were born at home and she came to the clinic only because of bleeding. Her husband

64

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

had beaten her. She says if she does not bear a son her husband will end their marriage and will abandon her with her daughters. (Nasima, 2014)

This individual case is not unique; Rizvi et al. conducted 30 focus groups of 250 women in neighboring Pakistan, finding that “many of the women indicated they become repeatedly pregnant due to pressure from their husband and his family, and are prohibited from using birth control. In addition, the women are kept in a perpetual state of malnourishment” (Rizvi et al., 2014). Because of the subordinative first political order, women are clearly having more children than they want and, arguably, more than they can support. The stories told by another Afghan woman, Zarghona, a 28-year-old with six children, makes this plain: Here in Afghanistan having children is not something a woman decides. Whatever men want is what happens. If my husband wants it, I will surely have more children. Most men usually want more than six children … In Afghanistan men don’t think about the future, they just want as many children as possible. They don’t worry about how the children live or whether they are healthy or not. (Zarghona, 2014)

In addition to these qualitative accounts, is there quantitative empirical support for the proposition that birth rates are higher where women are subordinated? Yes. In multivariate regression modeling controlling for seven covariates, our research team found that the afore-mentioned scale of the subordination of women was significantly and strongly associated with variables measuring birth rates and fertility. More specifically, using a p ≤ 0.001, the subordination of women was both significant and also the variable with the largest effect size in explaining the variables Demographic Pressure (Fund for Peace), Total Fertility Rate (World Bank), Contraceptive Prevalence (World Bank) and Bricker and Foley’s Youth Risk Factor (Bricker & Foley, 2013). The odds ratios were quite striking; for example, for every scale point worse on the index measuring the subordination of women, a country was 1.92 times as likely to have a high total fertility rate (above 3.67), and 1.68 times as likely to have a higher/worse score on the Youth Risk Factor scale. The effects on national security and stability of such higher birth rates are readily observable.2 The scholarly literature suggests that high fertility rates may threaten national security and stability; in the words of Potts and Hayden, “political instability and violence often follow hard on the heels of high birth rates”, noting that in 2008, the director of the CIA asserted that “rapid population growth [was] the number one problem in national security” (2008, p. 91). For example, high birth rates are associated with youth bulges which, in turn, are associated with higher levels of instability (Cincotta, 2011, pp. 188–9) such

SEX, DEMOGRAPHICS AND NATIONAL SECURITY

65

as was seen in the Arab revolutions of 2011 and 2012 (Bricker & Foley, 2013, pp. 190–91). Bricker and Foley conclude, “historical evidence shows some of the most violent periods of unrest in human history are linked to the presence of large youth cohorts” (2013, p. 181). Interestingly, the subordination of women not only brings higher birth rates, but also higher levels of death, disease and hunger. When a society disenfranchises those to whom it has also assigned the gendered role of ensuring the health and well-being of individuals – women – it cannot but reap the consequences. This is again borne out in statistical multivariate analysis of data from the World Health Organization: the subordination of women is a highly significant and strong predictor of variables such as life expectancy for both males and females, health expenditure per capita, maternal mortality, infant mortality, deaths of children due to diarrhea, prevalence of stunting, wasting and underweight status among children and the percentage of the overall population that is malnourished. The subordination of women is also strongly associated with food insecurity. For every step higher/worse on the scale of the subordination of women, a country is 1.8 times more likely to have a dire score on the Global Hunger Index. Gender, demography and national security are inextricably linked in this example. The subordination of women is clearly and strongly determinative of high fertility rates, high disease rates, high death rates and high hunger rates. In turn, these rates are connected to outcomes relevant to national security and stability.

How the treatment of women affects marriage markets, in turn affecting national stability A second example of the linkage between gender, demography and security can be found in the violence catalyzed through marriage market obstruction. While demographers have recognized the linkage between gender and security-related factors such as youth bulges, as documented above, there is need for further probing of this important linkage. In this section, we raise that bar, suggesting that the clearance or obstruction of marriage markets in any human society is a demographic characteristic strongly related to male grievance, which is in turn strongly related to levels of societal violence and conflict. In societies where the most powerful forces are extended male kin networks, the perpetuation of that kin group and its power through time is of utmost

66

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

concern. However, to this point in human history, men have not been able to produce new men without women. Indeed, most of the mechanisms of women’s subordination that we have already examined, such as few or no property and inheritance rights, early age of marriage, polygyny and others, are meant to subordinate female interests to male interests for the purposes of male reproduction of male kin. Marriage thus becomes an extremely important phenomenon of societies that subordinate women, for men must ensure that their offspring are in fact their kin. Thus, we find that in societies that subordinate women, young men are not considered fully adult men until they have married, with the potential of extending the male kin group into the future as well as accumulating assets and power that may be perpetuated through time. While this is the primary purpose of marriage in such societies, marriages are also used, of course, to create alliances between male kin groups, and can be seen as a form of exchange between men. The bride is offered in marriage by one man and received in marriage by another. Indeed, her will in the matter may be utterly irrelevant. Charrad cogently observes, “Women represent a potential source of rupture in the web uniting the men of the patrilineage”, which is why they are so strictly subordinated (2001, p. 55). As a result, the marriage market in which young men must participate to find wives is fraught with tension, not only for the individual young men themselves, but also for the society as a whole. Each young man will feel it absolutely necessary to marry for their own future in society, and the grievance associated with their inability to marry will also be felt in the larger society. The subordination of women through the mechanisms we have already outlined predisposes the society to marriage market obstruction, whereby young men – especially poorer ones – are unable to find wives. The mechanisms of women’s subordination create at least three structural goads which cause this obstruction to occur. The first goad to marriage market obstruction is by elimination of females from the childbearing and future childbearing population due to devaluation of female life as enacted by such phenomena as female infanticide, sex-selective abortion, passive neglect of females, active neglect of females, high levels of violence against women and high levels of female suicide. These practices significantly alter the sex ratio of the society. The second goad to marriage market obstruction is through polygyny, which mimics the effects of a highly masculinized sex ratio by allowing dominant males to accumulate women while non-dominant males may face a scarcity of women. In essence, this simulates an altered sex ratio among women of marriageable age. The third goad to marriage market obstruction is through

SEX, DEMOGRAPHICS AND NATIONAL SECURITY

67

high brideprice/wedding costs that substantially raise the age of marriage for young adult men – or preclude it altogether – due to inability to accumulate the demanded sums. The overall effect of all three structural goads is to delay marriage for men, especially men of low socio-economic status, or make it altogether unlikely for a significant percentage of such men in the society; in other words, the marriage market in the society cannot clear because it is obstructed. The result of such obstruction is destabilization, as men seek to better their market position in any way possible, legal or illegal. Abnormal sex ratios skewed toward males A significant imbalance between the number of young adult men and young adult women in society, whether through offspring sex selection, high mortality rates for girls compared to boys or women compared to men or polygyny, obstructs the marriage market. There are simply not enough women for each male patriline member to marry and produce offspring. These “surplus” men become the “bare branches” of their societies; branches of the patrilineal family tree that will never bear fruit. It is worth noting that for 2020, the estimate is that 12 percent of young adult men are surplus to the number of young adult women in India, and in China, the figure may be 15 percent (Hudson & den Boer, 2005). Of course, men with advantages will marry – it is men without advantages, who are already at risk of antisocial behavior, who will not. These non-elite men who find themselves pushed out of the marriage market will feel a deep sense of grievance, as well as a determination either to obtain, one way or the other, the advantages that will allow them to enter the market and be successful, or, alternatively, to circumvent or upend the system that relegates them to outsider status. Thus, it comes as no surprise that masculinized sex ratios correlate with property crime and violent crime rates in numerous empirical studies. For example, using annual province-level data, Edlund and colleagues find that for every 1 percent alteration of the sex ratio in favor of males, there is a corresponding 3.7 percent increase in violent and property crime rates in China. These are not white-collar crimes; these are bare branch crimes – the crimes of “bullies, bandits and rebels”. They hypothesize that high sex ratios create this violent reaction because they act much like polygyny does; the society is “wrecked by the revolt of the mass of inferior men who are condemned to celibacy by it” (2013, pp. 1520–34). Other empirical studies have also revealed similar results in other nations such as India and Korea: i.e. that violent crime increases in tandem with an increase in sex ratios (Cameron et al., 2016; Dreze

68

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

& Khera, 2000, pp. 335–52; Lu, 2012). Gottschall senses the Ourobouros-like nature of the equation: “parents bias their investment in favor of males largely because the world is violent; the world is violent largely because parents bias their investment in favor of males” (2008, p. 151). And crime is not the only trend that corresponds to sex ratios: protests and demonstrations also increase in size and frequency. Consider the Nien Rebellion in nineteenth-century China, which originated in a province where because of grotesquely abnormal sex ratios, 25 percent of men were unable to marry at all (Hudson & den Boer, 2005, p. 208). Daniel Little writes, “Families adopted the practice of female infanticide to increase family income and security, but the long-term aggregate result was a skewed demography in which there was a large surplus of young men. These young men became natural recruits for bandit gangs and local militia – thus providing resources for the emergence of collective strategies of predation and protection” (1992, p. 209). These rebels eventually had national aspirations for power, and were only put down by imperial forces with great loss of life. Prevalent polygyny A similar chain of events attends societies that select for prevalent polygyny instead of sex ratio alteration due to the value placed upon women’s productive labor, as in sub-Saharan Africa. Prevalent polygyny is the functional equivalent of an abnormal sex ratio. By marrying multiple women, elite men create an underclass of young adult men who are elbowed out of the marriage market because there are not enough women in the market: these women are alive – unlike societies with abnormal sex ratios, where they were never born in the first place – but they have been monopolized by men with far greater resources than these young men. For example, Tucker notes that in the 1950s in West Africa, due to polygyny, “at least a quarter of the male African population was permanently excluded from marriage, leaving a volatile class of unattached males” (2014, p. 83). And this does not bode well for peace within the society. Joseph Henrich and his co-authors assert: By expanding the pool of unmarried men and elevating the degree of intrasexual

competition, more polygynous marriages will increase men’s discounting of the future and risk-taking, resulting in more socially undesirable behaviours … In suppressing intrasexual competition and reducing the size of the pool of unmarried men, normative monogamy reduces crime rates, including rape, murder, assault, robbery and fraud, as well as decreasing personal abuses. (2012, p. 660)

SEX, DEMOGRAPHICS AND NATIONAL SECURITY

69

No wonder, then, that some studies have shown greater ease of recruitment into rebel groups in polygynous societies. For example, Mokuwa et al. demonstrate significantly greater ease of rebel recruitment in areas with higher rates of polygyny within Sierra Leone, compared to those areas with lower rates (2011, pp.  339–66). Anthropologists have found significant correlation between polygyny and the amount of warfare in which societies engage (Gleditsch et al., 2011, pp. 265–70; Harris, 1996, pp. 57–79; Kanazawa, 2009, pp. 25–34; Krieger & Renner, 2018). Some scholars have even asserted that polygynous societies are more likely to engage in expansionist warfare as a means of distracting low status mate-less males (Boone, 1983). In other words, polygyny has long been linked to societal instability and lack of resilience in the scholarly literature. One of the causal mechanisms involved is clearly the creation of a standing pool of marginalized and disaffected young adult men, in reaction to which authoritarianism appears useful as a counterweight. As Robert Wright puts it, “Extreme polygyny often goes hand in hand with extreme political hierarchy, and reaches its zenith under the most despotic regimes” (1995, p. 98). In an intriguing empirical study of 186 societies, Laura Betzig finds the correlation between despotism and polygyny to be 0.72, significant at the 0.01 level (1986). Korotayev and Bondarenko obtain similar results (2000). Furthermore, societies practicing polygyny were shown by McDermott and Cowden to also demonstrate higher per capita arms expenditures and significantly lower respect for political rights and civil liberties (2015). Brideprice and wedding costs In societies that highly subordinate women, it is common for a man to “buy a wife” by paying a brideprice to the bride’s father. Thus, another way to create the functional equivalent of a high sex ratio is a significant escalation in marriage/wedding costs for grooms in brideprice-practicing societies. For example, an entire ethnic group in South Sudan was destabilized by rising bride prices among the group around the time of independence. Cattle theft rose precipitously as desperate would-be grooms sought additional livestock to meet the rising price of brides, and revenge attacks followed these thefts (Richmond & Krause-Jackson, 2011). Jada Tombe, a young man in South Sudan, noted, “We risk our lives to raid other communities so we can pay bride price” (Aleu & Mach, 2016). A United States Institute of Peace (USIP) field study noted, “A government official provided an alternative motive for joining militia groups: ‘Some youth are joining the rebels [militias] to loot properties so they can marry.’ A recent Norwegian People’s aid report supports indications in the authors’ interview data that some young men in South Sudan join armed gangs, at least in part, because they believe it will help them

70

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

pay dowries”; note that they actually mean brideprice, not dowry (Sommers & Schwartz, 2011, p. 6). Societal stability and resilience are profoundly affected by the existence of a “brideprice economy.” Furthermore, rebel and terrorist groups are attuned to this source of discontent, and recruit by promising to solve the marriage asset problem for young men. It is fascinating to see just how many terrorist and rebel groups are concerned about the marriage prospects of the young men in their ranks. For example, Diane Singerman notes, “To mobilize supporters, there were many reports of radical Islamist groups in Egypt in the 1990s arranging extremely low-cost marriages among the group’s members” (2007, p. 49). More recently, “Hamas leaders have turned to matchmaking, bringing together single fighters and widows, and providing dowries and wedding parties for the many here who cannot afford such trappings of matrimony.” The Syrian government also provides brideprices and weddings for its soldiers, including those who are wounded and those who lost brothers in the civil war (Brennan, 2017). And, of course, ISIS was well-known for providing its foreign jihadis with the opportunity to marry they may not have had in their own country. In one such campaign, ISIS offered “its fighters a $1,500 bonus to go towards a starter home along with a free honeymoon in their stronghold city of Raqqa” (Baker, 2015). Another report found that “ISIS foreign fighters paid $10,000 dowries to the families of their brides”, suggesting that the group was attracting foreign fighters by promising resources (and available women) to marry (Ahram, 2015, pp. 57–78). Hudson and Matfess provide an in-depth case study of Boko Haram’s use of brideprice grievance to attract recruits through promises of facilitating marriage through the capture of young women (Hudson & Matfess, 2017). “In this crisis, these men can take a wife at no extra charge”, explained Kaka, a young woman orphaned, captured and raped by Boko Haram. “Usually it is very expensive to take a wife, very hard to get married, but not now” (Matfess, 2016b). Matfess met another young woman who told her story: Bawagana, a shy 15-year-old living in Sanda Kyarimi camp, one of the official IDP sites, said that a Boko Haram fighter had come to her home in Dikwa, 90 kilometres east of Maiduguri, and asked ‘Do you love me?’ ‘Of course I answered, “no!”’ she said, with her eyes fixed on the ground. ‘The boy got very angry and said: ‘If you do not come with me, I will kill your father, but if you come with me I will let him live.’ I followed to save my father.’ The boy left 10,000 naira (about $50) on the floor. It was a bride price in Boko Haram’s eyes. (Matfess, 2016a)

In sum, marriage markets may be obstructed by high marriage costs, prevalent polygyny or abnormal sex ratios favoring males. In societies that subordinate

SEX, DEMOGRAPHICS AND NATIONAL SECURITY

71

women, groups seeking to recruit young adult men to their cause can exploit these dynamics by preying on those interested in redressing the injustice they feel on a personal level. This, in turn, may seriously degrade the stability and security of the society. Furthermore, marriage market obstruction also affects governance in the society, pushing it toward greater authoritarianism. Empirical testing bears out these observations. In multivariate regression models, these three marriage market obstruction goads were found to be significantly associated with outcome variables such as the Global Terrorism Index (Institute for Economics and Peace) and the Political Terror Scale (Purdue University) (Gibney et al., 2019; Hudson & Hodgson, 2020). The broader scale of women’s subordination was also significantly associated in multivariate regression with measures of intra-state conflict, autocracy and levels of violence within the society (Hudson et al., 2020). For example, the odds of having a fragile state were 2.53 times higher for every scale point higher/worse on the measure of the subordination of women.

What is holding us back? We hope these examples have stimulated an interest in the many ways in which gender, demography and national security/stability are linked. At the same time, it is impossible not to notice that very few scholars are working in this tri-sectional research space. We suggest several reasons why this is the case. Lack of training in gender analysis Most national security graduate programs contain no requirement to undertake training in gender analysis. In fact, many of the top programs offer no classes whatsoever in gender analysis, for example MIT’s highly regarded graduate Security Studies Program (SSP Classes | MIT Security Studies Program (SSP), n.d.). National security is still a highly gendered field, coded male, and the idea that national security could be related to women is still a tough sell in certain circles. Astoundingly, there are likewise very few stand-alone courses in gender analysis in doctoral programs in demography. While one might argue that demography perforce includes a consideration of gender, that consideration does not normally probe the structuration of male/female relations except as those inhibit or allow access to birth control methods. An understanding of the nuances of brideprice, polygyny and so forth, are not normally part of the graduate curriculum in demography.

72

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Lack of data There is a profound lack of data for those wishing to investigate the nexus of gender, demography and security. Many important statistics are still not disaggregated by sex (Criado-Perez, 2020). In our own efforts, we discovered that key statistics, such as the prevalence of patrilocal marriage, have never been compiled before. While statistics on female literacy, or female labor force participation, or female representation in national legislatures are relatively easy to find in several excellent datasets, such as that of the World Economic Forum, or of the European Institute for Gender Equality, other critical data are completely absent. For example, even a basic statistic such as official rape rate is unavailable for approximately 80 countries. I am also involved in a project to quantify and map marriage market migration among the countries of Asia, and have found again that there is virtually no data to be had on this phenomenon. While Vietnam has reputedly one of the largest export markets for brides, it keeps no statistics on its women emigrating for such purposes. This is highly ironic, for Vietnam also has the worst birth sex ratio in the world, surpassing China’s, and thus an exodus of Vietnamese women to China, South Korea and Taiwan is very consequential. These challenges led several colleagues and me to create The WomanStats Project, a freely accessible online database that compiles, archives and also scales important information related to women. The WomanStats Database covers 176 nations (all those with at least 200,000 population), and has excellent coverage from 1995 onward. It contains “practice” variables, which encode accounts of what is actually happening on the ground, “law” variables, which specify official national as well as customary law concerning women, “data” variables stipulating quantified prevalences and “scale” variables which are our original ordinal scaling of important phenomena (The Womanstats Project, n.d.). We also map each of our scales (WomanStats Maps, n.d.). And yes, we have a scale of the prevalence of patrilocal marriage, if that is of interest to your research agenda. Lack of interdisciplinarity It is impossible to forward the gender/demography/security research agenda without being acquainted with the work of anthropologists, sociologists, ecologists, evolutionary biologists, historians, psychologists and political scientists. In our own work, each of these disciplines has theoretical frameworks and methodological skill sets that are invaluable to our research, besides those to be found in demography, gender and international relations. Unfortunately,

SEX, DEMOGRAPHICS AND NATIONAL SECURITY

73

graduate education – at least in the United States – is fairly narrowly focused, and so a willingness to bootstrap oneself up in these other fields is often a prerequisite. Lack of over-arching theory When we began our research, we discovered there was no over-arching theory linking what was happening with women to nation state-level outcomes. There were important insights to be found in philosophy and anthropology, such as in the work of Carole Pateman (1988) and Peggy Reeves Sanday (1981). There were also significant understandings to be found in the growing fields of feminist security studies and feminist international relations, such as the pioneering work of J. Ann Tickner (1993), Cynthia Enloe (1990), Christine Sylvester (1994), V. Spike Peterson (1992) and others. Linkages between the treatment of women and conflict have been established empirically by scholars such as Mary Caprioli (2000, 2003, 2005), Erik Melander (2005), Victor Asal et al. (2013, 2017), and others. In the fields of health and development, we find longstanding evidence of linkages between the treatment of women and outcomes, such as in the work of John Williamson (Williamson and Boehmer, 1997) and Ester Boserup (1970). In other words, we have the philosophical backdrop and scattered empirical pieces, but an explicit effort to gather these in a holistic theory-building enterprise has been needed. We have attempted to build that more over-arching theoretical framework ourselves, in the book The First Political Order: How Sex Shapes Governance and National Security Worldwide. We hope this volume will catalyze a new generation to develop their own theoretical frameworks, allowing them to ask questions that could not even be conceived of before, because we did not have the concepts or the propositions to do so.

What we could be asking If we could address these challenges, enabling us to move forward with a gender/demography/security research agenda, what fresh questions might we be able to ask, and answer? We believe that this research agenda would have profound implications for the policy community, not only for the academy.

74

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Policymaking could become more effective if we could answer questions such as: • Are we more effective in anticipating instability if we are tracking phenomena such as brideprice trajectories and prevalence of polygyny for the purposes of situational awareness? • Are we more effective in promoting stability if we examine the gender ideology of subnational actors before committing resources in their support? • Are we more effective in identifying agents of domestic instability if we are also tracking domestic household terror perpetration? • Are alliance patterns predicated on treatment of women in allying nations, and are alliances weaker when those patterns are mismatched? • Is the export of democracy less successful when women are more highly subordinated within society? • Will the toleration of enclaves of family and personal status law that subordinate women undermine the stability of a nation? • Would ending child marriage worldwide be more far more effective as an international stabilization initiative than more conventional foreign assistance programs? • What will be the effects on national and regional security of the almost 200 million missing women in Asia? • How will cross-border marriage migration patterns for the purposes of feeding the obstructed marriage markets of Asia affect both source and destination countries? • How are very high and very low fertility rates mirror images, as each are linked to the structuration of male/female relations within society? There are a whole host of subsidiary questions that could be asked and answered, as well. For example, are higher birth order young men more at risk for radicalization than first- or second-born sons? This would certainly make sense in brideprice societies where later-born sons may have to fend for themselves in terms of amassing resources for brideprice. Is the democratic peace a function of the treatment of women? That is, are conflicts better predicted using the degree of consonance of treatment of women rather than the level of democracy? With gender lenses attached to our security demographics viewpoint, we can see so much more than was possible previously.

SEX, DEMOGRAPHICS AND NATIONAL SECURITY

75

Conclusion After reading this chapter, it can be more easily understood why it is disconcerting to see discussion of demographics so often divorced from inquiry into the nature of relations between men and women within a society. The character of that relationship literally sets the stage for demographic sustainability or demographic disaster, with profound consequences for national security, stability and resilience. The situation, status and security of women must thus be regarded as critical variables in both demographic and security analysis. It is time for a new generation of scholars to move forward in identifying and describing the linkages between gender, demography and security. This research would not only be of academic interest, but would have real-world policy implications as well.3

Notes 1.

While we will not go into great detail in this chapter about the operationalization of this scale, it incorporates 11 phenomena which encode and signal these mechanisms of coercion and control: (1) prevalence of patrilocal marriage, (2) degree of inequity in family/personal status laws and practices favoring males, (3) prevalence of brideprice/dowry, (4) prevalence and legality of polygyny, (5) women’s age of marriage in law and practice, (6) prevalence and legality of cousin marriage, (7) women’s property and inheritance rights in law and practice, (8) son preference and sex ratio alteration, (9) overall level of violence against women, (10) degree of societal sanction for femicide, (11) legal exemption for rapists offering to marry their victims. This multivariate index is designed to capture the degree to which a first political order of women’s subordination is present within a given society. 2. There are also real national security effects of low, sub-replacement birth rates (see, for example, Nichiporuk, 2000; Nyce and Schieber, 2005; Stephen, 2011). Sub-replacement birth rates are also strongly linked to the situation of women; if women are in effect told they must choose between economic security and all it entails (such as significant personal autonomy and potential for meaningful public voice) and having a child, women may choose economic security. 3. This short chapter restates in a more focused form the broader argument posited in Hudson et al. (2020); certain sentences may be similar to those found in that larger work, and the book is cited when that happens. We acknowledge the US Department of Defense, which provided funds for our data collection and scalings through the US Army Research Laboratory and the US Army Research Office via the Department of Defense’s Minerva Research Initiative under grant number W911NF-14-1-0532. Hudson would also like to thank the Andrew Carnegie Corporation for their supportive fellowship in the social sciences, which allowed her release time to write.

76

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

References Ahram, A. I. (2015), ‘Sexual Violence and the Making of ISIS’, Survival, 57 (3), 57–78, https://​doi​.org/​10​.1080/​00396338​.2015​.1047251. Aleu, P. T. & P. Mach (2016, June 26), Bride-price Tradition is Destructive, but Strong, in Strife-torn South Sudan. Development and Cooperation, https://​www​.dandc​.eu/​en/​ article/​bride​-price​-tradition​-destructive​-strong​-strife​-torn​-south​-sudan. Asal, V., R. Legault, O. Szekeley & J. Wilkenfeld (2013), ‘Gender Ideologies and Forms of Contentious Mobilization in the Middle East’, Journal of Peace Research, 50 (3), 305–18. Asal, V., M. Schulzke & A. Pate (2017), ‘Why Do Some Organizations Kill While Others Do Not?: An Examination of Middle Eastern Organizations’, Foreign Policy Analysis, 13 (4) (October), 811–31. Baker, K. (2015, May 26), ‘ISIS offers fighters $1,500 starter home bonus and free honeymoon’, Mail Online, https://​www​.dailymail​.co​.uk/​news/​article​-3098235/​ISIS​ -offers​-fighters​-1​-500​-starter​-home​-bonus​-free​-honeymoon​.html. Betzig, L. L. (1986), Despotism and Differential Reproduction: A Darwinian View of History, Aldine Publishing. Boone, J. L. (1983), ‘Noble Family Structure and Expansionist Warfare in the Late Middle Ages: A Socioecological Approach’, in R. Dyson-Hudson & M. A. Little (eds), Rethinking Human Adaptation: Biological and Cultural Models, Westview Press, pp. 79–96. Boserup, E. (1970), Woman’s Role in Economic Development, Routledge. Brennan, S. (2017, February 6), ‘Couples prepare for a mass wedding in Syria to celebrate the ceasefire’, Mail Online, http://​www​.dailymail​.co​.uk/​~/​article​-4197474/​ index​.html. Bricker, N. Q. & M. C. Foley (2013), ‘The Effect of Youth Demographics on Violence: The Importance of the Labor Market’, Political Science, 7 (16), 179–94 Cameron, L. A., X. Meng & D. Zhang (2016), ‘China’s Sex Ratio and Crime: Behavioral Change or Financial Necessity?’, SSRN Electronic Journal, https://​doi​.org/​10​.2139/​ ssrn​.2737718. Caprioli, M. (2000), ‘Gendered Conflict’, Journal of Peace Research, 37 (1), 51–68. Caprioli, M. (2003), ‘Gender Equality and State Aggression: The Impact of Domestic Gender Equality on State First Use of Force’, International Interactions, 29 (3), 195–214. Caprioli, M. (2005), ‘Primed for Violence: The Role of Gender Inequality in Predicting Internal Conflict’, International Studies Quarterly, 49 (2), 161–78. Charrad, M. (2001), States and Women’s Rights: The Making of Postcolonial Tunisia, Algeria, and Morocco, University of California Press. Cincotta, R. (2011), ‘Africa’s Reluctant Fertility Transition’, Current History, 110 (736), 184–90. Criado-Perez, C. (2020), Invisible Women: Data Bias in a World Designed for Men, Abrams. Dreze, J. & R. Khera (2000), ‘Crime, Gender, and Society in India: Insights from Homicide Data’, Population and Development Review, 26 (2), 335–52. Edlund, L., H. Li, J. Yi & J. Zhang (2013), ‘Sex Ratios and Crime: Evidence from China’, Review of Economics and Statistics, 95 (5), 1520–34, https://​doi​.org/​10​.1162/​REST​_a​ _00356.

SEX, DEMOGRAPHICS AND NATIONAL SECURITY

77

Enloe, C. (1990), Bananas, Beaches, and Bases: Making Feminist Sense of International Politics, University of California Press. Gibney, M., L. Cornett, R. Wood, D. Arnon, A. Pisanò & B. Gray (2019), The Political Terror Scale 1976–2018, The Political Terror Scale, http://​www​.p ​ oliticalt​errorscale​ .org/​. Gleditsch, K. S., J. Wucherpfennig, S. Hug & K. G. Reigstad (2011), ‘Polygyny or Misogyny? Reexamining the “First Law of Intergroup Conflict”’, The Journal of Politics, 73 (1), 265–70, https://​doi​.org/​10​.1017/​s0022381610001003. Gottschall, J. (2008), The Rape of Troy: Evolution, Violence, and the World of Homer, Cambridge University Press. Harris, M. (1996), ‘The Evolution of Human Gender Hierarchies’, in B. D. Miller (ed.), Sex and Gender Hierarchies, reprint, Cambridge University Press, pp. 57–79. Henrich, J., R. Boyd & P. J. Richerson (2012), ‘The Puzzle of Monogamous Marriage’, Philosophical Transactions of the Royal Society B: Biological Sciences, 367 (1589), 657–69, https://​doi​.org/​10​.1098/​rstb​.2011​.0290. Hudson, V. M. & A. M. den Boer (2005), Bare Branches: The Security Implications of Asia’s Surplus Male Population, MIT Press. Hudson, V. M., D. L. Bowen & P. L. Nielsen (2020), The First Political Order: How Sex Shapes Governance and National Security Worldwide, Columbia University Press. Hudson, V. M. & K. B. Hodgson (2020), ‘Sex and Terror: Is the Subordination of Women Associated with the Use of Terror?’, Terrorism and Political Violence, 1–28, https://​doi​.org/​10​.1080/​09546553​.2020​.1724968. Hudson, V. M. & H. Matfess (2017), ‘In Plain Sight: The Neglected Linkage between Brideprice and Violent Conflict’, International Security, 42 (1), 7–40, https://​doi​.org/​ 10​.1162/​ISEC​_a​_00289. Kanazawa, S. (2009), ‘Evolutionary Psychological Foundations of Civil Wars’, The Journal of Politics, 71 (1), 25–34, https://​doi​.org/​10​.1017/​S0022381608090026. Korotayev, A. & D. Bondarenko (2000), ‘Polygyny and Democracy: A Cross-Cultural Comparison’, Cross-Cultural Research, 34 (2), 190–208, https://​doi​.org/​10​.1177/​ 106939710003400205. Krieger, T. & L. Renner (2018), ‘A Cautionary Tale on Polygyny, Conflict and Gender Inequality’ (Working Paper No. 2018–02), Diskussionsbeiträge, https://​ www​.econstor​.eu/​handle/​10419/​177367. Little, D. (1992), Understanding Peasant China: Case Studies in the Philosophy of Social Science, Yale University Press. Lu, C. (2012, May 3), Excess of Marriageable Males and Violent Crime in China and South Korea, 1970–2008, Population Association of America, San Francisco, California, http://​paa2012​.princeton​.edu/​abstracts/​121243. Matfess, H. (2016a, February 8), ‘Boko Haram Is Enslaving Women, Making Them Join the War’, Newsweek, https://​www​.newsweek​.com/​nigeria​-boko​-haram​-buhari​ -chibok​-girls​-424091. Matfess, H. (2016b, April 26), ‘Here’s Why So Many People Join Boko Haram, Despite Its Notorious Violence’, Washington Post, https://​www​.washingtonpost​.com/​ news/​monkey​-cage/​wp/​2016/​04/​26/​heres​-why​-so​-many​-people​-join​-boko​-haram​ -despite​-its​-notorious​-violence/​. McDermott, R. & J. Cowden (2015), ‘Polygyny and Violence Against Women’, Emory Law Journal, 64, 1767–814. Melander, E. (2005), ‘Gender Equality and Intrastate Armed Conflict’, International Studies Quarterly, 49 (4), 695–714.

78

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Mokuwa, E., M. Voors, E. Bulte & P. Richards (2011), ‘Peasant Grievance and Insurgency in Sierra Leone: Judicial Serfdom as a Driver of Conflict’, African Affairs, 110 (440), 339–66, https://​doi​.org/​10​.1093/​afraf/​adr019. Nasima (2014, July 11), The Herat Maternity Clinic, Afghan Women’s Writing Project, http://​awwproject​.org/​2014/​07/​the​-herat​-maternity​-clinic/​. Nichiporuk, B. (2000), The Security Dynamics of Demographic Factors, MR-1088-WFHF/ RF/DLPF/A, Rand. Nyce, S. A. and S. J. Schieber (2005), The Economic Implications of Aging Societies, Cambridge University Press. Pateman, C. (1988), The Sexual Contract, Stanford University Press. Peterson, V. S. (1992), Gendered States: Feminist (Re)Visions of International Relations Theory, Lynne Rienner Publishers. Potts, M. & T. Hayden (2008), Sex and War: How Biology Explains Warfare and Terrorism and Offers a Path to a Safer World, BenBella Books. Richmond, M. & F. Krause-Jackson (2011, July 25), ‘Cows-for-Bride Inflation Spurs Cattle Theft in South Sudan’, Bloomberg.com, https://​www​.bloomberg​.com/​news/​ articles/​2011​-07​-26/​cows​-for​-bride​-inflation​-spurs​-cattle​-theft​-among​-mundari​-in​ -south​-sudan. Rizvi, N., K. S. Khan & B. T. Shaikh (2014), ‘Gender: Shaping Personality, Lives and Health of Women in Pakistan’, BMC Women’s Health, 14 (1), 53, https://​doi​.org/​10​ .1186/​1472​-6874​-14​-53. Sanday, P. R. (1981), Female Power and Male Dominance: On the Origins of Sexual Inequality, Cambridge University Press. Singerman, D. (2007), The Economic Imperatives of Marriage: Emerging Practices and Identities Among Youth in the Middle East (SSRN Scholarly Paper ID 1087433), Social Science Research Network, https://​doi​.org/​10​.2139/​ssrn​.1087433. Sommers, M. & S. Schwartz (2011), Dowry and Division: Youth and State Building in South Sudan, Special No. 295, pp. 1–16, United States Institute of Peace, https://​www​ .usip​.org/​sites/​default/​files/​SR​_295​.pdf. SSP Classes | MIT Security Studies Program (SSP) (n.d.), https://​ssp​.mit​.edu/​about/​ subjects. Stephen, E. H. (2011), ‘Policy Concerns of Low Fertility for Military Planning in South Korea’, East-West Center, https://​www​.eastwestcenter​.org/​system/​tdf/​private/​ api102​.pdf​?file​=​1​&​type​=​node​&​id​=​33184. Sylvester, C. (1994), Feminist Theory and International Relations in a Post-Modern Era, Cambridge University Press. The Womanstats Project (n.d.), http://​www​.womanstats​.org/​. Tickner, J. A. (1993), Gender in International Relations: Feminist Perspectives on Achieving Global Security, Columbia University Press. Tucker, W. (2014), Marriage and Civilization: How Monogamy Made Us Human, Regnery Publishing, Inc. Williamson, J. & U. Boehmer (1997), ‘Female Life Expectancy, Gender Stratification, Health Status, and Level of Economic Development: A Cross-National Study of Less Developed Countries’, Social Science and Medicine, 45 (2), 305–17. WomanStats Maps (n.d.), http://​www​.womanstats​.org/​maps​.html. Wright, R. E. (1995), The Moral Animal: Evolutionary Psychology and Everyday Life (Vintage Books 1st edition), Vintage Books, a division of Random House, Inc. Zarghona (2014, July 10), Whatever Men Want Is What Happens, Afghan Women’s Writing Project, http://​awwproject​.org/​2014/​07/​the​-herat​-maternity​-clinic/​.

PART II Implications of demographic change: context and connections

6

Whiteshift: demographic change, populism and polarization in the West

Eric Kaufmann

Introduction There has been a dramatic increase in support for what Eatwell and Goodwin (2018) term national populism in the West, that is, a form of right-wing populism focusing on immigration and national identity issues. This is distinct from both left-wing populism, oriented toward economic disparities, or a religion-focused right populism, seeking religious revival and moral traditionalism. While the latter is not entirely absent from populist right discourse, it is a minor theme, and in some cases is entirely absent. The 2010s national populist surge began when the UK Independence Party, Danish People’s Party and French Front National won nearly 30 percent of the vote in the 2014 European Union parliament elections. The European migrant crisis peaked the following year – some 2 million people, many fleeing Syria’s civil war, entered the EU. During the migrant crisis, immigration shot up the list of Europeans’ electoral priorities, and support for a number of populist right parties – notably Germany’s Alternative for Germany (AfD) and the Sweden Democrats – soared. As the Eastern Mediterranean route closed due to an EU–Turkey deal to restrict crossings, numbers began rising in the western Mediterranean. In response, immigration concerns rose in Italy, and later in Spain. Italy’s Lega won 17.4 percent in the 2018 Italian elections, reaching 34 percent in the 2019 European elections. Moreover, Vox, a new Spanish populist party, won 15 percent of the vote in the November 2019 Spanish election. In the 2019 European elections, populist right parties did even better than in 2014.

81

82

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Meanwhile, Britain voted to leave the European Union in June 2016, a referendum in which immigration was the central issue for Leave voters (Clarke et al. 2017). In early 2016, Donald Trump, an outsider to the party establishment, won the US Republican primary, the only one of 17 primary candidates to make immigration central to his pitch. The same anti-immigration message helped him win the presidential election in November of that year. In New Zealand, a coalition between the Labor Party and the populist right New Zealand First came to power in 2017. One of their key planks was a pledge to halve immigration. Finally, in the 2019 Quebec provincial elections, the populist CAQ won a majority, promising to enact a historic cut in Quebec’s immigration intake. While the pro-immigration Justin Trudeau won the Canadian federal election in the same year, the country’s Conservative and Liberal voters had never been as polarized over questions of immigration and multiculturalism (‘Immigration’ 2019). Why is this happening now? A glance at the data shows that, in most cases, increasing immigration led to immigration rising up the list of voters’ election priorities, which created demand for the populist right and boosted their support. The connection between population change and politics is not automatic – ideology, media and national identity mediate it – but population dynamics are a key factor. More broadly, perceptions of demographic shifts have played a major role in world history, whether expressed as a fear of rising powers (such as Germany or the USSR) or of rising minority groups within countries, such as Catholics in Northern Ireland or Muslims in Côte d’Ivoire (Morland 2019; Nordås 2012). As the chapters in this volume show, the study of political demography is an attempt to understand the role of population change in politics. The focus is on the uneven impact of the demographic transition between different politically relevant units, whether civilizations, countries or ethnic groups. A major branch of political demography concerns demographic disparities between ethnic groups within countries, known as differential ethnic population growth (Teitelbaum 1998). In many countries, power is at least partly contested along ethnic lines. In democracies, where numbers translate more readily into power, the competition between ethnic groups, and especially their political representatives, to ‘win the census’ shows the importance of demographic change (Horowitz 1985: 194–6). In Lebanon, the plunging Christian share of the population led to the removal of a question about religion on the census in order to preserve a political settlement premised on a Christian majority. But even in the United States, politically charged questions have been raised over whether to charac-

WHITESHIFT

83

terize Hispanics as white, to create a separate category for Middle Easterners, and, more recently, to include a census question on citizenship status (Wang 2019). This volume contends that the study of population change is often overlooked in political science. In this chapter, I show how population shifts – and the heightened threat perceptions these give rise to – are major forces shaping contemporary politics. Those who study right-wing populism or ethnic conflict often ignore population dynamics. Even those who consider the role of immigration for populism fail to situate this discussion in the wider literature on differential ethnic population growth in particular, and political demography more generally. With demographic and economic disparities set to increase in the decades ahead, more attention needs to be paid to the way migration and ethnic change are processed by host societies. How rapid and culturally distinct are migration flows? To what extent are ethnic majorities and national identities absorptive or exclusive? How strong are cross-cutting, competing issues, which may distract political systems from focusing on problems of ethnic change? These are some of the questions that will need to be asked to better understand the politics of developed and developing societies in the twenty-first century.

Political demography: theory and group boundaries There is often a considerable subjective component to categorizing human groups, and boundaries between groups may be fuzzy and subject to social reconstruction. This includes ethnic categories – for example, at different times Germans have become Czechs and Czechs have assimilated into Germans (Deutsch 1966 [1953]). The Pashtun of Afghanistan and Pakistan are a classic case, with the Baluchi–Pashtun line based on criteria that differ by region (Barth 1969). At other moments, whole categories spring up, reducing the salience of older boundaries. The emergence of unhyphenated white Americans out of a social milieu where the Protestant–Catholic–Jewish divide was central, is one example (Alba 1990). Despite these examples of fluidity, proscriptions against intermarriage are often strong, especially where religion serves as an ethnic boundary marker and there is a history of violence. In Northern Ireland, there is little inter-ethnic marriage between Irish Catholics and British-Unionist Protestants; in India, Hindu–Muslim marriage is rare. Sometimes race is strongly tied to ethnic boundaries. In South Africa (as in the pre-civil rights US South), interracial

84

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

marriage is uncommon. In addition, Andreas Wimmer argues that matrilineal descent lines produce less rigid ethnic boundaries than groups where descent is traced through the paternal line (Wimmer 2008). In sum, ethnic changes in tight-bounded societies are more likely to result in tensions than in places where assimilation or boundary redefinition overwhelms demographic change. The latter is no guarantee of comity: intermarriage was common prior to violent conflict in the former Yugoslavia and in many sub-Saharan African countries that have experienced ethnic violence. Yet fluidity does help reduce frictions arising from demographic change. Theories of nationalism and ethnicity help frame these dynamics. One view, constructionism (sometimes referred to as modernism or instrumentalism), is that boundaries are fluid and under the sway of political or economic actors. The opposing view, ethnosymbolism, is that boundaries are relatively fixed, especially at any point in time, anchored by cultural understandings and traditions of where boundaries lie that change very slowly over time (Smith 1998). Primordialists go further, claiming that perceptions of difference are rooted in genetic distance and evolutionary psychology, though most allow that ethnic boundaries are set by social and political forces (van den Berghe 1979). The two main theoretical camps disagree on the role of demographic change in ethnic conflict. Constructionist approaches suggest that demographic change itself has a limited effect. Demographics are used instrumentally by political actors who are motivated by a desire for wealth, power or status. Alarm at demographic change is therefore something that is manufactured by the media and politicians, bearing little relationship to demographic reality. Ethnosymbolists would aver that demographic change does in fact matter: group boundaries are rooted in tradition and change relatively slowly. Thus demographic change translates fairly readily into perceptions of loss or gain. This gives rise to anxiety among the losing group, and thence to demographic engineering techniques such as pronatalism, territorial demarcation, immigration restriction or harsher methods including ethnic cleansing and genocide (Morland 2016).

Geographic differences in national populism The preceding theories try to help us understand the political and social consequences of demographic changes described elsewhere in this volume. Global migration is not simply a matter of material factors like demography, technology and economics. If it were, East Asia would likely be experiencing the same

WHITESHIFT

85

ethnic transformation – what David Coleman terms the ‘third demographic transition’ in which ethnic majorities transition to becoming minorities – as the West (Coleman 2006). Yet, while the foreign-born share of the USA and major western European countries is in the 10–20 percent range, and over 20 percent for Canada, New Zealand, Australia and Switzerland, the corresponding figure is just 1.75 percent in Japan and 3.4 percent in South Korea. This difference is even more marked when we compare the share of citizens that are foreign-born across the eastern and western parts of the developed world. Japan, for instance, offers five-year work permits and does not confer Japanese citizenship. While Singapore and the Gulf Arab states have large foreign-born populations, these generally do not gain access to citizenship and, by extension, membership in the national community. These politico-demographic facts largely explain why we do not see right-wing populism and cultural polarization in East Asia but find it in the West. Eastern Europe, which has few immigrants but strong populism, may appear anomalous, but comports with the wider pattern when bearing in mind that it is only East European states inside the European Union, and thus exposed to the ‘threat’ of migration and diversity, that are experiencing anti-immigration populism. This represents a larger-scale operation of the ‘halo effect’ in which relatively homogeneous places bordering on diverse ones respond with a heightened threat response (Rydgren and Ruth 2013).

The liberal revolution Nations, especially developed ones, have high levels of control over demographic shifts. Thus political demographers must be able to explain why some countries open up to change while others don’t. Political culture – the dominant ideology of a state – is an important variable here. Liberalism in the West emerged in a context of low migration, but has altered its form in response to demographic pressure from without. It therefore makes sense to draw attention to an important evolution of liberal ideas, in which ideas of liberty and equality come to be extended to non-nationals, and conceptions of nationhood recast from a model based on ethnic homogeneity to one based on diversity. The origins of these ideas arguably lie in the early twentieth-century United States, whose northern cities had become majority Catholic and Jewish due to large-scale non-traditional immigration in the 1895–1925 period. Liberal Progressives like John Dewey, Israel Zangwill and Jane Addams, in the 1905–10 period, urged the country to embrace this new diversity and to con-

86

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

struct a futuristic polyglot nation. Though the idea of America as a trans-ethnic melting-pot goes back to the eighteenth century, this always coexisted with an ethno-cultural understanding that the country was simultaneously an Anglo-Protestant construct (Kaufmann 2004). Central to this outlook was an opposition to the restrictive immigration laws of 1917, 1924 and 1952, and to the prohibition of the sale of alcohol from 1920 to 1933, measures supported by predominantly rural or working-class Protestant Americans. In the late 1960s, what I term ‘left-modernist’ cosmopolitanism expanded amongst a new knowledge class and via a rising strata of educated youth. These processes continued through the 2010s. Those animated by this outlook were disproportionately located in elite institutions like the media, universities, courts and the arts and culture industries. Once a critical mass of left-modernist decision-makers was in place, they could exercise moral leadership and policy change through an elite institutionalization strategy (Kaufmann 2004). Liberalism is thus a key enabler of politico-demographic change. An early expression of the new cosmopolitan-liberal outlook was the removal of the nationally selective, and in some cases racially exclusive, immigration systems of the USA, Canada and Australia. In the USA this took the form of the Hart–Celler Act of 1965 which abandoned a quota system based on nations’ share of the US 1920 population. In Canada, there was a gradual removal of preference for British immigration after 1962, and in Australia an end to the so-called ‘White Australia’ policy in 1966. These moves were not undertaken with the intent to remake the ethnic composition of western populations, as far right ‘replacement’ ideologues accuse. Rather, as Robert Kennedy’s testimony during deliberations over Hart–Celler makes clear, they expected family sponsorship provisions and established immigration patterns to keep the nation’s ethnic composition from changing whilst eliminating de jure ethnic preferences in policy. Anglo settler societies opened their immigration regimes to high-volume, culturally non-selective inflows in the 1960s, and Europe followed suit soon after. Moral suasion and sensitivities over accusations of racism prompted guestworker programs in northern and western Europe to transform into permanent settlement programs, which in turn became the basis for ‘nation of immigrants’-style multiculturalism (Joppke 1998). In Europe and America, in the 1980s and beyond, non-specific international human rights law concerning terms such as ‘non-discrimination’ were reinterpreted by Western courts to expand the definition of refugee, and the protections afforded to claimants and overstayers. Taboos made discussion of numerical restriction difficult due to

WHITESHIFT

87

the association between immigrants and nonwhites in a normative climate of heightened sensitivity to minorities. Together, the liberal-cosmopolitan drift of political norms made ethnic selection, and even numerical restriction, difficult to impossible despite public opinion (Freeman 1994). Britain, which technically allowed all imperial subjects to immigrate during the days of the British Empire, was suddenly faced with this potential becoming reality due to cheaper transport, the lower cost of imported labour and population growth in overseas possessions. In response, it restricted Commonwealth immigration – at the time mainly from the Caribbean and Indian subcontinent – in 1961 and led to comparatively low immigration from the late 1960s through the late 1990s. This is a departure from the pattern in North America and Australia, but is ultimately less of an exception to the liberal rule than one might presume. The ‘cultural turn’ of the left gradually began to break through the country’s class-dominated political discourse, embedding itself in the modernizing New Labour project in the 1990s (West 2013). This ‘cultural turn’ sought to take the party beyond its working-class trade union base to appeal to a rising segment of well-educated, capitalist-friendly metropolitan liberals. As soon as Tony Blair’s New Labour government took office, net immigration rapidly expanded from its historic limit of 50,000 per year to 150,000, approaching 300,000 during Blair’s second term. The well-established association between racism and immigration restriction made it difficult to criticize the surge of refugee claimants, with successive conservative leaders like William Hague and Michael Howard routinely attacked as racist for raising the subject of ‘bogus asylum seekers’ (BBC 2000). This again shows the power of the new liberalism to open a country to demographic changes that would be politically consequential. Labour acted precipitously in 2004, making Britain one of just three countries to waive all restrictions on migration from the A8 East European accession states that newly gained membership to the European Union. This resulted in 1.5 million new immigrants by 2010, nearly ten times the number the government predicted (West 2013). The arrival of large numbers of white Eastern European immigrants muddied the antiracism–immigration equation sufficiently for politicians to address it: first the extreme right British National Party in the mid-2000s and then the populist UK Independence Party in the later 2000s. The Tories under David Cameron, sensing pressure on their right flank, promised to cap numbers at the ‘tens of thousands’ in the 2010 election, winning that contest decisively over Labour. Despite the European increase, non-EU immigration was higher than EU migration in all years up to the Brexit referendum in 2016. The non-European

88

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

share of the population of England and Wales more than doubled between 2001 and 2011, from 3 million to 8 million (7–14 percent of the total). In combination, EU and non-EU immigration levels were by far the most important reason people gave for voting to leave the European Union in June 2016 (Clarke et al. 2017). The British case, like that of Europe more generally, illustrates how actual demographic change through immigration had serious effects on politics. Demographic change interacted with a pre-existing political divide over membership of the European Union. This Euroskeptic political grievance was invigorated by cultural threats arising from large-scale immigration, resulting in Brexit. Here we see the theoretical importance of marrying demographic and political perspectives, and how political demography often works through the interaction effects between demographic and political (or economic) variables.

Immigration politics outside the West Ethnosymbolist and instrumentalist theories can also explain the political demography of immigration outside the West. Some point to the relative openness of Latin American countries to Asian immigration during periods of restriction in North America, though the pull of Latin America was considerably weaker and numbers never reached a high level (FitzGerald and Cook-Martin 2014). More recently, Afro-Creole and anglophone Belize and Antigua have accepted significant Latin American immigration from Guatemala. Brazil and Colombia have accommodated Venezuelan refugees from the Maduro regime and Turkey has sheltered over 3 million Syrians within its borders. Even so, Latin American countries such as Chile and the Dominican Republic have taken a much tougher approach to more culturally distinct migrants from Haiti (Laing and Miranda 2018). It seems that, in many cases, tolerance for large-scale immigration or asylum-seeking extends only to immigrants of a similar culture and religion (i.e. ‘white’ Venezuelans in Colombia and Brazil, or Sunni Syrians in Turkey). In Africa, pan-African sentiment helped legitimate a generous immigration policy in countries such as South Africa. It also underlay Kenya and Tanzania’s welcoming approach to refugees from Somalia, Sudan and elsewhere. However, since the 1990s, pan-African sentiment has waned and the number of refugees has risen. Anti-immigration sentiment has become endemic in South Africa, leading to periodic pogroms against immigrant groups (Polakow-Suransky 2017: 246–7). In Kenya and Tanzania, the pan-Africanism that originally permitted refugees free movement has given way to a containment strategy in which refugees are

WHITESHIFT

89

barred, or not permitted to leave their camps without a permit (Maple 2016: 11; Rutinwa 1999: 2; see also Daley 2013). In addition to these examples, there is an established pattern of ‘Sons of the Soil’ migration from relatively advanced, densely populated parts of countries to their ethnic peripheries. Han Chinese became the majority in Xinjiang in the 1990s and have also migrated to Tibet. Javanese have migrated to West Papua, Kikuyu from coastal Kenya have moved inland to the (ethnic Luo and Kalenjin-dominated) Rift Valley, and Sinhalese have migrated to Tamil areas in the north and east of the country (Côté et al. 2018). These flows typically result in resentment or even violent resistance from local majorities, sometimes contributing to civil war, as in Côte d’Ivoire (Nordås 2012). Many of the dynamics found in South–North migration seem to be replicated in the South–South migration of ethnically distinct peoples across ethnic or national borders. Given rapid population growth in the developing world, and the fact most migration is intra- rather than inter-continental, national populism in the global South could eventually become more important than dynamics in the global North.

The decline of white majorities A central focus of this book is to foreground demographic developments in the study of politics. One of the political-demographic ‘megatrends’ noted by political demographer Jack Goldstone is ethnic change in the West, arising through immigration (Goldstone 2012). Though developments in the global North have their counterparts in the global South, the big change in the past 50 years has been the rise in South-to-North long-distance migration. Large Western societies remain important due to their economic clout, military power and cultural ‘soft’ power (even if these advantages are waning). They are also central to the post-2014 ‘populist moment’ which is reconfiguring the basis of politics and may even be altering alliance behavior along ‘nationalist-globalist’ lines. One of the most important demographic-perceptual dynamics in the West today concerns the rise of the non-European share of the population, which encompasses, but is not restricted to, concern over Islam. Research shows that minority, immigrant and Muslim share is generally overestimated by a factor of two or three – in some cases fivefold (Ipsos MORI 2018). Part of this reflects the fact that minorities stand out more and hence draw more attention from voters – including from minorities, who overestimate immigrant share

90

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

at higher rates than whites (Alba et al. 2005; Herda 2019). However, those who favor lower levels of immigration or vote for populists also overestimate minority share, following a more general observation in public opinion research that people tend to overstate things they care deeply about (crime, for example). This comports with constructionist theories that emphasize perceptions over demographic realities, but the story is not so simple. On the ethnosymbolist side of the ledge, there is an important distinction between concern over levels and worry over changes. People tend to overestimate minority levels, as noted, suggesting a role of social construction. Yet evidence is less clear that people misperceive changes over time: concern over immigration tends to correlate with actual shifts in the quantity of immigration over time, suggesting that people are able to accurately gauge shifts (Kaufmann 2017). In empirical work, changes in immigration almost always result in greater opposition to immigration and higher populist right voting, at all geographic scales. However, higher minority levels – especially where well-established for over a decade – exhibit a nonlinear effect that is moderated by geographic scale. That is, at the local level, a high minority presence can lead to more interethnic contact, reducing majority-group concern. Whites in diverse urban neighbourhoods are typically more tolerant than those in rural areas or exurbs. However, at the city, state or national scale, higher minority levels seem – all else equal – to be associated with heightened majority concern (Kaufmann and Goodwin 2018). Demographic indicators that capture both size and change in the Muslim share are especially useful in predicting the potential demand for the populist right. Figure 6.1, for example, plots the highest populist right poll or vote in a West European country against Pew’s medium projection for Muslim share in 2050. The correlation is 0.47, rising to 0.59 when we consider Muslim share in 2030. Populist right parties are highly volatile because their brands tend to be weak, not well established or institutionalized, and thus centered around a charismatic leader. Leadership succession and factionalism can lead the party to experience a sudden loss of popularity, but this is deceptive. The emergence of a viable leader or an immigration crisis can boost its support equally quickly. Too often, observers have counted a party like the Austrian Freedom Party out when it loses a popular leader. It is more useful to look at high water-marks as they give a sense of the strength of demand in the electorate. While the foregoing would suggest an ethnosymbolist demography-to-politics causal path, constructionism is also important, especially in Eastern Europe, which has little immigrant diversity but strong national populism. Yet I main-

WHITESHIFT

91

Source: Election results and Pew Forum on Religion in Public Life 2011.

Figure 6.1

Highest populist right poll vs medium Muslim share projection

tain that it is erroneous to include Eastern Europe, which has few Muslims, in this analysis. Eastern European countries inside the EU (not outside it), tend to manifest a ‘halo effect’ in which the presence of diversity and free movement is close enough to unsettle identities without bringing people into direct contact with Muslims so as to allay unreasonable fears. The term ‘halo effect’ stems from work on the local dynamics of populist right voting in which Rydgren and Ruth (2013) show that the highest local threat perceptions occur in relatively white ‘halo’ areas that immediately surround diversity, such as exurbs. Examples include the outer suburbs of cities as variegated as Amsterdam, Montreal, Stockholm and London. Donald Trump performed exceedingly well in the exurban belt of small communities in major US metropolitan areas (Orfield 2017). East European EU nations like Hungary, Poland or Slovakia also tend to be more conservative on a range of social issues because of their history of subordination and especially because the liberal value shifts of the 1960s – notably on race and immigration – largely bypassed the communist world. Immigrant share is correlated with Alternative for Germany (AfD) voting in West Germany but not in the East, where the Eastern European pattern of cultural conservatism interacts with the halo effect to boost the populist right (Schwander and Manow 2017: 12).

92

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Highly homogeneous nations tend to be xenophobic, a nonlinear effect. Initial fears pass when minority share rises above a low threshold of 1–2 percent minority, leading to habituation and comfort. However, as minority levels rise above 4–5 percent minority, concern appears to grow once more. We see this in Europe, where low and high national Muslim share both predict a more restrictionist value public than is true when a country has a moderately low share of Muslims, i.e. in the 2–4 percent range (Kaufmann 2018: 253). More diverse settings appear to result in more attitudinal polarization as liberal- and conservative-minded voters respond in diametrically opposite ways to ethnic diversity. A good example of this is the way attitudes to illegal immigration in the United States polarize by political party the closer one gets to the US border. Higher proportional Latino increase in a county is correlated with a larger opinion divide between Republican and Democratic voters over whether a path to citizenship or deportation is the best way to deal with the undocumented. At the border, partisans are nearly 50 points apart on whether the undocumented should be allowed to stay in the United States, while 2,000 kilometers to the north there is no partisan difference (Gravelle 2016).

Immigration attitudes or salience? One major finding in the literature is that more rapid immigration tends not to alter attitudes to immigration very much. A person’s views on whether immigration should be reduced, increased or stay the same are given by their psychology and ideology, as well as party affiliation. The increased social liberalism of the post-1960s West contributed to a liberalization of attitudes in immigrant-receiving countries like the United States, Canada and Australia. Thus, even as unprecedented levels of non-European immigration entered these countries (initially, more so North America), attitudes did not become more hostile. As the Baby Boom generation increased its presence in the electorate in the 1990s, immigration opinion was softening, even as higher diversity was stimulating threat. The two forces largely offset each other in America, but from the mid-1990s, liberalization was the order of the day. Based on an index of 72 indicators in US social surveys, liberalization of attitudes among whites took place exclusively among Democrats rather than Republicans (Goldberg 2020). Yet it is one thing to want less immigration, as a majority of Europeans and near-majority of North Americans do, and quite another to say immigration is the most important issue facing one’s country. This latter measure, known

WHITESHIFT

93

as immigration salience, is what creates the demand conditions for national populism in the West. Figure 6.2 shows how in Britain, the sharp rise in immigration after Tony Blair’s Labour government took office in 1997 broke with a longstanding period, since the late 1960s, of low net migration in a range between zero and 50,000. As immigration rose, the number of media stories increased and the share of Britons saying immigration was the top issue facing the country (i.e. immigration salience) climbed along with it.

Source: Ipsos MORI 2015: 5.

Figure 6.2

News stories about immigration, those mentioning immigration as an issue and net migration over time

In Europe, we see a very similar pattern. Immigration into Europe began to rise sharply in 2014, peaking in late 2015 with the migrant crisis of asylum seekers from the Syrian conflict and elsewhere, and national populism rose along with the flows. One study showed that in 9 of 10 major West European countries between 2005 and 2016, immigration, immigration salience and the populist right vote were significantly correlated. Nevertheless, demography, which affects the demand side for national populism, is not the only story. Political supply is also important, and here politicians and, to an extent, the media, can frame perceptions of ethno-demographic change and immigration. In the United States, rapid growth in the illegal immigrant population to around 12 million, and a quick increase in ethnic diversity due both to immigration and a considerably higher Hispanic total

94

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

fertility rate (2.84 compared to 2.01 for whites in the early 2000s), created the demand conditions for a politics of immigration as early as the 1980s. Even as net migration from Mexico largely ceased and Hispanic fertility converged toward that of the white ‘Anglo’ majority by the late 2010s, the demographic momentum and cultural response from the post-1990s period shaped attitudes on the right (Valentino et al. 2013). While immigration was an important issue for whites in specific locales (South Florida, Southern California), as exemplified by the 1994 Proposition 187 vote in California denying services to illegal immigrants, the issue did not gain traction at the federal level until the 2010s. This was mainly due to political supply constraints on the right. In both the right-wing media and in the Republican party, a universalist brand of conservatism blended conservative family values, libertarian economics and a muscular foreign policy together with a liberal approach to immigration and an appeal to Latino voters. This ‘neoconservative’ blend was, however, challenged by an older ‘paleoconservative’ tradition focused on conserving white, Anglo-Protestant ethnic traditions. This first emerged in Pat Buchanan’s relatively successful 1992 and 1996 primary campaigns, in which Buchanan pitched a tough message about border control. It subsequently gained ground with both Proposition 187 in California and local anti-immigration ordinances, beginning in the mid-2000s and later including Arizona’s Senate Bill 1070 of 2010. In 2014, a bipartisan immigration reform bill was scuttled by grassroots restrictionist pressure, and House Majority Leader Eric Cantor, a pro-immigration figure within the Republican establishment, was defeated in a local Republican primary. In his place rose upstart Dave Brat, a relative unknown whose attacks on Cantor’s immigration credentials helped win him the seat. In the middle of that year, a wave of tens of thousands of Central American mothers and children came across the border seeking asylum, raising the profile of the immigration issue and demonstrating how demographic changes can shape politics. So began a new chapter in US immigration history in which the issue came to be ranked as the top priority by 10–20 percent of Republican voters, an unprecedented development since records began in the 1930s. This set the stage for Donald Trump’s successful 2015 Republican primary bid. In his opening speech announcing his new campaign, on June 16, 2015, Trump broke a number of taboos that had been observed by both the Republicans and the right-wing media to that point. The first was to campaign explicitly on immigration, which none of the other 16 primary contenders did. The second was to use direct, sometimes racist, language: “When Mexico sends its people”, he

WHITESHIFT

95

began, “they’re not sending their best … They’re rapists. And some, I assume, are fine people” (Washington Post 2015). In November 2016, Trump stunned by winning the presidential election in a campaign characterized by blunt language on immigration. The immigration issue was the most statistically significant predictor of both Trump support in the Republican primary and vote-switching in the presidential election. It was central to explaining the shift of a number of blue-collar white Obama Democrats to the Republican banner – a key constituency of swing voters in battleground states in the Midwest such as Wisconsin and Michigan, with their large working-class white electorates. A number of tough immigration measures followed once Trump took office, including a travel ban targeting a number of Muslim-majority countries and a sharp cut in refugee admissions. Among white Americans, immigration opinion split dramatically along partisan lines between 2012 and 2016, a major change from previous decades, in which Republican and Democratic opinion on the question was relatively similar and the issue was not politicized to a major degree. In liberal societies, ethno-demographic change frequently leads to partisan divisions over immigration. This shifts the basis of politics away from more materialist left–right divisions to cultural cleavages based on attitudes to ethno-cultural change, immigration and national identity. The American case shows that demographic change is important, but that ethnic change and immigration rates do not always translate immediately into restrictionist populism. In Western Europe, by contrast, the relationship between migration and populist support was closer, perhaps because of the multiplicity of parties (aside from Britain) and diversity of media outlets (as in Britain) which served as a conduit for local discontent to break into the national conversation. In the USA, the two party system and the sway of neoconservatism within much of the right-wing media helped deflect opinion toward other issues prior to 2015. Political supply factors, including politicians and the media, therefore have some leeway in directing attention toward, or away from, demographic shifts, and framing how voters should respond.

Conclusion The unevenness of the demographic transition, both between and within countries, is altering global ethnic makeup. In Western nation states, the combination of relative wealth, demographic maturity and a cresting of post-1960s liberal-egalitarianism has created the conditions for rapid ethnic change. East

96

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Asia and East Europe are relatively rich and demographically mature, but have much weaker liberal-cosmopolitan movements, and are thereby opting for a path of population stagnation which preserves their nation’s historic ethnic composition. In the global South, state capacity is weaker, but a number of countries have acted to restrict or contain immigration and refugee flows after an initial period. Migration from co-ethnic sources, or from religiously similar populations, as in Turkey with Syrians, or in Colombia and Brazil with Venezuelan refugees, is considered more acceptable than inflows from more culturally distant groups, as with Haitians in Latin America. Liberalism interacts with demography in important ways in the West. First, liberalism has opened the doors of Western countries to larger inflows of culturally distinct immigrants than is the case in many other wealthy parts of the world. Second, liberal norms constrained mainstream parties and the media from responding to popular concerns over ethnic change. This process created unmet demand, which opened the way for the rise of populist parties or leaders. In many cases, mainstream parties have accommodated populists, which has eroded populist support but also led to liberal-cosmopolitan blowback. The result is polarization, partly along ethnic lines between white majorities and minorities, but mainly within the white population, between liberals and conservatives. Thus, demographic change is shifting the dominant cleavage of politics in Western countries from left–right economics to the ‘globalist–nationalist’ or ‘open–closed’ cultural dimension. This has even gained an international dimension, with nationalist actors in the West forming alliances with East European (or even Brazilian and Indian) counterparts as well as Putin’s Russia (Teitelbaum 2020). Meanwhile, globalist-liberals in Western countries forge connections with liberals in Eastern Europe and Asia. The rise of national populism is connected to immigration and ethnic change, what David Coleman terms the ‘third demographic transition’. Political supply factors mediate the relationship between demographic change and political response. This can take the form of discursive taboos or alternative framings which draw attention away from, or toward, immigration-led ethnic change. The fluidity or fixedness of group boundaries also affects how consequential demographic changes are. Having said this, demographic shifts act as conditioning factors, creating fertile soil on the demand side which make it increasingly likely that populist entrepreneurs can break through. Like most political-demographic phenomena, immigration and ethnic change tend to affect politics in interaction with other social variables, in second- or third-order ways, to produce political effects (Hout 2006; Kaufmann and Toft 2012: 9).

WHITESHIFT

97

References Alba, R. D. (1990), Ethnic Identity: The Transformation of White America. New Haven, CT: Yale University Press. Alba, R. et al. (2005), ‘A distorted nation: perceptions of racial/ethnic group sizes and attitudes toward immigrants and other minorities’, Social Forces 84 (2), 901–19. Barth, F. (1969), ‘Introduction’, in Ethnic Groups and Boundaries: The Social Organisation of Culture Difference, London: Allen & Unwin. BBC (2000), ‘Hague fuels asylum row’, BBC News, Apr. 30. Clarke, H. D. et al. (2017), Brexit: Why Britain Voted to Leave the European Union, Cambridge: Cambridge University Press. Coleman, D. (2006), ‘Immigration and ethnic change in low‐fertility countries: a third demographic transition’, Population and Development Review 32 (3), 401–46. Côté, I. et al. (2018), People Changing Places: New Perspectives on Demography, Migration, Conflict, and the State, London: Taylor & Francis. Daley, P. (2013), ‘Refugees, IDPs and citizenship rights: the perils of humanitarianism in the African Great Lakes region’, Third World Quarterly 34 (5), 893–912. Deutsch, K. (1966 [1953]), Nationalism and Social Communication: An Inquiry into the Foundations of Nationality, 2nd edition, Cambridge, MA: MIT Press. Eatwell, R. and M. Goodwin (2018), National Populism: The Revolt against Liberal Democracy, London: Penguin UK. FitzGerald, D. and D. Cook-Martín (2014), Culling the Masses: The Democratic Origins of Racist Immigration Policy in the Americas, Cambridge, MA: Harvard University Press. Freeman, G. P. (1994), ‘Can liberal states control unwanted migration?’, The ANNALS of the American Academy of Political and Social Science 534 (1), 17–30. Goldberg, Zach (2020), Georgia State University, Twitter, Mar. 22. Goldstone, J. (2012), ‘A theory of political demography: human and institutional reproduction’, in J. A. Goldstone, Eric Kaufmann and Monica Duffy Toft (eds), Political Demography: Identity, Conflict and Institutions, Oxford: Oxford University Press, pp. 10–30. Gravelle, T. B. (2016), ‘Party identification, contact, contexts, and public attitudes toward illegal immigration’, Public Opinion Quarterly 80 (1), 1–25. Herda, D. (2019), ‘Tracking ignorance: examining changes in immigrant population innumeracy in the United States from 2005 to 2013’, Migration Letters 16 (2), 329–39. Horowitz, D. (1985), Ethnic Groups in Conflict, Berkeley: University of California Press. Hout, M. (2006), ‘The demographic imperative in social change: political implications political demography: ethnic, national and religious dimensions’, paper presented at the London School of Economics. ‘Immigration: Half back current targets, but colossal misperceptions, pushback over refugees, cloud debate’, Angus Reid Institute, Oct 7, 2019, http://​angusreid​.org/​ election​-2019​-immigration/​. Ipsos MORI (2015), ‘Shifting ground’, London: Ipsos MORI. Ipsos MORI (2018), ‘The perils of perception’, Ipsos MORI, https://​www​.ipsos​.com/​ ipsos​-mori/​en​-uk/​perils​-perception​-2018. Joppke, C. (1998), ‘Why liberal states accept unwanted immigration’, World Politics 50 (2), 266–93.

98

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Kaufmann, E. (2004), The Rise and Fall of Anglo-America: The Decline of Dominant Ethnicity in the United States, Cambridge, MA: Harvard University Press. Kaufmann, E. (2017), ‘Levels or changes?: ethnic context, immigration and the UK Independence Party vote’, Electoral Studies 48, 57–69. Kaufmann, E. (2018), Whiteshift: Populism, Immigration and the Future of White Majorities, London: Penguin UK. Kaufmann, E. and M. J. Goodwin (2018), ‘The diversity wave: a meta-analysis of the native-born white response to ethnic diversity’, Social Science Research 76, 120–31. Kaufmann, E. and M. D. Toft (2012), ‘Introduction’, in J. A. Goldstone, Eric Kaufmann and Monica Duffy Toft (eds), Political Demography: How Population Changes are Reshaping National Politics and International Security, Oxford: Oxford University Press, pp. 3–9. Laing, Aislinn and Natalia A. Ramos Miranda (2018), ‘Chile sends 176 Haitian migrants home on criticized ‘humanitarian flight’’, Reuters, Nov. 7, https://​www​.reuters​.com/​ article/​us​-chile​-migrants​-idUSKCN1NC30V. Maple, N. (2016), ‘Rights at risk: a thematic investigation into how states restrict the freedom of movement of refugees on the African continent’, UNHCR New Issues in Refugee Research, Paper 281. Morland, P. (2016), Demographic Engineering: Population Strategies in Ethnic Conflict, London: Routledge. Morland, P. (2019), The Human Tide: How Population Shaped the Modern World, London: Hachette UK. Nordås, R. (2012), ‘The devil in the demography? Religion, identity and war in Côte d’Ivoire’, in A. Goldstone, Eric Kaufmann and Monica Duffy Toft (eds), Political Demography: How Population Changes are Reshaping National Politics and International Security, Oxford: Oxford University Press, pp. 252–67. Orfield, M. (2017), in T. Edsall, ‘White-on-white voting’, New York Times, Nov. 16, https://​www​.nytimes​.com/​2017/​11/​16/​opinion/​trump​-white​-voting​.html. Pew Forum on Religion in Public Life (2011), ‘The future of the European Muslim population’, V. Skirbekk, B. Barakat, A. Goujon, K. C. Samir, E. Kaufmann, E. Lundevaller and M. Stonawski, online report, https://​www​.pewforum​.org/​2011/​01/​ 27/​the​-future​-of​-the​-global​-muslim​-population/​. Polakow-Suransky, S. (2017), Go Back to Where You Came From: The Backlash Against Immigration and the Fate of Western Democracy, Oxford: Oxford University Press. Rutinwa, B. (1999), ‘The end of asylum? The changing nature of refugee policies in Africa’, UNHCR New Issues in Refugee Research, Paper 5. Rydgren, J. and P. Ruth (2013), ‘Contextual explanations of radical right-wing support in Sweden: socioeconomic marginalization, group threat, and the halo effect’, Ethnic and Racial Studies 36 (4), 711–28. Schwander, H. and P. Manow (2017), ‘It’s not the economy, stupid! Explaining the electoral success of the German right-wing populist AfD’, CIS Working Paper (94). Smith, A. (1998), Nationalism and Modernism: A Critical Survey of Recent Theories of Nations and Nationalism, London & New York: Routledge. Teitelbaum, B. (2020), War for Eternity: The Return of Radical Traditionalism and the Rise of the Populist Right, London: Penguin Allen Lane. Teitelbaum, M. S. (1998), A Question of Numbers: High Migration, Low Fertility, and the Politics of National Identity, New York: Hill and Wang. Valentino, N. A., T. Brader and A. E. Jardina (2013), ‘Immigration opposition among U.S. whites: general ethnocentrism or media priming of attitudes about Latinos?’ Political Psychology 34 (2), 149–66.

WHITESHIFT

99

Van den Berghe, P. (1979), The Ethnic Phenomenon, New York: Elsevier. Wang, H. L. (2019), ‘Why is the census bureau still asking a citizenship question on forms?’, Aug. 9, https://​www​.npr​.org/​2019/​08/​09/​743296249/​why​-is​ -the​-census​-bureau​-still​-asking​-a​-citizenship​-question​-on​-forms​#:​~:​text​=​The​ %20survey​%20asks​%20about​%20people's​,Census​%20Bureau's​%20manual​%20for​ %20interviewers. Washington Post (2015), ‘Full text: Donald Trump announces a presidential bid’, 16 June, https://​www​.washingtonpost​.com/​news/​post​-politics/​wp/​2015/​06/​16/​full​-text​ -donald​-trump​-announces​-a​-presidential​-bid/​. West, E. (2013), The Diversity Illusion : What We Got Wrong About Immigration and How To Set It Right, London: Gibson Square. Wimmer, A. (2008), ‘The making and unmaking of ethnic boundaries: a multilevel process theory’, American Journal of Sociology 113 (4), 970–1022.

7

Wombfare: the weaponization of fertility

Monica Duffy Toft

Introduction Fertility affects the future power and influence of ethnic and religious groups within countries. If you are trying to project the relative power balances among rival groups or predict whether political tensions between competing ethnicities or religions will lead to conflict, an obvious place to start is any discrepancy in birth rates. The raw size of a group helps shape its political clout. More people often translate into more votes in elections and greater political influence, with larger groups, all else equal, enjoying better access to government officials and input over government policies. Differential changes in the populations of groups therefore upset existing political balances, processes and structures. Ethnic or religious groups with shrinking populations in relative terms risk losing their political dominance – and frequently try to cling to whatever power they have left – while expanding groups are expected to demand more authority, sometimes violently. Political violence and civil conflicts are more likely as demographic transitions occur, when one group realizes that it is either growing or shrinking relative to the other. The importance of fertility is not lost on ethnic and religious groups. Groups will often adopt pronatalist policies, encouraging members to give birth to more children in the hopes that their populations will grow compared to their adversaries. As demographic balances in specific areas or across entire countries change, groups with high fertility seek to gain more leverage and influence. Both non-religious and religious groups have actively encouraged larger families to further their political goals. Pronatalist policies with a political agenda have been pushed by secular leaders including Adolf Hitler in Germany, Josef Stalin in the Soviet Union, Nicolae Ceausescu in Romania and Yasser Arafat in Palestine/Israel, and religious groups like Christians and 101

102 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Muslims in Lebanon, ultra-Orthodox Jews in Israel, fundamentalist Christians in the West and Mormons in the United States. There is a positive correlation between religiosity and fertility (Kaufmann 2010; Lehrer 2004). Traditional religious practices and ideals prove particularly effective in boosting fertility. Condemning practices that reduce fertility – like contraception, abortion, divorce and homosexuality – and advocating distinct gender roles and large families all contribute to relatively fast growth rates for religious groups (Norris and Inglehart 2004). While many of these practices are adopted without consideration of the political effects, there is often an active and conscious decision to encourage people to have more children to further longer-term political objectives. Globally, the most religious areas are experiencing the fastest population growth rates, with Muslims, Christians and Jews having higher fertility rates than atheists, agnostics or other unaffiliated populations (Pew Research Center 2015a). I coined the term “wombfare” to describe the use of fertility as a political weapon typically deployed by elites to defeat rival ethnic or religious groups (Toft 2012). The tactic explicitly or implicitly contributes to the long-term strategy by elites to gain political advantages for their group over competing groups, and it has been used by religious and secular elites alike. In some instances, high fertility rates are pursued for economic, cultural or religious reasons with little consideration of the political ramifications, while in others they are pushed specifically to strengthen the group’s numbers to advance their political agenda. Regardless of whether the motivation is directly political, changes in relative group sizes could trigger conflict. Thus, regardless of motivation, differential demographic growth rates can spark conflicts when a dominant group is about to lose or loses its majority status to a growing minority. Either the shrinking majority decides to crack down to protect its status, or the expanding minority wages war to gain control. This chapter describes how ethnic and religious groups use wombfare as a weapon. I first discuss the importance of religion, and how it affects the fertility rates of groups vying for power and influence. To explain how the tactic can contribute to the onset of political violence and state collapse, I then describe Power Transition Theory and how it relates to differential demographic growth rates within countries. To probe the full implications of wombfare and differential population growth rates, I include four qualitative studies: Christians and Muslims in Lebanon, Arab Muslims and ultra-Orthodox Israeli Jews in Israel/Palestine, the fundamentalist Christian Quiverfull movement in the West (most prominently in the United States) and Mormons in the United States. While wombfare is used for different reasons and in different

WOMBFARE: THE WEAPONIZATION OF FERTILITY 103

ways depending on the circumstances, in all cases higher fertility rates can both enhance a group’s long-term political influence and increase risk of war. This holds across a variety of states. Just consider the rise of populism in stable democracies, arising from the fear of some of the majority “white” natives being supplanted by non-natives. Rhetoric about boosting birthrates is part of the discourse to prevent the relative decline of the dominant group (Brostoff 2019). There is not a lot written on wombfare, therefore we do not know a lot about the dynamics. One of the main reasons is that demography itself is under-researched within the political science discipline, largely because political science prefers to study questions that capture direct causality, not longer-term, indirect factors such as population changes. Second, there are serious data limitations when it comes to studying the changing demographics of groups within and across states. We often do not know the relative size or the relative growth rates of groups, either because data are simply unavailable if a state does not conduct censuses or because the censuses are unreliable or manipulated. This explains why research into the dynamics of wombfare requires qualitative case studies that draw on an array of different types of sources.

The power of religion Wombfare is a weapon that is not only accessible to the devout. Secular groups and leaders continue to recognize the importance of fertility and demographics, often adopting pronatalist policies to increase their share of a country’s population and gain economic and political benefits. Pronatalist policies are common and may entail tax breaks, stipends and job security to boost fertility and births. Wombfare is different because one group is favored, typically over the interests of another. And although racial groups might call for segmented pronatalist policies to boost numbers, religious groups seem particularly adept at deploying the weapon of wombfare. In part, this is due to the traditions, practices and beliefs common among religions that encourage larger families regardless of the political implications. Support for early marriages and customary gender roles – with women at home taking care of the family – increase the average number of children a woman is likely to give birth to in her lifetime; and disapproval of non-heterosexual relationships, divorce, abortion and the use of contraception all tend to increase fertility rates (Kaufmann 2010). Furthermore, religious groups have access to institutions and resources that are in a position to welcome and support larger families. Religious groups

104 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

therefore might include fertility and children within their longer-term political strategy, consciously acknowledging the importance of demographics for influence and power. In Shall the Religious Inherit the Earth?, Eric Kaufmann argues that fundamentalist populations globally are expanding their share of the population at the expense of secular ones (Kaufmann 2010). Education – particularly for girls – birth control and urbanization have all contributed to falling fertility rates in developed countries, yet even here, higher birthrates prevail among the devout. Non-religious populations are simply not having enough children to replace themselves, while devout Muslim, Christian and Jewish families continue to have relatively high fertility rates and their absolute numbers are growing. Fertility rates have fallen among the most religious groups, but not as much as they have for their secular counterparts, and it does not take huge differentials to change demographic balances in the future. Fundamentalism is in many ways a backlash against modernity and secularism, and religious groups continue to push for larger families and growing political power. Population projections support Kaufmann’s argument globally, with the world as a whole becoming more religious. Muslims have the highest fertility rates of any religion, followed by Christians, Hindus and Jews – with all of them above the replacement rate, or the number needed to prevent the population from shrinking in absolute terms (Pew Research Center 2017a). People with no religion, however, have fertility rates that are too low to sustain their populations, and they will account for a shrinking proportion of global births in the coming decades. This is not only seen at the global level. For example, Muslims are expected to account for an increasing share of Europe’s population even if future immigration is prevented (Pew Research Center 2017b). To be sure, demographers expect large numbers of people to walk away from their religion, but the effects of religious switching will likely be overshadowed by fertility (Pew Research Center 2017b). Strong religions maintain decent retention rates, with powerful motivations to stay within the group and powerful disincentives to leave the faith behind (Kaufmann 2010). Demographers thus have a good sense of the ties between religiosity and fertility. The more faithful have more babies.

Power Transition Theory While higher fertility rates increase the relative size of a group’s population and expectations for political influence, there is extremely limited research

WOMBFARE: THE WEAPONIZATION OF FERTILITY 105

into whether differential rates will instigate violence. The prospect of an ethnic majority being overtaken by a minority can, in certain cases, be the cause of conflict and possibly civil war (see Levy and Krebs 2001; Strand and Urdal 2005; Toft 2002). The theory that best explains how wombfare can lead to kinetic war is Power Transition Theory (PTT). When A. F. K. Organski developed the theory to determine why major conflicts tended to arise when the parties in question were approaching or just beyond power parity, he highlighted the changes in relative power caused by a state that became powerful enough to jeopardize the status quo (Organski 1958; Organski and Kugler 1980). Such instances of power transition, according to Organski’s theory, could lead potential powers to seek to move the distribution of valuable assets that traditionally support sitting powers. This, in turn, might lead a rising power to threaten or assault the incumbent. PTT contains six central assumptions (see DiCicco and Levy 2001). First, the state is the basic unit of analysis. Second, states are rational actors. Third, the international system contains a dominant power (or coalition) and weaker powers. Fourth, domestic political systems can be compared to the international system. Fifth, international developments stem primarily from changes that occur within states. And sixth, alliances remain relatively stable. With these six assumptions, the theory emphasizes two key variables, including relative power and how satisfied states are with the existing international order, or status quo (DiCicco and Levy 2001). While most states value peace, this is not always the case, nor is peace always the highest priority. Bearing this in mind, one of the central contributions of the theory is that states anticipate that their “power” will eventually represent (or achieve) their individual interests – at times at the expense of interstate peace. While it was developed as a tool for studying interstate conflict, PTT is useful in predicting when shifts in majority and minority populations (for our purposes, majority and minority religious groups) would likely result in violence. In essence, the logic of PTT tells us that there is a high risk of war right before or right after a demographic shift. That is, such conflict would erupt around the time that a minority group gained relative majority status within the state. Sensing this, a sitting majority (soon to be a minority) might attack the growing group before it could make use of its newly acquired power.1 Or, conversely, the increasing minority group, feeling emboldened by its size and legitimated by its majority status, might challenge the declining majority group. According to this logic, the most striking scenario for states grappling with tension brought on by differential demographic growth is when a shrinking majority is coupled with a growing minority.

106 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

PTT can thus be readily applied to population dynamics between groups within states. With population serving as a chief factor in most social science indices of national capabilities (de Soysa et al. 1997), the rise or fall of group sizes within a state fails to correlate perfectly with power in terms of a pure majority or proportional rule system. (The industrial power of a state also fails to correlate precisely with its ability to engage in war.) What this means is that the population of one group relative to others in a multireligious state will hold consequences for control over that state – a situation corresponding to the way that a state’s industrial power will influence who maintains power internationally. Due in large part to the fact that the number of states with democratic forms of government has gone up since World War II, I would argue that demographic shifts play a more important role in today’s international system than in the past. Because majority rule is at the center of democratic legitimacy, and because identity groups – both ethnic and religious – within a state will tend to vote as blocs, the democratization of states has rendered demographic shifts among identity groups more significant (see Hegre et al. 2001; Reynal-Querol 2002; Toft 2003). To see how this plays out in practice, I highlight four cases when groups used wombfare to support their longer-term political strategies. Christians and Muslims in Lebanon The precarious balance among religious groups in Lebanon points to the importance of population numbers. Government positions are distributed among Christians and Sunni and Shia Muslims, but they are based on long outdated population estimates – the last census was in 1932. Differential fertility rates and faster Muslim population growth contributed to interreligious violence, a brutal civil war from 1975 to 1990 and a shift in the country’s Muslim political representation. While it is clear that the Christian population has been steadily shrinking for decades as the Muslim population has increased, there is an understandable fear of conducting a new census. Changing agreements to reflect current demographics is likely to inflame sectarian tensions and risks igniting another civil war. Christians are reluctant to cede power, but with relatively low fertility rates their political influence will be increasingly out of step with demographic realities, eventually making the situation untenable (if it is not already). Christians were once the majority in Lebanon. The Christian population largely welcomed France’s mandate after World War I, adopting a pro-Western stance. The demographic balance began to shift in the 1920s, however, as the country expanded into areas with large Muslim populations. While Christians

WOMBFARE: THE WEAPONIZATION OF FERTILITY 107

continued to favor a close alignment with Europe, the growing number of people who had a strong Arab and Muslim identity wanted to turn toward Syria instead. But the 1932 census solidified the Christians’ dominance in the country’s political institutions, giving Christians six members of parliament for every five Muslim members. In 1943, France established the state of Lebanon with the intention of establishing a strong Western ally in the Middle East. The 1948 Arab–Israeli War caused a mass exodus of Palestinian refugees into Lebanon, threatening to upset the country’s delicate religious balance and end the Christian majority. The Lebanese government attempted to forestall any problems by denying citizenship to most Palestinian Muslims while granting it to Palestinian Christians (Fargues 2001). In the 1960s and 1970s, foreigners and Lebanese citizens, including many Christians, began emigrating to other countries (Soffer 1986). While the economy began to pick up in the 1970s following increased industrialization and urbanization, these same trends also fueled underlying religious tensions. Large social movements with strong sectarian identities formed and competed for power and influence, contributing to an uptick in political violence and conflict between Christians and Muslims. The country descended into an all-out civil war in 1975 that pitted sectarian militias against one another for 15 years. The fighting wiped out integrated neighborhoods and hardened separate identities for different religious communities (Chamie 1977). The Taif agreement ending the war evenly split parliamentary seats between Christians and Muslims, curbed the power of the Maronite Christian president, and strengthened the powers of both the prime minister and the speaker of the parliament. Even though the negotiated deal reduced Christian control, Christians were still overrepresented in parliament. Unofficial censuses indicated a shrinking proportion of Christians in Lebanon. In the mid-1950s, Christians were an estimated 55 percent of the population, but that percentage fell significantly to around 43 percent in 1998 (Fargues 2001). Today, Christians make up only 37 percent of registered voters and an even smaller proportion of Lebanese youth (The Economist 2016). The falling numbers are due, in part, to higher fertility rates among Muslims. A study in 1971 found that Shia Muslim women had an average number of seven births, Sunni Muslims had slightly less than six and Catholic women had slightly less than five (Chamie 1977; Faour 1989). Muslims represented a growing majority in Lebanon, but their political authority did not match their expanding numbers. With little hope of drastically increasing their population in the near future, Lebanon’s Christians are left with only painful options. They can try to cling to the present configuration, but with differential growth rates it is only a matter

108 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

of time before the system will need to be reformed – and delaying the inevitable increases the likelihood of another war. While Christians could theoretically try to carve out a smaller autonomous enclave that more closely resembles the Lebanon of a century ago, it is unlikely that either side would be willing to make the sacrifices this would entail. Finally, Christians could accept the changing demographic realities and give up their perks, choosing instead to operate as a minority in a Muslim-majority state. Convincing Christians to voluntarily surrender their dominant position, however, is exceedingly difficult. For their part, Muslims will continue to demand more control. While higher Muslim fertility rates in Lebanon were originally due to cultural, economic and religious considerations, political calculations are increasingly intertwined with demographics. Given the differential growth rates among religious groups, Muslims are less and less likely to be satisfied with the status quo agreements curtailing their political power. At this point, the continued conflict and discord in Lebanon has only reinforced existing identities and their sway on political power. It would be surprising if future research found this not to be the case, but we will have to wait and see. Jews and Muslims in Israel and Palestine Wombfare is seen on both sides in Israel and Palestine. Both Arab Muslims and ultra-Orthodox Jews have higher fertility rates than secular Jews, altering the identity of the state and shaping possible solutions to the Israeli–Palestinian conflict. While the growing percentage of Muslims in Israel and the Palestinian territories pose a “demographic threat” to those demanding a state defined by its Jewish identity, the rising proportion of ultra-Orthodox Jews also threatens Israel’s ability to be a secure, Jewish and democratic state. Fertility rates directly affect politics and security in Israel and Palestine, and could contribute to another religious war between Jews and Muslims. Demography has always been important in Israel. Zionist leaders emphasized the importance of a strong Jewish majority before the state of Israel was even founded (Morris 2001). Zionist strategy relied on high Jewish birthrates and immigration, and Arabs were seen as a demographic threat from the beginning (Portugese 1998). The declaration of independence and the 1948 Arab–Israeli war proved a decisive demographic victory for the Jewish population. The exodus of Palestinian refugees meant only 12 percent of the new state’s citizens were Arab (Tal 2016). Israel’s leaders sought to maintain Jewish dominance through continued immigration and pronatalist policies, but Muslim Arabs

WOMBFARE: THE WEAPONIZATION OF FERTILITY 109

also viewed large families as an important component of their resistance and longer-term political strategies. Israel’s demographic challenges were heightened after the Six-Day War in 1967 with its acquisition of new territory in the West Bank, East Jerusalem, the Gaza Strip, the Golan Heights and the Sinai Peninsula. Arab Muslim populations dwarfed the Jewish populations in all of these areas. In response, the government promoted larger Jewish families through low-interest housing loans and media campaigns (Portugese 1998). But the pronatalist policies were not enough to counteract the high fertility rates among Palestinians. In the 1980s, Jewish women had fewer than three children on average – although the rate was much higher for ultra-Orthodox women – while Palestinian Arabs in the West Bank and Gaza had more than six (Fargues 2001). Arab children under the age of four outnumbered their Jewish counterparts in Israel and the Palestinian territories for the first time in 1985 (Friedman 1987). The collapse of the Soviet Union and the subsequent influx of Jewish immigrants provided a brief respite, but this only delayed the underlying trends. In 1993, Palestinian leader Yasser Arafat described high Palestinian birth rates as a “biological time-bomb” that threatens to blow up Israel from the inside, and wombfare contributed to Israel’s decision to build a security barrier in the early 2000s (Davis 2003). The Palestinian Census Bureau said the Arab population surpassed the Jewish population in Israel as a whole for the first time in 2016 (Berger and Khoury 2018; Pyzer 2016). Although the precision of these data is debated, what is not debated is that higher fertility rates indicate that the proportion of Muslims will keep rising and that this will largely be at the expense of secular Jews. The ultra-Orthodox Jewish population makes up around 9 percent of the population today, and it is projected to reach 29 percent by 2059, given its high fertility rates (Malach et al. 2016). The ultra-Orthodox are largely aligned with the political right in Israel, and their strict religious adherence, refusal to serve in the military and rejection of Western society and modernity will complicate Israel’s socioeconomic security. The use of wombfare and presentation of different data by groups on both sides of the religious divide in Israel and Palestine increasingly complicates leading proposals for one-state and two-state solutions to the conflict.2 Conservative Christians in the West There has been a rise of wombfare-like dynamics in the Western world as fundamentalist Christians seek to enhance their numbers against what they perceive to be threats to their livelihoods, notably non-Christians (in par-

110 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

ticular Muslims) and secularism (understood as the diminishment of faith in society more generally). Two groups stand out: the Quiverfull movement and Mormonism. The conservative Quiverfull movement aims to reverse the declining birthrates among Protestant Christians in Western countries, most actively in the United States. The movement resembles other endogenous growth sects like the Amish and ultra-Orthodox Jews, and it emphasizes the infallibility of the Bible, traditional patriarchal gender roles and large families (Kaufmann 2010). While the Quiverfull movement is still small, its teachings have influenced millions of Evangelicals. They view themselves as champions of the Christian culture against the growing menace of Islam, pointing to the disparity of birthrates between Muslims and Christians to demonstrate the urgency of the threat (Hagerty 2009). Wombfare is used to arrest any loss of political influence by fundamentalist Christians in the West. Yet, because of their size, they have limited political clout or influence. The interdenominational and grassroots movement began to take root in the 1980s, believing the use of birth control in marriage inevitably leads to an acceptance of abortion. Women’s lives instead should be a sacrifice to God and they should submit to their husbands by treating motherhood as their full-time occupation (Pride 1985). They view Islam as an existential threat, seeing the 9/11 terrorist attacks as a harbinger of the rise of Islamic extremism, and infuse their rhetoric with the specter of “race suicide,” pointing to declining birthrates in Europe, ethnic conflicts and immigration to the United States as indications of Western society’s imminent decline (Joyce 2006). Multiculturalism breeds anarchy, and members need to have more babies to maintain and expand Christendom. There is an infusion of militaristic language in their writings, and the Quiverfull movement’s ultimate goals go beyond raising children – they seek to reclaim Western society for conservative Christianity through wombfare. Mormons in the United States are widely known for their high fertility rates. Church teachings emphasize the importance of large families, clear gender roles and early marriage, which have all contributed to higher fertility rates than their Catholic and Protestant counterparts. But, despite the differential demographic growth rates, Mormon women use birth control and Mormons do not try to achieve clear political goals through their pronatalist policies (Heaton and Calkins 1983). Still, the high concentration of Mormons in Utah has given the group considerable political clout at both the local and national levels. Mormons have one of the most partisan voting records of any ethnic or religious group in the United States, with most identifying as Republican, and

WOMBFARE: THE WEAPONIZATION OF FERTILITY 111

the most religious overwhelmingly – 88 percent of respondents – declaring themselves Republican or leaning Republican (Pew Research Center 2015b). Mormons immigrated to rural Utah in the nineteenth century to escape persecution and establish a holy city where the involvement of the Church extended beyond weekly services to public affairs, social organizations and proselytizing (McQuillan 2004). Church teachings pervaded all aspects of life, and a high birthrate was to a certain extent necessary for the nascent Church’s survival (Hastings et al. 1972). Given the centrality of the family, pronatalist writing often included dramatic language, linking fertility to the “fate of the nation” and any attempts to curb it as a risk of “race suicide” (Musser 1904). Large families in many ways were considered indispensable to the success of the community. At the same time, Church doctrine has been pragmatic when facing political pressures, as it abandoned its support for polygamy to fulfill a condition of statehood for Utah and granted the priesthood to black men in response to the civil rights movement.

Weaponizing fertility Religious and ethnic groups encourage women to have more children to support their political strategy. Higher fertility rates can enhance a group’s long-term political influence at the expense of their rivals, and differential demographic growth rates can lead to civil conflict and state collapse. While wombfare is not an exclusively religious phenomenon, there is a strong relationship between religiosity and fertility. The world’s most devout tend to have more children, and religious communities are growing faster than secular ones around the world. The use of wombfare plays out in different ways depending on the context. Muslims in Lebanon have more children for largely religious, economic and cultural reasons, but differential growth rates increasingly shape political demands and have contributed to political violence and conflict. Israeli Jews consciously adopted pronatalist policies to avert a demographic defeat by Arab Muslims, and the danger of igniting another war looms large. The fundamentalist Quiverfull movement seeks to promote fertility to prevent the decline of Christianity in the face of a rising Arab populace. Mormons encourage large families for religious and cultural reasons, and their higher fertility rates have granted them considerable local and national political influence. While the weapon is deployed differently and the outcomes are shaped by the situation

112 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

and behavior of other groups, differential fertility rates do seem to affect domestic governments and the stability of states. Yet, despite the insight that these three cases provide, there is much research that needs to be done. We do not have widespread evidence of the conditions under which wombfare will emerge or when it will lead to instability and violence. Other cases might include a whole host of multinational states with differential growth including China, Fiji, Singapore and the United States. A cross-national and time series data set of one country such as India would be a tremendous asset here, as would an accounting of demographic changes of groups across all states. Such data are limited. The bottom line is that there are compelling logical arguments and some evidence, but the cases remain largely anecdotal. What is needed is a fair number of additional cases highlighting key dynamics and conditions isolating how wombfare leads to conflict and when it does not.

Notes 1.

2.

It should be noted that Power Transition Theory has been criticized for not being able to predict the precise moment in the demographic shift when civil war would flare up. Similarly, the theory cannot forecast which actor would, in all likelihood, instigate the fighting. The theory leaves open the possibility of war breaking out at one of three stages: just before a minority–majority transition, in the midst of transition (in other words, at parity), or just after a transition, when a former minority shifts into its role as a clear majority. There was much made of the fact that Muslim and Jewish rates were reported as near equal in 2019 (see https://​www​.jns​.org/​jewish​-fertility​-rate​-outstrips​-arab​ -one​-for​-the​-first​-time​-in​-israels​-history/​), yet these were not disaggregated, labeled simply Muslim and Jewish and included only Israeli citizens, lending credence to the importance of the ultra-Orthodox community and its relatively high fertility rates for the broader Jewish population. So long as the Arab and ultra-Orthodox rates remain higher than other groups (e.g. secular Jews), tensions will remain.

References Berger, Y. and J. Khoury (2018), ‘How Many Palestinians Live in Gaza and the West Bank? It’s Complicated’, Ha’aretz, https://​www​.haaretz​.com/​israel​-news/​how​-many​ -palestinians​-live​-in​-gaza​-and​-the​-west​-bank​-it​-s​-complicated​-1​.5956630. Brostoff, M. (2019), ‘How White Nationalists Aligned Themselves with the Antiabortion Movement’, Washington Post, August 27, https://​www​.washingtonpost​.com/​

WOMBFARE: THE WEAPONIZATION OF FERTILITY 113

outlook/​2019/​08/​27/​how​-white​-nationalists​-aligned​-themselves​-with​-antiabortion​ -movement/​. Chamie, J. (1977), ‘Religious Differentials in Fertility: Lebanon, 1971’, Population Studies, 31 (2), 365–82. Davis, D. (2003), ‘Biological Warfare’, The Spectator, September 5. de Soysa, I., J. R. Oneal and Y. Park (1997), ‘Testing Power-Transition Theory Using Alternative Measures of National Capabilities’, Journal of Conflict Resolution, 41 (4), 509–28. DiCicco, J. M. and J. S. Levy (2001) ‘The Power Transition Research Program: A Lakatosian Analysis’, in Colin Elman and Miriam Fendius Elman (eds), Progress in International Relations Theory: Appraising the Field, Cambridge, MA: MIT Press, pp. 109–57. Faour, M. (1989), ‘Fertility Policy and Family Planning in the Arab Countries’, Studies in Family Planning, 20 (5), 254–63. Fargues, P. (2001), ‘Demographic Islamization: Non-Muslims in Muslim Countries’, SAIS Review, 21 (2), 441–82. Friedman, T. (1987), ‘A Forecast for Israel’, The New York Times, October 19, https://​ www​.nytimes​.com/​1987/​10/​19/​world/​a​-forecast​-for​-israel​-more​-arabs​-than​-jews​ .html. Hagerty, B. B. (2009), ‘In Quiverfull Movement, Birth Control is Shunned’, NPR Morning Edition, March 25, https://​www​.npr​.org/​templates/​story/​story​.php​?storyId​ =​102005062. Hastings, D. W., C. H. Reynolds and R. R. Canning (1972), ‘Mormonism and Birth Planning: The Discrepancy Between Church Authorities, Teachings and Lay Attitudes’, Population Studies, 26 (1), 19–28. Heaton, T. B. and S. Calkins (1983), ‘Family Size and Contraceptive Use Among Mormons: 1965–75’, Review of Religious Research, 25 (2), 102–13. Hegre, H., T. Ellingsen, S. Gates and N. P. Gleditsch (2001), ‘Toward a Democratic Civil Peace? Democracy, Political Change, and Civil War, 1816–1992’, American Political Science Review, 95 (1), 33–48. Joyce, K. (2006), ‘The Quiverfull Conviction: Christian Mothers Breed “Arrows for the War”’, The Nation, November 26. Kaufmann, E. (2010), Shall the Religious Inherit the Earth?: Demography and Politics in the Twenty-First Century, London: Profile Books. Lehrer, E. L. (2004), ‘Religion as a Determinant of Economic and Demographic Behavior in the United States’, Population and Development Review, 30 (4), 707–26. Levy, J. S. and R. R. Krebs (2001), ‘Demographic Change and Sources of International Conflict’, in Myron Weiner and Sharon Stanton Russel (eds), Demography and National Security, New York: Berghahn Books, pp. 62–108. Malach, G., L. Cahaner and M. Choshen (2016), Statistical Report on Ultra-Orthodox Society in Israel, Jerusalem: The Israel Democracy Institute. McQuillan, K. (2004), ‘When Does Religion Influence Fertility?’, Population and Development Review, 30 (1), 25–56. Morris, B. (2001), Righteous Victims: A History of the Zionist-Arab Conflict, 1881–1999, New York: Vintage Books. Musser, A. M. (1904), ‘Race Suicide—Infanticide, Prolicide, Leprocide vs. Children— Letters to Messrs. Joseph Smith and William H. Kelley’, privately published pamphlet, Salt Lake City, UT. Norris, P. and R. Inglehart (2004), Sacred and Secular: Religion and Politics Worldwide, New York: Cambridge University Press.

114 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Organski, A. F. K. (1958), World Politics, New York: Knopf. Organski, A. F. K. and J. Kugler (1980), The War Ledger, Chicago, IL: University of Chicago Press. Pew Research Center (2015a), ‘The Future of World Religions: Population Growth Projections, 2010–2050’, April 2, https://​www​.pewforum​.org/​2015/​04/​02/​religious​ -projections​-2010​-2050/​. Pew Research Center (2015b), ‘Religious Landscape Study’, November 3, https://​www​ .pewforum​.org/​religious​-landscape​-study/​. Pew Research Center (2017a), ‘The Changing Global Religious Landscape’, April 5, https://​www​.pewforum​.org/​2017/​04/​05/​the​-changing​-global​-religious​-landscape/​. Pew Research Center (2017b), ‘Europe’s Growing Muslim Population’, November 29, https://​www​.pewforum​.org/​2017/​11/​29/​europes​-growing​-muslim​-population/​. Portugese, J. (1998), Fertility Policy in Israel: The Politics of Religion, Gender, and Nation, Westport, CT: Praeger. Pride, M. (1985), The Way Home: Beyond Feminism, Back to Reality, Wheaton, IL: Crossway Books. Pyzer, Z. (2016), ‘Lies, Damn Lies and Demographics: Are Jews now a Minority Between the River and the Sea?’ The Jerusalem Post, July 12, https://​www​.jpost​.com/​ israel​-news/​jews​-a​-minority​-between​-the​-river​-and​-the​-sea​-460205. Reynal-Querol, M. (2002), ‘Ethnicity, Political Systems, and Civil Wars’, The Journal of Conflict Resolution, 46 (1), 29–54. Soffer, A. (1986), ‘Lebanon: Where Demography is the Core of Politics and Life’, Middle Eastern Studies, 22 (2), 197–205. Strand, H. and H. Urdal (2005), ‘Differential Growth, Political Instability and Violent Conflict’, paper presented at the Annual International Studies Association Conference, Honolulu, HI. Tal, A. (2016), The Land is Full: Addressing Overpopulation in Israel, New Haven, CT: Yale University Press. The Economist (2016), ‘Census and Sensibility’, November 5, https://​www​.economist​ .com/​middle​-east​-and​-africa/​2016/​11/​05/​census​-and​-sensibility. Toft, M. D. (2002), ‘Differential Demographic Growth in Multinational States: Israel’s Two-Front War’, Journal of International Affairs (1) (Fall), 71–94. Toft, M. D. (2003), The Geography of Ethnic Violence: Identity, Interests, and the Indivisibility of Territory, Princeton, NJ: Princeton University Press. Toft, M. D. (2012), ‘Wombfare—Religious and Political Dimensions of Fertility’, in Jack A. Goldstone, Eric P. Kaufmann and Monica Duffy Toft (eds), Political Demography: How Population Changes are Reshaping International Security and National Policies, New York: Oxford University Press, pp. 213–25.

8

Health and demography

Jeremy Youde

Introduction Disease can affect age structure just as fertility can, with profound implications for political institutions, the international community’s global health agenda and our collective understanding of the short- and long-term effects of various health challenges. The Covid-19 pandemic is just the latest example of how a disease outbreak can pose serious and long-lasting effects on nearly every country in the world. Although global health and political demography are deeply interrelated, the connections between them are relatively underexplored in research. This chapter focuses on four such connections: the competing notions of the epidemiological and health transitions; the intersection between refugees, climate change and health; the linkages between maternal and child health and conflict; and whether the structure of global health governance institutions facilitates cross-national collaboration on these matters.

Health transitions in the modern era Disease and illness are not randomly distributed. Economics, geography, gender and a whole host of other societal cleavages shape an individual’s health trajectory. One of the still unresolved questions is how local, national and institutional forces have a positive effect on improving health outcomes for all and exactly how various elements contribute to personal and societal health. Two of the most prominent explanations for how health outcomes change within societies have been the epidemiological and health transitions. More than two generations ago, Omran (1971) described what he called the epidemiological transition: the notion that states progress from high fertility rates and most mortality and morbidity being caused by infectious 115

116 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

diseases, to low fertility rates and most death and illness being the result of non-communicable diseases. These changes, he argues, are largely attributable to increased food security, socioeconomic development and improvements in medicine and public health. Omran’s approach provides a useful starting point, but is limited by a number of factors. First, it suggests a linear progression unreflective of and oversimplifying the experience of most countries (McKeown 2009). Second, it inadequately accounts for the differentiations and cleavages within societies that lead to unequal outcomes. Non-communicable diseases cause a growing proportion of deaths worldwide, but Heuveline et al. (2002) find that infectious diseases still cause the vast majority of excess mortality among the poorest quintile. Third, the epidemiological transition pays too little attention to the social determinants of health and does not account for broader structural changes like macroeconomics and climate change (Santosa et al. 2014). Finally, the epidemiological transition has a hard time addressing the emergence of new health issues like HIV/AIDS and Covid-19, and diseases that re-emerge due to antimicrobial resistance, like tuberculosis (Mercer 2018). Scholars have developed a health transition model that can better account for the broadening array of diseases and health conditions affecting human health, qualitative and quantitative inequalities within a given society and the increasing complexity of the institutions governing health. While there exist various models of the health transition, what unites them is a shared understanding that changes in societal health depend on an appreciation of both epidemiological characteristics and the various ways in which societies respond to health situations (and health situations respond to society) (Vallin and Meslé 2004: 13–14). Authors differ in terms of which elements they prioritize in seeking to explain the health transition. Frenk et al. (1991) posit that the epidemiological transition is essentially a subset of the broader health transition – an important element, but only part of the equation. Caldwell (1993), on the other hand, highlights the social, cultural and behavioral determinants of health as central to a society’s health transition. The debates about the epidemiological transition and the health transition (and debates over what constitutes the health transition itself) have played out in a variety of global health initiatives. In 1978, representatives from nearly every country in the world converged in what is today Kazakhstan to sign the Declaration of Alma-Ata. The declaration embraced a holistic definition of health that moved beyond the absence of illness, decried health inequalities between rich and poor states, described economic and social development as prerequisites for health, emphasized the affirmative role of the state in provid-

HEALTH AND DEMOGRAPHY 117

ing for societal health and pledged to make primary health care accessible to all (World Health Organization 1978). These initiatives fell under the banner of “Health for All by 2000”, and national governments and the World Health Organization (WHO) all pledged to work toward making these ideas a reality. Unfortunately, the optimism of the Declaration of Alma-Ata ran into political, economic and social fights happening more broadly within the geopolitical landscape of the late 1970s and 1980s. The emergence of conservative political figures in the United States, United Kingdom and Germany; the ongoing tensions of the post-détente Cold War era; and the widespread economic recessions of the 1980s all sapped the movement toward universal primary health care (Hall and Taylor 2003). As a result, the failure of Health for All by 2000 directly intersected with the underlying assumptions of the epidemiological transition and health transition. In particular, the declaration raised serious questions about the role of government in providing health care, the unequal distribution of health care access within societies and the willingness of wealthy states to challenge their own political and economic assumptions in order to improve the health of people in other countries. More recently, there have been other efforts to incorporate health into broader goals to improve the lives of the world’s citizens. The Millennium Development Goals and Sustainable Development Goals both recognized that political, economic and social development could not happen in any meaningful way without prioritizing improved health (Sachs 2012). These efforts take the underlying assumptions of the health transition seriously, and they have fostered improvements in a number of areas, but debate continues about the willingness of wealthy and powerful states to foster better health outcomes in other countries.

Climate change, health and demography Climate change can catalyze new challenges around health and demography. Climate change and the movement of peoples (both voluntarily and involuntarily) are intimately linked, though it is not always clear that intergovernmental bodies recognize this interdependence – in May 2019, the WHO approved two separate action plans: one to address climate change and another to address refugees. A growing body of evidence, though, demonstrates that addressing either requires addressing both simultaneously, and alongside health and nutrition.

118 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Climate change will have direct effects on human health. First, changing weather patterns will have negative consequences for our health as disease vectors expand their range. In addition, if temperatures rise and rainfall decreases or comes inconsistently, agricultural production will suffer. The Midwestern United States experienced its wettest 12 months ever in 2018 and 2019, with massive flooding throughout the region (Sullivan 2019). These weather shifts have serious consequences that extend far beyond any region’s borders. One study estimated that the damage to corn crops from climate change over the next 30 years could reach $3 billion annually (Rosenzweig et al. 2002). Lower crop yields lead to surging global grain prices and increased food insecurity (Brown and Funk 2008; Schmidhuber and Tubiello 2007). At the other extreme, East Africa has experienced prolonged droughts throughout the 2010s, which have reduced food production and increased rates of severe malnutrition, particularly among young children. Kenya, for example, declared a national drought emergency in early 2019 after maize production in coastal regions was 99 percent lower than normal (Reliefweb 2019). Adequate nutrition and reliable access to food is absolutely key for human health, and climate change is making it increasingly difficult to achieve even basic standards. Second, excessive heat from rising temperatures causes death and illness – and heat waves are becoming more common. A summer 2018 heat wave in Japan killed nearly 150, hospitalized more than 70,000 due to hyperthermia and saw the country record its highest-ever temperature – 41.1 C or 106 F in the city of Kumagaya (Japan Times 2018). Officials cited the lack of air conditioning and other cooling infrastructure for increasing the death toll, and lamented that rising temperatures are “the new normal” (St. Michel 2018). That same year, more than 60 people died in Pakistan when a heat-induced power outage made it difficult to cool off (Sayeed 2018). Meteorologists expect this trend to intensify as climate change alters weather patterns and leads to higher temperatures. Indeed, in the United States, heat waves are now responsible for more deaths than any other natural disasters – and the problem is particularly bad in urban areas (Walker 2019). Part of what makes these heat waves so deadly is how they interact with existing social structures. Increasing heat waves raise questions about not just the availability of cooling centers and medical services, but also about our collective ability to address the pressing challenges facing populations writ large. Social isolation, especially among the elderly and those who have health challenges, is a major contributor to mortality during heat waves and has led Rome and other Italian cities to set up social intervention programs to reduce heat-related mortality among the elderly (Liotta et al. 2018). It is not simply how the climate

HEALTH AND DEMOGRAPHY 119

is changing; rather, it is the interaction between climate change and the larger structures in which society exists, and how those structures can respond to climate changes. Third, climate change alters where and how infectious diseases circulate. Mosquitoes are a common disease vector, spreading illnesses like malaria, dengue and yellow fever. As their habitats change and expand because of climate change, it is highly likely that an expanding proportion of the world’s population will be at an increased risk of contracting mosquito-borne diseases, and such illnesses will spread into areas where they are not currently present. Climate change will make more of the planet hospitable to Aedes aegypti and Aedes albopictus, two of the most common disease carriers. Recent analyses suggest that these two species’ expanded habitats will threaten nearly half of the world’s population by 2050 (Kraemer et al. 2019), meaning that we will likely see more cases of mosquito-borne diseases in new areas. This issue is not unique to malaria and mosquitoes, though. High temperatures and excessive rainfall allow cholera to take root in areas with poor sanitation (Jutla et al. 2013). Climate change can also alter animal habitats, making it easier for diseases to leap from animals to humans. Scientists worry that Hendra virus, a disease that spreads from flying foxes to horses to humans in the Australian states of Queensland and New South Wales, is likely to increase as climate change alters the living conditions for flying foxes and brings them into closer contact with both horses and humans (Martin et al. 2018). How will any of this matter for demography? One way to see this linkage is to consider that, while refugee and migrant flows are the result of complex factors, climate-induced health changes will add to these stressors. Two issues are key for understanding the link between climate, health and migrant health. First, forced displacement due to natural disasters can upend health services (McMichael et al. 2012). When Hurricane Maria hit Puerto Rico in September 2017, people were forced from their homes and faced a range of health challenges in the aftermath of the storm. Overcrowding, poor sanitation and a lack of health care have all posed long-term threats to the island’s population (Sciubba and Youde 2017). Second, the demographic trend of urbanization introduces a number of health challenges – both on a collective and a personal level. As Chapter 4 described, people often move to urban areas with the hope of better services and opportunities, but unplanned or unanticipated urban growth can lead to higher levels of air pollution, increased vehicle traffic and inadequate sanitation and health care services. On an individual psychological level, urbanization can

120 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

also lead to higher stress levels (Moore et al. 2003). As people stream into cities to flee the effects of climate change, those cities may be overwhelmed by them and their health needs. Dhaka, the capital of Bangladesh, is one of the most populous and most densely populated cities in the world. During the first decade of the twenty-first century, the city’s population doubled to more than 12 million by 2009. Over the next decade, the city’s population nearly doubled again to more than 20 million. More than 400,000 people move to the city annually, and the International Organization for Migration finds 70 percent are climate refugees due to increased storms and decreased access to freshwater (McPherson 2015). Such a high rate of urbanization has negatively affected the city’s residents. A study by the International Institute for Global Health at the United Nations University found that viral and communicable diseases are spreading within Dhaka at an increasing rate, with the under-five-year-old mortality rate double that of rural areas in Bangladesh. The increase in air pollution also causes 15,000 premature deaths annually (Haque et al. 2014). This is not a pre-ordained outcome, but it is one that governments must proactively address in order to prevent poor outcomes and further marginalization of new arrivals (Kjellstrom and Mercado 2008).

Gender, demography and health If we want to understand global health and its interplay with political demography, we also need to take gender seriously. O’Manique writes, “Social and political life is profoundly gendered and feminist scholarship has a crucial role to play in illuminating both the foundations of health insecurities and the effects of insecurities on differently gendered and located bodies” (2015: 48). Gender powerfully affects the ways in which a person experiences health and health care, and access to health and health care is mediated by the social, cultural and power relationships that are inextricably linked to gender. Paying attention to gender and health also promotes security and stability. Previous research has demonstrated that states with lower levels of women’s political, economic and social equality are less stable and more prone to conflict (Caprioli 2005; Caprioli and Boyer 2001; Hudson et al. 2012; see also Chapter 5, this volume). When women have less of a voice in political, economic and social structures, they end up with fewer rights – and society as a whole is worse off. This link suggests that improving women’s rights and health would contribute to larger improvements in national and international stability, while also addressing the underlying social determinants of societal health. That is, improving women’s rights not only improves women’s health,

HEALTH AND DEMOGRAPHY 121

but also improves health outcomes for the rest of society since caretaking is a strongly gendered role in most societies. We do see some efforts to better incorporate a gender perspective into the global health agenda. In 2000, the international community committed to improving maternal health by reducing maternal mortality and achieving universal access to reproductive health as part of the UN’s Millennium Development Goals (MDGs). The successor to the MDGs, the Sustainable Development Goals (SDGs), features an even more expansive dedication to women’s health, pledging to reduce poverty through gender-sensitive strategies, double the agricultural productivity of women farmers to increase food security, reduce maternal mortality, ensure access to sexual and reproductive health care, improve access to quality health care services, end all forms of discrimination and violence against women and girls, expand educational and economic opportunities for women and promote gender equality at all levels. Taken together, these strategies demonstrate at least a rhetorical interest in improving women’s physical and mental health as part of an overall strategy to promote peace and prosperity. There is some reason for optimism that this commitment to women’s health is more than rhetoric and has led to actual concrete action. In 2000, the MDGs called for reducing maternal mortality ratio (the number of maternal deaths of women while pregnant or within 42 days of the end of pregnancy due to any cause related to or aggravated by pregnancy and its management per 100,000 live births) by 75 percent between 1990 and 2015 (United Nations 2000). This goal built upon earlier international efforts to reduce maternal mortality, such as the 1987 Safe Motherhood Conference in Nairobi, the World Summit for Children in 1990, 1994’s International Conference on Population and Development and the Fourth World Conference on Women in 1995 (Zurleck-Brown et al. 2013: 32). UNICEF data shows a maternal mortality ratio in 1990 of 385 deaths per 100,000 live births. By 2000, the ratio had dropped to 341 deaths per 100,000 live births. Data from 2015 show a maternal mortality ratio of 216 per 100,000 live births (UNICEF 2018). While this outcome fell short of the stated MDG target, and much of the improvement came from notable progress in China, it nonetheless represents a significant decrease in maternal mortality. Can the international community harness this apparent momentum to do better going forward? The UN’s SDG 3, Good Health and Well-Being, calls for a reduction in the global maternal mortality ratio to less than 70 per 100,000 live births by 2030 – that would be just over 66 percent lower than 2015 levels (United Nations Development Programme 2018). Achieving such

122 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

an audacious goal would require both substantial funding and deep political commitment. Current efforts cast doubt on whether either is plentiful enough to realize this goal. Funding for maternal and child health has not kept pace with the stated ambitions; although it has stayed above $3 billion since 2009, development assistance for maternal and child health dropped 12 percent in constant 2017 US dollars between 2013 and 2017 (Institute for Health Metrics and Evaluation 2018). Similarly, policy efforts have been stymied by a lack of resources, an inability of government officials and advocates to cohere on a common understanding and ongoing tensions between national governments and state and local officials in highly federalized systems like Nigeria’s, where the lines of responsibility for public health matters remain unclear (Shiffman and Okonfua 2007). Global strategies to prioritize safe motherhood have fallen victim to disagreements over the definition of key terms, advocates who have relatively little power within political systems and issue-framing that has not resonated with key policymakers and donor states (Shiffman and Smith 2007). More broadly, we need to understand the complex ways in which health, gender and globalization interact to produce significant demographic effects. Few global health programs take gender equality and equity seriously, women are over-represented in informal caretaking roles and, despite some progress at the WHO, women are consistently under-represented in leadership and decision-making roles within global health institutions. Rectifying these imbalances requires that we “challenge structural and social power inequalities within patriarchal societies that produce inequalities that disadvantage women” (Davies et al. 2019: 601). Davies et al. (2019) stress that we need to understand the intersectionality between gender and other social cleavages like race, class, religion, disability and ethnicity. They also note that mainstream research methods pay far too little attention to gender and, as such, can overlook important health dynamics. We need to use a variety of research methods if we are to develop a more comprehensive knowledge base for researchers and develop a fuller understanding of the roots and causes of inequality and inequity (Davies et al. 2019: 601–2). We can see concrete manifestations of this gender blindness. Gender is a key health determinant around the world, with gender inequality and norms directly linked to violence, traffic fatalities, alcohol consumption and sexually transmitted infections (Davies and Bennett 2016). More women than men died from Ebola during the 2014–16 outbreak in West Africa, largely because women are expected to act as caregivers in formal and informal settings – yet the WHO did not provide data about case numbers and deaths that were disaggregated by gender until nearly a year into the outbreak (Harman 2016).

HEALTH AND DEMOGRAPHY 123

Four times as many women died in the 2004 Boxing Day tsunami as men because women were expected to stay behind to look for family members and were more likely to be home (and thus closer to the seafront) when the tsunami struck. These outcomes are directly tied to gendered social and economic expectations (MacDonald 2005). Gender equity also has direct effects on fertility rates, maternal mortality and infant and child mortality (Morgan et al. 2017). Consider the Zika outbreak in Central and South America: responding to the outbreak and the associated increase in microcephaly in children born to infected mothers intersected powerfully with gender, religion and class. Women in a number of countries were advised to simply avoid getting pregnant during the outbreak, yet many of the affected states placed strong restrictions on birth control, abortion and family planning services due to the influence of the Roman Catholic Church (Davies and Bennett 2016; Harris et al. 2016). When the institutional arrangements designed to address global health matters are unable to address these unequal effects, efficacy suffers.

Getting global health institutions right The global health governance landscape can be confusing because of the large number of actors involved and the challenges of coordinating among a wide range of organizations. As a result, it can be difficult to assess how well the current institutional arrangement is working to promote global health or to introduce effective changes. This means that the international community’s ability to address the intersection of health and demographics may be limited, and it may take longer to respond to newly emerging issues. The WHO’s experience and institutional structure are emblematic of many of these challenges. Founded in 1948 as a specialized agency of the United Nations, the WHO has a constitutional mandate “to act as the directing and co-ordinating authority on international health work” (World Health Organization 1948: 2). This includes organizing collective efforts to address disease outbreaks, serving as a repository and disseminator of best practices and data and facilitating international negotiations and forums on pressing health matters. That is a large and broad mission, and the WHO faces significant challenges fulfilling it for several reasons. First, rather than a single organization, the WHO encompasses both the main organization with its headquarters in Geneva and six regional organizations. These regional organizations can respond to more local conditions, but their high degree of autonomy means that their operations may conflict with the WHO’s priorities (Hanrieder 2014). Even within the WHO headquarters itself, the organization

124 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

functions more as a collective agent, with different parts responsive to different principals and oversight structures (Graham 2013). Second, the WHO may be the most prominent global health organization, but it is hardly the leader. It does not have a large number of service-providing staff that it can deploy into a health emergency like Doctors Without Borders. It does not have the financial resources of the World Bank, which is actually the largest source of development assistance for health (Leon 2015: 96–104). It can have formal relationships with non-state actors, but the process for opening such relations remains contentious and there remain serious debates over whether non-state actors are gaining too much influence over the organization (Richter 2014). In a world where a single philanthropic organization can (and does) spend as much on global health as an organization with a 70-plus year history and near-universal membership, it can be hard for the WHO to find its footing. The lack of an explicit hierarchy among global health governance organizations and the relative ease with which organizations can engage on these issues from both a policy and a service delivery perspective allows for a greater diversity, but it can also make it harder to coordinate responses and increases the risk of both duplication of services and overlooking key areas (Ng and Ruger 2011). Third, the WHO has a remarkably small budget and little control over how it allocates its resources. The organization’s biennial budget for 2018–19 is $4.42 billion – about 30 cents per person per year worldwide. To put that in perspective, Americans spent just over $6 billion on pet grooming and boarding in 2017 alone (Woien 2018). The WHO’s budget is also roughly what the Bill & Melinda Gates Foundation spends each year on its health-related initiatives in its Global Health and Global Development programs. This amount of money does not give the WHO much in terms of resources, and it leaves little extra for emergencies. Compounding the problem, the WHO lacks control over the vast majority of its budget. The WHO can decide for itself how to allocate the annual dues it receives from member states, but that only accounts for about 20 percent of its overall budget and member states have largely resisted calls to increase their dues. The remaining funds come from voluntary contributions from states and non-state sources – but those funds are predominantly pledged for specific projects. That means that the donor, rather than the WHO, is deciding how the WHO can spend the money, and donor priorities may or may not align with the WHO’s own assessment of where those funds would be most useful (Gostin et al. 2015). Finally, development assistance for health as a broad category has plateaued at a time when needs are increasing. In 2018, the total amount of development

HEALTH AND DEMOGRAPHY 125

assistance for health was $38.9 billion – a decrease of more than 3 percent from the previous year. This continues a trend over the past five years of generally steady or slightly decreasing funding, and reverses the double-digit annual increases the global community witnessed during the first decade of the twenty-first century (Institute for Health Metrics and Evaluation 2019). These figures capture a broader array of organizations and institutions than just the WHO, but they also point to a common shortcoming in funding. These issues weaken the ability of the global health governance architecture to respond to new crises or cooperate in a meaningful manner. The range of actors involved in global health is expanding, but that has not necessarily come with better structures for fostering cooperation. Global health is receiving unprecedented levels of attention from policymakers, but funding and collaboration have not always followed. This all matters when thinking about political demography because disease outbreaks do not strike populations equally. Given that we are seeing more new diseases and more diseases popping up in new areas, the need for a more robust institutional response is only getting stronger – at a time when the institutions themselves are not. The WHO has identified at least 40 new infectious diseases in humans that have emerged since the 1970s (World Health Organization 2007: x), and this represents a serious and significant increase in the number and frequency of human infectious disease outbreaks – even when controlling for factors like disease surveillance capabilities and geography (Smith et al. 2014). This increase appears to be related to factors like increasing urbanization, changes in land usage patterns that bring humans and animals into closer contact, population growth, growing antimicrobial resistance and the rate of international travel (O’Dowd 2007). These risks are not equally distributed, either globally or within individual societies, but current global health institutions are not well-equipped in the current moment to address these differential effects. The Covid-19 pandemic is exposing the differences in pandemic preparedness among different states, questions about which groups in society get access to testing and treatment and the weaknesses in our global institutions due to under-funding and restricted organizational agency.

Conclusion Health has obvious and profound implications for political demography, but the exact nature of these connections in the contemporary era remains underexplored. Complicating our collective understanding of these connections are

126 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

the non-linear and indirect relationships between disease and demography. The outbreak of a particular disease does not automatically mean that the government will fall, but it can exacerbate existing tensions, undermine previous weaknesses and place one burden too many on already-stretched services. The relationship between health and demography also intersects with some of the other most pressing challenges within the global community: climate change; the continued displacement of people from their homes; the role of women in political and social spheres; the changing nature of warfare and conflict; and the efficacy of international institutions created in the immediate aftermath of World War II. As a result, developing a fuller appreciation of the intersection between health and demography provides a fascinating window for understanding modern international relations more fully.

References Brown, M. E. and C. C. Funk (2008), ‘Food security under climate change’, Science, 319 (5863), 580–81. Caldwell, J. C. (1993), ‘Health transition: the cultural, social, and behavioral determinants of health in the Third World’, Social Science and Medicine, 36 (2), 125–35. Caprioli, M. (2005), ‘Primed for violence: the role of gender inequality in predicting internal conflict’, International Studies Quarterly, 49 (2), 161–78. Caprioli, M. and M. A. Boyer (2001), ‘Gender, violence, and international crisis’, Journal of Conflict Resolution, 45 (4), 503–18. Davies, S. E. and B. Bennett (2016), ‘A gendered human rights analysis of Ebola and Zika: locating gender in global health emergencies’, International Affairs, 92 (5), 1041–60. Davies, S. E., S. Harman, R. Manjoo, M. Tanyag and C. Wenham (2019), ‘Why it must be a feminist global health agenda’, Lancet, 393 (10171), 601–3. Frenk, J., J. L. Bobadilla, C. Stern, T. Frejka and R. Lozano (1991), ‘Element for a theory of the health transition’, Health Transition Review, 1 (1), 21–38. Gostin, L. O., D. Sridhar and D. Hougendobler (2015), ‘The normative authority of the World Health Organization’, Public Health, 129 (7), 854–63. Graham, E. R. (2013), ‘International organizations as collective agents: fragmentation and the limits of principal control at the World Health Organization’, European Journal of International Relations, 20 (2), 366–90. Hall, J. J. and R. Taylor (2003), ‘Health for all beyond 2000: the demise of the Alma-Ata Declaration and primary health care in developing countries’, Medical Journal of Australia, 178 (1), 17–20. Hanrieder, T. (2014), ‘The path-dependent design of international organizations: federalism in the World Health Organization’, European Journal of International Relations, 21 (1), 215–39.

HEALTH AND DEMOGRAPHY 127

Haque, S. E., A. Tsutsumi and A. Capon (2014), Sick Cities: A Scenario for Dhaka City, accessed April 8, 2020 at https://​iigh​.unu​.edu/​publications/​articles/​sick​-cities​-a​ -scenario​-for​-dhaka​-city​.html. Harman, S. (2016), ‘Ebola, gender, and conspicuously invisible women in global health governance’, Third World Quarterly, 37 (3), 524–41. Harris, L. H., N. S. Silverman and M. F. Marshall (2016), ‘The paradigm of paradox: women, pregnant women, and the unequal burdens of the Zika virus pandemic’, American Journal of Bioethics, 16 (5), 1–4. Heuveline, P., M. Guillot and D. R. Gwatkin (2002), ‘The uneven tides of the health transition’, Social Science and Medicine, 55 (2), 313–22. Hudson, V. M., B. Ballif-Spanvill, M. Caprioli and C. F. Emmett (2012), Sex and World Peace, New York: Columbia University Press. Institute for Health Metrics and Evaluation (2018), Financing Global Health 2017: Funding Universal Health Coverage and the Unfinished HIV/AIDS Agenda, Seattle, WA: Institute for Health Metrics and Evaluation. Institute for Health Metrics and Evaluation (2019), Financing Global Health 2018: Countries and Programs in Transition, Seattle, WA: Institute for Health Metrics and Evaluation. Japan Times (2018), ‘Heat wave sends mercury to all-time high of 41.1 in Saitama’, July 23, accessed August 12, 2019 at https://​ www​ .japantimes​ .co​ .jp/​ news/​ 2018/​ 07/​23/​national/​temperature​-tokyo​-hits​-nationwide​-high​-year​-mercury​-rises​-40​-8​ -western​-city​-ome/​#​.XVHPUi2ZNTY. Jutla, A., E. Whitcombe, N. Hasan, B. Haley, A. Akanda, A. Huq, M. Alam, R. B. Sack and R. Colwell (2013), ‘Environmental factors influencing epidemic cholera’, American Journal of Tropical Medicine and Hygiene, 89 (3), 597–607. Kjellstrom, T. and S. Mercado (2008), ‘Towards action on social determinants for health equity in urban settings’, Environment and Urbanization, 20 (2), 551–74. Kraemer, M. U. G., R. C. Reiner, Jr., O. J. Brady, J. P. Messina, M. Gilbert, D. M. Pigott, D. Yi, K. Johnson, L. Earl, L. B. Marczak, S. Shirude, N. D. Weaver, D. Bisanizio, T. A. Perkins, S. Lan, X. Lu, P. Jones, G. E. Coelho, Roberta G. Carvalho, Wim Van Bortel, Cedric Marsboom, Guy Hendrickx, Francis Schaffner, Chester G. Moore, Heinrich H. Nax, Linus Bengtsson, Erik Wetter, Andrew J. Tatem, John S. Brownstein, David L. Smith, Louis Lambrechts, Simon Cauchemez, Catherin Linard, Nuno R. Faria, Oliver G. Pybus, Thomas W. Scott, Qiyong Liu, Hongjie Yu, G. R. William Wint, Simon I. Hay and Nick Golding (2019), ‘Past and future spread of the arbovirus vectors Aedes aegypti and Aedes albopictus’, Nature Microbiology, 4, 854–63. Leon, J. K. (2015), The Rise of Global Health: The Evolution of Effective Collective Action, Albany, NY: SUNY Press. Liotta, G., M. C. Inzerilli, L. Palombi, O. Madaro, S. Orlando, P. Scarcella, D. Betti and M. C. Marazzi (2018), ‘Social interventions to prevent heat-related mortality in the older adult in Rome, Italy: a quasi-experimental study’, International Journal of Environmental Research and Public Health, 15 (4), 715. MacDonald, R. (2005), ‘How women were affected by the tsunami: a perspective from Oxfam’, PLoS Medicine, 2 (6), e178. Martin, G., C. Yanez-Arenas, C. Chen, R. K. Plowright, R. J. Webb and L. F. Skerratt (2018), ‘Climate change could increase the geographic extent of Hendra virus spillover risk’, Ecohealth, 15 (3), 509–25. McKeown, R. (2009), ‘The epidemiologic transition: changing patterns of mortality and population dynamics’, American Journal of Lifestyle Medicine, 3 (Suppl. 1), 19S–26S.

128 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

McMichael, C., J. Barnett and A. J. McMichael (2012), ‘An ill wind? Climate change, migration, and health’, Environmental Health Perspectives, 120 (5), 646–54. McPherson, P. (2015), ‘Dhaka: the city where climate refugees are already a reality’, The Guardian, December 1, accessed April 8, 2020 at https://​www​.theguardian​.com/​ cities/​2015/​dec/​01/​dhaka​-city​-climate​-refugees​-reality. Mercer, A. J. (2018), ‘Updating the epidemiological transition model’, Epidemiology and Infection, 146 (1), 680–87. Moore, M., P. Gould and B. S. Keary (2003), ‘Global urbanization and impact on health’, International Journal of Hygiene and Environmental Health, 206 (4–5), 269–78. Morgan, R., M. Tetui, R. M. Kananura, E. Ekirapa-Kiracho and A. S. George (2017), ‘Gender dynamics affecting maternal health and health care access and use in Uganda’, Health Policy and Planning, 32 (Suppl. 5), v13–v21. Ng, N. Y. and J. P. Ruger (2011), ‘Global health governance at a crossroads’, Global Health Governance, 3 (2), 1–37. O’Dowd, A. (2007), ‘Infectious diseases are spreading more rapidly than ever before, WHO warns’, BMJ, 335 (7617), 418. O’Manique, C. (2015), ‘Gender, health, and security’, in Simon Rushton and Jeremy Youde (eds), Routledge Handbook of Global Health Security, New York: Routledge, pp. 48–59. Omran, A. R. (1971), ‘The epidemiologic transition: a theory of the epidemiology of population change’, Milbank Quarterly, 49 (4), 509–38. Reliefweb (2019), ‘Kenya: drought, 2014–2019’, accessed July 11, 2019 at https://​ reliefweb​.int/​disaster/​dr​-2014​-000131​-ken. Richter, J. (2014), ‘Time to turn the tide: WHO’s engagement with non-state actors and the politics of stakeholder governance and conflicts of interest’, BMJ, 348, g3351, https://​doi​.org/​10​.1136/​bmj​.g3351. Rosenzweig, C., F. N. Tubiello, R. Goldberg, E. Mills and J. Bloomfield (2002), ‘Increased crop damage in the US from excess precipitation under climate change’, Global Environmental Change, 12 (3), 197–202. Sachs, J. (2012), ‘From Millennium Development Goals to Sustainable Development Goals’, Lancet, 379 (9832), 2206–11. Santosa, A., S. Wall, E. Fottrell, U. Högberg and P. Byass (2014), ‘The development and experience of epidemiological transition theory over four decades: a systematic review’, Global Health Action, 7, 23574. Sayeed, S. (2018), ‘Pakistan heatwave kills 65 people in Karachi – relief organization’, Reuters, May 21, accessed July 11, 2019 at https://​ www​ .reuters​ .com/​ article/​us​-pakistan​-heatwave/​pakistan​-heatwave​-kills​-65​-people​-in​-karachi​-welfare​ -organization​-idUSKCN1IM2AU. Schmidhuber, J. and F. N. Tubiello (2007), ‘Global food security under climate change’, Proceedings of the National Academies of Science, 104 (50), 19703–8. Sciubba, J. and J. Youde (2017), ‘Puerto Rico’s troubles are far from over: the population’s health is at risk’, Washington Post, October 13, accessed July 22, 2019 at https://​www​.washingtonpost​.com/​news/​monkey​-cage/​wp/​2017/​10/​13/​puerto​-ricos​ -troubles​-are​-far​-from​-over​-the​-populations​-health​-is​-at​-risk. Shiffman, J. and F. E. Okonofua (2007), ‘The state of political priority for safe motherhood in Nigeria’, BJOG: An International Journal of Obstetrics and Gynecology, 114 (2), 127–33. Shiffman, J. and S. Smith (2007), ‘Generation of political priority for global health initiatives: a framework and case study of maternal mortality’, Lancet, 370 (9595), 1370–79.

HEALTH AND DEMOGRAPHY 129

Smith, K. F., M. Goldberg, S. Rosenthal, L. Carlson, J. Chen, C. Chen and S. Ramachandran (2014), ‘Global rise in human infectious disease outbreaks’, Journal of Royal Society Interface, 11 (101), 20140950. St. Michel, P. (2018), ‘Soaring temperatures force Japan to confront entrenched ideas on handling the heat’, Japan Times, July 28, accessed August 12, 2019 at https://​www​ .japantimes​.co​.jp/​news/​2018/​07/​28/​national/​media​-national/​soaring​-temperatures​ -force​-japan​-confront​-entrenched​-ideas​-handling​-heat/​#​.XVHRGC2ZNTY. Sullivan, B. K. (2019), ‘Unending Midwest rains add to wettest 12 months ever in US’, Bloomberg, May 24, accessed July 11, 2019 at https://​www​.bloomberg​.com/​news/​ articles/​2019​-05​-24/​unceasing​-midwest​-rains​-add​-to​-wettest​-12​-months​-ever​-for​ -u​-s. UNICEF (2018), ‘Maternal health: current status and progress’, accessed July 22, 2019 at https://​data​.unicef​.org/​topic/​maternal​-health/​maternal​-mortality/​. United Nations (2000), ‘Goal 5: improve maternal health’, accessed July 22, 2019 at http://​www​.un​.org/​millenniumgoals/​maternal​.shtml. United Nations Development Programme (2018), ‘Goal 3 targets’, accessed July 22, 2019 at http://​www​.undp​.org/​content/​undp/​en/​home/​sustainable​-development​ -goals/​goal​-3​-good​-health​-and​-well​-being/​targets/​. Vallin, J. and F. Meslé (2004), ‘Convergences and divergences in mortality: a new approach to health transition’, Demographic Research, Special Collection 2 (2), 11–44. Walker, A. (2019), ‘Our cities are getting hotter – and it’s killing people’, Curbed, June 21, accessed July 11, 2019 at https://​www​.curbed​.com/​2018/​7/​6/​17539904/​heat​-wave​ -extreme​-heat​-cities​-deadly. Woien, S. (2018), ‘Americans spend $70 billion on pets, and that money could do more good’, The Conversation, October 15, accessed July 24, 2019 at https://​ theconversation​.com/​americans​-spend​-70​-billion​-on​-pets​-and​-that​-money​-could​ -do​-more​-good​-102467. World Health Organization (1948), Constitution of the World Health Organization, Geneva: World Health Organization. World Health Organization (1978), Declaration of Alma-Ata, accessed April 10, 2020 at https://​www​.who​.int/​publications/​almaata​_declaration​_en​.pdf. World Health Organization (2007), World Health Report 2007: A Safer Future, Global Public Health Security in the 21st Century, Geneva: World Health Organization. Zurleck-Brown, S., H. Newby, D. Chou, N. Mizoguchi, L. Say, E. Suzuki and J. Wilmoth (2013), ‘Understanding global trends in maternal mortality’, International Perspectives on Sexual and Reproductive Health, 39 (1), 32–41.

9

Population, rebellion and revolution

Jack A. Goldstone

Introduction: population as a driver of social change From the struggling peasantry of the French countryside in 1789, to the industrial workers crammed into the workers’ quarters of Moscow and St. Petersburg in 1917, to the unemployed youth in Cairo in 2010, changes in the numbers, location and conditions of population have combined to produce history’s major revolutions and rebellions. To be sure, revolutions do not arise from demographic factors alone. There is no critical level of population change at which things fall apart, no critical population density or growth rate that consistently determines revolutionary potential (de Sherbinen 1995). Revolutions have occurred throughout history in states with widely varying demographic trends, even in some cases when population has been stable for long intervals. When population changes have led to political crises, it has been because the interaction of population changes with existing political, social and economic institutions undermined the operation of those institutions. Both population and institutions matter. The resilience or vulnerability of institutions – that is, whether or not they can continue to channel resources and produce desired outcomes under changing demographic conditions – determines whether a given pattern of population change leads to crisis. In addition, the specific demographic changes, and the kinds of institutions with which they interact, will determine what kind of revolution or rebellion occurs. Leadership and contingencies matter as well. Even if population changes are undermining the viability of existing political institutions, such changes do not determine when a revolution will finally break out, or how successful it will be (Goldstone 2014). A clumsy incumbent who shreds his own legitimacy and 131

132 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

credibility will bring a more radical outcome; more determined and united revolutionary leaders are more likely to enjoy success. The timing of a revolution often depends on contingent events: the aging or death of a fragile ruler; a military defeat or economic setback; the exposure of egregious corruption; an act of extreme arrogance or excessive force by the ruler; or a small act of resistance that snowballs into mass protests, like a dust particle that seeds a cloud that rises into a thunderstorm. In sum, population change is but one factor that plays into the complex dynamics of revolutions and rebellions. Nonetheless, in the history of political upheavals, population change has often played a significant, even critical, role.

Classic population-driven revolutions and rebellions: Demographic Structural Theory Economic, political and social institutions do not go on forever of their own accord. Each new generation must start businesses and find jobs; if they do not, the economy falters. Each new generation must recruit new leaders and select those who will hold office; without widely accepted and followed rules for succession, political conflict ensues. Each new generation must grant elite social status to groups who will be thought and cultural leaders, business leaders and high military and bureaucratic officials. If the flow of qualified aspirants to elite positions is too small relative to available positions, organizations lose their capacity and anarchy follows; if the flow is too large, social mobility is blocked and brutal factional conflicts over status-conferring positions are likely to arise. To maintain stability over time, institutions need to be balanced with population change, so that the available social positions are filled in each new generation without excessive conflicts or shortfalls (Goldstone 2011). In the pre-industrial world – that is, in most human societies everywhere before 1800 – institutions were relatively static because they were designed for a world of very modest population change. With very high rates of mortality due to disease and injury, even the high birth rates of pre-industrial societies produced very slight population growth: on average less than one-tenth of one percent per year (Livi-Bacci 2017, Table 1.1). Under such conditions, heredity was a perfect tool for regulating access to jobs and to elite positions: as each man had, on average, only one surviving son, that son could take his place in his job or inherit the father’s rank or title. Surviving younger sons could marry someone from a family who had only daughters, and inherit from their father-in-law in like fashion. Over time, this very slow growth of population

POPULATION, REBELLION AND REVOLUTION 133

could be matched by an equal and slow growth of jobs, or of elite positions, as land was improved or towns grew and rulers gradually expanded their armies and courts. Taxes could be kept at a stable rate as well, as low population growth meant stable prices and relatively fixed expenditures. This was the situation for countless centuries, as princes replaced kings, emperors succeeded emperors, aristocratic elites reproduced with minor gains and losses and the common people toiled, mostly following their parents into farming or crafts. But every once in a while, favorable weather and a pause in the cycles of pandemic disease would allow the population to grow faster for longer periods: growth rates might rise to 1 percent per year, and sustain that rate for fifty to a hundred years. That would bring three to four generations, with each significantly larger than the one before. Total population might double, or nearly so, in just a few generations, a striking contrast to the normally imperceptible rate of change. Such population growth, given the initially favorable conditions, at first brought increased prosperity. More workers allowed a greater division of labor; a surplus of labor in the countryside led to migration to cities, which produced more high value and specialty goods. Trade between city and country increased as well. But by the second or third generation, the numbers of youth started to bulge beyond what the existing institutions were designed to manage. Larger youth cohorts add to political volatility in several ways. Youth are generally more available for mobilization for protest, given their lack of attachments, usually having no families to support or investments to protect. Youth are also generally more idealistic, and so more easily drawn to extreme or radical views. Youth frequently are concentrated in educational institutions or labor settings, encouraging the sharing of new ideas and facilitating joint actions. They are also most acutely aware of changing social mobility conditions, and more likely to be deeply resentful of any obstacles to fulfilling their aspirations for jobs, family and higher status positions. Large youth cohorts are also likely to drive demand for an expansion of education, including higher education, leading to a surplus of educated youth. For all of these reasons, large youth cohorts have historically brought higher risks of political violence and revolution (Moller 1968; O’Boyle 1970; Urdal 2006, 2007; Madsen et al. 2010). In these periods of relatively rapid population growth, land may grow scarce, pushing up rents. In mainly rural societies, such as pre-revolutionary France or late Tsarist Russia, population growth led to the division of properties into smaller parcels, and higher prices for families to rent additional land. In the

134 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

cities of developing countries, as across North Africa in the decades before the Arab Spring, as the supply of labor surges, real wages are likely to stagnate or fall, while unemployment (especially youth unemployment) spreads. Among elite families, more numerous surviving sons will strain status transmission by simple inheritance; they thus will agitate for an expansion of university places, government positions or even chances for military glory. Prices of food and other essential goods will rise; but if taxation is fixed in nominal terms, the finances of states, and of elites who depend on fixed rent rolls, will suffer. The burden is even worse for states because, as a population grows, so too do administrative expenses, demands for state-supported positions and the costs of military expansion. Whether we look at France and Russia struggling with the rising costs of war or at Egypt, Libya and Syria supporting large armies while also seeking to fund popular subsidies for bread and fuel, population growth can unbalance state budgets. The squeeze on state finances leads governments to try to raise taxes, seize assets or slash spending, all of which lead to resentment against the ruler, and questioning of the legitimacy of these actions. If the government cannot gain adequate revenues to meet its needs, then these problems are made worse by declining administrative capacity and by declining loyalty among those elites who feel left out or insufficiently rewarded. Population growth also leads cities to grow even faster, concentrating the ambitious and the poor. Struggling peasants and workers, as well as elites who feel shaken by factional conflicts or doubts about the regime, may turn to alternative religions or movements to make sense of these changes. Puritanism, Communism, Nationalism and radical Islamism all thrived among fast-growing populations whose youth saw little or no future in existing institutions. Popular alternatives to the conservative state-supported faiths are salvation religions that promise a better life to the poor, and ascetic religions that promise moral superiority in a chaotic world to those who practice discipline in their lives. This syndrome – governments struggling with debts and revenue shortages; elites divided and competing over status and rewards; workers facing falling real wages and peasants beset by rising rents; and a combination of elites and popular groups, many concentrated in cities, being drawn to heterodox faiths – provides fertile grounds for revolution and rebellions (Goldstone 2016). In these conditions, any number of events can trigger a mobilization of popular groups behind aggrieved members of the elite seeking to create a new political or social system with greater order, justice and stability. If the ruler is stubborn,

POPULATION, REBELLION AND REVOLUTION 135

unprepared or simply unable to marshal sufficient allies and resources, rebellion can grow and turn into revolution. This is the pattern observed in the two great waves of revolution and rebellion that swept across Europe and Asia in the wake of two distinct periods of population growth, first in the seventeenth and then in the late eighteenth and nineteenth centuries. A third great wave of revolutions arose in the mid-twentieth and early twenty-first centuries in the developing world, spanning Asia, the southern hemisphere and the Middle East, as rapid population growth came to these regions somewhat later but with similar effects. The cycles of population growth and revolution began to unfold as the world recovered from the ravages of the Black Death, which swept across Europe and Asia in the 1300s. From 1400 to 1600, population growth returned only slowly. But with recovery, most states grew richer and stronger up to 1600: the cities expanded and new trade routes linked Europe to the Old World and the New. Agriculture became more intensive, commercial farms and fishing fleets expanded and both craft guilds and landed elites (at least those astute at management or able to rise at court) became more numerous and wealthier. In Europe, the Renaissance in arts and literature spread northwards from Italy to Germany, France and Britain, accompanied by the rise of stronger monarchs with more literate and powerful administrations. In the Middle East, the Ottoman Empire reached the height of its expansion and glories. And in China the Ming Dynasty became the largest and most successful native dynasty in centuries, producing exquisite Ming “china” and silks and tea for much of the world. But from 1600 onwards, decay set in everywhere. Population began to press on land and resources, causing prices to rise sharply. With expenditures steadily exceeding their incomes, kings and emperors debased their currencies, sought to raise taxes and cut new deals with their elites. Elites, now overgrown in numbers, divided into factions seeking their own advancement and sought to protect their resources from grasping courts. Urban elites sought privileges and protection or asserted their independence; while urban workers and peasants drifted into wage protests or banditry, and were readily recruited into movements for religious and/or political change. The result, triggered in part by a sharp downturn in climate and in part by a series of religious wars, was a wave of revolutions and rebellions in the second quarter of the seventeenth century that even contemporaries remarked upon as astounding in their severity and breadth (Parker 2014). From the Jelali rebellions and Janissary revolts in the Ottoman Empire, to revolutions against

136 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Spanish rule in Portugal, Naples and Sicily, to the Frondes in France and the Puritan Revolution in England, then east to the Khmelnitsky revolt in Poland– Lithuania, governments were shaken all across Europe from the 1620s through the 1640s. In China, the Ming Dynasty was overthrown by bandit revolts that turned into a national revolution, and when the Ming armies were driven back to Beijing, Manchu forces from the north swept in to take over the empire. After 1660, population growth ended all across Eurasia. Plague returned in the sixteenth and seventeenth centuries, with particularly severe epidemics after 1660. Smallpox added to the devastation. Southern Europe was particularly hard hit. However, the effect of this slowdown was to allow economic growth to catch up to, or even overtake, population. Prices flattened or fell, especially for grains; state treasuries returned to health as tax revenues grew; the number of those who felt entitled to elite positions stabilized; and the conditions of working people and peasants stopped declining. States were generally successful in re-imposing state religions and orthodoxy. A century of political stability ensued. But, from the early eighteenth century, population growth returned. By the late eighteenth and early nineteenth centuries, the revolutionary syndrome had returned as well: prices were rising, states were falling into debt, elite factionalism had revived and intensified as channels of social mobility grew clogged and declines in real wages and increases in rents had undermined the position of workers and peasants. Starting in the 1770s, revolutions and rebellions again spread all across the continent, from Pugachev’s Revolt in Russia and the French Revolution to the Greek Independence Revolution in the 1820s and the revolutions of 1830 and 1848 in Hungary, Romania, Germany, France and Italy. In China, the Taiping Rebellion of the 1850s launched the largest civil war in world history in terms of casualties, starting a century of revolts and revolutions that destroyed the last imperial dynasty. These two waves are examined in detail in Goldstone (2016), where I laid out this “Demographic Structural Theory” of how population interacted with pre-industrial institutions to bring periodic political crises. The theory was given a more formal mathematical structure by Turchin (2018) and Turchin and Nefedov (2009), who showed how it could be generalized. Yet the effect of population on politics did not fade away after the nineteenth century. The same patterns can be found in the wake of sustained population growth in modern developing societies, especially since population growth in the twentieth and twenty-first centuries can be far more rapid than ever before (Acemoglu et al. 2020). Modern growth rates can lead to population doubling

POPULATION, REBELLION AND REVOLUTION 137

in twenty-five to thirty years. Where such growth collides with government institutions that are fragile due to corruption, administrative failures or succession crises, the result is often revolutions and rebellions. After World War II, in dozens of European colonies, the rapid growth of indigenous populations since 1900 had produced large cohorts of youth who were readily mobilized, and large numbers of aspiring elites (many educated in Europe or European-founded schools) who chafed at their exclusion from political and economic leadership in their countries. Facing colonial governments that had been weakened by the war, they started revolutions for independence that spread across Asia, Africa and Latin America. Anti-colonial revolutions took place in India, Indonesia and Vietnam in Asia; and in Ghana, Kenya, Tanzania and Algeria in Africa. During the Cold War, local dictators, often backed by foreigners, were overthrown by communist revolutions in China, Cambodia, Cuba and Nicaragua, among others. These revolutions thrived on the enthusiasm of youth and the resentments of educated elites against their rulers in a world in which the population of developing regions mostly doubled from 1900 to 1950, and tripled from 1950 to 2000 (Frankema and Jervin 2014). In the early twenty-first century, the Arab Revolts that spread rapidly from Tunisia to Libya, Egypt, Syria and Yemen were similarly underlain by a half-century of strong population growth across the region (Korotayev and Zinkina 2011). By the early 2010s, despite good economic performance, governments were financially stressed by their inability to continue providing subsidized bread, cooking oil and jobs for their burgeoning populations; the universities were overflowing with graduates who expected more opportunity than closed autocratic regimes could provide, and working classes and professionals alike were finding it hard to afford housing or build a better future for their families. Once revolution in Tunisia suggested that Middle Eastern autocrats were vulnerable, revolutions and rebellions broke out across the Arab world. In sum, waves of population growth undermined regimes whose institutions for access to political power, wealth, jobs and elite status worked reasonably well under stable or slow growth, but which could not keep up with a series of generations each notably younger and larger than the last.

138 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Negative growth revolutions: Black Death, Meiji restoration While cumulating population growth is usually the cause of major revolutions and rebellions, there are circumstances in which the reverse trend – a major fall in population – can also bring political upheaval. Such outcomes can arise when the fall in population is so great as to shift the balance of power between landlords and peasants, or affect the distribution of resources or the revenues of rulers. For example, in Western Europe before the Black Death, population growth had started to create falling real wages and land scarcity. But before a political revolt could occur, the Black Death more than restored the balance, making labor the scarce factor in production and weakening the power of landlords. By the late fourteenth century, peasant uprisings in Britain and France succeeded in changing the relations between lords and peasants, effectively ending serfdom and creating an economy based on peasant farming and long-term stable rents. In China, the Black Death so disrupted the economy of the Mongol empire that the dynasty collapsed, providing an opportunity for a native Chinese leader to overthrow the Mongols and establish the Ming dynasty. In Japan, there was little or no population growth from the sixteenth to the eighteenth centuries. While normally that would conduce to political stability, Japan’s Shogunal government had kept government tax collections fixed in physical payments of rice, unusual for a relatively urbanized and commercial society. As the economy’s output grew while population remained stable, rice became relatively more plentiful and thus less valuable, shrinking government revenues and leaving the ruler to become ever more deeply indebted to urban merchants. This weakened the Shogun’s government, leaving it vulnerable to overthrow by provincial governors in the nineteenth century, when incursions by Western powers undermined the legitimacy of the traditional regime (Huber 1990). Thus, in some cases, a major shift from the normal pattern of slow but positive population growth in the direction of no growth or population decline could also undermine key institutions and bring down political regimes.

POPULATION, REBELLION AND REVOLUTION 139

Color revolutions: the outcome of more mature populations The most common revolutions in history have been ideologically radical social revolutions. Such events – like the French, Russian and Chinese Communist Revolutions – have generally followed the pattern described above by the Demographic Structural Theory. That is, they have generally followed a period of cumulating population growth that produced a youth bulge, heightened factional conflicts, states with declining revenues and rising debts and peasants and workers with declining real incomes. Yet there has been another pattern as well. In countries with a stable population and more mature age structure (median age 33 years or higher), which are still under autocratic rule, pressures for political change also build. Cumulative strong population growth usually occurs because mortality declines, so that more children survive and grow up to form ever larger youth and adult cohorts. By contrast, a more stable and mature population arises when fertility declines to low levels, at about two children per woman. Such low fertility usually only arises when societies have experienced a period of improved education, urbanization and rising incomes. Societies that have gone through these changes usually have produced a stronger civil society, as educated people form networks and associations among themselves. Such societies become increasingly uncomfortable under the strong hand of an autocrat, preferring to allow the marketplace and their own choices to determine their lives. Cincotta (2017; see also Cincotta and Doces 2011) has found that almost 80 percent of societies with median age over 35 are democracies, usually as a result of a transition by major reform or revolutions occurring sometime after the society’s median age passed 25. Since 1980, the Soviet Union, most of Eastern Europe, Georgia, Armenia and Ukraine are examples of countries where more mature populations have rebelled against authoritarian rule. In such countries, even if youth play a leading role, the population as a whole is less prone to follow radical ideologies or turn to violence. Instead, the focus of the rebellion is on demanding accountability from governments, with large but peaceful protests to seek greater democracy. Such relatively non-violent but dramatic “color revolutions” have become more common in the twenty-first century, and they are found mainly (though not exclusively) in societies with more mature age-structures. As global population growth slows, and more countries develop mature populations (median age 35 or more), we should

140 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

expect to see fewer radical social revolutions. However, mature populations that remain under autocratic rule – as is the case today in Russia, Belarus, Cuba and China – are still likely to produce non-violent democratic revolutions in the future. To be sure, democratic revolution is not the only possible outcome in more mature, slow-population-growth societies. We have also seen such societies (including Russia, Hungary, Poland and, to some degree, the United Kingdom and United States) turn to nationalist populism/authoritarianism. This is driven by two factors: the slowdown in economic growth and productivity brought by population aging and stagnation, which has brought heightened inequality as business and elites have captured a larger portion of the modest GDP growth; and the surge in immigration as globalization and conflicts in developing countries have encouraged more job and asylum seekers to move abroad. Such surges of immigration have aroused economic and cultural anxieties in more developed countries with already high inequality and slow-growing real incomes, giving an opportunity for politicians with a nationalist and populist message to draw support (Goldstone and Diamond 2020). Whether more mature aged societies tilt toward democracy or populist authoritarianism in the coming years thus depends both on the degree to which such societies see themselves as under economic and cultural assault, and on whether democratic movements or populist authoritarians are best able to provide a sense of hope and security to their populations.

Population trends and future revolutions There are not one but two major population trends in the world today. One trend – of slow or no population growth – is found in the more developed countries of Europe, East Asia, most of South Asia and the Americas. In these countries, fertility has fallen to near or under 2.0 children per woman, and population growth has slowed greatly. Indeed, by mid-century most of these countries will have stable or shrinking populations. These societies are also rapidly aging, with their median age moving into the low to upper 30s. Many of these countries are already democracies, but if they are currently under autocratic rule, their maturing age structures suggest that they are likely to experience color revolutions at some point in the next one or two decades. Such revolutions will not arise until a trigger event occurs that demonstrates the vulnerability of the regime. But the overall trend is clear: autocratic regimes – including those like Russia’s and Hungary’s, which have become

POPULATION, REBELLION AND REVOLUTION 141

more autocratic in the recent surge of ethno-nationalist populism – are likely to change as their populations grow older and better educated. Indeed, all the remaining mature-population autocracies, including Belarus, Russia, China, Hungary and Cuba, seem likely to have color revolutions that will reshape their politics to become more democratic. As countries like Vietnam, North Korea, Algeria, Myanmar, Iran and other autocracies experience population aging, and their median age shifts into the upper 30s, they too will become strong candidates for peaceful protests to transform autocratic rule into more accountable regimes. The second major population trend is found in the less developed countries of sub-Saharan Africa, Central America, the Andean countries of South America, parts of the Middle East and North Africa (notably Egypt, Syria, Iraq and Yemen) and parts of Central Asia (Pakistan and Afghanistan). In these countries, fertility remains high, at three or more children per woman, and mortality has fallen, so that population growth is rapid, in excess of 2 percent per year. These countries are also experiencing rapid urbanization, expanding education and large youth cohorts. These countries are thus on a similar course to those North African and Middle Eastern countries that experienced the wave of Arab Revolts in 2010–11. These countries are developing toward the syndrome of revolutionary vulnerability observed in the Demographic Structural Theory above. In the coming decades they are thus likely to experience more radical and violent revolutions that produce more ideological regimes. The spread of Islamist radicalism, under the banner of al-Qaeda and other extremist groups, that has spread from the Middle East to the Sahel in the west and to South and Southeast Asia in the east, and which briefly established the radical Islamic State in Iraq and Syria, is a clear manifestation of this tendency. The Covid-19 pandemic of 2020 is unlikely to have a demographic impact on the scale of the Black Death, or even HIV-AIDS, as it appears that mortality is very low overall (5 percent or less) and young people are only weakly affected. However, the economic effects of the Covid-19 pandemic could be much greater. If the impact of the disease on the world economy is to slow globalization, reduce demand for imports in the richer countries and halt tourism, then the economic impact on the developing world could be devastating. Economic crises that undercut state finances, lead elites to scramble for wealth and position in a shrinking market and halt any gains for working men and women, are likely to lead to political upheavals in those states with young and growing populations. There is also a risk of interaction between these two trends. That is, if rapid population growth in Africa and parts of the Middle East, South Asia and

142 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Central America continues, and leads to growing waves of conflict and disorder (possibly amplified by periodic climate crises (Lustgarten 2020)), that could produce recurrent surges of job seekers and refugees attempting to migrate to the richer, more mature societies. This could in turn maintain or increase the tendency in the latter to nationalist populism, and thus tilt the trend in these countries away from democracy and toward populist authoritarian regimes (Goldstone and Diamond 2020). To sum up, population trends have been a major factor driving waves of revolution and rebellion across history. Such dynamics seem far from over. Indeed, the next few decades could see another global wave of revolutions. In the slow-growing richer autocracies, with fast aging and more mature populations, we are likely to see a number of color revolutions demanding a shift to democracy. In the faster-growing poorer countries, with large youth cohorts and rapid urbanization, we are likely to see a number of radical revolutions driven by frustrations with choked mobility, stalled or declining progress in living standards, and financially troubled states.

Future research The preceding pages may have made it appear that we are fairly certain of the effects of population change on the onsets and outcomes of revolutions and rebellions. We are not; many questions remain. Here, I list just a few. First, we have much to understand about how past downturns or interruptions to population growth affected societies, and why. In some countries, the Black Death had very different effects than in Western Europe, where societies shook off serfdom and progressed to the Renaissance after the pandemic. For example, in Egypt, where prosperity was built in part upon complex irrigation works to productively channel the Nile’s annual floods, the drastic loss of population made it impossible to devote enough labor to maintaining the system. The result was a sharp decline in productivity and prosperity, from which Egypt did not recover for centuries (Borsch 2009). Questions regarding how pandemics and population loss did or did not contribute to the collapse of the Roman Empire are many (Harper 2017; Haldon 2019). Second, we need to investigate how and whether the relatively young and fast-growing countries of today, especially in sub-Saharan Africa and the Middle East, can avoid a future of revolutionary and civil wars. Can fertility be brought down in time to moderate the impact of growing youth cohorts? Can

POPULATION, REBELLION AND REVOLUTION 143

education, jobs and decent opportunities for improving lives be provided for the existing very young population? Can fast-expanding cities be made safe, healthy and productive? Humanity has never before seen such rapid population growth and urbanization in low-income countries. Can political institutions and capacities be supported or adapted to cope with existing population patterns before such growth diminishes? Third, the nature and causes of color revolutions, and the apparent stability of authoritarian regimes in the remaining mature-age autocracies, need much study. Is the stability of the regimes in Russia, China, Belarus and Cuba as illusory as the stability of the autocracies of Tunisia, Libya and Egypt before the Arab Revolts? Or are they institutionally different and more durable? Fourth, we need to develop fresh ideas on how to maintain economic growth and political stability in aging countries. Economic slowdowns and repeated recessions have produced anger and distrust of government. When combined with cultural anxieties driven by immigration, these conditions have spurred a widespread growth of populism and right-wing extremism. Can aging countries manage immigration to help spur their economies instead? Or do they need to turn to robots or other social reorganization as they shift to a permanent new condition of older, slow-growing populations? Finally, how will future population changes interact with other global trends? How will climate change, if it produces steadily higher temperatures and worse droughts or floods, affect the stability of governments in the already vulnerable regions of Africa and the Middle East (Raleigh and Urdal 2007; Urdal 2008)? How will global and local migration, if it continues to expand, affect national politics? The world has always seen episodes of migration, many of which provoked political conflicts; but these were mostly local in scale (Weiner and Russell 2000). The exceptionally rapid population growth now occurring in poorer regions, and consequent political upheavals, perhaps in concert with climate disasters, may spur massive surges of refugees across borders, and from the global South to the global North. Such surges stemming from the Arab Revolts in 2011–15 have already given impetus to Islamist radicalism in the Middle East, and to anti-democratic populism in the West (Kaufmann, Chapter 6 in this volume; Raleigh et al. 2008; Goldstone and Diamond 2020). We should also note that Islam is now, in terms of population, the fastest growing religion in the world, while Christianity is losing followers in Europe (Pew Research Center 2015). Will this lead to new political clashes? The world is heading into new frontiers of climate, urbanization, population aging, African population growth and religious alignments. Each of these, and even more their combination, has potential to bring political disruption.

144 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Understanding how these dynamics have operated in the past, and whether they are likely to be similar or different in the future, will be critical to foreseeing and responding to revolutions and rebellions to come.

References Acemoglu, D., L. Fergusson and S. Johnson (2020), ‘Population and Conflict’, The Review of Economic Studies, 87 (4), 1565–604. Borsch, S. (2009), The Black Death in Egypt and England: A Comparative Study, Austin, TX: University of Texas Press. Cincotta, R. (2017), ‘The Age-Structural Theory of State Behavior’, Oxford Research Encyclopedia of Politics, accessed July 9, 2020 at https://​ oxfordre​ .com/​ politics/​ politics/​view/​10​.1093/​acrefore/​9780190228637​.001​.0001/​acrefore​-9780190228637​-e​ -327​?rskey​=​41bM5W​&​result​=​15. Cincotta, R. and J. Doces (2011), ‘The Age-Structural Maturity Thesis: The Impact of the Youth Bulge on the Advent and Stability of Liberal Democracy’, in J. A. Goldstone, E. P. Kaufmann and M. D. Toft (eds), Political Demography: How Population Changes are Reshaping International Security and National Politics, Oxford: Oxford University Press, pp. 98–116. de Sherbinen, A. (1995), ‘World Population Growth and US National Security’, Environmental Change and Security Project Report of the Woodrow Wilson Center, 1, 24–9. Frankema, E. and M. Jervin (2014), ‘Writing History Backwards or Sideways: Towards a Consensus on African Population, 1850–2010’, Economic History Review, 67 (4), 907–31. Goldstone, J. A. (2011), ‘A Theory of Political Demography: Human and Institutional Reproduction’, in J. A. Goldstone, E. P. Kaufmann and M. D. Toft (eds), Political Demography: How Population Changes are Reshaping International Security and National Politics, Oxford: Oxford University Press, pp. 10–28. Goldstone, J. A. (2014), Revolutions: A Very Brief Introduction, New York: Oxford University Press. Goldstone, J. A. (2016), Revolution and Rebellion in the Early Modern World: Population and Population Change and State Breakdown in England, France, Turkey and China 1600–1850, London: Routledge. Goldstone, J. A. and L. Diamond (2020), ‘Demography and the Future of Democracy’, Perspectives on Politics, 18 (3), 867–80. Haldon, J. (2019), ‘War, Food, Climate Change and the Decline of the Roman Empire’, Journal of Late Antiquity, 12, 422–65. Harper, K. (2017), The Fate of Rome: Climate, Disease and the End of an Empire, Princeton, NJ: Princeton University Press. Huber, T. (1990), The Revolutionary Origins of Modern Japan, Stanford, CA: Stanford University Press. Korotayev, A. V. and J. V. Zinkina (2011), ‘Egyptian Revolution: A Demographic Structural Analysis’, Entelequia: Revista Interdisciplinaria, 13, 139–69. Live-Bacci, M. (2017), A Concise History of World Population, 6th edn, Chichester: Wiley-Blackwell.

POPULATION, REBELLION AND REVOLUTION 145

Lustgarten, A. (2020), ‘The Great Climate Migration has begun’, New York Times Magazine, July 26, 2020. Madsen, E. L., B. Daumerie and K. Hardee (2010), The Effects of Age Structure on Development: Policy and Issue Brief, Washington, DC: Population Action International. Moller, H. (1968), ‘Youth as a Force in the Modern World’, Comparative Studies in Society and History, 10, 237–60. O’Boyle, L. (1970), ‘The Problem of an Excess of Educated Men in Western Europe, 1800–1850’, Journal of Modern History, 42, 471–95. Parker, G. (2014), Global Crisis: War, Climate Change and Catastrophe in the Seventeenth Century, New Haven, CT: Yale University Press. Pew Research Center (2015), ‘The Future of World Religions: Population Growth Projections, 2010–2050’, accessed July 9, 2020 at https://​www​.pewforum​.org/​2015/​ 04/​02/​religious​-projections​-2010​-2050/​. Raleigh, C., L. Jordan and I. Salehyan (2008), Assessing the Impact of Climate Change on Migration and Conflict, Washington, DC: Social Development Department, The World Bank. Raleigh, C. and H. Urdal (2007), ‘Climate Change, Environmental Degradation and Armed Conflict’, Political Geography, 26, 674–94. Turchin, P. (2018), Historical Dynamics: Why States Rise and Fall, Princeton, NJ: Princeton University Press. Turchin, P. and S. A. Nefedov (2009), Secular Cycles, Princeton, NJ: Princeton University Press. Urdal, H. (2006), ‘Clash of Generations? Youth Bulges and Political Violence’, International Studies Quarterly, 50 (3), 607–29. Urdal, H. (2007), ‘The Demographics of Political Violence: Youth Bulges, Insecurity and Conflict’, in Lael Brainard and Derek Chollet (eds), Too Poor for Peace? Global Poverty, Conflict and Security in the 21st Century, Washington, DC: Brookings Institution, pp. 90–100. Urdal, H. (2008), Demographic Aspects of Climate Change, Environmental Degradation and Armed Conflict, New York: United Nations Expert Group Meeting on Population Distribution, Urbanization, Internal Migration and Development. Weiner, M. and S. S. Russell (2000), Demography and National Security, Providence, RI: Berghahn.

10

A research agenda for youth policies and investments

John F. May

Introduction The youth represent the future workforce and human capital of a society. Therefore, it is important to examine what new research would be needed regarding this group in order to improve youth policies and investments. This research agenda would enable scholars and policymakers to assess the weight of the youth when studying the relationship between population change and politics, which is the focus of political demography. Youth hold tremendous weight in the world population, as 3.2 billion people – close to 41 percent of the world population – were under the age of 25 in 2020. Almost 90 percent of these young people live in less developed countries (United Nations 2019). This large demographic group is expected to continue to increase during the next decades because fertility levels are still high in many less developed countries, particularly in sub-Saharan Africa. In 2020, adolescents and young adults between the ages of 10 and 24 (a sub-group of the population under the age of 25) were estimated at 1.85 billion persons. This sub-group will also continue to increase during the next decades because fertility decline has been slower in less developed countries (United Nations 2019). However, the absolute number of young people as well as their percentage of the total world population will eventually decrease with the completion of the demographic transition – the shift from high crude birth and death rates to low crude birth and death rates – especially in less developed countries. Differentials in the age structure between more and less developed countries are striking. In 2025, according to the Medium variant of the 2019 United Nations projections, the age group 0–19 in more developed countries will represent only 3.4 percent of the world population. However, the same age group 147

148 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

0–19 in less developed countries will then represent 29 percent of the world population (United Nations 2019). Fertility decline during the last phase of the demographic transition will also change the age structure toward relatively fewer young dependents and relatively more active and employed adults, again mostly in less developed countries. This shift would enable these countries to possibly capture a first demographic dividend, which is an economic surplus generated by a change in the age structure. Given the current and future weight of youth in the global population, youth issues have mobilized the attention of policymakers and researchers around the world. However, new challenges are emerging that will also determine the prospects of young people, namely the destruction of traditional jobs due to new technologies, the disruptions brought about by the emergence of new epidemics (e.g., the coronavirus Covid-19) and the potential impact of climate change. Consequently, it appears necessary to reassess past research on youth and young people and chart out a future youth research agenda. This chapter will first examine the definitions and demographics of youth and youth sub-groups as well as data access. Thereafter, the chapter will turn to the challenges that young people must confront, namely the building of their human capital, their access to jobs, their contribution to capturing a first demographic dividend and their linkages with national security issues. Finally, the chapter will analyze the assets and opportunities that youth could benefit from, namely the transition from youth to adulthood and citizenship, as well as the achievement of the Sustainable Development Goals (SDGs). Each of the three main sections of the chapter will examine the research gaps and discuss the additional and/or complementary research topics needed to inform future youth policies and investments.

Youth definitions and data access Which people belong to the group that is labeled “youth”? The definition of youth may vary with circumstances, especially with changes in the demographic, economic and sociocultural settings. For statistical purposes, the United Nations defines as youth persons aged 15 to 24 inclusive (United Nations n.d.). The World Health Organization (WHO) defines adolescents as people between 10 and 19 years of age. Combining adolescents and youth, the

A RESEARCH AGENDA FOR YOUTH POLICIES AND INVESTMENTS 149

WHO defines as young people persons between the ages of 10 and 24 (World Health Organization n.d.). More traditionally, demographers have defined the youth group as people from age 0 to 15 exactly, i.e., below the age of 15. However, this definition appears to be too narrow because more and more young people are engaged in secondary studies beyond age 15, pushing the boundaries of the youth group to at least age 18 or 19. Furthermore, the insertion of young people into the workforce may well go beyond the age of 20 and drag on until age 25 or even longer. This means that the age boundaries of the youth group could possibly range from age 0 to around age 30. The African Youth Charter pushes the upper limit of the youth group to age 35 (United Nations n.d.). Young people evolve through different crucial development stages, going from early childhood to puberty, and thereafter from critical years of formation and training to young adulthood. Therefore, the broad group of young people from 0 to 30 or 35 may be split into several youth sub-groups, although these sub-groups are not always well defined. The transitions between children, adolescents and young adults remain rather blurry. Are adolescents reaching the status of young adults at, say, age 18? Would the group of young adults extend well beyond age 18 or 20, and comprise individuals up to the age of 25 or even 30 or 35? How we define youth age groups matters for political demography because various age groups are connected to particular social, economic or political outcomes. A case in point is the definition of “youth bulges”, when a high proportion (40 percent or more) of the population is 15–29 years old, relative to the adult population above age 15 (May 2012). However, even definitions of the youth bulge may vary, which is an issue because such definitions undergird political demography theories. In several regions of the world (e.g., sub-Saharan Africa and the Middle East), some youth sub-groups may be more at risk of being involved in armed conflicts (Sciubba 2010). Another important factor affecting youth outcomes is the weight of some youth sub-groups relative to the rest of the population. For instance, when they were young, the US baby boomers experienced declining job prospects linked to the large size of their cohort relative to their parents’ generation (Easterlin 1980). Adolescent girls in less developed countries may require specific attention. Mensch and colleagues (1998) have proposed to focus research on girls at menarche and then again on girls married before or around age 18, the age

150 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

when they might have their first child. However, the World Bank (2006), in the seminal World Development Report 2007: Development and the Next Generation, has placed emphasis on the age group 12–24, seen as the crucial years of formation leading to young adulthood and citizenship. Finally, a number of studies have focused recently on child marriage,1 particularly in sub-Saharan Africa (Wodon et al. 2018). As can be seen, various definitions of young adults are proposed in the literature, but most data are available with rigid upper boundaries at 15 or 20 years, making it challenging to have a more nuanced assessment of youth and young adult issues across various age groupings and socioeconomic contexts. When looking at demographic and health indicators, one notices that there is a dearth of information on young people, especially girls, before age 15. Unfortunately, the Demographic and Health Surveys (DHS) and the Multiple Indicators Cluster Surveys (MICS) do not capture information on individuals before this age, despite the fact that many individuals may be sexually active between age 10 and 14. In some countries, girls are married and sometimes even give birth before age 15 (Mensch et al. 1998). Better and more detailed data are clearly needed. In this respect, the Population Council’s Girl Innovation, Research, and Learning (GIRL) Center launched the open access Adolescent Data Hub (ADH) in 2018. It is the first and largest data catalog specifically developed to focus on data on adolescents in low- and middle-income countries (LMICs), including more than 750 data sources. This should enable researchers to answer questions related to the lives of adolescents and also help donors and policymakers to identify gaps in existing data to inform future youth policies and investments (White et al. 2020). In addition, as researchers reexamine and possibly improve the definitions of youth sub-groups, they should consider inputs from various disciplines (e.g., anthropology, sociology, demography, economy) and combine quantitative and qualitative data. Moreover, such a youth grouping framework would probably need to be adapted for both the more and the less developed countries, as well as for the different regions of the world. The framework should also be differentiated according to the various socioeconomic situations (e.g., the wealth quintiles) and cultural settings.

A RESEARCH AGENDA FOR YOUTH POLICIES AND INVESTMENTS 151

Youth training and challenges The preceding section discussed the importance of evaluating youth definitions and of data access. In the following sections, we turn to implications of youth populations for economic development and national security. However defined, adolescence and transitions to adulthood and citizenship typically bring challenges and opportunities for the state and society, even across cultural and regional contexts. Human capital formation and access to jobs Human capital is necessary for countries to develop, achieve their socioeconomic goals and remain innovative. Human capital can be defined as the knowledge, skills and health (among other assets) that young people accumulate during their training years and beyond. This foundation enables young people to perform labor and produce economic value, thereby realizing their potential as economically productive members of the society. Developing and preserving human capital can also facilitate the eradication of extreme poverty and the creation of more inclusive societies. Human capital formation is crucial during the early years of life but the process of human capital accumulation and preservation plays out during the entire life. This requires that governments invest in their young people through nutrition, health care, quality education, jobs and skills, social services and women’s empowerment. Moreover, formation and preservation of human capital are necessary to prepare youth for the more highly skilled jobs of the future, in order to compete more effectively in the global economy.2 The task of youth formation and the challenges related to human capital formation and job creation are crucial given the current weight and the expected increase of the youth population in coming decades. A large body of research has been devoted to the formation of human capital needed for the capturing of a demographic dividend. In the case of sub-Saharan Africa, for instance, research has shown that the region will need to overcome gaps in education and training in order to benefit from a demographic dividend. Sub-Saharan Africa is lagging behind the other regions of the world (e.g., East Asia) particularly in terms of young adult educational attainment and productivity (Eberstadt 2017). However, more research is needed to assess the youth educational mixes and thresholds that are necessary for capturing a demographic dividend.

152 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Another critical challenge for youth is to acquire the skills necessary for the economy of tomorrow. Educating the future workforce will require increasing school enrollment at all levels, including technical and vocational education and training (TVET), and enhancing the quality of education. Therefore, countries will need to reassess and strengthen their education strategies to equip young people with the right skills and training, including exposure to new technologies, especially the digital technologies of the Fourth Industrial Revolution (4IR).3 So far, too little is known about the ways and means to translate education curricula into skills that are applicable for the day-to-day jobs of the economy of tomorrow. Future research will need to be based on more extensive fieldwork and grassroots studies. To be successful, this research will also need to address the issues of women’s empowerment and gender equity. Finally, considerable efforts will be needed in the area of data collection. A major challenge will be how to measure the efforts needed to build human capital. In this respect, the World Bank (2018) has proposed a new Human Capital Index (HCI). This tool is built around five indicators: child survival to age 5, expected years of schooling, harmonized learning outcomes, adult survival rate from age 15 to 60 and percentage of children under 5 not stunted.4 Four of these five indicators pertain specifically to youth. The HCI value ranks between zero and one. For example, a value of 0.5 for a country means that productivity in that country is only 50 percent of what it could be. Countries can use the HCI to assess how much income they are missing because of human capital gaps, and how much faster they can turn these losses into gains, if they act right away. To conclude, the creation of new jobs will remain a daunting challenge in many less developed countries. In these countries, many young adults enter jobs in the informal sector because opportunities are not readily available in the formal sector (Filmer and Fox 2014). Labor markets are linked to various levels of economic development (Sullivan 2019). However, labor markets can be improved and/or made more flexible with appropriate policy reforms. Contribution to capturing a first demographic dividend Governments of less developed countries will need to muster the potential contribution of youth to capture a first demographic dividend. In short, a first demographic dividend can be defined as an economic surplus generated during a period of accelerated economic growth.5 In the 1960s and 1990s, several countries in East Asia experienced a first demographic dividend. This East Asian economic boom was triggered to a large extent (estimated at about 40 percent) by a significant and rapid fertility decline, which modified the age

A RESEARCH AGENDA FOR YOUTH POLICIES AND INVESTMENTS 153

structure. Young dependents became relatively less numerous whereas active adults became relatively more numerous (Groth and May 2017). Today, a key public policy question is how policymakers and their development partners could replicate this process in the less developed countries, particularly in sub-Saharan Africa (World Bank 2015; Groth and May 2017). The African Union (AU) decided to adopt the theme “Harnessing the Demographic Dividend through Investments in Youth” for the year 2017, demonstrating both an overall policy objective for the region and the preeminence of youth in the AU development agenda (African Union 2016; see also Chapter 11, this volume). The theory of the first demographic dividend posits that it is necessary to decrease (relative to adult workers) the number of young dependents below age 15 and build and preserve the human capital of young adults aged 10 to 19. Some researchers have highlighted the great potential for young populations to capture a first demographic dividend. Arab and African youth is often seen as “making development possible” (Momani 2015; Sippel et al. 2011). However, the capturing of a first demographic dividend appears to be more complex than just having a large youth population. In particular, it requires improving the demographic, employment and socioeconomic dependency ratios.6 This goal cannot be achieved without a rapid and sharp fertility decline, which will transform the bottom of the population pyramid from a pyramidal to a rectangular shape. Nonetheless, youth appears to be a central element in the process of capturing a first demographic dividend, provided it can benefit from the human capital formation and preservation (especially education and health) that will enable them to access remunerated jobs in the modern economy. The linkages between the youth and the capturing of a first demographic dividend will need more attention because there are still many research gaps. One key question is how to effectively mobilize the contributions of youth to capture a first demographic dividend (African Union 2016). Another research area would be to explore the multiple benefits of a first demographic dividend beyond economic growth. The demographic dividend also helps improve child survival prospects, educational outcomes and political stability. However, the chain of causality between these factors and the demographic dividend needs more analysis (Population Reference Bureau 2018). Youth and national security Large youth populations in low-income countries often imply high dependency ratios and saturated labor markets, with soaring un- and under-employment rates. These situations, known as youth bulges (as mentioned), are especially

154 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

challenging in many less developed countries with deteriorating economies that are dependent on primary commodities, along with low incomes and unequal wealth distribution. Having a high proportion of discontented youths with extensive grievances and few alternatives, especially in an unequal society, can motivate violence and insurgency (Collier et al. 2003). Overall, young populations can then function as a stressor of grievances and potential conflict (Sciubba 2010). Recent research has highlighted the impact of youth on the security situation of societies; in particular, it has examined the linkages between the proportion of youth, on the one hand, and the possibilities for countries with numerous youth to experience unrest and violence, on the other (Cincotta et al. 2003; Urdal 2012). Leahy and colleagues (2007) have established that, between 1970 and 1999, 80 percent of all civil conflicts, defined as events causing at least 25 deaths, occurred in countries in which 60 percent or more of the population was under age 30. They distinguished four main types of age structures, namely very young, youthful, transitional and mature. During the 1990s, countries with a very young age structure were three times more likely to experience civil conflict than countries with a mature age structure. Moreover, the potential negative impact of the youthfulness of the population is also related to the countries’ autocratic or weakly democratic governments (Leahy et al. 2007). Although researchers have not been able to establish a causal link between a young age structure and civil unrest outcomes – and more empirical research is needed on this aspect – they claim that “age structures yield insights into many of the political, economic and security challenges that countries face, now and in the future” (Leahy et al. 2007: 9). More generally, researchers have attempted to explore the linkages between conflicts and the demographic transition. Rapid demographic transitions in less developed countries have generated situations of youth bulges, which, as mentioned, have occasionally led to conflicts. The swiftness of the demographic transition appears to be a crucial factor here because it creates an upsurge of youth in the society that cannot be absorbed rapidly enough through traditional socioeconomic policies and programs (Rustad et al. 2017; see also Chapter 9, this volume).

Youth assets and opportunities To accomplish their training, address their challenges and fulfill their potential, young people can rely on their assets and benefit from opportunities.

A RESEARCH AGENDA FOR YOUTH POLICIES AND INVESTMENTS 155

Appropriate policies can help them strengthen their decision-making capabilities and transition smoothly to the status of adults and citizens. In addition, international development frameworks, such as the Sustainable Development Goals (SDGs), can help support building young people’s human capital, which should enable societies to avoid national security pitfalls and capture a first demographic dividend. Transition from youth to adults and citizens In recent decades, international declarations and resolutions have recognized the importance of having children and youth engaged in the decision-making process pertaining to their own future. This was highlighted in the United Nations Convention on the Rights of the Child, adopted in 1989 and echoed in the International Conference on Population and Development (ICPD) held in Cairo, Egypt in 1994. The ICPD Conference specifically focused on provisions addressing young women’s sexual and reproductive rights, including the access of adolescents to contraceptives (Mburu et al. 2013). The Sexual and Reproductive Health and Rights (SRHR) agenda has been reaffirmed over time, more recently at the ICPD+25 Conference in Nairobi, Kenya in 2019. The transition of young people from dependents to citizens and decision-makers is perhaps one of the most complex aspects of the process of personal development. Young people between the ages of 12 and 24 seek identity and independence. They make decisions that affect not only their own present and future well-being, but also that of others. Moreover, they often do this in rapidly changing demographic and socioeconomic environments, especially in less developed countries. As youth transition to adulthood, they also transition to work, citizenship, marriage and parenthood, assuming of course that they have benefited from education and health services prior to undergoing these four transitions (World Bank 2006). For each dimension and/or transition from youth to adults and citizens, there are risks and opportunities. For all dimensions and transitions, policies, institutions and investments do matter. Development policies need to support young people’s transitions to adulthood. Education systems need to be designed to prepare young people to cope with the increasing demands of changing economies. Access to remunerated jobs is the overarching policy goal that must inform all youth policies and investments. With respect to the transition from youth to adulthood and work life, a number of questions warrant further research. Among these are: What can be done to help youth avoid serious consequences of risky behavior, such as

156 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

death from HIV/AIDS, drug abuse and accidents from risky sports? What is the influence of peers, intergenerational relationships and sociocultural settings on youth reproductive outcomes? Can the youth’s creative energy be directed productively to support development thinking? What kind of support do young people get as they enter the labor market? Can they move freely to where jobs are available? In particular, the transitions to adulthood will require more explanatory work, especially on the determinants of these trends. This will be achieved by “building multidisciplinary teams, following cohorts over prolonged periods of time, and measuring a full range of social, psychological, health, and economic outcomes while deploying a mix of research methods” (National Research Council and Institute of Medicine 2005: 12). International development frameworks have integrated specific interventions to support the formation and empowerment of youth. Within these frameworks, youth-oriented interventions have been linked to all development sectors in a holistic and synergistic way. This includes both the Millennium Development Goals (MDGs), which covered the years 2000 until 2015, and the SDGs, encompassing the years 2015 until 2030. Out of the eight MDGs, seven targeted youth, at least indirectly (World Bank 2006). In contrast to the MDGs approach, the SDGs shifted the policy focus to secondary education, the goal being that all boys and girls complete a full course of primary and secondary education. The goal of SDG 4: Provide Quality Education is to ensure inclusive and equitable quality education and promote lifelong learning opportunities for all (see particularly Targets 4.1, 4.3 and 4.4).7 In addition, several SDG goals (e.g., Goal 1: Eliminate Poverty; Goal 2: Erase Hunger; and Goal 3: Establish Good Health and Well-Being) are also designed to help improve youth outcomes (World Bank 2017). An important research topic would be to examine the youth-related outcomes of the MDGs in order to design and implement policies and investments to better address the needs of youth. An examination of the weight of the youth in the achievement of the SDGs is also warranted, because the fulfillment of the SDGs would help several less developed countries to build their human capital, capture a first demographic dividend and eventually reach the status of emerging market economies.

Conclusions The large and growing number of youth and young people has rekindled interest in the research agenda for youth policies and investments at

A RESEARCH AGENDA FOR YOUTH POLICIES AND INVESTMENTS 157

both the national and global levels. Addressing issues affecting youth has become a global concern beyond the specific situation of girls and young women. Adolescents have also become a key policy priority group (May 2012; Rotenberg forthcoming). Adolescence and the transition to young adulthood are important times during which defining experience is gained through education, peers and relationships and employment. Yet it is also a period when youth are susceptible to unsafe sex, unplanned pregnancy, high un- and underemployment, violence and the inability to meet their potential or aspirations (Gribble 2010). In recent decades, there has been a push for policies and investment in youth and adolescents as a way of improving development outcomes (Gribble 2010). Fundamental elements to empower youth and adolescents include: education, ensuring sufficient and adequate employment for the young population, protecting adolescents from gender inequity and harmful practices and ensuring access to satisfactory health services. In some less developed countries, the young population structure and rapidly growing population make these conditions more difficult to meet. In these countries, there is often an insufficient supply of jobs for large cohorts steadily entering the workforce (International Labour Organization 2004) and jobs are usually more difficult to create in the formal sector (Filmer and Fox 2014). An enhanced research agenda on youth will need to look more closely at the determinants of a number of youth and young adults’ outcomes and features. Required analyses should also examine cross-cutting issues, for example the linkages between youth and the capturing of a first demographic dividend as well as the overarching policy goal to create jobs for the numerous young adults that will enter the job market. Last, but not least, future youth and adolescent research will need to be multi-disciplinary and combine quantitative and qualitative approaches.

Notes 1. 2. 3.

Child marriage is a custom whereby parents, guardians or tutors entice boys and girls to enter into a formal marital union before the age of 18. Child labor remains a major problem in many less developed countries. In addition to several other negative impacts, child labor deprives very young people of key education opportunities (Kielland and Tovo 2006). The concept of the 4IR covers the new technologies (for instance, artificial intelligence, 3D printing, quantum computing, robotics, internet, telecommunications,

158 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

biotechnology, nanotechnology and energy storage) and how these new technologies transform the way we live, work and interact with each other (Schwab 2017). 4. The World Bank is currently considering adding other indicators to its HCI. 5. In theory, it is possible to capture a second demographic dividend if the benefits of the first demographic dividend have been properly saved and invested (Lee and Mason 2006). 6. The demographic dependency ratio is the number of people aged 0 to 14 and 65+, multiplied by 100 and divided by the number of people aged 15 to 64; the employment dependency ratio is calculated similarly, but takes into account employed or unemployed status; and the socioeconomic dependency ratio is the number of people whose income is lower than consumption times 100, divided by the number of people whose income is greater than consumption (Groth et al. 2019). 7. See https://​sdgs​.un​.org/​goals/​goal4.

References African Union (2016), AU Roadmap on Harnessing the Demographic Dividend through Investments in Youth. In response to AU Assembly Decision (Assembly/AU/Dec.601 (XXVI) on the 2017 theme of the year, Addis Ababa: African Union Commission. Cincotta, R. P., R. Engelman and D. Anastasion (2003), The Security Demographic: Population and Civil Conflict after the Cold War, Washington, DC: Population Action International. Collier, P., L. Elliot, H. Hegre, A. Hoeffler, M. Reynol-Querol and N. Sambanis (2003), Breaking the Conflict Trap: Civil War and Development Policy, Washington, DC: World Bank/Oxford University Press. Easterlin, R. A. (1980), Birth and Fortune: The Impact of Numbers on Personal Welfare, New York: Basic Books. Eberstadt, N. (2017), ‘Manpower, Education, Skills and Jobs in Sub-Saharan Africa: Past Trends and Future Outlook’, in H. Groth and J. F. May (eds), Africa’s Population: In Search of a Demographic Dividend, Cham: Springer, pp. 225–50. Filmer, D. and L. Fox (2014), Youth Employment in Sub-Saharan Africa, African Development Series, Washington, DC: World Bank. Gribble, J. (2010), Investing in Youth for National Development (Policy Brief January 2010), Washington, DC: Population Reference Bureau. Groth, H. and J. F. May (eds) (2017), Africa’s Population: In Search of a Demographic Dividend, Cham: Springer. Groth, H., J. F. May and V. Turbat (2019), ‘Policies Needed to Capture a Demographic Dividend in Sub-Saharan Africa’, Canadian Studies in Population, 46 (1): 61–72. International Labour Organization (2004), ‘Youth Unemployment at All-Time High’, World of Work: The Magazine of the ILO, 52: 21–3. Kielland, A. and M. Tovo (2006), Children at Work: Child Labor Practices in Africa, Boulder, CO: Lynne Rienner. Leahy, E., R. Engelman, C. Gibb Vogel, S. Haddock and T. Preston (2007), The Shape of Things to Come: Why Age Structure Matters to a Safer, More Equitable World, Washington, DC: Population Action International. Lee, R. and A. Mason (2006), ‘What is the Demographic Dividend?’, Finance and Development, 43 (3): 16–17.

A RESEARCH AGENDA FOR YOUTH POLICIES AND INVESTMENTS 159

May, J. F. (2012), World Population Policies: Their Origin, Evolution, and Impact, Dordrecht: Springer. Mburu, G., I. Hodgson, A. Teltschik, M. Ram, C. Haamujompa, D. Bajpai and B. Mutali (2013), ‘Right-based Services for Adolescents Living with HIV: Adolescents Self-Efficiacy and Implications for Health Systems in Zambia’, Reproductive Health Matters, 21 (41): 176–85. Mensch, B. S., J. Bruce and M. E. Greene (1998), The Uncharted Passage: Girls’ Adolescence in the Developing World, New York: The Population Council. Momani, B. (2015), Arab Dawn: Arab Youth and the Demographic Dividend They Will Bring, Toronto: University of Toronto Press. National Research Council and Institute of Medicine (2005), Growing Up Global: The Changing Transitions to Adulthood in Developing Countries, Panel on Transitions to Adulthood in Developing Countries, Cynthia B. Lloyd (ed.). Committee on Population and Board on Children, Youth, and Families, Division of Behavioral and Social Sciences and Education, Washington, DC: The National Academies Press. Population Reference Bureau (2018), The Four Dividends: How Age Structure Change Can Benefit Development, accessed July 14, 2020 at https://​www​.prb​.org/​the​-four​ -dividends​-how​-age​-structure​-change​-can​-benefit​-development/​. Rotenberg, S. (forthcoming), ‘Priority Groups in Population Policies’, in J. F. May and J. A. Goldstone (eds.), International Handbook of Population Policies, Cham: Springer. Rustad, S. A., G. Østby and H. Urdal (2017), ‘Conflicts and the Demographic Transition: Economic Opportunity or Disaster?’, in H. Groth and J. F. May (eds), Africa’s Population: In Search of a Demographic Dividend, Cham: Springer, pp. 483–96. Schwab, K. (2017), The Fourth Industrial Revolution, New York: Crown Business. Sciubba, J. D. (2010), The Future Faces of War: Population and National Security, Santa Barbara, CA: Praeger. Sippel, L., T. Kiziak, F. Woellert and R. Klingholz (2011), Africa’s Demographic Challenges: How a Young Population Can Make Development Possible, Berlin: Berlin Institute for Population and Development. Sullivan, T. A. (2019), ‘Demography of the Labor Force’, in D. L. Poston, Jr. (ed.), Handbook of Population, 2nd edition, Cham: Springer, pp. 263–78. United Nations (n.d.), Definition of Youth, New York: United Nations Department of Economic and Social Affairs (UNDESA), accessed July 14, 2020 at https://​www​.un​ .org/​esa/​socdev/​documents/​youth/​fact​-sheets/​youth​-definition​.pdf. United Nations (2019), World Population Prospects: The 2019 Revision, New York: United Nations, Department of Economic and Social Affairs, Population Division. Urdal, H. (2012), ‘Youth Bulges and Violence’, in J. A. Goldstone, E. P. Kaufmann and M. D. Toft (eds), Political Demography: How Population Changes are Reshaping International Security and National Politics, Oxford: Oxford University Press, pp. 117–32. White, C., C. Misunas, E. Chuang, M. de Almada, S. Psaki and T. D. Ngo (2020), ‘The Adolescent Data Hub: The Largest Catalog of Open Access Data on Adolescents Living in Low- and Middle-Income Countries’, Studies in Family Planning, 51 (1): 103–15. Wodon, Q. T., C. Male, C. E. Montenegro, H. Nguyen and A. O. Onagoruwa (2018), Educating Girls and Ending Child Marriage: A Priority for Africa, The Cost of Not Educating Girls Series, Washington, DC: World Bank. World Bank (2006), World Development Report 2007: Development and the Next Generation, Washington, DC: World Bank.

160 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

World Bank (2015), Africa’s Demographic Transition: Dividend or Disaster? Africa Development Forum, Washington, DC: World Bank. World Bank (2017), Atlas of Sustainable Development Goals 2017: World Development Indicators, Washington, DC: World Bank. World Bank (2018), The Human Capital Project, Washington, DC: World Bank. World Health Organization (n.d.), Health for the World’s Adolescents. Geneva: World Health Organization, accessed July 14, 2020 at https://​apps​.who​.int/​adolescent/​ second​-decade/​section2/​page1/​recognizing​-adolescence​.html.

11

Demography and democracy

Hannes Weber

Introduction After 14 different chancellorships in 14 years and numerous attempted coups d’états by radical political factions, democracy finally collapsed in Weimar Germany in 1933. Today, whenever political commentators feel that German democracy is in danger, they warn about “Weimar conditions”. But, while some “Weimar conditions” could possibly re-emerge – e.g., high unemployment, strong political polarization and support for populist parties – others clearly cannot. One of the factors that have irreversibly changed since the 1920s and 1930s is the demographic makeup of Germany and most other European countries. In 1933, the largest age cohort in Germany was young people aged 25–29 years old, followed by the age group 20–24 years (cf. Weber 2019, p. 31). Those were the people born between 1900 and 1914, when the total fertility rate (TFR) was still around 4.5 children per woman, and these cohorts did not fight in the First World War. They were, both in absolute numbers as well as relative to the total population, the largest generation in German history. Compare this to present-day Germany. By far the largest cohort is the age group 55–59 years. The 20–24-year-olds are only almost half as numerous. Without immigrants and their children, who already make up a third of the population in this age group, the share of young adults would be even lower. Weimar Germany’s demographic profile – an abundance of young and unmarried people with little prospects on the labor market – was central to exacerbating the country’s problems. Today, by contrast, the largest cohort is about to retire and German politicians worry about a shortage of skilled labor that could fill gaps being left by the baby boomers (Figure 11.1). From what we know about violent overthrows of democratic regimes, it is usually not the age group 55–59 years who participate in riots, terrorist movements, organized crime and attempted coups. Around the globe, extremist 161

162 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Figure 11.1

Population pyramids for Germany, 1925 and 2018

organizations and rebel movements instead primarily recruit among young males who are not yet integrated into social institutions such as marriage and the labor market (Hirschi and Gottfredson 1983; Wilson and Daly 1985; Lee 2011). Among this group, opportunity costs of taking part in collective violence are lowest (Collier and Hoeffler 2004, p. 569). This can be observed in an aging society like Germany’s as well. The violent riots by right-wing soccer fans in Cologne in 2014, clashes between left-wing autonomists and police in Hamburg in 2017 and Islamist terrorist attacks in Berlin and elsewhere in 2016 were all carried out predominantly by young males. But most of the population is of advanced age and has other things to do than participating in organized violence. Older people show a much lower disposition to violence as a form of political action, or to risk their lives for higher causes such as country, justice or religion (Weber 2013, p. 349). Even if support for populist political parties or extremist movements has increased, the gray-haired baby boomers are unlikely to march on the Reichstag. “Weimar conditions” will not return. This chapter looks at how demographic profiles and trends, specifically the proportion and growth in a country’s youth population, affect the prospects of liberal democracy. A liberal democratic regime is a system of majority rule enhanced by liberal institutions such as the rule of law and the protection of minority rights (Diamond 1999). Like all modern societies, democracies today are characterized by social antagonisms and cleavages on ethnic, geographic, socio-economic, religious or other grounds (Lipset and Rokkan 1967). A democratic system must ensure that the interest and identity conflicts arising out of these cleavages are balanced and compromises are negotiated in the political arena in a non-violent way. Therefore, political extremist movements which reject any compromise and resort to violent means to achieve political goals are a threat to liberal democratic systems. If young people, and especially young males, are more often involved in extremist movements and political violence, the chances of liberal democracy surviving might at least partly depend on the demographic makeup of a country.

DEMOGRAPHY AND DEMOCRACY 163

The notion that demography affects democracy is not new. Aristotle, in his Politics (Book 7, 1326a) assumed “that it is difficult and perhaps impossible for a state with too large a population to have good legal government” (Aristotle 1995, p.  53). For Malthus (1986 [1798]), it was the exponential growth rate rather than total population size that was responsible for economic and ecological crises and hence instability. Historian Herbert Moller (1968, p.  241) specifically saw the growth rate among the youth population as a destabilizing source: “The explosion of revolutionary, warlike, nationalist […] activity toward the end of the eighteenth and in the early nineteenth centuries received its social fuel from the large youth cohorts of ‘Europe’s initial population explosion.’” And, according to Nazli Choucri (1974, p. 72), youthfulness interacts with the economy: “The higher the proportion of youthful population, and the greater the unemployment, the greater are the possibilities of dissatisfactions, instabilities, and violence”. Political instability and violence, in turn, threaten democratic regimes because, at its core, democracy implies a peaceful compromise of conflicting ideologies and decision-making by mechanisms such as majority rule rather than by force (Lipset 1959). More recent research has linked the presence of a “youth bulge” to a higher risk of civil conflict and war (Mesquida and Wiener 1999; Cincotta et al. 2003; Urdal 2006), to revolutions and political regime change (Nordås and Davenport 2013; Goldstone et al. 2014), and to lower chances for liberal democracy (Cincotta and Doces 2012; Dyson 2012; Weber 2013). The term “youth bulge” refers to an age pyramid where the cohorts aged 15–29 years are disproportionately large compared with the older (and also younger) age groups. This typically occurs in early and intermediate stages of the demographic transition when infant mortality has already declined, fertility rates may have started declining as well, but high birth rates in the recent past are causing the young adult population to grow rapidly for three more decades. There are, of course, other demographic phenomena than youth bulges that might challenge liberal democracy. Examples include the gender ratio (see Chapter 5, this volume) or the growth in the old-age dependency ratio of a society (see Chapter 2, this volume). Since research on the youth bulge (particularly the male part of it) has generated the most evidence so far, we will focus on this demographic aspect in this chapter. At a certain point during the transition – as in Weimar Germany – the cohorts aged 15–29 years are much larger in size compared with both the older as well as the children’s cohorts. This is precisely the constellation that proponents of the “demographic dividend” (see Chapter 13, this volume) perspective deem conducive to economic growth (Bloom et al. 2003): both the young-age as well as the old-age dependent populations are small compared with the eco-

164 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

nomically active population. This means that young people entering the labor market have fewer children or elderly to care for, leading to greater investments in human capital. However, this demographic dividend can also become a security risk: generational crowding of large youth cohorts causes reduced employment opportunities among young people (Bloom et al. 1988), and this “societal bottleneck” can spur grievances among the numerous young people (Urdal 2006). A fertility rate of four children per woman means that every generation is twice as large as its predecessor. If the number of jobs and attractive positions in society does not grow at a similar pace, frustrations among young people and social unrest might ensue. Consider the case of Tunisia in 2008 right before the “Arab Spring”, when the number of young people aged 15–29 years (3.0 million) was greater than both the children’s population (2.5 million 0–14-year-olds) and the preceding cohort of 30–44-year-old citizens (2.2 million). This youth bulge coincided with expansions in secondary and tertiary education at a time when the labor market was not able to provide the large number of better-educated youths with adequate jobs and promising careers. Studies find that this is a combination that can lead to social unrest and political violence (Weber 2019). The descriptive correlation between young age and political violence is well documented. However, it is not yet clear whether this means that the prospects for liberal democracy are also diminished in youthful countries. Several theoretical mechanisms have been proposed, specifying how a youth bulge leads to greater political instability and diminished chances for liberal democracy. Most of them involve some interactions with other factors such as unemployment or level of education, but these have rarely been tested empirically. Moreover, it is not straightforward to aggregate findings about individual associations between demographic characteristics and political outcomes onto the level of regions or countries (McCall et al. 2013). As a consequence, the existing body of empirical research is far from consensual, with several studies finding no significant association between the presence of a youth bulge and the persistence of liberal democratic regimes (Goldstone et al. 2010; Lutz et al. 2010; Urdal and Hoelscher 2012). As an additional problem, several methodological issues have to be overcome that are challenging this field, as well as quantitative political science in general. Future research will have to address these methodological issues in order to arrive at reliable results, which might then allow researchers to derive useful forecasts and political recommendations from the theoretical and statistical models.

DEMOGRAPHY AND DEMOCRACY 165

Theoretical mechanisms Some theoretical links between demography and democracy are undisputed, but some are contested. On the one hand, it is widely accepted that young males are over-represented in groups that seek to impose their political goals onto society by violent means (Wilson and Daly 1985; McCall et al. 2013). A number of socio-economic and socio-biological arguments as well as theories of socialization explain this finding (Sherrod 2006, pp. 599–600). Socio-biological accounts and theories of socialization focus on how youth partly detach from the cultural values and belief systems of their parents during the transition from adolescence to adulthood. As a result, many young people are more susceptible to political extremist or religious fundamentalist views and intergroup hate as compared with older people, giving them the motives to take part in political violence (see, e.g., Weber 2013). From a socio-economic account, young and unmarried males are not yet fully integrated into societal institutions such as jobs and marriage, which means they have more means (and lower opportunity costs) to take part in organized violence and rebel movements. On the other hand, the consequences for liberal democracy of having a large share of young males as a proportion of the population are less obvious. A number of theoretical mechanisms are discussed in the literature, explaining how age structure could affect the prospects for liberal democracy. These theoretical accounts can broadly be divided into theories of democratization (Dyson 2012) versus theories of democratic stability (e.g., Weber 2013; Cincotta 2017). With regard to the former, Dyson (2012) assumes that higher life expectancy leads to more savings and greater investment in the future and hence a greater demand for political participation and equal rights. Moreover, fertility decline empowers women and enables parents to invest more in their education and human capital which in turn also increases the demand for rule of law and political responsiveness (Wilson and Dyson 2017). Mechanisms explaining democratic stability usually depart from the individual-level relationship between youth and propensity to political violence. Democracy – the “peaceful ‘play’ of power” (Lipset 1959, p.  71) – is threatened by actors that seek to impose their interests and values onto society by violent means. Indeed, for Lijphart (1977, p. 4), “a low level of potential or actual civil violence” is a “prerequisite or an indicator of [a stable democratic regime]”. Not every act of violence threatens to overthrow democratic rule, though, and political violence is of course not confined to youthful countries.

166 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Another mechanism describing how age structure affects liberal democracy focuses on leaders’ reactions to grievances in a youthful population. In general, it can be posited that changes in the demos part of democracy likely affect the kratos part: “Because the foundation of democracy is the principle of majority rule, states adopting democratic forms of government find themselves keenly interested in the proportions of politically active groups that inhabit their territories” (Toft 2005, p.  25). In countries with a low median age, political leaders might be more responsive to the needs and desires of young people as compared with the elderly. If young people are more susceptible to violent and extremist ideologies, political leaders might appeal to these ideas during mobilization campaigns (Brooks et al. 2019, p.  65). The resulting policies might undermine liberal democracy; e.g., in the form of prosecution of minority groups (Brooks et al. 2019, p. 65). This might be especially the case if young people suffer from high youth unemployment and other factors causing grievances (Brickner and Foley 2013). This mechanism thus expects political incumbents to adopt illiberal policies in order to get the approval of young and frustrated voters. As a further theoretical mechanism, political elites can launch authoritarian policies if they fear a destabilizing threat posed by youth movements (e.g., Krawatzek 2018). For instance, leaders might ban opposition parties or media outlets, or remove checks and balances to governmental decision-making. Citizens as well as elites in the economic or military sectors may accept these illiberal measures, fearing disorder and the destruction of property (Cincotta and Doces 2012). As a related mechanism, the military can take over control from the government, claiming to restore order or to prevent extremist (e.g., communist, nationalist or Islamist) parties from seizing power. This has happened many times in Latin America and Turkey or, more recently, in Egypt, when President Mohamed Morsi of the Muslim Brotherhood was removed from office by the military after violent protests in 2013. While military coups and authoritarian reversals happen in more mature societies as well (e.g., Thailand), proponents of the youth bulge theory contend that the social conditions (such as social unrest and disorder) that usually precede coups d’états against elected governments are more likely to emerge in youthful countries (Cincotta 2017). A final mechanism focuses on state capacity to provide education, jobs and infrastructure (Cincotta and Weber forthcoming). In countries where young cohorts entering adulthood are twice or three times as large as their parental cohorts, governments can face difficulties providing these services. This causes the government’s perceived legitimacy among the population to decline and opens opportunities for other actors such as local clan leaders, religious organ-

DEMOGRAPHY AND DEMOCRACY 167

izations or paramilitary factions that challenge the authority of the central government and portray themselves as caretakers for ordinary citizens. The democratic process may be depicted as slow or indecisive by these actors, who then campaign for authoritarian rule. Needless to say, all these mechanisms are specified in a non-deterministic way such that the presence of a youth bulge may or may not lead to social unrest, diminished state capacity and authoritarian reversal. The biggest need for further theory building certainly pertains to the conditions under which liberal democracy fails in youthful states. The interactions of a youth bulge with other factors such as unemployment or level of socio-economic development remain unclear to date. For instance, many theorists expect the impact of a youth bulge to be strongest when coinciding with high youth unemployment (Choucri 1974, p. 72). However, more recent research finds that, despite being employed, young people might engage in political protests if their income falls short of what could be expected in light of their educational attainment (Campante and Chor 2014). This suggests that relative deprivation, rather than unemployment or the absolute level of income, drives violent protests (Gurr 1970). The latter finding also questions the role of educational expansion, because expectations about lifestyle and income rise with educational attainment. Rapid expansion in education, if not coupled with equally strong expansions in job opportunities for highly educated people, can produce large numbers of college graduates who might fail to find adequate positions on the labor market, leading to frustration and relative deprivation (Østby and Urdal 2010). Indeed, historical research finds that rebellions and revolutions were often preceded by an expansion in education (Goldstone 1991; see also Chapter 9, this volume). It is often not the less-educated and impoverished youths that challenge the authority of government, rather the highly educated but marginalized young people that campaign for an alternative social order. Indeed, a recent study finds that, if increases in post-primary education within a country coincide with a large youth bulge, the number of deaths from political violence tends to rise in the same country (Weber 2019). An open question for future research is thus whether educational expansions in youthful countries can also be a threat to liberal democratic rule via these mechanisms. There are, however, not only interactions but also feedback effects that need better theorization and modeling in future research. For instance, it is well known that lower fertility rates and a decline in the young-age dependent population cause rising income levels through greater investments in human capital (Headey and Hodge 2009; Lee and Mason 2014). But higher income

168 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

and better education are also determinants of lower fertility rates (Lutz and Qiang 2002). Thus, there exists a vicious circle of higher stocks of capital and higher wages on the one hand and fewer births on the other hand (at least for a certain period of time). Smaller youth populations can also contribute to lower youth unemployment, further decreasing the risk for youths to engage in political violence. There is a time lag of two to three decades, however, between the onset of fertility decline and a flattening of the youth bulge. Thus, strong expansions in education may trigger fertility decline and rising income levels, but the youth bulge will still be present for a generation, as is currently the case in countries such as Iran (Lutz et al. 2010). This is arguably the constellation that is both an opportunity for growth, socio-economic modernization and eventual democratization – but also implies an increased risk for political instability and authoritarian reversal (Weber 2019). In sum, there is clearly a need for more rigorous theoretical work on how demographic profiles and trends affect the prospects for liberal democracy in a country. The crucial micro–macro link, i.e., how young people’s actions affect aggregate political outcomes (such as executive or legislative actions towards a more authoritarian system of government), is not clear, although numerous theoretical accounts exist. Theorists also disagree on the socio-economic circumstances that are most detrimental for liberal democracy in countries with youth bulges. One way to move forward might be to study the complex interplay between (levels and changes in) education, employment, income or fertility rates with computational approaches (such as agent-based models). Informed with insights from social psychology and microeconomics, such models could be used to arrive at theoretical hypotheses on how well-known individual correlations translate to aggregate outcomes. The resulting hypotheses on thresholds or interaction effects can then be tested empirically.

Empirical findings There is now abundant evidence on the macro-level correlation between a more mature age structure and the chances for liberal democracy in a country. Most of this knowledge comes from aggregate country-level data from the period since 1950. Much less is known about other historical periods or about other social aggregates than the national level, e.g., regarding sub-national differences between regions or ethnic groups and the effects on local or central governments. Furthermore, empirical studies attempting a more causal interpretation of the bivariate association between demography and democracy are rare. Research designs that exploit, e.g., (natural) quasi-experiments which are

DEMOGRAPHY AND DEMOCRACY 169

common in other fields of research, are hard to discover in this particular area, especially given the high intercorrelation between demographic trends and many other variables. This points to a number of methodological challenges that empirical research has to overcome, as will be discussed below. Figure 11.2 shows the geographic distribution of youth ratio across the world in 2015, where youth ratio is defined as the share of young males aged 15 to 29 years as a proportion of all males aged 15 or older (United Nations 2016). Throughout most of Europe, young adults make up less than a quarter of the total adult population. By contrast, in many sub-Saharan African countries and a few others such as Afghanistan and Yemen, half of all adults are in the young age group; 15–29-year-olds dominate the economically and politically active population in these countries. In the Americas, the demographic transition has progressed at a fast rate in many countries during the past decades. Guatemala and Honduras have the relatively largest youth populations, followed by Haiti. Falling birth rates cause the youth bulge to level off in most world regions, but youth ratio increased in 28 countries between 1996 and 2015, owed mostly to continuously high fertility rates in sub-Saharan Africa.

Figure 11.2

Youth ratio in 182 countries of the world, 2015

If we compare Figure 11.2 to Freedom House (2016) ratings for “political rights” and “civil liberties”, it is obvious that there is some geographic overlap between youthful and undemocratic countries. But there are also some cases that do not match the pattern, in particular Russia and China. Figure 11.3 explores this relationship in greater detail, grouping countries into five cate-

170 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

gories, roughly corresponding to quintiles of the distribution of youth ratio. The two sub-scales for “political rights” and “civil liberties” are combined to an index running from 2 (free) to 14 (not free) on the y axis. The black solid lines mark the median value in each group, box boundaries correspond to the 25 percent and 75 percent percentiles, and each gray dot is a country in 2015. A steady increase in the average liberal-democratic score can be noted as one moves from the more youthful to the more mature countries. Similar patterns can be observed using different measures for liberal democracy other than Freedom House.

Figure 11.3

Youth ratio vs Freedom House Index in 182 Countries, 2015

Between 1996 and 2015, Freedom House changed their overall rating from “free” to either “partly free” or “not free”, or from “partly free” to “not free” in 66 instances. Let us call these instances of authoritarian reversal, i.e., a country became less liberal-democratic compared with the previous year. If we look at a country’s risk of experiencing authoritarian reversal in a given year as a proportion of all countries that were rated “free” or “partly free”, there was no country in the very mature age category (youth ratio below 23 percent) that experienced a setback in Freedom House’s ratings throughout this time period. By contrast, the risk increases from 0.8 percent in the second quintile to 6.2 percent each year in the highest quintile (youth ratio greater than 47

DEMOGRAPHY AND DEMOCRACY 171

percent). Clearly, the countries with the largest youth bulge had the highest risk of becoming less democratic in a given year. Within the more mature age group (second quintile), there were only four countries (out of 75) whom Freedom House gave lower summary scores at some point between 1996 and 2015: Bosnia and Herzegovina (2003), Russia (2004), Thailand (2005, 2006 and 2014) and Ukraine (2014). Contrast this with 26 countries in the group with the largest youth bulge who experienced authoritarian reversal at least once during this period (out of 52 countries that at some point belonged in this group and had at least one period of “free” or “partly free” ratings). This means that half of all very youthful countries who had reached an at least intermediate stage in terms of liberal democracy experienced a downturn in democracy in a window of 20 years. It is thus obvious that there exists a rather strong bivariate connection between age structure and democracy. In youthful countries, liberal democracy is much less likely to survive over the course of two decades as compared with more mature societies. Countries that completed the demographic transition such that middle-aged and older people dominate over young adults in their age pyramids are hardly ever experiencing authoritarian reversals once they install liberal democratic rule. But a bivariate association does of course not suffice in order to judge whether age structure is really a relevant causal factor influencing regime type. The observational correlation could as well be spurious, meaning that other factors might be causally responsible for democratic downturn and also happen to be correlated with age structure. To this end, researchers have specified multivariate models which also control for other variables that are often associated with democratic stability such as level of income, education, economic growth and ethnic and religious fractionalization (e.g., Cincotta and Doces 2012; Dyson 2012; Weber 2013). Using various measures of age structure (e.g., youth bulge or median age) as well as different operationalizations of democracy, these studies conclude that net of the effects from socio-economic and other variables, liberal democracies appear to be indeed less stable in youthful societies. For instance, Weber (2013) finds that for countries with a youth bulge above the world median, the odds of democratic breakdown are seven times higher compared to less youthful countries if per capita income, growth rate and some other variables are taken into consideration. Thus, the descriptive findings (Figure 11.3) are largely confirmed by multivariate analyses. However, controlling for a handful of third variables is not enough to arrive at causal interpretations in this setting. Due to the complex

172 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

nature of demographic trends, their interrelatedness with other factors, and general issues with analyzing highly aggregated data, some questions remain even though the empirical patterns look very clear on first sight.

Methodological challenges The standard approach to assessing the effect of demography on liberal democracy is via some form of regression analysis using (yearly) time series of country-level data. Arguably, the biggest challenges for such an approach are presented by the following issues: 1. Multi-collinearity between predictor variables: Norway is in the left-most group in Figure 11.3, while Afghanistan is in the right-most group. But these two countries surely not only differ with regard to age structure and democracy; there are also differences in education, history, culture, geography and the economy. Ideally, to assess the effect of youth bulge on democracy net of socio-economic and other factors, one would want to have two groups of equally poor and culturally similar countries, but one group has a youthful age structure (the “treatment”) whereas the other is more mature. But in reality, the latter control group is very small, as virtually all poor countries are youthful. In the presence of high multi-collinearity, the partial regression coefficient for youth bulge is strongly influenced by outlier cases that do not conform to the general pattern (see, e.g., Schrodt 2014). The inclusion or exclusion of a few specific cases or variables might thus significantly affect the overall findings, as Weber (2019) has shown for the effect of youth bulge on political violence. If we indeed assume that there are causal connections and feedback effects between demography and socio-economic development, an approach that treats age structure as the predictor and income only as a “control” variable is problematic. 2. Slow changes in path-dependent variables: Although some countries have witnessed quite drastic drops in their fertility rates over a short period of time, the year-to-year changes in median age or youth bulge are still quite small in most cases. From a counterfactual understanding of causality, this is problematic. Something that does not change much over time cannot be causally responsible for disruptive changes in the outcome (e.g., sudden democratic breakdown in a certain year). 3. Interaction effects between demography and other factors such as unemployment or educational attainment: Although often theorized, such interactions are rarely tested in empirical models. And if they are, the

DEMOGRAPHY AND DEMOCRACY 173

above-mentioned problems with highly collinear variables become ever more acute. There are several potential remedies to these issues to consider in future research. First, while virtually all previous studies use some form of regression model, there are numerous alternative methods which can deal better with some of the issues mentioned. For instance, tree-based machine learning algorithms such as random forests or gradient boosting machines are less affected by multi-collinearity, have no linearity assumption, and automatically map all relevant interactions between predictor variables. However, there is a caveat to using such tools in applied political research: These algorithms were developed for predictive models, meaning that they optimize predictive accuracy for the outcome (e.g., democratic stability) in out-of-sample tests. They do not, however, deliver reliable estimates for a specific parameter of some predictor variable of interest (e.g., youth bulge) (see Mullainathan and Spiess 2017). Thus, if the main focus is on predictive modeling, research on the future prospects for liberal democracy can benefit from adopting approaches such as cross-validation and out-of-sample testing that are common in machine learning (Cranmer and Desmarais 2017) and considering different methods apart from the “monoculture” of linear and logistic regression (Schrodt 2014). If the main focus is on theory testing – does a youth bulge affect liberal democracy, and if so, under which conditions? – black-box machine learning algorithms will help little, and we will instead need more carefully constructed research designs to identify causal effects and interactions. Cross-country studies with panel data could benefit from within-country designs over longer periods of time (see, e.g., Cincotta and Weber forthcoming). Case-control matching designs could be applied to systematically select countries that can potentially provide evidence on the youth effect (Ho et al. 2007). To study the formation of rebel movements or the genesis of social unrest, a cross-national perspective could be combined with data on a sub-national level in order to exploit within-country variation and isolate the youth effect from country-specific peculiarities. The evidence accumulated by this research might then in turn inform predictive models of phenomena such as liberal democratic longevity.

Outlook The start of the 2020s continues a wave of democratic reversal and a return to authoritarianism around the world. From a demographic transition perspective, however, some theorists believe the chances for stable liberal democracy

174 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

might improve over the coming decades, at least in some world regions. Many countries outside Europe and North America have experienced a substantial decline in fertility rates and will enter a phase of considerable population aging between 2020 and 2040. This is particularly the case in many Latin American and Southeast Asian countries, and, to a lesser extent, also in the Middle East and North Africa. The cohorts of young adults entering the labor market will not be twice as large as their parental generation anymore, but rather approximately equal in size. This will, as proponents of the youth bulge theory hold, lead to less competition on the job market and thus better prospects for young people, leading to less pressure on governments to provide services. Fewer young people will be left out of the loop and the potential recruiting base for extremist movements will decline. Middle-aged and older people who are embedded in various social institutions such as marriage and job will dominate the age structure, and these groups typically refrain from taking part in political violence. However, there are a few caveats to this optimistic outlook. First, there are many countries where youth bulges will continue to be prominent for decades. In most sub-Saharan African countries, fertility rates are still very high and youth populations will grow to unprecedented numbers in the next years. The widespread misconception that large youth cohorts per se constitute a “demographic dividend” contributes to lower awareness of the continued issues of high fertility and population growth (Coole 2013), despite the evidence that a substantial reduction in birth rates is usually a prerequisite for better socio-economic development and eventual democratization. Thus, waiting for youth bulges to flatten by themselves might not be enough; rather, more efforts to improve reproductive health and female education are arguably needed to achieve this end. Even after the onset of an eventual decline in fertility, however, the relative dominance of youth cohorts in the age pyramid will increase for two to three decades due to the time lag between falling birth rates and declining youth bulges. Consider, again, Weimar Germany in 1925: fertility rates had already fallen to just above two children per woman, but the youth bulge grew to a historical high and political stability was all but imminent in subsequent years. Another problem for the optimistic outlook is presented by the growing group of demographically mature countries (such as Cuba, China or Russia) that are not liberal democracies. Whereas in most empirical studies these countries are depicted as outliers to the general trend, as a group these cases certainly challenge existing theory and research. Readers can refer to Chapter 2, on population aging, for a more extended discussion of these cases.

DEMOGRAPHY AND DEMOCRACY 175

In sum, research on how demography affects liberal democracy is far from being conclusive to date. The various interactions and feedback effects between progression in the demographic transition, socio-economic development, education and employment are still largely unexplored. Various mechanisms for how political actors react to social stress posed by large and growing youth populations have been proposed in the literature, but these often remain in the black box. Qualitative research and process tracing with case studies might shed light on the perceptions and actions among elites leading to authoritarian revivals in liberal democracies (Krawatzek 2018 is an insightful example). There are several avenues for quantitative research as well in order to accumulate more reliable evidence. In the end, researchers may get to the ground of what is needed to prevent “Weimar conditions” once and for all.

References Aristotle (1995), Politics, translated by Ernest Barker, Oxford: Oxford University Press. Bloom, D. E., Freeman, R. B. & Korenman, S. D. (1988), ‘The labour-market consequences of generational crowding’, European Journal of Population/Revue Européenne de Démographie 3 (2): 131–76. Bloom, D. E., Canning, D. & Sevilla, J. (2003), The Demographic Dividend: A New Perspective on the Economic Consequences of Population Change, Santa Monica et al.: RAND. Bricker, N. Q. & Foley, M. C. (2013), ‘The effect of youth demographics on violence: the importance of the labor market’, International Journal of Conflict and Violence 7 (1): 179–94. Brooks, D. J., Brooks, S. G., Greenhill, B. D. & Haas, M. L. (2019), ‘The demographic transition theory of war: why young societies are conflict prone and old societies are the most peaceful’, International Security 43 (3): 53–95. Campante, F. R. & Chor, D. (2014), ‘“The people want the fall of the regime”: schooling, political protest, and the economy’, Journal of Comparative Economics 42 (3): 495–517. Choucri, N. (1974), Population Dynamics and International Violence: Propositions, Insights, and Evidence, Lexington: Lexington Books. Cincotta, R. (2017), ‘The age-structural theory of state behavior’, in Oxford Research Encyclopedia of Politics, retrieved Aug. 7, 2019 from https://​oxfordre​.com/​politics/​ view/​10​.1093/​acrefore/​9780190228637​.001​.0001/​acrefore​-9780190228637​-e​-327. Cincotta, R. P. & Doces, J. (2012), ‘The age-structural maturity thesis: the impact of the youth bulge on the advent and stability of liberal democracy’, in Goldstone, J. A., Kaufmann, E. P. & Toft, M. D. (eds), Political Demography: How Population Changes are Reshaping International Security and National Politics, Oxford: Oxford University Press, pp. 98–116. Cincotta, R. P., Engelman, R. & Anastasion, D. (2003), The Security Demographic: Population and Civil Conflict After the Cold War, Washington, DC: Population Action International.

176 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Cincotta, R. P. & Weber, H. (forthcoming), ‘Youthful age structures and the risks of revolutionary and separatist conflicts’, in Goerres, A. (ed.), Global Political Demography: How Population Change and Politics/Policies Influence one Another. Collier, P. & Hoeffler, A. (2004), ‘Greed and grievances in civil war’, Oxford Economic Papers 56 (4): 563–95. Coole, D. (2013), ‘Too many bodies? The return and disavowal of the population question’, Environmental Politics 22 (2): 195–215. Cranmer, S. J. & Desmarais, B. A. (2017), ‘What can we learn from predictive modeling?’, Political Analysis 25 (2): 145–66. Diamond, L. (1999), Developing Democracy Toward Consolidation, Baltimore/ London: Johns Hopkins University Press. Dyson, T. (2012), ‘On demographic and democratic transitions’, Population and Development Review 38 (Supplement): 83–102. Freedom House (2016), Freedom in the World 1972–2015, Washington, DC: Freedom House. Goldstone, J. A. (1991), Revolution and Rebellion in the Early Modern World, Stanford: University of California Press. Goldstone, J. A., Bates, R. H, Epstein, D. L., Gurr, T. R., Lustik, M. B., Marshall, M. G., Ulfelder, J. & Woodward, M. (2010), ‘A global model for forecasting political instability’, American Journal of Political Science 54 (1): 190–208. Goldstone, J. A., Marshall, M. G. & Root, H. (2014), ‘Demographic growth in dangerous places: concentrating conflict risk’, International Area Studies Review 17 (2): 120–33. Gurr, T. R. (1970) Why Men Rebel, Princeton: Princeton University Press. Headey, D. D. & Hodge, A. (2009), ‘The effect of population growth on economic growth: a meta‐regression analysis of the macroeconomic literature’, Population and Development Review 35 (2): 221–48. Hirschi, T. & Gottfredson, M. (1983), ‘Age and the explanation of crime’, American Journal of Sociology 89 (3): 552–84. Ho, D. E., Imai, K., King, G. & Stuart, E. A. (2007), ‘Matching as nonparametric preprocessing for reducing model dependence in parametric causal inference’, Political Analysis 15 (3): 199–236. Krawatzek, F. (2018), Youth in Regime Crisis: Comparative Perspectives from Russia to Weimar Germany, Oxford: Oxford University Press. Lee, A. (2011), ‘Who becomes a terrorist? Poverty, education, and the origins of political violence’, World Politics 62 (2): 203–45. Lee, R. & Mason, A. (2014), ‘Is low fertility really a problem? Population aging, dependency, and consumption’, Science 346 (6206): 229–34. Lijphart, A. (1977), Democracy in Plural Societies: A Comparative Exploration, New Haven/London: Yale University Press. Lipset, S. M. (1959), ‘Some social requisites of democracy: economic development and political legitimacy’, American Political Science Review 53 (1): 69–105. Lipset, S. M. & Rokkan, S. (1967), Party Systems and Voter Alignments, New York: Free Press. Lutz, W., Cuaresma, J. C. & Abbasi-Shavazi, M. J. (2010), ‘Demography, education, and democracy: global trends and the case of Iran’, Population and Development Review 36 (2): 253–81. Lutz, W. & Qiang, R. (2002), ‘Determinants of human population growth’, Philosophical Transactions of the Royal Society 357: 1197–210. Malthus, T. R. (1986) [1798], ‘An essay on the principle of population’, The Works of Thomas Robert Malthus, London: Pickering & Chatto Publishers, vol. 1, pp. 1–139.

DEMOGRAPHY AND DEMOCRACY 177

McCall, P. L., Land, K. C., Dollar, C. B. & Parker, K. F. (2013), ‘The age structure-crime rate relationship: solving a long-standing puzzle’, Journal of Quantitative Criminology 29 (2), 167–90. Mesquida, C. G. & Wiener, N. I. (1999), ‘Male age composition and severity of conflicts’, Politics and the Life Sciences 18 (2.1): 181– 9. Moller, H. (1968), ‘Youth as a force in the modern world’, Comparative Studies in Society and History 10 (3), 237–60. Mullainathan, S. & Spiess, J. (2017), ‘Machine learning: an applied econometric approach’, Journal of Economic Perspectives 31 (2), 87–106. Nordås, R. & Davenport, C. (2013), ‘Fight the youth: youth bulges and state repression’, American Journal of Political Science 57 (4): 926–40. Østby, G. & Urdal, H. (2010), ‘Education and civil conflict: a review of the quantitative, empirical literature’, Paper commissioned for the EFA Global Monitoring Report 2011. Schrodt, P. A. (2014), ‘Seven deadly sins of contemporary quantitative political analysis’, Journal of Peace Research 51 (2), 287–300. Sherrod, L. R. (2006), Youth Activism: An International Encyclopedia (Vol. 2), Westport/ London: Greenwood Publishing Group. Toft, M. D. (2005), ‘The state of the field: demography and war’, ECSP Report 11, pp. 25–8. United Nations (2016), Demographic Yearbook, New York: United Nations. Urdal, H. (2006), ‘A clash of generations? Youth bulges and political violence’, International Studies Quarterly 50 (3): 607–29. Urdal, H. & Hoelscher, K. (2012), ‘Explaining urban social disorder and violence: an empirical study of event data from Asian and sub-Saharan African cities’, International Interactions 38 (4): 512–28. Weber, H. (2013), ‘Demography and democracy: the impact of youth cohort size on democratic stability in the world’, Democratization 20 (2): 335–57. Weber, H. (2019), Der demographische Wandel: Mythos – Illusion – Realität, Stuttgart: Kohlhammer. Wilson, B. & Dyson, T. (2017), ‘Democracy and the demographic transition’, Democratization 24 (4): 594–612. Wilson, M. & Daly, M. (1985), ‘Competitiveness, risk taking, and violence: the young male syndrome’, Ethology and Sociobiology 6: 59–73.

12

Demographic engineering and strategic demography

Michael S. Teitelbaum

Introduction The chapters in this volume identify numerous ways in which demographic and political change interact. Notwithstanding the salience of such interactions, many modern scholars have tended to shy away from the topic, and indeed they have had good historical reasons to do so. Since medieval times at least, arguments and assumptions about demographic change have played central roles in the often tendentious assumptions and analyses of mercantilists, utopians, free-market physiocrats, Malthusians and neo-Malthusians, socialists of various persuasions, Marxians and Marxist-Leninists, French Republicans, nationalists, ethnic and religious leaders, racists, eugenicists and fascists. During the interwar years of the twentieth century, some researchers were personally tainted by their involvement in the excesses of eugenics, “scientific racism”, Nazism, Stalinism, nationalism and other -isms. More recently, politicized arguments about demographic issues were promoted by the New Right, resurgent nationalists of diverse commitments, a few feminist thinkers and some among the conservative religious (including Christian, Muslim, Hindu and Jewish). Taken together, the often-passionate beliefs that for centuries have littered the intersection of politics and demography led many postwar researchers to see it as a subject incompatible with objective analysis. While such concerns are understandable, they do not obviate the reality of interactions between political and demographic developments, both domestic and international, that are too significant and numerous for scholars to ignore. As this volume details, promising research subjects include the global spread of very low fertility and the effectiveness and legitimacy of diverse efforts to reverse it; demographic aging attributable to declining fertility and its implications for economic and social policy; changing demographic com179

180 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

position as it relates to identity politics and social cohesion; the growing use of out-migration as instruments of foreign policy; the impacts of disorderly international migration upon domestic politics in destination countries; and the implications of these demographic shifts for long-established regimes relating to human rights. The focus of this chapter is on policies designed to affect the size, composition, distribution and growth rate of a population – “demographic engineering” (Weiner and Teitelbaum, 2001, pp. 54–5). Such efforts have been undertaken by governments, political parties and a variety of interest groups that believe that the demographic changes they support will further their interests. The strategies involved may be explicit or implicit, more commonly the latter. Demographic engineering efforts may be addressed to any of the three primary drivers of demographic change – mortality, fertility or migration. The first – mortality – obviously is an explicit goal of the wars, massacres, genocides and intentional famines and epidemics common over the past centuries, and need not be discussed here. Our focus instead is upon demographic engineering policies intended to affect either fertility or migration.

Engineering national fertility declines Where fertility rates are high, numerous governments and non-governmental organizations around the world have embraced explicit or implicit policies to reduce high fertility to levels they deem more conducive for national stability and prosperity. Many such policies emerged in the years following World War II, when it became clear that population growth in many newly independent countries had accelerated as high rates of infant and child mortality declined while fertility rates remained high. Some of these policies proved effective and constructive; others had only limited impacts. Explanations for such divergent outcomes are still being debated. In some cases, fertility rates declined to levels far lower than had been anticipated or intended, leading to other demographic challenges such as rapid shifts in age composition and awkward reversals of policy. One of the earliest fertility reduction policies was legalized voluntary abortion legislation adopted in 1948 by the Diet (parliament) of occupied Japan. The legislative initiative was taken by liberal members of the Diet, in the context of expert analyses suggesting that unless Japan’s then high fertility rates moderated, the country could experience sustained mass unemployment, poverty,

DEMOGRAPHIC ENGINEERING AND STRATEGIC DEMOGRAPHY 181

famine and epidemics, and could not be expected to evolve into a stable and prosperous society. Unmet demand for effective control of fertility proved to be substantial among the Japanese public. With little or no promotion by the government, high fertility rates during the postwar years declined rapidly. Most observers agree that this rapid fertility decline, coupled with political stability, successful economic policies and high standards of innovation and entrepreneurship in Japanese industry, contributed substantially to the Japanese economic miracle from the 1950s onward and transformed poverty-stricken postwar Japan into one of the world’s most prosperous countries. In 1952, only four years after the Japanese Diet’s legalization of abortion, recently independent India adopted its National Programme for Family Planning. This policy was designed to reduce national fertility rates by providing voluntary contraceptive and sterilization services to enable couples to limit their family size to the desired number of children. Though it proved to be less effective in lowering fertility rates than its proponents had hoped, this model of government-supported voluntary family planning programs was followed a decade later by four East and Southeast Asian countries later collectively described as the Asian Tigers – South Korea, Taiwan, Hong Kong and Singapore – with far stronger fertility-decreasing effects. The impressive economic successes of these countries led many others in the region to follow, including Indonesia, Vietnam and China. Others in the region demurred for internal political reasons, most prominently the Philippines. During the 1970s, both India and China shifted toward less voluntary and sometimes crudely coercive policies to lower national fertility rates, with the most dramatic being China’s “one-child policy” adopted around 1980. More recently, such national population policies have been promoted as ways for high fertility countries to capture what is called the “demographic dividend”. The underlying argument of this attractively alliterative formulation is that fertility declines will lead to a time “window” of several decades during which the proportion of the population in working ages rises in ways that will accelerate economic growth. (For a fuller discussion, see Chapter 13 in this volume by Patierno, Madsen and Gaith.)

Engineering national fertility increases The most time-honored and obvious efforts to engineer fertility increases are “pro-natalist” policies (nataliste in French) intended to increase a state’s economic power and military potential relative to those of its strategic com-

182 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

petitors. (For a fuller discussion, see Chapter 7 in this volume by Toft.) Such pro-natalist policies were especially popular among governments of industrialized countries during the first half of the twentieth century, most explicitly so in France vis-à-vis Germany. One example is the extraordinary intervention by French Premier Georges Clemenceau during the 1919 French Senate debate on ratification of the Treaty of Versailles, which codified victory by France and her allies over Germany: The treaty does not say that France must undertake to have children, but it is the first thing which ought to have been put in it. For if France turns her back on large families, one can put all the clauses one wants in the treaty, one can take all the guns from Germany, one can do whatever one likes, [but] France will be lost because there will be no more Frenchmen. (Clemenceau, 1919)

Such views were widely shared among French elites. Only five years later another French Premier and former President of France, Raymond Poincaré, once again pointed to the “military problems” arising from French demographic decline (Teitelbaum and Winter, 1985, p. 36). During the Great Depression of the 1930s, fertility rates in most European countries declined to previously unheard-of levels, evoking concerns and pro-natalist strategies through much of Europe. These high-level political concerns about low fertility ranged across the political spectrum from left to right. The most explicit were those promulgated by autocratic governments led by Benito Mussolini, Adolph Hitler and Joseph Stalin, but even Swedish Social Democrats such as Alva and Gunnar Myrdal (both Nobel Prize winners) shared similar concerns. Though many believed that the economic crisis underlay fertility declines, some very prominent economic thinkers, including John Maynard Keynes, argued that low fertility was driving the economic decline (Keynes, 1937).1 As alarm about declining fertility during the 1930s spread widely in France, Italy, Britain, Sweden and elsewhere, a variety of prominent socialists, social democrats and liberals took the opportunity to mobilize political support among their more conservative colleagues for significant reform measures (family and child allowances, subsidized housing, national healthcare and childcare systems) as essential for survival of the nation. Such political strategies contributed substantially to the widespread establishment of what came to be called the European welfare state (Teitelbaum and Winter, 1985, pp. 58–62). Concerns about low fertility were muted by the widespread fertility increases after World War II but began to reappear during the 1960s, especially in Warsaw Pact states as fertility rates again declined. Many of the governments

DEMOGRAPHIC ENGINEERING AND STRATEGIC DEMOGRAPHY 183

of these states embraced pro-natalist policies, the most extreme of which were those of Romania under its longtime President Nicolae Ceauşescu, who in 1966 peremptorily banned abortion (which was legal and widely used) and most forms of contraception (Teitelbaum, 1972).

Engineering differential fertility changes within states There are numerous cases in which leaders of sub-national groups have sought to gain political power over competitor sub-national groups by encouraging higher fertility among their adherents. Such efforts are most obvious in states (or empires) with multi-ethnic and/or multi-religious populations that are vying with one another for power, especially if such groups are geographically concentrated or of comparable size. A valuable and detailed discussion of this phenomenon has been presented in Chapter 7 in this volume by Toft, and due to space constraints will not be addressed in any detail here. In some cases, there have been efforts to deploy the power of the state to incentivize increased fertility rates among populations that are at risk of losing their majority status due to low fertility. A particularly interesting case arose during the final years of the USSR in the 1980s, when there were active debates about proposals intended to increase fertility in the slow-growing but politically dominant Slavic republics, while at the same time lowering fertility rates in the USSR’s high-fertility republics in Central Asia. Such proposals proved highly controversial, and had not been agreed and implemented before the Central Asian Republics became independent states following the dissolution of the USSR in 1991 (Teitelbaum and Winter, 1998, pp. 96–108).

Engineering international migration With respect to migration, nearly all governments have formal policies that explicitly seek to restrain inward migration by non-citizens, though some of these policies are poorly enforced. A few modern states have also sought to control out-migration, most notably those in the former Warsaw Pact. At the same time, many governments also have adopted policies to affirmatively stimulate either inward or outward migration.

184 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Engineering in-migration to increase population size or modify age structure There are multiple examples of governments adopting policies designed to accelerate economic growth by expanding their workforces or modifying their age structures via increases in immigration. Other governments have adopted expansive immigration policies for strategic reasons, such as those intended to populate sparsely settled territories to deter incursions or settlement from neighboring countries, or to expand the population base from which military personnel can be drawn. As governments today try to find ways to adapt to or mitigate the economic effects of population aging, the past may be instructive to help us understand the conditions under which political leaders may see increased immigration as one of those strategies, and the likely sociopolitical effects if they choose to do so. During the nineteenth century, the US government facilitated large-scale in-migration as a means of expanding its workforce and populating the massive territories it had acquired through the Louisiana Purchase in 1803, the admission of Texas to the Union in 1845 and the peace treaty ending the Mexican–American War in 1848. Even more explicit policies of this type were adopted by Australia and Canada during the twentieth century. Australia combined an enormous landmass with a small and sparse population and a geographic location close to densely populated Asian regions. The strategic thinking involved was best exemplified by the Australian post-World War II political slogan “Populate or Perish”, promoting as a matter of national security the active recruitment of large-scale immigration from Europe while excluding most migration from Asia. Similar, if less explicit, strategies may underlie Canada’s longstanding expansive immigration policies, beginning under British rule and continued to the present day. Like Australia, Canada had an enormous landmass but a small and sparse population, and during the eighteenth and nineteenth centuries was troubled by threats of invasions from its then-expansionist neighbor to the south, the United States. In recent decades, increased in-migration has been seen by some as an effective way to engineer age structure by counteracting demographic aging – the shift in age structure toward higher proportions of older persons described in Chapter 2 – that they fear will slow economic growth and increase the fiscal burdens of official retirement and health policies. Governments continue to pursue such efforts even though most demographic analyses have shown that population aging is driven primarily by declining fertility rates, and that only

DEMOGRAPHIC ENGINEERING AND STRATEGIC DEMOGRAPHY 185

small countervailing effects can be expected from politically plausible increases in immigration (Teitelbaum, 2004; United Nations, 2000). In the energy-rich states of the Persian Gulf, large-scale in-migration of temporary workers with no rights of residence has been a demographic strategy for many decades. Since most of these states have quite small indigenous populations, these imported temporary workers (mostly from elsewhere in the Middle East and from Asia) have often outnumbered native-born populations; in some cases foreign workers have constituted more than three-quarters of the total resident population. There have been numerous cases of employers abusing these temporary migrants, and convulsive episodes of mass expulsions during periods of economic crises or perceived threats to national security. Both have led to humanitarian dilemmas and to foreign policy tensions with the source countries of those being expelled. Given the widespread use of these immigration strategies, this topic is conducive to comparative scholarship on both the causes and effects of such policies and there is much room in the field for such studies. Engineering out-migration by nationals Meanwhile, other governments have long pursued policies that facilitate or incentivize out-migration of their nationals to other countries. Often they see such policies as a way to reduce the potential for internal political opposition by providing employment prospects not available domestically, while also generating large return flows of hard currency (remittances) that overseas workers send back to their extended family members (World Bank, 2019). Again, such policies can be both explicit and implicit – including creation of organizations charged with promoting out-migration of citizens for employment abroad, education policies that knowingly produce graduate numbers that far exceed domestic demand in occupations for which there is international demand (such as healthcare and information technology), or governments’ (usually implicit) facilitation of irregular migration by their nationals to other more prosperous countries. Policies in one or more of these forms have been pursued by many countries, including most notably the Philippines, India, Pakistan, Bangladesh, Vietnam, Malaysia, Turkey, Egypt, Lebanon, Cuba, Haiti, Dominican Republic, Venezuela, Mexico, Guatemala, El Salvador and Nicaragua. The outcomes of such policies are mixed. In general, they have succeeded in providing employment opportunities that were more attractive than those in the source country, and have generated substantial remittance flows from overseas workers. At the same time they often have subjected overseas workers

186 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

to difficult-to-control exploitation and abuse, and have exposed the economies of source countries to damaging external shocks when economic and political developments in destination countries result in mass expulsions or large-scale voluntary returns by migrant workers. More malign examples include actions by some governments and private groups, especially those that identify themselves with particular ethnicities or religions, to “unmix” the multi-ethnic/multi-religious populations often inherited from past imperial regimes. These include encouragement of both in-migration by co-ethnics and co-religionists, as well as coercive out-migration of other groups. (For a fuller discussion, see Chapter 3 in this volume by Chu). One of the most dramatic examples occurred during the 1920s following the collapse of the Ottoman Empire. Large ethnic and religious groups that had long resided together in imperial regions that became parts of modern Turkey and Greece were compelled to depart under international agreements that established national borders for these two newly independent states. The numbers were large, with about 1.8 million people expelled in both directions – an estimated 1.3 million Greek Christians from Turkey to Greece, and about 500,000 Muslim Turks from Greece to Turkey. While it is known that many expelled Greeks and Turks died in the process, the numbers remain unclear. Even earlier, beginning around 1914, some 1.5 million Armenian and mostly Christian residents of the collapsing Ottoman Empire died (once again the numbers are in dispute) during relocations or deportations by Ottoman Turkish forces. Most historians of the period describe this massive mortality event as the Armenian Genocide – a description that is forcefully rejected by many Turks (for recent examples, see Bloxham, 2005; Dadrian, 1995; De Waal, 2015; Kevorkian, 2011). Turkish President Recep Tayyip Erdoğan has said that these actions by Ottoman Turkish forces in the early twentieth century were “reasonable” at the time (Koc, 2019). The seemingly irresolvable conflicts between Israelis and Palestinians also have their origins in the Ottoman Empire, which held imperial sway in the region for six centuries beginning around 1300, but began a slow decline around 1800 until it finally collapsed after World War I. The government of Israel officially has encouraged immigration by world Jewry, while Palestinian organizations (formally not “governments”, but with some of their attributes) assert a “right of return” to Israeli territory for the millions of descendants of the Palestinians who departed during and since the first Arab–Israel War in 1947. As the Israeli–Palestinian example shows, such demographic engineering continues to make and remake states today, with consequences that spill over international borders. Following the collapse of the Socialist Federal Republic

DEMOGRAPHIC ENGINEERING AND STRATEGIC DEMOGRAPHY 187

of Yugoslavia in 1992 (a region that earlier was also part of the Ottoman Empire), ethnic and religious “unmixing” not seen for half a century in Europe reappeared. Nearly a decade of violent attacks within the successor republics of Yugoslavia resulted in the “ethnic cleansing” of minority population groups in different regions of the former Yugoslavia, most notably Bosnian Muslims, Serbs, Croatians and Albanians. Beginning in 2016 in Myanmar, similarly violent actions led by the Buddhist-dominated military and by politicized Buddhist monks coerced the mass departure of nearly one million Rohingya Muslim residents. Multiple similar examples have been seen during the chaotic events that continue within the borders of modern Syria and Iraq. These national security and humanitarian crises with demographic roots demonstrate the continued need for scholarship in this area.

Engineering immigration at sub-national levels Efforts at demographic engineering can also be seen at sub-national levels, often led by non-governmental interest groups seeking to influence national policies related to demography. In countries too numerous to mention, such interest groups – economic, political, ethnic, religious, racial, national, linguistic – have promoted migration policies that serve their interests. Scholars interested in research on civil conflict or national and sub-national identities would be well-served by attention to the impacts of demographic change in their cases, particularly when demographic engineering is involved. The most obvious groups are the employers and employer associations in high-income countries that have invested heavily in lobbying for laws and practices that allow them to freely recruit workers from low-wage countries, thereby depressing or limiting increases in their labor costs and weakening the ability of their workforces to unionize. Employers operating in low-wage and labor-intensive industries are especially active here – fruit and vegetable agriculture, meat and fish processing, hotels and restaurants, construction – as are employers in higher-wage industries such as information technology who recruit large numbers of temporary workers from low-wage countries (Hira and Hira, 2005). We even see these dynamics in party politics: some political parties and their supporters have attempted to expand the size of those demographic categories they consider most likely to support them. In some cases, such political interests may conflict with economic interests, as when wealthy employers who normally support center-right parties promote migration policies that lower

188 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

their labor costs but over time lead to increased constituencies for center-left parties. Meanwhile, in settings characterized by active politics of “identity” (e.g. ethnicity, race, nationality, religion, language, gender), interest groups have pursued strategies to increase their influence by encouraging demographic growth of the categories they claim to support. While some encourage larger families (see Chapters 6 and 7 in this volume, by Kaufmann and Toft, respectively), a more common approach seeks more immediate impact by expanding immigration of foreign nationals with related identities. Such advocacy has been actively pursued in the United States over the past decades, directed toward a wide range of immigration policies including permanent immigration; temporary visas; refugees and asylum-seekers; and migration variously described as irregular/unlawful/illegal/undocumented.2 Among the most active interest groups have been some claiming to represent minority populations including Hispanic/Latino/Latinx, Asian American, African American, Cuban and Haitian. Religious organizations have also been active, including, for example, the National Conference of Catholic Bishops, the Lutheran Immigration and Refugee Service and some Jewish organizations such as HIAS. Although such groups typically focus especially upon co-religionists, their arguments often invoke theological and moral principles that apply broadly. Some challenge the very legitimacy of national borders in limiting in-migration by people from poor countries seeking economic opportunity, or those fleeing autocratic governments. The latter argument was prominent and politically successful during the Cold War, especially on the part of some Jewish American organizations promoting preferential admission of co-religionists from the former Soviet Union,3 and by Cuban American groups lobbying for favorable admission of migrants from Castro’s Cuba. Advocacy focused on such identities has not been limited to efforts to increase in-migration. In some countries, advocates have supported harsh measures to encourage or coerce the out-migration of members of competing ethno-linguistic, religious, racial and tribal groups. These can range from discriminatory practices related to education or economic activities, to disproportionate allocation of public resources, to outright violence. Recent examples of the latter include violent attacks by Islamist radical groups upon Christians and other non-Muslim groups in Iraq, Syria, Egypt, Lebanon and Sri Lanka. In India, there have been violent attacks upon Indian Muslims by extremist Hindu nationalist groups, and a long history of violence against mostly Bengali migrants by indigenous groups in the northeastern “tribal” states. In Myanmar, United Nations organizations and human rights groups report that Buddhist monks joined the Myanmar military in acts of ethnic

DEMOGRAPHIC ENGINEERING AND STRATEGIC DEMOGRAPHY 189

cleansing and genocide that forced a million Rohingya Muslims to flee (United Nations, 2018).4

Engineering demographic data In some countries, interest groups have sought to engineer the collection, classification and reporting of demographic data in ways they believe will expand their political power and influence. One flagrant example is that of modern Lebanon, created after the League of Nations allocated to France its Mandate for Syria and Lebanon following the collapse after World War I of the multiethnic and multi-confessional Ottoman Empire. This case is discussed in some detail in Chapter 7 of this volume. A similar case is that of Nigeria, whose efforts even to conduct a national census have been mired in controversy. Census data determine the distribution of power and central government funding between the northern states, which are predominantly Muslim, and the predominantly Christian south and east where most of nation’s oil resources and much of its business activities are located. Most census efforts since Nigerian independence in 1960 have been canceled, discredited or had their data withheld (Metz, 1991). Indeed the controversies over the 1962 and 1963 censuses and related alleged manipulation of the 1964 national elections exacerbated longstanding political tensions that eventually led to the Nigerian civil war in 1967–70, which devastated the population of oil-rich Biafra (Campbell, 1976). The last census of Nigeria was in 2006, and is the first to be seen as reasonably credible. Contesting claims about population data also loom large in the seemingly intractable conflicts of the Middle East. The borders of the new state of Israel, carved out of previously Ottoman territory by the UN General Assembly in 1947, were never fully accepted by many Arab states in the region. These original borders were expanded following the ensuing wars with neighboring Egypt, Syria and Jordan, which also left Israel in occupation of the West Bank, the Golan Heights and the Gaza Strip. To this day, dueling claims prevail about the size of the Israeli Jewish population as compared with the Palestinian Arab population of the region. In general, Palestinian organizations claim far larger Palestinian populations than are reported by the statistical agencies of Israel. In the United States, political efforts over many decades have dramatically enlarged the number of “race” and related categories in the US Census. For example, beginning in the 1970s, Asian American interest groups and leg-

190 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

islators succeeded in mandating the US Census Bureau to disaggregate its previous Asian “race” category into 7+ new “race” categories that, as may be seen from the following list, are based primarily on nationality and ethnicity: Chinese, Filipino, Asian Indian, Vietnamese, Korean, Japanese and Other Asian. As part of the same recategorization, the previous Census “race” category “Native Hawaiian or Other Pacific Islander” was disaggregated into four more detailed “race” categories: Native Hawaiian, Samoan, Chamorro and Other Pacific Islander (US Census Bureau, 2020a). Meanwhile, an additional Census question was added, asking respondents of all “races” to report if they are of “Hispanic or Latino origin”, a Census category that can be of any “race”. The 2020 US Census questionnaire lists 14 categories of “race”, plus three additional write-in fields (US Census Bureau 2020c, Question 9). The Census Bureau states that “all of the changes were made in order to monitor compliance with laws, regulations, and policies related to anti-discrimination, civil rights and affirmative action, and to plan and fund government programs that provide funds for particular racial and ethnic groups” (US Census Bureau, 2020b). In his historical review of the Census “race” categories, Kenneth Prewitt, Director of the Census Bureau from 1998 to 2000, noted that beginning with the 2000 Census the “race” question invited respondents to “mark one or more”, leading to new reliability problems due to individuals providing different answers at different times. He added that the Census classification of “race” is typically cross-tabulated with the “Hispanic/non-Hispanic” question, leading to well over 100 ethno-racial groups in the 2000 decennial census data – a categorization that could become much larger were additional ethnic or multi-ethnic categories to be added in the future (Prewitt, 2005). Objective research could provide useful understanding about the extent to which these successful efforts to engineer official data on racial and ethnic groups were driven by the Census Bureau’s professionals or by political considerations, whether they have enhanced understanding or increased confusion about racial and ethnic categories, and their effects upon enforcement of civil rights laws and policies. Census Bureau data on “race” are widely used in political campaigns, most of which pay close attention to the relative size and distribution of what they call the “key demographics,” and to changes underway in these. Their focus is upon potential and actual voters broken down into categories such as race, ethnicity, religion, language, region, urban/rural, education, age, income/ wealth and so on. Parties and politicians retain political consultants to massage census and other demographic data to gain insights into these key voting blocs. Demographic groups that are growing relative to others often receive special

DEMOGRAPHIC ENGINEERING AND STRATEGIC DEMOGRAPHY 191

attention from politicians even if they are not (yet) large, and especially if the margins of electoral success are narrow such that differentially large growth of even small voting blocs might have leveraged impacts upon political outcomes. Meanwhile, US legislation and court rulings have enabled, and in some cases required, states to deliberately configure the boundaries of election districts to ensure that some have voting majorities of under-represented racial minorities.

Engineering migration for foreign policy: “strategic demography” Many governments have seen their migration policies as useful strategies of foreign policy. Trade negotiators often offer visa preferences as part of trade agreements. Special temporary work visas of this type, with alphabet-soup designations such as H-1B, H-1B1, TN and E-3 were agreed to by the United States in its trade agreements with Canada, Mexico, Australia, Singapore and Chile. Meanwhile, special visas for refugee resettlement are offered by many countries in support of foreign policies promoting international human rights and social justice. During the Cold War, the refugee policies of the USA and other Western countries were especially accommodating to citizens of the USSR, other Warsaw Pact countries, and Cuba, in part in support of human rights, in part to discredit those regimes by demonstrating that even their own citizens did not want to reside there. Special access for such refugees proved sustainable only so long as the numbers able to emigrate were small. Once Communist states began to allow large-scale departures, these open-admission policies attracted criticism and were increasingly constrained. The case of Cuban migration to the USA became especially fraught after the Cuban government encouraged and even coerced the out-migration of significant numbers of inmates from Cuban prisons and mental hospitals during the chaotic Mariel boatlift of 1980. In Weapons of Mass Migration, Greenhill (2010) identified over 50 cases since 1950 involving the intentional creation, manipulation or exploitation of real or threatened mass population movements as tools of asymmetric foreign policies. Most commonly, such strategies are deployed by poorer and weaker countries against more prosperous liberal democracies nearby. Greenhill concludes that the goal is to threaten either to overwhelm their capacity to accommodate large and disorderly migratory flows, or as a kind of “normative blackmail” that exploits the targeted countries’ domestic politics and expressed

192 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

commitments to protect migrants fleeing violence, persecution or poverty. Her assessment is that well over half of these cases of mass migration deployed as a foreign policy weapon have proved successful (Greenhill, 2010). A recent and unusually explicit use of out-migration for such foreign policy goals has been that by the Turkish government since 2015 in enabling irregular departures of small boats organized by smuggler networks from Turkish territory to nearby Greek islands. This contributed greatly to the divisive “European migration crisis” of 2015–16, following the disorderly entry of more than one million migrants trekking into and across the territory of multiple EU countries. Most were ultimately admitted to Germany by the mainstream coalition government led by Chancellor Angela Merkel. Major foreign policy initiatives ensued. In fraught negotiations with the European Union in 2016, Erdoğan demanded and ultimately received promises of €6 billion of financial support, along with commitments to facilitate entry visas for Turkish nationals and to restart lapsed negotiations about Turkish admission to the EU. In return, Erdoğan promised to restrain irregular migratory movements from Turkish territory to Greece and other European countries.5 These arrangements have proved unstable however, leading to a continuing diplomatic saga as the Turkish government has repeatedly threatened to suspend such migration restraints on grounds that the EU has not fulfilled its commitments. President Erdoğan himself has been direct and outspoken in warning European countries that he can once again facilitate large-scale migrations toward Europe from among the millions of Syrian, Afghan and other migrants who have been admitted to Turkey on humanitarian grounds but wish to move on to economically more attractive EU countries. This crisis – which remains far from resolution as of this writing – has in turn had major impacts on domestic politics in many EU countries. In general, it has weakened public support for mainstream center-right and center-left parties and strengthened left-wing and especially right-wing parties. It has also led to sharp challenges to fundamental elements of the European project, including the foundational 1957 Treaty of Rome, the 1992 Maastricht Treaty, the 1997 Treaty of Amsterdam, the 2004 Directive on Free Movement of Citizens, the 1990 Schengen Convention and the 2014 Dublin Regulation (Teitelbaum, 2015, 2017). Many provisions of these important agreements have been suspended or ignored by member states and, in general, the subject of migration continues to be a highly divisive one in the multilateral politics of the European Union and in the domestic politics of its members. These developments suggest that Greenhill’s book, published in 2010, offered pres-

DEMOGRAPHIC ENGINEERING AND STRATEGIC DEMOGRAPHY 193

cient foresights into the potential of facilitating or constraining migratory movements as instruments of asymmetric foreign policy. More militant uses of migration have been deployed by other states. In 2006, the Libyan dictator Muammar Gaddafi famously forecast that large-scale Islamic migration and eastward expansion of the EU would eventually transform Europe into an Islamic rather than predominantly Christian continent (Gaddafi, 2006). Violent Islamic militant groups such as al-Qaeda and ISIS are also alleged to be using the disorderly flow of asylum seekers to Europe as cover for inserting their operatives into European countries (Europol, 2017). Meanwhile, other non-governmental Islamist groups have taken advantage of Western countries’ provision of visas for religious personnel to dispatch militant Islamist clerics as religious guides for Muslims resident in European countries, while Saudi Arabia has provided substantial funding for the building of mosques in European countries, some of which are led by fundamentalist Salafi or Wahhabi clerics. In a somewhat different format, directed migrations have been deployed by governments as an unconventional means to assert their sovereignty over regions outside their borders, or to establish more effective control in areas over which central government control is contested. Recent examples include directed migration from Morocco to former and current territories of Spain, and of Han Chinese to Tibet. In 1975, the government of Morocco organized the “Marche Verte” by 350,000 unarmed Moroccan civilians into the contiguous territory of the longtime Spanish colony of Spanish Sahara, as the Spanish government moved toward its de-colonization. In doing so, the Moroccan government asserted its claims to historical sovereignty over the territory, claims that were and continue to be contested by indigenous tribal groups and other countries in the region. Continuing conflict over this territory may also explain why on 17–18 May 2021, Moroccan border guards allowed up to 12,000 irregular migrants to cross its border into the small neighboring Spanish enclave of Ceuta. This occurred only weeks after the Moroccan foreign ministry had warned the Spanish government that there would be consequences for its April 2021 decision to admit for COVID-19 hospitalization a leader of the rebels who have long challenged Moroccan sovereignty over Spanish Sahara (Casey and Bautista, 2021). With respect to China and Tibet, the Chinese government has long sought to integrate Tibet into the Chinese motherland by incentivizing the migration by large numbers of ethnic Han Chinese (who comprise well over 90 percent

194 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

of the Chinese population of 1.4 billion) into Tibet, with its relatively tiny population of about 3 million. The Tibetan government in exile in India claims that Han Chinese living in Tibet now outnumber ethnic Tibetans, and that the malign goal of the migration policy is to marginalize Tibetans and Tibetan culture in the region. The Chinese government claims that its benign goal is to modernize a backward region and spur local economic development (AsiaNews (Italy), 2015). Demographic engineering and international human rights regimes One of the most important international achievements since World War II has been the adoption of hundreds of international agreements related to human rights, some of which address either fertility or migration. The list begins with the foundational Universal Declaration of Human Rights adopted by the UN General Assembly in 1948. Article 16 of the Universal Declaration describes the family as “the natural and fundamental group unit of society”, on the basis of which later international declarations affirmed that coercive measures to affect fertility decisions by families constitute violations of basic human rights. (For example, “any choice and decision with regard to the size of the family must inevitably rest with the family itself, and cannot be made by anyone else”.)6 Over the ensuing half century, most national governments have respected such provisions. However, some continued to prohibit certain methods by which families can most effectively implement their fertility decisions, and in a few notable cases governments have breached human rights principles by excessive responses – both pro-natalist and anti-natalist – to concerns that fertility levels were either too low or too high. As noted earlier, during the 1960s and 1970s the government of Romania prohibited nearly all modern methods of contraception and delegalized abortion, with the declared goal of increasing fertility rates in Romania (Teitelbaum, 1972). Meanwhile, beginning in the 1970s, the governments of the world’s two largest countries, India and China, adopted anti-natalist policies that limited family size, leading in some localities to harsh and even coercive enforcement. The stringency of such policies in India proved to be politically unpopular and short-lived, but China’s vigorous legal limits on family size were sustained for more than three decades. To their credit, most international demographic researchers were critical of such measures, although those in the countries engaged in such policies often were understandably reticent. With respect to migration, another foundational UN human rights agreement was adopted only three years after the UN’s Universal Declaration of Human

DEMOGRAPHIC ENGINEERING AND STRATEGIC DEMOGRAPHY 195

Rights. The 1951 UN Refugee Convention (relating to the status of refugees)7 was grounded in the Universal Declaration’s Article 14, but more explicitly recognized the right of all persons to seek asylum from persecution in their country of nationality and provided specific criteria for such claims. Many provisions of this Convention were subsequently incorporated into national and multinational legislation, including those of the United States and the European Union. Recent actions to engineer migration, such as those deployed by the governments of Turkey, Syria, Myanmar and others (see above), have posed serious challenges to these refugee agreements and legislation, none of which appears to have anticipated that states might deploy out-migration as instruments of foreign policy. In response, well-established principles and agreements related to refugees and asylum-seekers have been suspended in many countries that have long been supporters of human rights. A number of governments have acted to deny entry to would-be migrants on the high seas, in alleged violation of other longstanding international agreements. There also are allegations that governments such as those of Italy and Greece have facilitated the movement of asylum-seekers in their territories toward the borders of other EU member states. Until recently, the government of Mexico allowed, and sometimes tacitly facilitated, the movement through its territory of irregular migrants from third countries to the weakly regulated southern border of the United States. In recent years, most such migrants were Central American nationals, who were allowed to cross Mexico’s southern border with Guatemala and travel across Mexico in order to enter the United States unlawfully or to claim asylum at its border with Mexico. Other would-be migrants from around the world were able to do the same by taking advantage of Mexican government policies that made short-term visas readily available. These policies were modified only after vigorous diplomatic representations from the United States that included threats to close its border with Mexico and to impose stiff tariffs on the large volume of Mexican exports to the United States. Meanwhile, the United States government has sharply constrained its previous handling of asylum claims at its borders, adopting measures that human rights advocates assert violate its international agreements and domestic legislation. The domestic politics are fraught, emotional and increasingly partisan, and have been complicated by numerous cases of conflicting court decisions in response to lawsuits claiming that these actions are unlawful or unconstitutional.

196 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

These and other recent experiences illustrate the challenges to international human rights regimes now being posed by out-migration and transit migration practices embraced by some states’ foreign policies. They also demonstrate how states are increasingly deploying foreign policy negotiations, including those related to trade, to influence policies of other states related to migration. Together with Chapter 3 in this volume by Chu, these examples point to a need for further research on the global and national migration decisions being adopted by various states in response to such challenges.

Conclusion Interactions between political and demographic change, both domestic and international, are of increasing salience. Ill-informed or shortsighted policies by some national governments, coupled with efforts by interest groups to engineer demographic patterns and data, have led to rising tensions in domestic politics and international relations, and stimulated troubling challenges to hard-won gains related to international human rights. In view of the checkered and often polemical history at the intersection of politics and demography, it is understandable that some political scientists and demographers have preferred to shun the subject in favor of less contentious research topics. The result of such avoidance is counterproductive, as it surrenders the terrain to committed interest groups and ideologues who cannot be expected to be objective while depriving political leaders of important evidence and understanding derived from objective data and research.

Notes 1. For further discussion, see Teitelbaum and Winter, 1985, pp. 36–9, 56–82. 2. Even the terminology used to refer to such categories is politically contentious; they are used interchangeably here, with no negative connotations intended. 3. In one extraordinary case during the Cold War, the Jewish American group HIAS allegedly dispatched staff members to Vienna airport to encourage Soviet Jewish migrants making flight connections on their way to Israel to instead travel on to the United States (Nocera, 1982). 4. These allegations are rejected by the Government of Myanmar on grounds that Rohingya do not hold Myanmar citizenship and that the military actions were in response to violent attacks by Rohingya militants (BBC News, 2020). 5. Similar negotiations have been undertaken by the EU with North African migrant transit countries such as Libya and Morocco.

DEMOGRAPHIC ENGINEERING AND STRATEGIC DEMOGRAPHY 197

6.

“Statement on population by the UN Secretary General, U Thant,” Human Rights Day, December 10, 1967. See also Final Act of the International Conference on Human Rights, Tehran, Republic of Iran, April 22 to May 13, 1968, Resolution XVIII: Human Rights Aspects of Family Planning, https://​www​.un​.org/​en/​ development/​desa/​population/​theme/​rights/​index​.asp. 7. See https://​www​.unhcr​.org/​en​-us/​3b66c2aa10. The 1951 Convention was originally limited to the European region but was later expanded globally via its 1967 Protocol.

References AsiaNews (Italy) (2015). “Beijing sends a new flood of Han migrants to Lhasa: Tibetans risk disappearing,” January 27, http://​www​.asianews​.it/​news​-en/​Beijing​-sends​-a​ -new​-flood​-of​-Han​-migrants​-to​-Lhasa:​-Tibetans​-risk​-disappearing​-33294​.html. BBC News (2020). “Myanmar Rohingya: Government rejects ICJ ruling,” January 23, https://​www​.bbc​.com/​news/​world​-asia​-51229796. Bloxham, D. (2005). The Great Game of Genocide: Imperialism, Nationalism, and the Destruction of the Ottoman Armenians. Oxford: Oxford University Press. Campbell, Ian (1976). “The Nigerian census: an essay in civil‐military relations,” The Journal of Commonwealth & Comparative Politics, 14(3), 242–54. Casey, N. and J. Bautista (2021). “‘Come on in, boys’: A wave of the hand sets off Spain–Morocco migrant fight,” New York Times, June 2, https://​www​.nytimes​.com/​ 2021/​06/​02/​world/​europe/​spain​-ceuta​-migrants​-morocco​.html. A print version was published with the headline “Morocco Sends Spanish Outpost a Migrant Influx,” New York Times, June 3, Section A, Page 1. Clemenceau, G. (1919). Journal Officiel: Débats du Sénat, October 11, pp. 1625–6, cited in Tomlinson (1985). Dadrian, V. (1995). The History of the Armenian Genocide: Ethnic Conflict from the Balkans to Anatolia to the Caucasus. Oxford: Berghahn Books. De Waal, T. (2015). Great Catastrophe: Armenians and Turks in the Shadow of Genocide. Oxford: Oxford University Press. Europol (2017). European Union Terrorism Situation and Trend Report 2017. Brussels: European Union Agency for Law Enforcement Cooperation. Gaddafi, M. al- (2006). “Islam will conquer Europe without firing a shot,” speech in Timbuktu, AlJazeera TV, April 10, translated into French and English, https://​www​ .youtube​.com/​watch​?v​=​WCGYKSEsYFM. Greenhill, K. M. (2020). Weapons of Mass Migration: Forced Displacement, Coercion, and Foreign Policy. Ithaca and London: Cornell University Press. Hira, R. and A. Hira (2005). Outsourcing America: What’s Behind Our National Crisis and How We Can Reclaim American Jobs. New York: AMACOM. Kevorkian, R. (2011). The Armenian Genocide: A Complete History. London: I. B. Tauris. Keynes, J. M. (1937). “The economic consequences of a declining population,” Eugenics Review, 29, 13–17. Koc, C. (2019). “Erdogan says deporting Armenians was ‘appropriate’ at the time,” Bloomberg, April 24, https://​www​.bloomberg​.com/​news/​articles/​2019​-04​-24/​ erdogan​-says​-deporting​-armenians​-was​-appropriate​-at​-the​-time.

198 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Metz, H. C. (1991). Nigeria: A Country Study. Washington: GPO Library of Congress. Nocera, J. (1982). “Tales of the Vienna Airport”, Harper’s Magazine, May, https://​ harpers​.org/​archive/​1982/​05/​tales​-of​-the​-vienna​-airport/​. Prewitt, Kenneth (2005). “Racial classification in America: where do we go from here?” Dædalus, 134(1), 5–17. Teitelbaum, M. S. (1972). “Fertility Effects of the Abolition of Legalized Abortion in Romania,” Population Studies, 26 (November), 405–17. Teitelbaum, M. S. (2004). “The media marketplace for garbled demography,” Population and Development Review, 30(2), 317–27. Teitelbaum, M. S. (2015). “The Truth about the Migrant Crisis,” Foreign Affairs, September, https://​www​.foreignaffairs​.com/​print/​1115410. Teitelbaum, M. S. (2017). “Europe’s migration dilemmas: unavoidable and unresolved,” Foreign Affairs, May, https://​www​.foreignaffairs​.com/​articles/​europe/​2017​-05​-11/​ europe​-s​-migration​-dilemmas. Teitelbaum, M. S. and J. Winter (1985). The Fear of Population Decline. New York and London: Academic Press. Teitelbaum, M. S. and J. Winter (1998). A Question of Numbers: High Migration, Low Fertility, and the Politics of National Identity. New York: Hill & Wang, a division of Farrar, Straus, Giroux. Tomlinson, R. (1985). “The disappearance of France, 1896–1940: French politics and the birth rate,” Historical Journal, 28(2), 405–15. United Nations (2000). Replacement Migration: Is It a Solution to Declining and Ageing Populations? Department of Economic and Social Affairs, Population Division, http://​www​.un​.org/​esa/​population/​publications/​migration/​migration​.htm. United Nations (2018). Report of the Independent International Fact-Finding Mission on Myanmar, https://​www​.ohchr​.org/​Documents/​HRBodies/​HRCouncil/​FFM​ -Myanmar/​A​_HRC​_39​_64​.pdf. US Census Bureau (2020a). “2020 Census Questions: Race,” https://​2020census​.gov/​en/​ about​-questions/​2020​-census​-questions​-race​.html. US Census Bureau (2020b). American Community Survey, “Why We Ask Questions About … Hispanic or Latino Origin,” https://​www​.census​.gov/​acs/​www/​about/​why​ -we​-ask​-each​-question/​ethnicity/​. US Census Bureau (2020c). “US Census 2020, Informational copy,” Question 9. OMB No. 0607-1006, https://​www2​.census​.gov/​programs​-surveys/​decennial/​2020/​ technical​-documentation/​questionnaires​-and​-instructions/​questionnaires/​2020​ -informational​-questionnaire​.pdf. Weiner, M. and M. S. Teitelbaum (2001). Political Demography, Demographic Engineering. New York and Oxford: Berghahn Books. World Bank (2019). Migration and Remittances, September 26, https://​www​.worldbank​ .org/​en/​topic/​labormarkets/​brief/​migration​-and​-remittances\.

13

The demographic dividend: positive prospects, unclear path

Kaitlyn Patierno, Elizabeth Leahy Madsen and Smita Gaith

Introduction In January 2016, the African Union (AU) Assembly announced the decision of African heads of state to establish the AU’s theme for 2017 as “Harnessing the Demographic Dividend through Investments in Youth”. The announcement was the culmination of a decade of growing enthusiasm around a compelling objective: attaining the demographic dividend (DD), or the accelerated economic growth countries can achieve as their age structure matures through the demographic transition from high to low fertility and mortality. Attention to the relationship between population dynamics and socioeconomic development was not new. Faced with rapid population growth that often outpaced the government’s capacity to provide infrastructure and services, many African heads of state had adopted policy frameworks on population and development beginning with the first African Population Conference in 1971. Between the 1970s and early 2000s, global, regional and country-led initiatives on population and development were largely rooted in Malthusian arguments about population growth and resource scarcity. In contrast, the DD concept is optimistic and appealing, offering the prospect of economic benefit following demographic change. Yet, while enthusiasm remains high, many DD-oriented development agendas are both broad and vague, and progress towards achievement of the DD’s potential economic boost remains elusive for many countries.

199

200 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

In response to the need for targeted guidance for countries that wish to attain a DD, this chapter identifies the demographic and broader health, human capital and economic policy prerequisites, drawing on a review of the surprisingly scant literature on specific policy interventions across sectors that motivate or inhibit attainment of the dividend.

Explaining the dividend Population age structure, or the share of the population in key age groups, has important social and economic implications. Age structure is often examined through the dependency ratio, which measures the size of the working-age population (frequently defined as ages 15–64) relative to the size of the non-working age, or dependent, population. The dependency ratio construct assumes that the income generated by the working-age population supports those who are economically dependent, such as children and seniors. The demographic transition is the long-term shift in birth and death rates from high to low levels in a population. The transition begins with mortality declining while fertility remains high, resulting in rapid population growth during the early transition period. Later, once there is a substantial and sustained drop in fertility rates, the dependency ratio also declines. With fewer young dependents to support, countries experience a decades-long window of opportunity to attain the first demographic dividend, which is driven by the growth in the working-age population. As the working-age population matures, a productive workforce can increase household savings and investments. As a result, national incomes can rise further and countries may attain a second demographic dividend sustained even longer than the first (Mason et al., 2017). Countries can attain the first DD through two broad pathways. First, all else being equal, the shift toward a larger proportion of the population in the working ages will automatically yield higher per-capita incomes. This phenomenon is referred to as the “mechanical effect” of age structure change, as the resulting income growth is driven automatically by the change in population age structure, without any further public policy investments – but its effects are relatively moderate and short-lived. Second, the mechanical effect can be compounded if labor productivity also increases through investments in human capital and the macroeconomic climate. The combination of these two pathways yields the full DD; among the East Asian “Tigers”, the most well-documented examples, successful attainment of the DD boosted gross domestic product (GDP) growth by up to two percentage points annually, or

THE DEMOGRAPHIC DIVIDEND: POSITIVE PROSPECTS, UNCLEAR PATH 201

one-third of their total growth in the late twentieth century (Bloom & Sachs, 1998; Bloom & Williamson, 1998; Mason, 2001). The potential benefits of age structure change are not limited to the economy. Recent research expands the DD concept, showing how age structure change opens a window of opportunity to attain benefits – or dividends – in four sectors: health, education, the economy and political stability. Using median age as a measure of age structure change and applying statistical analysis to a panel of over 100 countries, Cincotta (2018) projected that the likelihood of countries achieving specific development benchmarks in each of those sectors increases as fertility declines. The results for each of the “Four Dividends” (The PACE Project, Population Reference Bureau, 2019) show the following: • Political Stability: At a median age of 27, countries become 50 percent likely to be free of revolutionary conflict for at least five years. • Health: At a median age of 29, countries become 50 percent likely to attain the Sustainable Development Goals target for child survival – an under-five mortality rate at or below 25 deaths per 1,000 live births. • Education: At a median age of 25, countries become 50 percent likely to attain a secondary school enrollment rate of 60 percent. Advancing from educational enrollment to attainment becomes likely later in the age structure transition: A country becomes more likely to achieve late secondary attainment rates at or above 60 percent among young adults at a median age of 31 years. • Economy: At a median age of 31, countries become 50 percent likely to attain upper-middle income status. While the Four Dividends analysis does not predict countries’ age structural trajectories, it can estimate the probability of countries with a given median age having attained or not attained each benchmark. Figure 13.1 plots the likelihood of Nepal attaining each of the Four Dividends benchmarks in 2030 and 2050 based on its projected median age if current fertility trends persist. As demonstrated in the case of Nepal, while the odds of attaining these benchmarks are low for youthful countries, they accelerate quickly once the demographic window of opportunity opens at a median age of 25 to 26 years. Median age is a relatively crude demographic measure; any given median age can encompass several diverse age structures. Still, no country has reached a median age of 26 until its fertility rate has declined below three children per woman. And once the window is open, appropriate policy structures must already be in place, as countries only have a few decades to benefit from the first dividend. When median age reaches 40 to 41 years, the size of

202 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Figure 13.1

Percent likelihood of achieving four dividends by median age in Nepal

THE DEMOGRAPHIC DIVIDEND: POSITIVE PROSPECTS, UNCLEAR PATH 203

the working-age population shrinks relative to dependent older adults. At this point in the demographic transition, the first DD is no longer possible, although if its benefits have been attained, a second DD can offset the potential economic drains of an aging population.

The search for policy implications While the benefits of the DD are proven and can be experienced across policy sectors, they are not straightforward to attain. The number of countries that have successfully achieved a DD is few – less than ten according to the body of economic research, and mostly concentrated in East Asia. Further, for age structure change to yield more than a simple mechanical economic benefit, the DD requires complex interactions between demography, human capital, labor economics and governance variables, many of which are difficult to measure. A large body of literature incorporates policy levers into analyses of the relationship between demographic factors, including fertility and population growth, and economic change. While various models project the potential for a future DD and incorporate policy variables, very few studies explicitly quantify the effect of defined policy interventions, within any sector, to attainment or failure to attain economic growth in the context of specific past examples of age structure change. Even studies that analyze the contribution of policy interventions to a DD in a specific country or setting without quantifying their impact on economic growth are limited. Therefore, efforts to translate economic and demographic research findings into actionable evidence have largely stalled around a broad and fairly general set of recommendations: age structure change through fertility decline motivated policies influencing women’s fertility choices, including increased adoption of family planning; investments in human capital, including health and education; and improvements to employment and governance. Below, we summarize findings from the few studies that do analyze specific policy variables on the DD, grouped according to sector.

Fertility The theoretical link between fertility, age structure and economic growth is both intuitive and well-documented. Interventions that contribute to fertility decline have two potential channels to promote a DD: by stimulating age

204 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

structure change and by contributing to increased female labor force participation. The contribution of age structure change to growth in income per capita through the mechanical effect (sometimes termed the compositional or accounting effect) is well-documented in the DD literature (Bloom et al., 2003; Bloom et al., 2010; Bloom & Sachs, 1998; Bloom & Williamson, 1998; Cuaresma et al., 2014). When fertility falls, the size of the working-age population increases relative to the population of young-age and old-age dependents who do not generate income. Assuming new entrants can find gainful employment in the labor market, the shift toward a larger proportion of the population in the working ages will automatically lead to increases in income per capita. The legalization of contraception and expansion of family planning programs have been linked to economic growth in several countries. After contraception was legalized in Ireland in 1979, age structure change had become the largest driver of per capita income growth by 1995, comprising nearly three percentage points, or approximately one-half, of total annual growth (Bloom and Canning, 2003). In Japan, the removal of legal barriers to contraception in 1947 and resulting decline in the child dependency ratio stimulated rapid economic growth, but this period passed quickly (Inoue, 2001). Japan’s demographic window was open for an abbreviated period (less than 40 years, from the early 1960s to 2000) due to precipitous fertility decline, which created rapid population aging and heavy demands on the pension system (Bloom et al., 2003). In Indonesia, the creation of the State Ministry on Population and the National Family Planning Coordinating Board facilitated integration of population issues and services into broader development planning. Indonesia’s family planning program initially focused on service delivery and outreach to the most populous islands and communities, and then expanded to harder-to-reach coastal and remote areas. Ultimately, these policy and program investments produced widespread knowledge and use of contraceptives and a shift to smaller families (Pasay and Wongkaren, 2001). The link between access to and use of contraception and fertility decline is demonstrated in a broad literature (Bongaarts, 1984; Miller & Babiarz, 2016; Tsui, 2001). Interventions that contribute to fertility decline, including expanded access to contraception and abortion, have also been linked to economic growth through female labor force participation. In Ireland and South Korea, increases in female participation in the formal labor market driven by fertility decline reinforced age structure changes by further expanding the size of the paid workforce (Bloom et al., 2009; Bloom & Canning, 2003). Across countries, women’s tenure in the labor force is reduced by an average of 1.9 years per

THE DEMOGRAPHIC DIVIDEND: POSITIVE PROSPECTS, UNCLEAR PATH 205

woman for every one child increase in the total fertility rate. The same analysis suggests that, over the course of the demographic transition, a four-child reduction of the total fertility rate is associated with an 18-percentage point increase in the female labor supply, ultimately leading to an estimated 7 percent increase in income per capita (Bloom et al., 2009).

Human capital development No country has achieved the DD without investing in its people. Improvements in human capital – people’s health and well-being, education and skills – can drive sustainable development by generating a more productive, competitive workforce. As child survival improves, couples’ motivations to have large families to provide “insurance” against potential child mortality diminishes, and as fertility declines, parents can invest more in each child’s schooling, nutrition and health care. Household investments in the “quality” of each child ultimately contribute to improved labor productivity, as healthier and better-educated children grow into more economically productive adults. Likewise, lower fertility can facilitate improvements in educational attainment among girls and young women, promoting higher female participation in the formal workforce. Governments can make policy choices to facilitate these mutually reinforcing benefits in the context of age structure change. The Republic of Korea is one of the four East Asian Tigers celebrated for extraordinarily rapid economic growth rates between the 1960s and 1990s. In 1960, the Republic of Korea had a GDP per capita of just $158, but three decades later, GDP per capita had reached $6,516 (The World Bank, n.d.). This rapid economic growth was fueled by purposeful investment in human capital development. In 1962, the government initiated the National Family Planning Program, which promoted the benefits of smaller families and made contraception widely available and accessible. Through these investments, the total fertility rate fell rapidly from 6.3 to 2.1 children per woman between 1960 and 1983. The early 1960s also saw a concerted effort by the government to restructure the educational system (Gribble, 2012). Applying a “sequential bottom-up approach”, Korea first instituted compulsory primary education in 1954, achieving near-universal enrollment by 1959 (Lee et al., 2012). The government then addressed growing demand for secondary education by instituting policies to promote access, including substantial public investment in schools. Subsequent policy changes in the 1970s and 1980s reduced regulations on tertiary education enrollment. This program of reform ultimately produced one of the most educated populations in the world. Simultaneously,

206 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

investments in universal health coverage starting in the 1970s led to increased life expectancy and rapid reductions in infant mortality (Lee et al., 2012). Today, the Republic of Korea ranks second in the world on the World Bank’s Human Capital Index and enjoys a GDP per capita of $31,263 (The World Bank Human Capital Project, 2018). Health There is little research that both quantifies the contributions of increased life expectancy to economic growth in the context of age structure change and explicitly identifies policy investments undergirding improved health outcomes. Bloom et al. (2010) explored the demographic and policy factors influencing economic growth in China and India and found that increased life expectancy was the biggest driver of higher growth rates in both countries. In the case of China, these gains in life expectancy are attributed to government investments in disease prevention and public health. Some studies find a small, positive effect of improved health on economic growth in the context of age structure change, but do not assess underlying policy interventions (Acemoglu & Johnson, 2007; Bloom et al., 2004; Cervellati & Sunde, 2009; Shastry & Weil, 2003; Weil, 2007). These limited findings suggest that further research quantifying the contribution of policies that improve population health to attainment of the DD is needed. Investments in preventive health measures, including family planning and immunization services, were qualitatively identified as an impetus for fertility decline in seven East Asian countries, thereby contributing to attainment of the DD (McNicoll, 2006). All seven countries invested in building systems that expanded access to health care across the country, including in remote villages. Education While many studies of the DD include variables measuring education, most do not make any reference to related policy interventions. Bloom and Canning (2003), discussed above, did not find a significant association between average years of school of the workforce and income per capita in Ireland. However, they noted that public education was free beginning in the 1960s, which led to increased school enrollment. Ssewamala (2015) suggested that education can both shift the labor force from informal to formal employment and increase competitiveness in global markets. Using examples from the Middle East and Africa, Ssewamala argued that countries with an undereducated or unskilled workforce will not be able to realize the DD.

THE DEMOGRAPHIC DIVIDEND: POSITIVE PROSPECTS, UNCLEAR PATH 207

It is important to note that girls’ education has a known bidirectional association with contraceptive uptake and reduced fertility. Secondary education, in particular, has transformative health and economic benefits for girls, including delaying age at first marriage and age at first birth. Therefore, educational outcomes and age structure outcomes driven by changes in reproductive health are not mutually exclusive, and separating the impact of these interventions in the literature is challenging. Education also acts as a potential driver of increased workforce size and productivity, as women with secondary or tertiary education are more likely to participate in the formal labor market and can expect to earn more compared to women with primary or no education (Wodon et al., 2018). There is broad global commitment to closing gender gaps in education, and data reflect substantial progress. In most regions of the world, the gender enrollment ratio in secondary education is equitable or favors girls; in the Middle East and North Africa and sub-Saharan Africa, girls continue to face moderate inequity in secondary enrollment, at ratios of 0.94 and 0.88, respectively, relative to boys (The World Bank, 2019). Globally, women complete 0.8 fewer years of education over the course of their lifetimes than men (Jakiela & Hares, 2019). Human capital development and labor force participation Many studies of the effect of investments in human capital suggest that to maximize the DD, the economic benefits of the mechanical effect must be at least partially reinvested in the human capital of future workers. Declines in fertility initially introduce increases in gross national product (GNP) per capita, through the mechanical effect of increasing the ratio of the working to non-working population (Lee & Mason, 2010). However, without investments in human capital that enhance the productivity of the working-age population, this brief first dividend tapers out within one generation. In such settings, the dividend ultimately reverses into a negative effect on GNP per capita as a relatively low-productivity population ages. In contrast, if investments in human capital are made while fertility declines, GNP per capita eventually stabilizes at around 40 percent above its pre-fertility transition level. One such example is Brazil, which was unable to fully capitalize on its demographic transition due to its policy environment. Although the large working-age population was initially a boon to Brazil’s economic growth, rather than invest those returns in human capital development of its young people, policies channeled them to the national transfer system, which a large elderly population continues to draw upon. This ultimately suppressed the magnitude of Brazil’s DD (Queiroz et al., 2006).

208 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

The gender dividend Cutting across all areas of human capital investment, the intersection of fertility, girls’ education and female labor force participation underscores that investment in women and girls must be at the center of efforts to attain the DD. Recognition of women’s critical contributions to the DD has given rise to the emerging concept of the “gender dividend”, which refers to the increased contribution of women’s work to the economy as fertility falls and greater gender parity is achieved in both education and formal workforce participation. Estimates suggest that halving the gender gap in labor income by 2050 would quadruple annual growth of India’s support ratio, or the number of economic producers per consumer (Donehower, 2016). Countries aiming to capture the benefits of a DD will benefit from gender-focused interventions that facilitate women’s greater participation and equity within the formal labor market. While an array of policy interventions are available to governments to address gender gaps in the labor force, it is critical that governments prioritize efforts to reduce the double burden of work that employed women face in balancing jobs and gendered expectations around household responsibilities, including care for children and the elderly (Pignatti, 2016). When women have equitable access to education and employment opportunities but face inequitable expectations of work performed at home, many choose paid work without the encumbrances of marriage and childbearing, leading to sharp declines in fertility rates such as those recently seen in East Asia (Anderson & Kohler, 2015).

Macroeconomic climate Analyses of the impact of the economic policy within the context of age structure change are few, but they consistently identify positive effects from openness to trade and foreign investment. Closed economic policies account for 32 percent of the growth gap in Africa between the period of independence and the late 1990s compared to East Asia, where many countries had open trade and pro-industrial policies that promoted manufacturing and exports, and less dependence on agricultural commodities (Bloom & Sachs, 1998). In Ireland, policies encouraging direct foreign investments and promoting exports were well-timed with policies that reduced cost barriers to education and enabled contraceptive uptake (Bloom & Canning, 2003). Together, these policies positioned Ireland to benefit from age structure change, ultimately yielding a dividend.

THE DEMOGRAPHIC DIVIDEND: POSITIVE PROSPECTS, UNCLEAR PATH 209

The economic “miracles” to which age structure change contributed in East Asia were not replicated in Latin America, despite the latter region’s own fertility decline. Between 1975 and 1995, annual per capita GDP growth in Latin America was one-eighth the level enjoyed by East Asia. Policy priorities of governments in the region directly hindered the potential to capitalize on age structure change, compared to those established by the East Asian Tiger economies. Barriers to trade, corruption and unstable financial systems that disincentivized savings inhibited several countries in Latin America from reaping the DD (Bloom et al., 2003). These findings are consistent with a robust literature documenting the impact of policy interventions that promote foreign trade and investment on economic growth. However, as with other sectors, few such studies incorporate age structure change in their analysis. Greater effort to quantify the combined effect of age structure change and economic policy interventions is needed.

Governance While several studies address the role of effective governance on attainment of the DD, there is scant effort to quantify the effect of specific policy interventions. Composite variables measuring quality of institutions do appear periodically. Factors such as rule of law, quality of bureaucracy, corruption, property rights and contract regulation may be associated with accelerated economic growth in the context of age structure change, but the findings are limited (Bloom & Sachs, 1998). Previous research has identified a strong link between the demographic transition and governance as measured by regime type, but has not addressed the relationship with economic growth. Countries are unlikely to achieve and sustain democracy until their age structure has significantly matured (Cincotta & Doces, 2012). At a median age of 29, a hypothetical country has a 50 percent likelihood of being classified as “free” in its political rights and civil liberties, and as median age rises further, so does the likelihood of freedom (Cincotta, 2017). However, many of the East Asian Tigers were not fully democratic during the years when they attained the DD – and many of the macroeconomic policies identified as necessary, especially reducing social security benefits for retirees, could receive an unfavorable reaction from the voting public in a democratic system. Further research to document the contribution of specific policy interventions to enhance good governance to attainment of the DD is warranted.

210 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

The elusive roadmap to a dividend As countries continue to develop demographic dividend roadmaps and incorporate the DD into national development and vision strategies, it is important to further build and clearly articulate the evidence base for policy recommendations. While evidence around specific policy interventions in the literature is scant, some takeaways do emerge. First, investments in voluntary family planning can help countries attain the dividend by accelerating age structure change and contributing to increased female labor force participation. Evidence around the contribution of fertility decline to the DD through the mechanical effect on age structure change is reinforced by the smaller body of literature that incorporates policy interventions, which finds a positive effect of policies and programs that enable couples to plan the timing and spacing of their families. Second, an economy open to trade and foreign investment is key to capitalize on demographic change. While evidence around the contribution of economic policies to the DD is limited, the literature consistently indicates that policies which open the economy to foreign trade and investment contribute to economic growth in the context of age structure change. An extensive body of literature in the field of economics provides guidance on specific policy strategies to accomplish growth and development goals, but further research on the relative impact of those policy choices in the context of age structure change is warranted. Finally, policy interventions that improve health, advance educational attainment and ensure early dividend returns are reinvested toward the workforce are likely to be beneficial. Investments that increase life expectancy, boost school enrollment and expand access to financial services can promote higher productivity among the workforce. Further, to ensure that the benefits of the first dividend last beyond a single generation, it is key for returns to growth to be reinvested in building the human capital of current and future workers – not largely channeled toward retirees. Where the DD has been embraced as a guiding objective for development planning, the dearth of specific, actionable, evidence-based policy recommendations that outline specific actions that can be taken across multiple sectors prevents the DD from being fully relevant to countries that are entering or yet to enter the demographic window. Arguably, this absence of clear, specific and contextualized policy guidance contributes to the elusiveness of the DD for many countries. Limited analysis also makes it difficult to compare the relative impact of different policy interventions, inhibiting countries’ ability to

THE DEMOGRAPHIC DIVIDEND: POSITIVE PROSPECTS, UNCLEAR PATH 211

prioritize evidence-based policy interventions that promote attainment of the dividend in resource-constrained environments. The limited pool of studies likely reflects the small number of countries in which the DD has been successfully attained – there is little to draw upon for research – as well as the methodological challenges of disentangling and quantifying the contributions of a multiplicity of policy investments. Existing studies have demonstrated the feasibility of measuring the effects of policy interventions empirically in the DD context, using variables ranging from the legalization of contraception to indices of institutional quality. However, while certain policy-related variables, such as those in the health sector, have datasets with extensive historical coverage, datasets in other sectors may be more recent or inconsistently collected. For example, governance-related indicators often only extend back to the 1990s or, at best, the 1980s, limiting their utility for statistical modeling examining countries that had attained (or missed attaining) the DD by the same period. The World Bank’s Worldwide Governance Indicators, for example, provides annual ratings of six domains of governance across more than 200 countries, but only extends back to 1996. This limitation should become a less significant constraint over time, now that at least two decades of data are available in most sources. Measurement of policy interventions is also subject to critique on the question of data quality. Governance indicators, such as those published by the World Bank and the World Economic Forum, rely on the aggregation of surveys of “experts” identified by the dataset authors, which unavoidably introduces an element of subjectivity. Other variables that can be treated as proxies for policy choices, such as educational attainment, have been questioned over their reliability relative to measures such as test scores or schooling quality (Delgado et al., 2013; Hanushek, 2013), for which data coverage may be weaker. Despite these methodological challenges, more research quantifying the contribution of specific policies and programs in the context of age structure change is critical. A number of promising areas for further analysis are apparent. First, the number of countries that have successfully achieved the first DD is much smaller than the number that achieved its prerequisite: fertility decline and a temporary bulge in the working-age population. Efforts to market the successes of a handful of East Asian economies more than 30 years ago have become increasingly stale, begging the question of why there are so few more recent DD examples. Further research pinpointing the policy failures of countries whose favorable age structures met with only tepid economic outcomes could steer others toward success.

212 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Second, increased female labor force participation is often touted as a significant benefit of fertility decline. However, in many current high-fertility countries, female labor force participation is already relatively high, albeit concentrated in the informal sector, where women do not benefit from long-term savings plans or other safety nets. Further, intriguing research indicates that efforts to promote women’s educational attainment and workforce participation without addressing deeply rooted gender barriers in household and care work can lead to unexpected demographic consequences, namely women choosing to opt out of marriage and childbearing in favor of economic independence. Analysis of the effects of fertility decline on women’s educational attainment, transitions to formal employment and changes in women’s income may improve the case for gender-specific drivers of the DD. Finally, further analysis of policies that inflate or extend the dividend is warranted. Often, significant declines in fertility precipitate handwringing among policymakers about the economic consequences of population aging. Given the potential for human capital investments to yield a second demographic dividend, potentially greater than the first, such anxieties may be overwrought. As countries move through the demographic transition and begin to experience the economic benefits of fertility decline, guidance around policy investments that maximize and lengthen the size and duration of the first dividend, and promote attainment of the second, is critical.

References Acemoglu, D. & Johnson, S. (2007), ‘Disease and Development: The Effect of Life Expectancy on Economic Growth’, Journal of Political Economy, 115 (6): 925–85. Anderson, T. & Kohler, H.-P. (2015), ‘Low Fertility, Socioeconomic Development, and Gender Equity’, Population and Development Review, 41 (3): 381–407. Bloom, D. E. & Canning, D. (2003), ‘Contraception and the Celtic Tiger’, Economic and Social Review, 34 (3): 229–47. Bloom, D. E. & Sachs, J. D. (1998), ‘Geography, Demography, and Economic Growth in Africa’, Brookings Papers on Economic Activity, No. 2, 207–95, available at https://​ www​.brookings​.edu/​bpea​-articles/​geography​-demography​-and​-economic​-growth​ -in​-africa/​. Bloom, D. E. & Williamson, J. (1998), ‘Demographic Transitions and Economic Miracles in Emerging Asia’, World Bank Economic Review, 12 (3): 419–55. Bloom, D. E., Canning, D. & Sevilla, J. (2003), The Demographic Dividend: A New Perspective on the Economic Consequences of Population Change, Santa Monica: RAND. Bloom, D. E., Canning, D. & Sevilla, J. (2004), ‘The Effect of Health on Economic Growth: A Production Function Approach’, World Development, 32 (1): 1–13.

THE DEMOGRAPHIC DIVIDEND: POSITIVE PROSPECTS, UNCLEAR PATH 213

Bloom, D. E., Canning, D., Fink, G. & Finlay, J. E. (2009), ‘Fertility, Female Labor Force Participation, and the Demographic Dividend’, Journal of Economic Growth, 14: 79–101. Bloom, D. E., Canning, D., Hu, L., Liu, Y., Mahal, A. & Yip, W. (2010), ‘The Contribution of Population Health and Demographic Change to Economic Growth in China and India’, Journal of Comparative Economics, 38 (1): 17–33. Bongaarts, J. (1984), ‘Implications of Future Fertility Trends for Contraceptive Practice’, Population and Development Review, 10 (2): 341–52. Cervellati, M. & Sunde, U. (2009), ‘Life Expectancy and Economic Growth: The Role of the Demographic Transition’, IZA Discussion Paper, available at http://​ftp​.iza​.org/​ dp4160​.pdf. Cincotta, R. (2017), ‘The Age-Structural Theory of State Behavior’, in Oxford Research Encyclopedia of Politics, Oxford: Oxford University Press. Cincotta, R. (2018), Does Demographic Change Set the Pace of Development? Available at New Security Beat, December 3, https://​ www​ .newsecuritybeat​ .org/​ 2018/​ 12/​ demographic​-change​-set​-pace​-development/​. Cincotta, R. P. & Doces, J. (2012), ‘The Age-Structural Maturity Thesis: The Impact of the Youth Bulge on the Advent and Stability of Liberal Democracy’, in J. A. Goldstone, E. P. Kaufmann & M. D. Toft, Political Demography: How Population Changes Are Reshaping International Security and National Politics, Boulder, CO: Paradigm, pp. 98–116. Cuaresma, J. C., Lutz, W. & Sanderson, W. (2014), ‘Is the Demographic Dividend an Education Dividend?’, Demography, 51 (1): 299–315. Delgado, M. S., Henderson, D. J. & Parmeter, C. F. (2013), ‘Does Education Matter for Economic Growth?’, Oxford Bulletin of Economics and Statistics, 76 (3): 334–59. Donehower, G. (2016), ‘Gender, Age and Economic Activity’, UN Expert Group Meeting on Changing Population Age Structures and Sustainable Development, New York: United Nations. Gribble, J. (2012), ‘South Korea’s Demographic Dividend’, Population Reference Bureau, November, available at https://​www​.prb​.org/​south​-korea​-population/​. Hanushek, E. A. (2013), ‘Economic Growth in Developing Countries: The Role of Human Capital’, Economics of Education Review, 37: 204–12. Inoue, S. (2001), ‘Population Policies and Programs in Japan’, in A. Mason (ed.), Population Policies and Programs in East Asia, Honolulu: East–West Center, pp. 23–38. Jakiela, P. & Hares, S. (2019), Mind the Gap: 5 Facts About the Gender Gap in Education, June 17, available at Center for Global Development, https://​www​.cgdev​.org/​blog/​ mind​-gap​-5​-facts​-about​-gender​-gap​-education. Lee, C., Kim, Y. & Byun, S. (2012), ‘The Rise of Korean Education from the Ashes of the Korean War’, Prospects, 42 (3): 303–18. Lee, R. & Mason, A. (2010), ‘Fertility, Human Capital, and Economic Growth over the Demographic Transition’, European Journal of Population, 26: 159–82. Mason, A. (2001), Population Change and Economic Development in East Asia: Challenges Met, Opportunities Seized, Stanford: Stanford University Press. Mason, A., Lee, R., Abrigo, M. & Lee, S.-H. (2017), Support Ratios and Demographic Dividends: Estimates for the World, New York: United Nations. McNicoll, G. (2006), ‘Policy Lessons of the East Asian Demographic Transition’, Population and Development Review, 32 (1): 1–25. Miller, G. & Babiarz, K. S. (2016), ‘Family Planning Program Effects: Evidence from Microdata’, Population and Development Review, 42 (1): 7–26.

214 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Pasay, N. H. & Wongkaren, T. S. (2001), ‘Population Policy and Programs in Indonesia’, in A. Mason (ed.), Population Policies and Programs in East Asia, Honolulu: East– West Center, pp. 135–57. Pignatti, N. (2016), ‘Encouraging Women’s Labor Force Participation in Transition Countries’, IZA World of Labor, available at https://​wol​.iza​.org/​articles/​encouraging​ -womens​-labor​-force​-participation​-in​-transition​-countries. Queiroz, B., Turra, C. & Perez, E. (2006), ‘The Opportunities We Cannot Forgo: Economic Consequences of Population Changes in Brazil’, XV Encontro Nacional de Estudos Populacionais, ABEP, available at https://​ideas​.repec​.org/​p/​anp/​en2006/​ 32​.html. Shastry, G. K. & Weil, D. N. (2003), ‘How Much of Cross-Country Income Variation Is Explained by Health?’, Journal of the European Economic Association, 1 (2–3): 387–96. Ssewamala, F. M. (2015), ‘Optimizing the “Demographic Dividend” in Young Developing Countries: The Role of Contractual Savings and Insurance for Financing Education’, International Journal of Social Welfare, 24 (3): 248–62. The PACE Project, Population Reference Bureau (2019), Four Dividends: How Age Structure Change Can Benefit Development, available at http://​thepaceproject​.org/​ 4dividends/​. The World Bank (2019), ‘World Development Indicators’, October 11, available at https://​databank​.worldbank​.org. The World Bank (n.d.), ‘GDP per capita (current US$)’, available at https://​data​ .worldbank​.org. The World Bank Human Capital Project (2018), ‘Republic of Korea’, available at https://​databank​.worldbank​.org/​data/​download/​hci/​HCI​_2pager​_KOR​.pdf. Tsui, A. (2001), ‘Population Policies, Family Planning Programs, and Fertility: The Record’, Population and Development Review, 27: 184–204. Weil, D. N. (2007), ‘Accounting for the Effect of Health on Economic Growth’, The Quarterly Journal of Economics, 122 (3): 1265–306. Wodon, Q., Montenegro, C., Nguyen, H. & Onagoruwa, A. (2018), Missed Opportunities: The High Cost of Not Educating Girls, Washington, DC: The World Bank.

14

Forecasting in age-structural time

Richard Cincotta

Introduction Originally the product of research in the US National Intelligence Council, age-structural modeling was intended to serve as a quantitative, exploratory, long-range forecasting tool for foreign affairs, defense and intelligence analysts, a tool that could operate alongside a number of social science theories. Designed to run on publicly available data and rely on any off-the-shelf statistical package, it can graphically display conclusions in ways readily understood by analysts and policymakers. Age-structural modeling is about timing. Each of these models’ predictive curves operates in age-structural time, which measures progress along the path of the age-structural transition – the long-term shift from a youthful set of population distributions, to distributions that encompass much larger proportions of adults who are middle-aged or seniors. The goal has been to provide analysts with an expectation of when a social, economic or political shift is likely to occur, not to indicate how it happens or explain why. Rather than endeavor to unravel the Gordian knot of latent effects and feedbacks, within which demography plays a part, the intent of generating age-structural functions is to inform analysts of probabilistic shifts that are associated with the movement of states across the age-structural transition. Essentially, age-structural models transform data representing a categorical segment of a social, economic or political transition, from their positions in the chronological time domain, T, into positions in the age-structural domain, M, an axis, measured in years of median age, m, that represents the continuous path of the age-structural transition. Rather than being concerned about how states perform during a specific year in chronological time, analyses in 215

216 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

age-structural time predict their likely performance (or behavior), described in broad categories, at specific median ages – generally from 15.0 to 55.0 years – as their age structures mature and, thus, as they advance through the age-structural domain (Cincotta 2017). There are advantages to working in this alternative time scale. Whereas in chronological time, analysts are frequently blindsided by the apparent abruptness of political transitions, or unexpected stalling/backtracking of indicators of political or development progress, these discontinuities largely disappear in age-structural time. And, unlike chronological time, age-structural time need not stop at the present. Once probability curves (called age-structural functions) are generated in age-structural time, they can then be applied several decades into the chronological future using the UN Population Division’s (UNPD) biennially revised set of demographic projections. This chapter is devoted to applying the age-structural modeling technique to forecast the timing of transitions in social, economic and political development. The objectives are threefold: (1) to review accepted theory concerning the dynamics of social, economic and political behavior of states during the course of the age-structural transition and to identify the most significant gaps in this pursuit; (2) to review age-structural modeling’s most striking findings and forecasts; and (3) to discuss the relevance of these findings and current forecasts in light of ongoing and expected global trends. The Global Trends four-phase schema To facilitate testing, standardize maps and other graphic means of communicating results and forecasts, and clarify narratives and discussions among analysts and policymakers, age-structural modeling divides the age-structural domain into four discrete phases. Developed within the National Intelligence Council (NIC) Global Trends program (NIC 2012, 2017) from schemas published by Leahy et al. (2007) and Malmberg and Lindh (2006, p. 68), each of the four phases is defined in terms of the country-level median age, m (the age of the person for whom precisely half of the population is younger), a scalar measure used to crudely characterize and compare the age distributions of populations. These four phases (with standard three-letter abbreviations) are: • • • •

youthful phase [YTH], m ≤ 25.5 years; intermediate phase [INT], 25.6 ≤ m ≤ 35.5 years; mature phase [MAT], 35.6 ≤ m ≤ 45.5 years; and post-mature phase [PMT], m ≥ 45.6 years.

FORECASTING IN AGE-STRUCTURAL TIME 217

The use of single decimal digit precision (tenths) is clearly more precise than the concept warrants, and the categorical boundaries are unrealistically abrupt. Moreover, the representations of population profiles using horizontal bar graphs (see Figure 14.1, profiles aligned at the top of the figure) miss the variation within phases. Nonetheless, the standardization of these categorical boundaries and their iconic representation help keep maps, graphics and discussions consistent among analysts and technicians who are, otherwise, unfamiliar with demography, and it facilitates analysis using simple bar graphs and non-parametric statistical tests. What we know, and don’t know Today, country-level age structures are more diverse than ever. According to the UNPD, they range in median age (the age of the person for whom one-half of the population is younger) from just above 15 years in Niger, the most youthful country, to more than 48 years in Japan, the world’s most mature. Because the most mature country-level age structures have evolved over the course of fertility decline from very youthful age structures, demographers have a detailed record of their performance and behavior along this transition. Thus, researchers generally understand more about the youthful and intermediate phases of the age-structural transition, less about the mature phase and very little about the post-mature phase – a phase that, as of the year 2020, only four countries have entered (Japan, Germany, Italy and Portugal). Empirically, demographers understand a great deal about what to expect of states in the early phases of the age-structural transition, the youthful phase – a portion of the transition through which roughly two-thirds of the world’s states have passed. They may know even more about the transition’s intermediate phase. The site of an unusual amount of economic, social and political change, considerable research has been focused on it. However, demographers understand less about the adaptive evolution of states in the mature phase, which lately has been yielding more of its secrets, and they know almost nothing about how states will adapt to advanced population aging in the post-mature phase. Demographers are aware that in the youthful phase, where populations are numerically dominated by children, adolescents and young adults, countries typically experience very high rates of workforce growth, rapid growth of cohorts entering the school-age population, low per-capita levels of human capital and low levels of financial support for childhood dependents (Lee and Mason 2011). These conditions have presented obstacles to attaining high levels of educational attainment, child survival, per-capita income (Cincotta

218 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Note: The plotted cumulative proportion (points) and smoothed cumulative distribution function are shown for under-age-five mortality rate of less than 25.0 deaths per 1000 live births, 1970 to 2010, across the age-structural transition (measured in median age). The curve, π(m), was generated using logistic regression. Its first derivative, π’(m), is expected to peak where π(m) equals 0.50, shown as π’max. The age-structural transition is divided into four phases (shown at top of graph with iconic population profiles): youthful (YTH), intermediate (INT), mature (MAT) and post-mature (PMT).

Figure 14.1

Logistic curve fitting across the age-structural transition

2017), human capital spending (Lee and Mason 2011) and youth employment (Easterlin 1968). Youthful populations typically feature locally powerful extended families, clans and patronage networks (Wusu and Isiugo-Abanihe 2006). They rarely reach high levels of democracy (i.e., liberal democracy) (Cincotta 2008/09; Dyson 2013; Weber 2012; see also Chapter 13 of this volume), and when they do, they typically lose those levels within a decade

FORECASTING IN AGE-STRUCTURAL TIME 219

(Cincotta 2008/09; Weber 2012). Moreover, due to the general instability of their regimes, states with youthful populations are substantially more likely to abrogate a military alliance than those that are age-structurally more mature (Kim and Sciubba 2014). Following a decline in the total fertility rate (TFR), generally below 2.8 children per woman (Cincotta 2017), the subsequent shift into the transition’s intermediate phase, considered the most advantageous of the four phases, initiates a surge in the proportion of adults in the youngest portion of the prime working ages. This surge boosts the effective producers per dependent consumer, augments the number of producers and adult consumers and vastly increases the number of potential taxpayers (Lee and Mason 2011). Coincident with low fertility and high levels of support for children, this phase – sometimes referred to as the demographic window – typically heralds the rapid rise of educational attainment, human capital spending, per-capita income and, among some states, the accumulation of assets and/or savings among a broad public (Lee and Mason 2011). It also has been the most likely place for the rise of a stable liberal democracy (Cincotta 2008/09). Economic growth rates tend to slow as states leave the demographic window and enter the transition’s mature phase. Despite an aging workforce and a growing group of retirees in age-structurally mature states, favorable economic and stable political conditions typically prevail – conditions that make these states ultimate destinations for migrants fleeing a lack of opportunity and criminal and political violence. There has been an expectation that states that amassed considerable human capital and infrastructure during their intermediate phase will continue to facilitate savings and asset accumulation broadly among their citizens. However, recent research identifies a tendency toward rising income and asset inequality among most mature states (Fine et al. 2019) and a notable shift toward political nationalism and democratic back-sliding. Because of the novelty of the transition’s post-mature phase, scant empirical evidence – or even observational anecdote – can yet be applied to understand how state institutions might be mobilized to mitigate these risks. Demographic projections indicate that states in this phase will be faced with larger-than-ever proportions of retirees and dependent elderly, and declines in the proportion of those in the prime working ages. Such conditions risk fiscal imbalances which could be addressed by higher levels of taxation, higher rates of immigration and higher levels of debt (Eberstadt and Groth 2010). Some analysts speculate that, in the future, these states will find their capacity to extend military power far beyond their borders substantially weakened and their participation in the international system constrained (Brooks et al. 2018/19; Haas 2007),

220 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

while other analysts find the same pressures can induce adaptive responses that favor bilateral alliances and greater participation in regional economic and defense organizations (Sciubba 2016). Modeling Age-structural modeling’s standardized protocol (discussed in more detail in Cincotta 2017) is structured to generate logistic curves in the age-structural domain. It requires that: • cases include only those for independent states with populations over 500,000; • the development transition under analysis is divided into a meaningful sequential series of dichotomous outcome categories (WHO and World Bank development maps are useful sources for categories), or an available standard series of outcome categories (e.g., the World Bank’s income classes, or Freedom House’s freedom scores from Freedom in the World reports); • each outcome category, except the least advanced category, is individually modeled; • each observation in the outcome category being studied is coded as either a zero (0), indicating a value less advanced than those encompassed by the outcome category; or as a one (1), representing a value within the outcome category or within a more developmentally advanced category; • age-structural modeling’s domain variable, country-level median age (the age of the individual, in years, for whom precisely half of the population is younger), appears as the only continuous independent variable in the logit equation; and • all other independent variables, including standard controls – e.g., resource reliant states (oil and mineral rents greater than 15 percent of GDP), and non-resource reliant states with a small population (less than 5.0 million) – and optional experimental variables are coded dichotomously (0,1). When graphed in the age-structural domain, the resultant probability function (known as an age-structural function) that is generated by the logistic regression algorithm is described by a simple logistic curve – an S-shaped curve (shown in Figure 14. 1) with its inflection point, π’max (the maximum of its first derivative), at a probability, π, of 0.50 – or by a partial segment of a simple logistic curve where the inflection point is not visually apparent. Where the inflection point is apparent, it serves as a useful guide for forecasters, who should expect a large proportion of states to advance into this category in the vicinity of this point.

FORECASTING IN AGE-STRUCTURAL TIME 221

Forecasting models: results and conclusions Age-structural forecasting begins with statistical expectations generated by age-structural functions. Age-structural research has yielded six sets of age-structural functions that predict the timing of the following six categorical transitions, three socioeconomic and three political: • child survival transition (from high to low levels of under-age-five mortality); • late-secondary educational attainment transition (from low to high levels of late-secondary attainment among 20- to 29-year-olds); • per-capita income transition (from low to high levels of per-capita gross national income); • political liberalization transition (from autocracy and partial democracy, to liberal democracy); • civil political stability transition (from high to low prevalence of non-territorial armed conflict, including revolutions and other non-separatist warfare); and • ethnic stability transition (from high to low prevalence of territorial armed conflict, which is primarily composed of ethnoreligious separatist rebellion). Socioeconomic development transitions The child survival transition has been categorized using the World Health Organization’s (WHO) global maps of country under-age-five mortality (U5M) levels (WHO 2015). This transition is measured in deaths of children under age five per 1,000 live births, drawn from the UN Population Division’s most current revision (UNPD 2019): low child survival (≥ 100.0), lower-middle (50.0 to 99.9), middle (25.0 to 49.9), upper-middle (10.0 to 24.9) and high (< 10.0). The child survival transition (Figure 14.2a) is largely an irreversible transition that typically commences and makes substantial progress prior to significant fertility decline. However, the achievement of the upper-middle category appears constrained to the intermediate phase of the age-structural transition. While reversals in child survival are relatively infrequent, setbacks have been associated with the AIDS epidemic (mortality associated with mother-to-child HIV transmission), high-intensity warfare (principally in sub-Saharan Africa and Asia) and punctuated catastrophic conditions. The late-secondary educational attainment transition follows the proportion of young adults, aged 20 to 29 years, who have attained late-secondary educational level or higher (Figure 14.2b). The five proportional, late-secondary

222 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Note: Categorical age-structural functions comprising four fundamental transitions: (a) child survival; (b) late-secondary educational attainment; (c) per-capita income; and (d) political liberalization. Each graph shows the categories within each transition, and their stacked probabilities (and 0.95 confidence intervals) across four age-structural phases: youthful (YTH), intermediate (INT), mature (MAT) and post-mature (PMT).

Figure 14.2

Categorical age-structural functions

educational attainment categories, expressed as percentages, are: low (< 20.0), lower-middle (20.0 to 39.9), middle (40.0 to 59.9), upper-middle (60.0 to 79.9) and high late-secondary attainment (≥ 80.0). Data are drawn from the Wittgenstein Center Data Explorer (WCDGHC, 2015; an annual average of about 10 percent of cases are unavailable). This transition is largely irreversible,

FORECASTING IN AGE-STRUCTURAL TIME 223

yet more susceptible to stalling than the child survival transition. Educational systems vary in terms of quality and length, and some European programs encourage early transfers to apprenticeships and employment. The per-capita income transition follows the inflation-adjusted estimates of gross national income (GNI) per capita (Atlas Method) in current US$, using the World Bank’s FY2018 revised schema: low ( $12,055). For most states, advances in per-capita GNI appear largely irreversible. While there is evidence of periodic regional and global economic setbacks (e.g., 1998–2000; 2008–09), they are less observable in the categorical representation of the transition. Moreover, control variables that identify oil- and mineral-reliant economies isolate the effects of state income derived from rents and sovereign wealth funds, as well as separating the effects of commodity price fluctuations from other more human-capital dependent economies (Figure 14.2c). Each transitional series of age-structural functions can be put to work to help analysts make a first approximation of a country’s expected progress along any of these transitions. For a country that is over 5 million population and not a GCC or other highly resource-dependent state, analysts can individually assess countries – now and in the future – by extending a vertical line on the graph, perpendicular to the country’s median age. Such a line cuts across the probabilistic regions associated with each of the transition’s categories, giving analysts a visual indication of which categories in the transitional set are most likely to be realized at a given median age. If a calculated statement of these probabilities is needed, these can be generated mathematically (as follows). For example, according to the UN Population Division’s 2019 estimates, the 2018 median age of Bangladesh is 26.8 years. Extending a line perpendicular to 26.8 years, through the transitional series associated with the World Bank’s income classes (Figure 14.2c), we find the three categorical age-structural models divide that line into four segments (the full line equaling 1.0). Measuring those segments, mathematically, indicates that Bangladesh’s probability of being in the low-income class in 2018 is 0.14, in the lower-middle income class is 0.62, in the upper-middle income class is 0.19 and in the high-income class is 0.06. As it turns out, the World Bank reported that, in 2018, Bangladesh was in its lower-middle income class. Likewise, one can perform this exercise for 2030, when the UNPD’s medium fertility variant projects Bangladesh will have a median age of 31.6 years. For this age structure, the income transition’s age-structural models yield the following set of probabilities: low income, 0.03; lower-middle income, 0.39; upper-middle income, 0.38; and high income, 0.20. Thus, looking

224 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

ahead a decade, there appears to be an almost equal probability that, in 2030, Bangladesh will still be in the World Bank’s lower-middle income class, or have shifted to its upper-middle income class. Political development transitions To model the political liberalization transition (see Chapter 12 this volume), a liberal democratic regime is assumed to be indicated by an assessment of “free” status in Freedom House’s annual survey of political rights and civil liberties (see FH 2020). Further analyses (Cincotta 2017) indicate that: • stable “free” status is most often achieved in the vicinity (±2 years) of a median age of 29 to 30 years; • at any time, a relatively small proportion of age-structurally youthful states (from 15 to 20 percent) are assessed as “free”, but generally are unable to sustain this status for more than a decade if they remain in the youthful phase; • the least populous states (less than 5 million residents) have been able to attain “free” in their youthful phase and remain remarkably stable, unlike more populous states; • single-party authoritarian regimes (identified by the Authoritarian Regime Database 2015; see also Hadenius and Teorell 2007) are generally autocratic political monopolies that have been consolidated using an exclusive ideology, and can be expected to advance through the intermediate and mature phases without attaining “free”; and • military regimes (identified by the Authoritarian Regime Database) are vulnerable to the advancement of age-structural time; they are stable while the population is youthful, but typically yield to more liberal regimes as the population moves through the transition’s intermediate phase. With inequality and migration looming as larger political issues for countries in the age-structural transition’s mature phase, democratic backsliding could be on the way – particularly in post-communist states. In 2018, Hungary and Serbia declined from “free” to “partly free”. Moreover, in 2018, Poland, while still assessed as “free”, lost a full point on Freedom House’s scoring system. Declines from “free” among youthful democracies represent the most common form of categorical democratic backsliding (Figure 14.2d). Nonetheless, since 1972, three states in the transition’s intermediate or mature phase have dropped from “free” to “partly free” without yet recovering their “free” status: Thailand in 2005; Ukraine in 2010; and Indonesia in 2013.

FORECASTING IN AGE-STRUCTURAL TIME 225

The civil political stability transition hypothesizes an age-structurally timed decline of non-territorial intra-state conflict (revolutions and other non-separatist intra-state conflicts) – i.e., the transition toward less frequent incidents of non-territorial conflict. Drawn from the Uppsala/PRIO Conflict Data Set (UCDP/PRIO 2017), this type of armed intra-state conflict is defined as a “contested incompatibility resulting in at least 25 battle-related deaths in a calendar year” (UCDP/PRIO 2017, p. 1), in which the opposition actor intends to change or modify the political system or regime (rather than a territorial conflict, also known as a separatist conflict, where the opposition actor intends to change or modify the status of a territory). The analysis by Cincotta and Weber (forthcoming) employs a single outcome variable: the occurrence (1), or absence (0), of at least one conflict-year of revolutionary conflict during the next five years. Three age-structural models, each differentiated by conflict history – i.e., by the number of revolutionary conflicts that a state has experienced during the prior four years, include: • conflict absence (0 conflicts in the past 4 years); • intermittent conflict (1 or 2 conflicts in the past 4 years); and • persistent conflict (3 or 4 conflicts in the past 4 years). Where revolutionary conflict has been absent, the annual probability of an onset of revolutionary conflict is relatively low (Figure 14.3a), yet still most elevated among the very youthful states – those with a median age of less than 20 years. Youthful countries with a recent history of revolutionary conflict run very high risks of another incident of conflict during the next five years. However, progress into the intermediate phase of the age-structural transition tends to dramatically reduce that risk. Together, these three conflict-history models suggest that: • countries in the youthful phase of the transition are more likely to be engaged in a non-territorial conflict (see Cincotta and Weber forthcoming); and • the risk of intermittent non-territorial (revolutionary) conflict declines precipitously during the age-structural transition’s intermittent phase. The ethnic political stability transition hypothesizes an age-structurally timed decline of intra-state territorial conflict (separatist conflict) – i.e., the transition toward less frequent incidents of territorial conflict, defined as a conflict where the opposition actor intends to change or modify the status of a territory (UCDP/PRIO 2017). This analysis (Cincotta and Weber forthcoming) employs the same categories as the civil political stability transition. However, the trends appear weaker and less certain (Figure 14.3b).

226 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

Note: Trends in the five-year probability, π5, of experiencing a (a) non-territorial conflict (revolutions and other non-separatist attempts to violently destabilize the central government) and experiencing a (b) territorial conflict (separatist conflict) as states traverse the age-structural transition, measured in median age. Curves (solid lines) and their 0.95 confidence limits (dashed lines) are read as probabilities for three different conflict histories: absent conflict (0 conflict years during the past 4 years); intermittent conflict (1 or 2 conflict years during past 4 years); and persistent conflict (3 or 4 conflict years during past 4 years).

Figure 14.3

Probability of conflict

These functions impart several lessons. Their trajectories suggest that the risk of a new emergence of a territorial conflict among states that have not experienced them recently (absence) is low, even for youthful states, and very

FORECASTING IN AGE-STRUCTURAL TIME 227

gradually declines as countries move through the age-structural transition, never reaching zero. The trajectories also imply that, once territorial conflicts become intermittent or persistent, the risk that the state will experience another conflict incident over the next five-year period is very high, even at advanced phases of the age-structural transition. The emergence of a territorial intra-state conflict is rare, and nearly non-existent in the least populous countries (population less than 5 million) (Cincotta and Weber forthcoming). However, the persistence and re-emergence of separatist conflict is common. Apparently, intermittent and persistent separatist conflicts are difficult to settle (Denny and Walter 2014). Moreover, separatist conflicts tend to persist through age-structural time. The method used to generate probabilities for the three categories in the political liberalization transition (Figure 14.2d) is identical to the vertical-line method previously described for the development transitions (Figures 14.2a, b, c). However, unlike the prior transitional sets of curves (shown in Figure 14.2), each conflict-history curve (Figures 14.3a, b) is, itself, a transition. In other words, erecting a vertical line, perpendicular to a median age, cuts through only two conflict categories. Above each curve (from the curve upward to 1.0), the length of the line segment represents the probability that a country will be engaged in a similar type of conflict during the next five years. Below the curve (from the curve downward to 0.0), the distance represents the probability that this country will not be engaged in a similar type of conflict during the next five years. Key Findings Development and political stability accompany the age-structural transition The pattern of the three development transitions – child survival, late-secondary educational attainment and per-capita income – are strikingly similar. Youthful age-structures, and the high levels of fertility that maintain them, consistently appear as powerful constraints to reaching moderately high levels of development. With the exception of some of the least populous states and those with high levels of oil and mineral endowments (discussed in the following section), youthful states are confined almost exclusively to the low and lower-middle categories of these development transitions. In contrast, the upper-middle categories of socioeconomic development are consistently properties of states that have attained the age-structural transition’s intermediate phase, passing a median age of 26 years. Crossing that threshold typically involves states declining to a total fertility rate below 2.8 children per woman (Cincotta

228 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

2017), plus additional years for the bulge of adolescents in the age-structure to mature. The political advantages inherent in attaining this demographic window are made graphically obvious by the intersection of upper-middle categories and the clustering of their inflection points near a median age of 30 years. Notably, the inflection point on the political liberalization transition’s “free” function (Figure 14.2d), and the civic political stability transition’s intermittent conflict function (Figure 14.3a) also appear within this cluster. Given the coincidence of these trends, it seems obvious why state capacity appears to coalesce in the intermediate phase, and why countries are able to build their infrastructure and institutions during this demographic window. Territorial intra-state conflicts have a weak relationship with the age-structural transition Among youthful countries, non-territorial conflicts – particularly an insurgency that seriously contests the legitimacy of the central government – is increasingly likely to be settled, to fade to very low levels of intensity or to emerge only briefly as states enter the intermediate phase and traverse the age-structural transition (Cincotta and Weber forthcoming). The exceptions among non-territorial conflicts are principally comprised of intermittent attacks occurring in the stable economies of age-structurally mature states – including the al-Qaeda-organized 9/11 attacks in 2001, the 2004 Madrid train and 2005 London Underground bombings that may have been assisted by al-Qaeda operatives and the Oklahoma City bombing by members of a US-based neo-Nazi militia. These have typically entailed terror against largely symbolic state targets or against civilians in capital cities or commercial hubs. This difficult-to-classify set of conflicts, which often features foreign actors, social media, international travelers and resident discontents, harbors no serious potential of overturning the central regime, but has had significant economic and social impacts. On the other hand, territorial intra-state conflicts – which generally involve ethnic separatist insurgents – tend to be highly persistent or intermittent across the age-structural transition. Separatist conflicts rarely begin where they have previously been absent, and they are nearly non-existent in small-population states (fewer than 5 million inhabitants). However, among more populous states, once they begin, it is difficult to completely eradicate a separatist conflict at any time in the transition (e.g., the separatist conflicts composing the Yugoslav Civil War, 1991–2001). Thus, it is little wonder that some multi-ethnic states with mature populations and distinct regional enclaves

FORECASTING IN AGE-STRUCTURAL TIME 229

have opted for peacefully dismembering their union (e.g., the dissolution of Czechoslovakia, 1993; and the Soviet Union, 1991). In some states, wide and sustained gaps in fertility and age-structure signal prolonged tensions between majority and minority groups – particularly when the minority represents a large proportion of the provincial or state population – and when they are politically organized (Blomquist 2016). Thus, median age at the country level is generally not indicative of these conflicts (Cincotta and Weber forthcoming; Yair and Miodownik 2016). Beware of exceptional conditions Among the most useful features of age-structural modeling has been its ability to identify exceptional conditions, and states experiencing those conditions. Exceptional conditions are those that permit or constrain states to not fit age-structural modeling’s most basic assumption – that development progress is associated with advancement across the age-structural domain. Once exceptional conditions have been observed or suspected, they can be deductively tested as dichotomous explanatory variables. Those that have repeatedly deviated from the larger pack of countries, across several development transitions, are now used in age-structural modeling’s logit equations as standard dichotomous controls. These include: (a) substantial levels of oil and mineral rents (when greater than 15 percent of GDP), and (b) the least populous states (fewer than 5.0 million residents). Most small-population states are age-structurally precocious performers in that they have regularly achieved the upper-middle levels of the development transitions before the rest of the pack, and a significant portion have both achieved “free” in the youthful phase and held on to it (e.g., Costa Rica, Jamaica, Botswana and Namibia). Moreover, fewer of these least-populous states are in non-territorial conflict as youthful states, and none have been in conflict since the 1970s while in the intermediate or mature phases. Oil- and mineral-reliant states are also precocious development performers. However, they generally lag in their political transitions – fewer than expected have achieved “free” and their risk of non-territorial and territorial conflict is generally greater. Four additional conditions appear to deviate from expectations. However, none of these have been added to age-structural models because they are represented by relatively few cases or they arise and decline erratically. Nonetheless, encountering any of these conditions in a focus country or its regional neigh-

230 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

borhood should prompt forecasters to temper the certainty of their statistical predictions or adjust its predicted outcome. These are: • the presence of a single-party political regime or theocracy (identified by the Authoritarian Regime Data Set 2015) deters political liberalization, and may impede other transitions. The most implacable of these regimes (e.g., China, North Korea and Iran) maintain elite means to smoothly accomplish leadership transitions capable of sustaining the regime’s political monopoly. States with these types of regimes have traversed the demographic window (the intermediate phase) without regime change or dramatic reforms and have successfully suppressed all challenges. • Ongoing high-intensity territorial (separatist) or non-territorial conflicts (revolutionary and other non-separatist conflicts) appear to deter progress in some political and development transitions, particularly if conflicts are intense and not territorially contained. • The influence of powerful neighboring states or their within-state proxies can deter political transitions. Examples include chilling effects of the Soviet Union on the neighboring Eastern European states during the Cold War, and Iran’s financial and material support of Hezbollah in Lebanon. • Disparate rates of majority and minority demographic change – from gaps in fertility or migration – can undermine political transitions (Blomquist 2016). Conditions appear to be most exceptional when that minority becomes sufficiently populous, and when that minority is politically organized. Relevance to ongoing and future trends By its nature, age-structural modeling is only capable of describing the timing of recognized political and development transitions that (a) are paced by the age-structural transition and (b) are, or can be, broken down into a meaningful series of discrete categories for which time-series data are abundantly available. Because age-structural modeling reaches into past data to build models that can be used to guide analysts’ perspectives on the future, its categorical models should be expected to most accurately theorize the timing of state behaviors during the early, data-rich phases of the age-structural transition, and be less knowledgeable about when categories of behavior arise, are reached, or retreated from, during the transition’s latter stages. Thus, the statistical characterizations of trends and their timing that are produced by age-structural modeling will be most relevant to the future of states that will remain in the youthful, intermediate or mature phases, and those that traverse between these phases.

FORECASTING IN AGE-STRUCTURAL TIME 231

To 2050, roughly three-quarters of all states are projected to remain in these three phases. By identifying clusters of youthful states, analysts should have an idea where non-territorial conflicts will most likely break out, and where all types of ongoing intra-state conflict will likely persist or frequently recur. The most persistently youthful clusters are likely to become crisis hotspots, generating warfare that spills across borders, pockets of severe food insecurity and significant out-migration. In contrast, among the handful of states that are projected to leave the youthful phase and pass deeply into the intermediate phase, analysts should look for newly emerging economies and hints of political liberalization. The remaining quarter of the world’s states are heading into an unchartered age-structural future. According to the UNPD’s medium fertility variant, more than two-thirds of all European states and nearly half of East Asia’s (including China) could reside in the post-mature phase in 2050 – about 40 states; up from only four, today. As a means to discern state behavior and performance under such novel demographic conditions, age-structural modeling is, admittedly, unfit. Until a larger, more varied set of countries affords a proper statistical view of post-maturity, our perceptions of this future will continue to be influenced by some combination of narrative theories, mechanistic models, comparative analyses, case studies and ideologically tinged personal opinion. States may, indeed, find institutional means to adapt smoothly to advanced population aging. Some may not. Whatever future unfolds in post-maturity, only the most disciplinary and ideologically constrained analysts can continue to remain oblivious to age structure’s influence on the evolution of modern states.

Acknowledgments Elements of this research have been supported, in part, by the Population Institute, Population Reference Bureau (PRB) and the National Intelligence Council. I also thank Elizabeth Leahy Madsen and Kaitlyn Patierno, both at PRB, for working on aspects of these relationships with me.

232 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

References Authoritarian Regime Data Set (2015), ‘Authoritarian Regime Data Set, 1972–2014’, Retrieved May 20, 2017 from https://​sites​.google​.com/​site/​authori​tarianregi​ medataset/​. Blomquist, R. (2016), ‘Ethno-Demographic Dynamics of the Rohingya-Buddhist Conflict’, Georgetown Journal of Asian Affairs, 3 (1), 94–117. Brooks, D. J., Brooks, S. G., Greenhill, B. D. & Haas, M. L. (2018/19), ‘The Demographic Transition Theory of War: Why Young Societies Are Conflict Prone and Old Societies Are the Most Peaceful’, International Security, 43 (3), 53–95. Cincotta, R. (2008/09), ‘Half a Chance: Youth Bulges and Transitions to Liberal Democracy’, Environmental Change and Security Program Report, 13, 10–18. Cincotta, R. (2017), ‘The Age-structural Theory of State Behavior’, in W. Thompson (ed.), Oxford Research Encyclopedia of Politics. Oxford: Oxford University Press. Cincotta, R. & Weber, H. (forthcoming), ‘Youthful Age Structures and the Risk of Revolutionary and Separatist Conflict’, in Goerres, A. & Vanhuysse, P. (eds), Global Political Demography. London: Palgrave Macmillan. Denny, E. K. & Walter, B. F. (2014), ‘Ethnicity and Civil War’, Journal of Peace Research, 51 (2), 199–212. Dyson, T. (2013), ‘On the Democratic and Demographic Transitions’, Population and Development Review, 38 (suppl.), 83–102. Easterlin, R. A. (1968), Population, Labor Force, and Long Swings in Economic Growth: The American Experience. New York: National Bureau of Economic Research & Columbia University. Eberstadt, N. & Groth, H. (2010), ‘Demography and Public Debt: Time for a “Demographic Stress Test” for the Western Economies. What Does It Mean for Switzerland?’ WDA-HSG Letters on Demographic Issues, No. 2010/1. St. Gallen, Switzerland: University of St. Gallen. Fine, D., Manyika, J., Sjatil, P. E., Tacke, T., Tadjeddine, K. & Desmond, M. (2019), ‘Inequality: A Persisting Challenge and Its Implications’, Discussion Paper. New York: McKinsey Global Institute. [FH] Freedom House (2020), Freedom in the World, 2020: A Leaderless Struggle for Democracy. Washington, DC: Freedom House. Haas, M. L. (2007), ‘A Geriatric Peace? The Future of U.S. Power in a World of Aging Populations’, International Security, 32 (1), 112–47. Hadenius, A. & Teorell, J. (2007), ‘Pathways from Authoritarianism’, Journal of Democracy, 18 (1), 143–56. Kim, T. & Sciubba, J. D. (2014), ‘The Effect of Age Structure on the Abrogation of Military Alliances’, International Interactions, 41 (2), 279–308. Leahy, E., Engelman, R., Vogel, C. G., Haddock, S. & Preston, T. (2007), The Shape of Things to Come: Why Age Structure Matters to a Safer, More Equitable World. Washington, DC: Population Action International. Lee, R. & Mason, A. (eds) (2011), Population Aging and the Generational Economy: A Global Perspective. Cheltenham, UK and Northampton, MA, USA: Edward Elgar Publishing. Malmberg, B. & Lindh, T. (2006), ‘Forecasting Global Income Growth Using Age-structural Projections’, in Pool, I., Wong, L. R. & Vilquin, E. (eds), Age-structural Transitions: Challenges for Development (pp.  60–82). Paris: Committee for International Cooperation in National Research in Demography.

FORECASTING IN AGE-STRUCTURAL TIME 233

[NIC] National Intelligence Council (2012), Global Trends 2030: Alternative Worlds. Washington, DC: Office of the Director of National Intelligence. [NIC] National Intelligence Council (2017), Global Trends: Paradox of Progress. Washington, DC: Office of the Director of National Intelligence. Sciubba, J. D. (2016), ‘Willing and (Somewhat) Able: Japan’s Defense Strategy in a Graying Asia’, Georgetown Journal of Asian Affairs, (Fall), 50–56. [UNPD] United Nations, Dept. of Economic and Social Affairs, Population Division (2019), ‘World Population Prospects: The 2019 Revision’, New York: United Nations, Dept. of Economic and Social Affairs. [UCDP/PRIO] Uppsala Conflict Data Program, & Center for the Study of Civil Wars, Peace Research Institute at Oslo (2017), UCDP/PRIO Armed Conflict Dataset Codebook: Version 4-2017, Oslo: UCDP/PRIO. Weber, H. (2012), ‘Demography and Democracy: The Impact of Youth Cohort Size on Democratic Stability in the World’, Democratization, iFirst, 1–23. [WCDGHC] Wittgenstein Centre for Demography and Global Human Capital (2015), ‘Wittgenstein Centre Data Explorer’, Version 1.2, retrieved April 17, 2018 from http://​www​.wittgensteincentre​.org/​dataexplorer. World Bank (2018), World Development Indicators, Online. Washington, DC: World Bank, retrieved September 20, 2018 from https://​databank​.worldbank​.org/​reports​ .aspx​?source​=​world​-development​-indicators. [WHO] World Health Organization (2015), Map: Under-5 Mortality Rate, 2015, retrieved April 17, 2018 from http://​gamapserver​.who​.int/​mapLibrary/​Files/​Maps/​ Global​_UnderFiveMortality​_2015​.png. Wusu, O. & Isiugo-Abanihe, U. C. (2006), ‘Interconnections Among Changing Family Structure, Childrearing and Fertility Behaviour Among the Ogu, Southwestern Nigeria: A Qualitative Study’, Demographic Research, 14 (8), 139–56. Yair, O. & Miodownik, D. (2016), ‘Youth Bulge and Civil War: Why a Country’s Share of Young Adults Explains Only Non-ethnic Wars’, Conflict Management and Peace Science, 33 (1), 25–44.

15

A twenty-first century agenda for policy-relevant demographic research

Suzanne E. Fry

Introduction Analysts and decision-makers use demography to evaluate the effectiveness of past decisions, to gauge the status and implications of current conditions and to anticipate broad, structural contours of the future. Even though demography is critically important for understanding our past, present and future, the second- and third-order implications of population size, structure and distribution are far from self-evident. Moreover, the slow-moving and long-term nature of demographic change can discourage analytic and policy attention to demography and its implications, leaving both analysts and policymakers vulnerable to strategic surprise. Demographic research can help us prevent strategic surprise and apprehend the future – especially if we understand how demographic trends are likely to interact with other trends or variables, such as the economy and environment, to produce political and geopolitical outcomes. Taking an explicitly strategic and geographic approach to demographic questions is one way to make demographic trends and their implications relevant to analysts and policymakers. A strategic approach to demography research is interdisciplinary, future-focused and systematically examines the national security implications of demographic change. A geographic approach disaggregates global and regional demographic patterns and examines how the observed or estimated population trends for a locality are likely to interact with other trends to produce local and geopolitical effects.

235

236 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

This chapter reviews how analysts use demographic research, suggests ways for maximizing the policy relevance of such research and concludes with a discussion of research questions related to aging and declining populations and urbanization that will help analysts anticipate and policymakers manage the future strategic landscape.

How analysts use demographic research Intelligence and policy analysts help policymakers answer two basic questions: Will some development or discontinuity occur? If it does, what will it mean for national security? Demography typically fits into such analysis at a strategic level, and often helps answer questions about a country’s potential future economic performance, its risk for large-scale instability or its ability to honor fiscal commitments, among others. Because, outside of war or disease, demographic developments are slow-moving, the bulk of analytic effort rests with assessing implications of demographic trends and detecting if, when and how long-ago forecasted developments will occur and how they will interact with other conditions. Doing such analysis well means grappling with: (1) time, accurately anticipating when and how lagged effects will manifest; (2) other trends, understanding how demography, the economy, technology and other dimensions interact with one another; (3) geography, understanding how demography and other trends manifest in specific regions and localities; and (4) policy or national security relevance, understanding how the demographic features impact the state’s interests. These steps help analysts maintain perspective, understand how demographic developments will play out locally and geopolitically and place near-term developments in a larger strategic context. This can be difficult when analysts lack interdisciplinary knowledge or track a country so closely on a day-to-day basis that they lose sight of a structural shift that has been underway for decades but is just now manifesting in second- and third-order implications. Contributing to this challenge are decision-makers with short time horizons, often elected officials, who are laser-focused on a few priorities and under pressure to demonstrate near-term success, and therefore expect analysts to provide warning of a near-term risk or opportunity. Strategic decision-makers, by contrast, typically have wider apertures and longer-time horizons and look to analysis for opportunities to create strategic advantage in the coming months, years or decades – or laterally on other issues. Given these realities,

A TWENTY-FIRST CENTURY AGENDA 237

intelligence and policy analysts generally seek demographic research that helps them address one or more of the following types of questions: • How might long-present demographic trends combine with other features to create political or economic shocks in specific regions, countries or cities? What do we know about the timing of such shocks? • How might current demographic trends relate to top near-term priorities of decision-makers? • How can such trends be shaped today for more advantageous conditions in the future?

Maximizing policy relevance of demographic research Policy-relevant demographic research, therefore, identifies the most important political, security, economic and other consequences of demographic features – and does so by placing demographic trends and relationships in interdisciplinary, temporal and geographic context. Policy-relevant research is also objective and clearly understandable by a range of audiences. The following are important considerations for effective policy-relevant research: • Most important for whom? Developing policy-relevant research requires knowing who your audience might be, his or her time horizon for policy action and decision-making, and a clear understanding of the national interests the policymaker is trying to advance. • Interdisciplinary. Researchers should ask how demographic features impact non-demographic variables or conditions. Building a body of knowledge concerning implications of a demographic feature allows for more systematic anticipation of the effects of that feature. Such research also provides a better understanding of the conditions under which those effects occur. • Temporal. Consider the time-span for when demographic features will be consequential. Are there past demographic features whose effects are about to manifest? Are there current features whose effects can be reliably forecasted? What intervening factors or variables would render such temporal judgments less reliable? • Geographic. Where will specific demographic features manifest? How will the demographic features play out in specific regions or cities? Do these local trends track with broader global trends? Why or why not? • Reliable. Analysts depend on the demographic researcher to not only perform the basic social and natural science to professional standards,

238 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

but also to convey findings in a way that can be interpreted and used by professionals in other fields. • Accessible. Communicating findings in a clear, accessible and objective manner should help inform analysts, policymakers, the media and the general public as they address thorny policy questions, such as: Should societies pursue more or less immigration? Larger or smaller families? Rural or urban investment? More or less continuing education? More or less investment in technology, defense, healthcare, eldercare, education, the environment or other policy areas?

Research for the future Most assessments of the future strategic context rely on analysis of key trends and uncertainties and a range of futures methodologies to anticipate the types of challenges and opportunities states, organizations and others are likely to face in the next 10 to 20 years. Recent futures studies warn about the ability of advanced industrial states to meet their welfare and defense commitments and about the future of work due to advances in information technology, computing and automation (“Global System” 2015; “Global Trends” 2015; “Global Trends” 2017; Hariri 2018; Morris 2016). They have assessed as well that tensions within countries will likely increase as key trends converge to make governance more difficult, while the risk of interstate conflict will increase to levels not seen since the end of the Cold War due to the possibility of a rising China displacing the United States, the erosion of norms against conflict and human rights abuses and increased nationalism and populism in response to domestic challenges (“Global Trends” 2017). Such warnings receive significant policy attention globally. Embedded in such assessments, however, are additional concerns linked to demographic trends that have received less attention – probably because their implications are potentially radical and it is not clear what should be done about them. First, as Chapter 11 in this volume described, aging and declining populations are challenging the sustainability of how societies and economies have been organized since the Industrial Revolution. Second, as multiple other chapters have touched on, urbanization and its ecological consequences are triggering questions about the earth’s carrying capacity and how to manage the politics and economics of such conditions. Both shifts are rooted in demography and its intersections with the economy, technology and the environment, and have prompted speculation and commentary, but systematic, policy relevant research remains underdeveloped. How, when and where these shifts and

A TWENTY-FIRST CENTURY AGENDA 239

their political and geopolitical effects play out – and possible solutions – is a twenty-first-century research agenda for the social sciences, including demography.

Aging and declining populations The aging populations of Northeast Asia, Europe and other members of the OECD are a bellwether for the rest of the planet beyond Africa, where age structure is forecasted to remain youthful. Contemporary demographic research has long warned of the likely consequences and unsustainability of past economic and social welfare choices. The political consequences are evident in North America and Europe and suggest not only a guns-versus-butter tradeoff in social policy but also a generational tradeoff in healthcare-versus-education. The austerity battles in Greece following the 2008/09 global financial crisis, pitting retirees against youth, could well be a sign of things to come for other parts of the rich, aging world. The realities of deferred, skipped or only-child parenthood in much of the OECD, while not yet as severe as that observed in China or Japan, are likely to compound labor shortages. The literature on population decline is somewhat less developed than on aging societies, despite a centuries-long trend of depopulation of rural communities in wealthy countries. Such dynamics recently have become politically consequential in the United States, contributing to a surge of populism and a political realignment. Important near-term research is underway as well on the economic, mental health, addiction and other sociological drivers of potential status reversal among white men in Canada, the United Kingdom and the United States. Meanwhile, China in the next decade will confront the social and economic costs of both a rapidly aging and declining overall population – and is betting on technology to overcome labor shortages. These dynamics in the rich West and East will almost certainly bring debates over immigration and technology to the fore, and probably in ways that will differ dramatically across contexts. As societies grapple differently with the political and economic costs of aging and declining populations, emerging before us is a natural experiment concerning which policy choices (and in which contexts) are most likely to ease these fiscal strains and social burdens while advancing national security. A key research priority moving forward will be tracking the performance of these choices over time, particularly among the states that first confronted these conditions. Systematic study of these choices, outcomes and contexts could

240 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

well provide a new toolkit of policy insights for countries that will face similar challenges in ensuing decades – and again underscore the importance of a strategic and geographic approach to demographic research. A second key research area related to aging and declining populations is determining which of these societies will question, if not reject, a key assumption that has underpinned economic development and social welfare since the eighteenth century: There will be enough labor to drive economic growth and pay for welfare and defense commitments. Demographic and related social science research is urgently needed on which societies are likely to abandon this economic and social model and which will try to augment it with alternative revenue choices – perhaps by taxing wealth differently or by taxing revenue from technology or data use – or alternative spending choices, like universal basic income. In which types of contexts will the traditional assumption of a larger working-age population than dependent populations – young and old – remain appropriate? Given likely future age structures, are there more realistic assumptions, policy choices and social models? New research on this topic would straddle both comparative political demography and social and political theory. Other key research questions concern technology and economics: To what extent can technology substitute for labor, and under what economic and political development conditions? Which countries or political systems (authoritarian or democratic) are best positioned to effect technology-driven economic growth, to direct tech-driven wealth back into society and to adjust to a post-labor economy? What would a post-labor economy look like? Related big picture research questions concern the security implications of aging and declining populations. China’s geopolitical rise, according to some forecasters, is now, not ten years from now – when its population decline accelerates (Eberstadt 2019a, 2019b). How will Chinese foreign policy and global economic engagement shift when Beijing becomes increasingly focused on domestic challenges? To what extent does Russia’s demographic decline of recent decades suggest geopolitical implications that would be relevant for anticipating the effects of Chinese demographic decline? Similarly, East Asia has long been cast as the region of the future, and yet it is the only world region projected to have a smaller population in 2100 than today (“World Population Prospects” 2019). Does a smaller East Asian population make the region more or less consequential economically or geopolitically than it is today? Much almost certainly depends on whether technology can substitute for labor and whether the adjustment of a post-labor society and politics can be made. Similarly, much depends as well on whether technology can substitute

A TWENTY-FIRST CENTURY AGENDA 241

for favorable demography and military force size in national power capacity and outcomes. Finally, how might the United States maintain its relative demographic advantages and manage the demographic disadvantages of its European and Northeast Asian allies? A separate set of questions related to aging and declining populations concerns the assumptions demographers make when generating their estimates. We face a future of 9 billion or 13 billion people by the end of the twenty-first century – but which is it (Tucker 2019)? Those are two vastly different futures for which to plan. What assumptions are made with respect to fertility rates? Will women in wealthy contexts return to having two children despite decades of consistently lower realized fertility rates (“World Population Prospects” 2019)? What economic, technological and political conditions would need to be present for women in the rich world to have two children? The environmental impact of urbanization and economic development Recent and future trends of rapid urbanization – historically coincident with the Industrial Revolution and greater use of fossil fuels, improved diets and longer life spans – are likely to exacerbate environmental degradation and climate change, absent very different political and economic choices than those common today or yesterday. The food, water and energy nexus critical for human life will concentrate even more in urban areas, many of which are in coastal zones vulnerable to sea-level rise and extreme weather. Some of the most demanding urban and environmental planning decisions will occur in just nine countries – the Democratic Republic of the Congo, Egypt, Ethiopia, India, Indonesia, Nigeria, Pakistan, Tanzania and the United States1 – where more than half of all population growth is set to occur in the next 50 to 100 years (“World Population Prospects” 2019). How will cities in these countries absorb urban growth? What should they do to minimize destruction of natural habitats and natural resilience? A further question beginning to get attention is how will climate change impact migration flows? War, crime and poverty have long driven mobility – and to that list we now add sea-level rise, extreme heat, drought, and poor air, water and soil quality. To what extent can we anticipate these conditions and map and plan for population movement? Whether sea-level rise in Bangladesh or cities of other parts of Southeast Asia, extreme heat in the Persian Gulf region or poor air and water quality making some cities of South Asia uninhabitable – such conditions are foreseeable. The key research questions are

242 A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

timing, geography and impacts on interests: When is population movement likely to occur and under what conditions, where will populations go, which other interests will be negatively or positively impacted, are there general patterns of migration that can be used to understand the consequences of environmentally-driven movement? A key assumption of modernity – and certainly of the Industrial Revolution and colonial eras before it – is that human beings and their consequences can and should be understood separately from their natural context. Rising human rights, feminist, environmental and climate consciousness has converged with arguments by religious figures like Pope Francis, indigenous leaders and others who call for an economy of ecological stewardship and human equality. Add to this new affronts to human dignity and privacy posed by technology titans – and we have a diverse set of stakeholders sharing the perspective that a proper understanding of humanity is one that is inseparable from, and therefore has obligations to, one another and the natural world. So what does this have to do with demography? It is possible, should environmental degradation continue apace and the earth’s carrying capacity to sustain human and natural life be threatened, that traditional moral arguments surrounding population – right to life versus access to birth control, for instance – could shift in light of other dire, immediate threats to human life: food security and the capacity of the natural world to support human life. Such questions are unthinkable and almost certainly too late once they have political traction – but they are critically important for demographic researchers to consider. Several research priorities include: What is the actual capacity of the planet to sustain human and natural life? At what point and for what reasons does the carrying capacity become threatened? To what extent can human actions – and which ones – improve the planet’s capacity to sustain human and natural life?

Conclusion This chapter has argued for a strategic and geographic approach to demographic research to increase its policy and national security relevance for the twenty-first century. Such an approach is interdisciplinary, sensitive to time and geographic space and poses and answers policy-relevant questions. Substantively, it would focus on the problems of too few people to sustain modern economic and social welfare arrangements, and too many people concentrated in cities and generating unsustainable environmental effects. The stakes are enormous for clear and better insights about what population size

A TWENTY-FIRST CENTURY AGENDA 243

and structure portend for social and economic policy, geopolitical competition and planetary survival. These are big, audacious topics – and they could not be more important for scholars to take seriously as they craft research agendas for the twenty-first century. Making such research understandable is a key step toward informed policy choices.2

Notes 1.

That the United States is the only developed country in this list is telling – highlighting the uniqueness of US demographic advantages among the countries of the rich world. 2. All statements of fact, opinion or analysis expressed are those of the author and do not reflect the official positions or views of the US government. Nothing in the contents should be construed as asserting or implying US government authentication of information or endorsement of the author’s views.

References Eberstadt, N. (2019a), ‘China’s Demographic Outlook to 2040 and Its Implications’, American Enterprise Institute, https://​www​.aei​.org/​research​-products/​report/​ chinas​-demographic​-outlook​-to​-2040​-and​-its​-implications​-an​-overview/​. Eberstadt, N. (2019b), ‘With Great Demographics Comes Great Power’, Foreign Affairs, https://​www​.foreignaffairs​.com/​articles/​world/​2019​-06​-11/​great​-demographics​ -comes​-great​-power. ‘Global System on the Brink: Pathways Toward a New Normal’ (2015), Atlantic Council and the Primakov Institute of World Economy and International Affairs (IMEMO), Washington, DC. ‘Global Trends: Paradox of Progress’ (2017), National Intelligence Council, Office of the Director of National Intelligence, Washington, DC. ‘Global Trends to 2030: Can the EU Meet the Challenge?’ (2015), European Strategy and Policy Analysis System (ESPAS), Brussels: EU. Hariri, Y. N. (2018), 21 Lessons for the Twenty-first Century, New York: Spiegel and Grau. Morris, L. (2016), Foresight and Extreme Creativity: Strategy for the twenty-first Century, Walnut Creek, CA: FutureLab. Tucker, C. (2019), A Planet of 3 Billion: Mapping Humanity’s Long History of Ecological Destruction and Finding Our Way to a Resilient Future, New York: Atlas Observatory Press. ‘World Population Prospects 2019: Highlights,’ United Nations, https://​population​.un​ .org/​wpp/​Publications/​Files/​WPP2019​_Highlights​.pdf.

Index

Abe, Shinzo 25 adolescents 6, 147–50, 155, 157, 217, 228 age-structural modeling 215, 216, 220, 229–31 age-structural time 215–31 age-structural transition development and political stability 227–8 exceptional conditions, beware 229–30 ongoing and future trends 230–31 territorial intra-state conflicts 228–9 age structure 6, 7, 12, 147, 148, 154, 165, 166, 171, 172, 174, 184, 200, 201, 239, 240 change 200, 201, 203–6, 208–11 aging 4, 6, 11, 17–26, 140, 143, 174, 184, 239, 240 Arab Spring 164 Aristotle 163 armed conflict 30, 51, 221 asylum 30–32, 35, 36, 39, 87, 88, 93, 94, 193–5 asylum-seekers 36, 188, 195 authoritarianism 7, 69, 71, 140, 173 Black Death 138 Bloom, D. E. 206 Bosnia 171 Brexit 26, 87, 88 brideprice 61, 67, 69–71, 74, 75 Caldwell, J. C. 116 Cali, M. 50 Canning, D. 206 census 82, 83, 106, 107, 109, 189, 190

child labor 157 child marriage 157 child mortality 12, 123, 180, 205 child survival transition 221 Choucri, Nazli 163 Christians 101, 102, 104, 106–8, 110, 179, 188, 189 Cincotta, R. 139, 201 civil conflict 2, 7, 10, 12, 36, 52, 101, 111, 154, 163, 187 civil political stability transition 221 civil war 12, 24, 31, 34, 52, 81, 89, 105–7, 112 climate change 1, 8, 9, 45–7, 49, 53–6, 115, 117–19, 241 health and demography 117–20 Coleman, David 26, 96 color revolutions 139–43 Concise History of World Population, Massimo Livi-Bacci 4 conflict 8, 12, 23, 24, 25, 34, 37, 49, 51–4, 56, 84, 105, 108, 111, 225–7, 229 country-level age structures 217 COVID-19 9, 115, 116, 125, 141, 148 Davies, S. E. 122 democracy 74, 142, 161–8, 171–3, 175, 218, 221 and demography 161–75 empirical findings 168–72 methodological challenges 172–3 theoretical mechanisms 165–8 democratic institutions 7 democratic stability 165, 171, 173 demographic change 2, 3, 8, 11, 12, 81–96, 179, 180, 235 demographic dependency ratio 158 245

246

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

demographic dividend 6, 12, 148, 151–3, 155–8, 163, 164, 199–212 elusive roadmap to 210–12 explaining 200–203 demographic engineering 179–96 demographic data 189–91 differential fertility changes 183 immigration, sub-national levels 187–9 in-migration, population and age 184–5 and international human rights regimes 194–6 international migration 183–7 migration, foreign policy 191–5 national fertility declines 180–81 national fertility increases 181–3 out-migration, nationals 185–7 demographics 61–75 demographic structural theory 132–7 demographic transition 17, 18, 95, 96, 147, 148, 154, 169, 171, 173, 175, 199–200, 203, 205 demography 61, 62, 65, 71, 72, 117–20, 126, 163, 165, 172, 175, 235, 236 and democracy 161–75 empirical findings 168–72 methodological challenges 172–3 theoretical mechanisms 165–8 dependency ratio 200 development 4, 6, 25, 26, 49–51, 116, 117, 121, 124, 150, 153, 155, 156, 199, 205, 210, 220, 227, 229, 230 differential ethnic population growth 82 domestic politics 7, 180, 191, 192, 195, 196 durable solutions 39 Dyson, T. 165 economic crises 2 economic development 6, 18, 19, 21, 25, 26, 49–52, 151, 152, 174, 175, 240, 241 environmental impact 241–2 economic growth rates 219 economics, of population aging 21–2 economy 201 education 133, 151, 152, 155–7, 164, 167, 168, 201, 205–8, 238

elites 68, 102, 134, 135, 140, 141, 166, 175, 230 environment 8, 11, 207, 235, 238 epidemiological transition 115–17 Erdoğan, Tayyip 186, 192 ethnic change 83, 84, 89, 95, 96 ethnic cleansing 187 ethnic conflict 83, 84 ethnic groups 11, 82, 111, 168, 190 ethnicity 84, 122, 188, 190 ethnic stability transition 221 ethno-national groups 7, 9 familialism 20 far-right political parties 1, 86 female infanticide 66, 68 fertility 3, 4, 6, 7, 10–12, 25, 64, 101–4, 111, 174, 180–83, 194, 203–5, 212 decline 147, 148, 152, 153, 165, 168, 203, 204, 206, 209–12 rates 102–4, 108, 163, 164, 168, 172, 174, 180, 182, 200, 201 weaponizing 111–12 first demographic dividend 148, 152–3, 155–8, 200 first demographic transition 18 forced displacement 32, 33, 35, 36, 39, 119 theories and findings on 32–5 forced migration 11, 29, 31–3, 35–9 best practices, research 38 gaps, research 35–6 methodological advances, displacement research 36–7 progressing research on 29–39 forecasting 235–43 in age-structural time 215–31 global trends, four-phase schema 216–17 models 221–31 political development transitions 224–7 socioeconomic development transitions 221–4 foreign policies 180, 191, 195, 240 Fourth Industrial Revolution (4IR) 152, 157 Freedom House 169–71, 224

INDEX 247

gender 6, 10, 11, 25, 65, 71, 72, 74, 110, 120–23, 207, 208, 212 demography and health 120–23 gender analysis 71 data, lack of 72 interdisciplinarity, lack of 72–3 over-arching theory, lack of 73 training, lack of 71 geography 37, 115, 125, 172, 236, 242 gerontocracy 18–22, 24 global governance 123 global governance institutions 123–5 global health 115, 120, 123–5 global health governance 115, 123–5 global health institutions 8, 122–5 globalization 1, 122, 140, 141 global population 4, 6, 45, 46, 148 Goerres, A. 19 Goldstone, J. 136 governance 11, 12, 19, 71, 73, 123–5, 203, 209, 211 gray social movements 22 great powers 24 Greenhill, K. M. 33, 191 health 2, 3, 4, 6, 8, 9, 10, 11, 12, 65, 115–26, 150, 151, 155, 201, 206 health care 119, 120, 151, 205, 206 health transitions, modern era 115–17 Herzegovina 171 higher fertility rates 64, 65, 102–4, 107–11, 115, 169, 180, 181 Hudson, V. M. 70, 75 human capital 147, 148, 151–3, 155, 156, 164, 165, 200, 203, 205–7, 219 human capital development 205–8 education 206–7 gender dividend 208 health 206 and labor force participation 207 human capital formation 151–2 human rights 22, 86, 180, 188, 191, 194–6, 238, 242 identity 11, 53, 54, 81, 82, 106–8, 155, 162, 180 ideology 74, 82, 85, 92, 224 immigration 25, 26, 81–3, 88–90, 92–6, 108, 140, 143

attitudes 92–5 citizenship 8, 92, 107, 148, 150, 151, 155 law 86 politics 88–9 industrialization 107 institutions 7, 8, 11, 12, 19–21, 24, 115, 125, 131–3, 137, 162 intelligence 13, 157, 215, 216, 236, 237 intergenerational relations 19–22 internally displaced people (IDPs) 33, 36, 53 interstate conflict 23, 105, 238 investments 68, 147–51, 153, 155–7, 199, 200, 205, 206, 210 Ireland 204, 206, 208 jobs 132, 133, 137, 140, 143, 148, 151, 152, 156, 157, 164–6 access 151–2 labor market 156, 161, 162, 164, 167, 174, 204, 207, 208 late-secondary educational attainment transition 221 Leahy, E. 154, 216 Lesthaeghe, Ron 18 liberal democracy 162–8, 170–73, 175, 218, 219, 221 liberal revolution 85–8 life expectancy 3, 4, 11, 21, 165, 206, 210 at birth 3, 4 Lijphart, A. 165 Lindh, T. 216 Livi-Bacci, Massimo 2, 4 macroeconomic climate 208–9 male/female relations, societal structuration 62–3 Malthus, Reverend T. 2, 163 marriage 7, 12, 63–72, 74, 75, 83, 110 marriage markets 7, 65–8, 70 obstruction 61, 63, 65, 66, 71 maternal and child mortality 12 maternal health 121 maternal mortality 65, 121, 123 Matfess, H. 70 mature populations 139, 140, 142, 228

248

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

mechanical effect, age structure change 200 Meiji restoration 138 Merkel, Angela 192 migration 3, 9, 10, 11, 32, 35, 38, 39, 48, 50, 56, 89, 180, 183–5, 188, 191–3, 195–6 Millennium Development Goals (MDGs) 121, 156 Moller, Herbert 163 mortality 3, 4, 11, 12, 65, 121, 123 Muslims 91, 92, 102, 104, 106–12 nationalism 84, 134, 179, 219, 238 national populism, geographic differences 84–5 national power 241 national security 23–5, 34, 61–75, 184, 185, 235, 236 and population aging 23–4 natural disasters 35, 39, 54–6, 118, 119 negative growth revolutions 138 Nien Rebellion 68 non-communicable diseases 116 non-territorial conflicts 225, 226, 228–31 one-child policy (China) 6, 181 Organski, A. F. K. 105 out-migration 180, 188, 191, 192, 195, 196 per-capita income transition 221 polarization 81–96 policy 32, 86, 153, 157, 180, 181, 191–3, 200, 203, 205, 206, 209–11, 236–8, 240 policy implications 203 policymaking 74 policy-relevant demographic research analysts and 236–7 future, research 238–9 policy relevance, maximizing 237–8 twenty-first century agenda 235–43 political demography 2, 3, 6, 8–13, 18, 26, 82, 83, 88, 120, 125, 149, 240 challenges, evolving 8 domestic politics, relationship 7–8 global power, relationship 8 institutions, importance 8

problematic 9–10 relationship to development 6–7 research in 3, 6–9 theory and group boundaries 83–4 see also individual entries political economy 21–2 political forecasting 13 political institutions 20, 24, 107, 115, 143 political liberalization transition 221 political power 8, 18, 19, 51, 108, 137, 189 political rights 69, 169, 170, 209, 224 political stability 136, 138, 143, 153, 174, 181, 201, 221, 225, 227, 228 political transitions 215, 216, 229, 230 political violence 101, 102, 107, 111, 162, 164, 165, 167, 168, 172, 174 population size 3, 4, 10, 46, 163, 184, 235, 242 population age structure 200 population aging 6, 11, 17–19, 21–6, 140, 141, 174, 184 additional areas, research 25–6 gerontocracy 19–22 gray social movements 22 institutions 19–22 intergenerational relations 19–22 and national security 23–4 political economy 21–2 research in 17–27 research limitations 24–6 populations 2, 3, 6, 8, 9, 17, 18, 21, 26, 47, 106, 131–44, 200 aging and declining 239–42 changes 2, 8, 9, 82, 83, 131, 132, 142, 143, 147 color revolutions 139–40 composition 2–4, 86, 180 density 1, 131 driven revolutions and rebellions 132–7 driver of social change 131–2 dynamics 3, 6, 82, 83, 106, 199 future research 142–4 future revolutions and 140–42 growth 132–8, 140–42, 174, 199, 203 politics, relationship 3–6 trends 140–42 populism 81–96, 140–43, 238, 239 poverty 8, 12, 45, 47, 49–53, 55, 56, 180, 181

INDEX 249

and economic development 49–52 power 8, 19, 55, 82, 89, 101, 102, 104–8, 122, 138, 183, 189 power transition theory (PTT) 102, 104–12 Christians and Muslims, Lebanon 106–8 Conservative Christians, West 109–11 Jews and Muslims, Israel and Palestine 108–9 prevalent polygyny 68, 70 pronatalism 84 pronatalist policies 101, 103, 108–10 public opinion 34, 87, 90

strategic foresight 193, 236–7 strategy 33, 53, 86, 88, 102, 104, 108, 111, 121, 185 Sustainable Development Goals (SDGs) 117, 121, 148, 155, 201

Quiverfull movement 110

Ukraine 139, 171, 224 uncharted territory 17–27 UN Demographic Yearbook 46 United Nations High Commissioner for Refugees (UNHCR) 29 urbanization 45–56, 119, 143, 241 climate change 54–6 conflict 52–4 environmental impact 241–2 poverty and economic development 49–52 sources of 48–56 urban migration 49–51, 54, 55

rebellion 68, 131–44, 221 refugees 29–37, 39, 88, 107, 108, 142, 143 global count of 30 region of origin 31 refugees and asylum-seekers 36, 188, 195 regime type 13, 18, 21, 171, 209 religion 11, 81–3, 88, 91, 102–4, 122, 123, 188, 190 power of 103–4 repatriation 32, 34, 39 research agenda, youth policies and investments 147–58 revolutionary conflict 201, 225 revolutions 131–44, 225, 226, 238, 241, 242 color 139–40 Russia 4, 5, 7, 17, 22, 134, 136, 140, 141, 143, 169, 171, 174 salience 92–5 second demographic dividend 200 second demographic transition 17, 18 separatist conflicts 225–8 sex 61–75 sex ratio 61, 66–9, 72, 75 slums 11, 50 social change 131–2 societal health 115–17, 120 state behavior 230 strategic demography 179–96 migration, foreign policy 191–5

technology 8, 9, 23, 84, 185, 187, 236, 238–40, 242 Teitelbaum, Michael 9 territorial conflict 225–7, 229 Thailand 10, 166, 171, 224 third demographic transition 26, 85, 96 total fertility rate (TFR) 18, 64, 161, 205, 219, 227 Trump, Donald 91

variables 64, 65, 72, 88, 169, 171, 172, 206, 211, 235, 237 voter turnout 19 weaponization 101, 103, 105, 107, 109, 111 Weapons of Mass Migration, K. M. Greenhill 191 Weber, H. 171, 172 wedding costs 69 Weiner, Myron 2 white majorities, decline of 89–92 whiteshift 81–96 window of opportunity 6, 200, 201 wombfare 12, 101–12 women 10, 11, 25, 62–8, 71, 73–5, 121–3, 207, 208, 212 abnormal sex ratios 67–8 brideprice and wedding costs 69–71

250

A RESEARCH AGENDA FOR POLITICAL DEMOGRAPHY

marriage markets 65–71 national stability 65–71 national stability and 63–5 nation’s birth and death rates 63–5 prevalent polygyny 68–9 working-age population 6, 12, 25, 26, 200, 203, 204, 207, 211 World Bank 64, 150, 152, 153, 155, 156, 158, 205, 207, 211, 223, 224 World Health Organization (WHO) 4, 65, 117, 122–5, 148–9, 221

youth 10, 133, 147–51, 153–7, 163–4, 167–71, 174 adults and citizens, transition 155–6 assets and opportunities 154–6 bulges 64, 65, 139, 149, 153, 154, 163–4, 166–9, 171–4 definitions and data access 148–50 and national security 153–4 policies 147–51, 153, 155–7 training and challenges 151–4 youthful populations 6, 12, 163, 166, 218, 219