No. 06-01 “Traders, Teachers, and Tyrants: Democracy, Globalization, and Public Investment in Education” by Ben Ansell Wo r k i n g Pa p e r S e r i e s 1737 CAMBRIDGE STREET • CAMBRIDGE, MA 02138 • TEL 617.495.4420 • FAX 617.495.8292 [email protected] • http://www.wcfia.harvard.edu Traders, Teachers, and Tyrants: Democracy, Globalization, and Public Investment in Education by Ben Ansell Paper No. 06-01 January 2006 About the Author: Ben Ansell, Graduate Student Associate, Ph.D. Candidate, Department of Government: International Relations and Comparative Politics, Harvard University. [email protected] Published by the Weatherhead Center for International Affairs, Harvard University. Copyright by the author. The author bears sole responsibility for this paper. The views expressed here are those of the author and do not necessarily represent the views of the WCFIA or Harvard University. Publications Chair, Weatherhead Center for International Affairs Robert Paarlberg Director of Publications, Weatherhead Center for International Affairs Amanda Pearson Submission procedures: Weatherhead Center affiliates are encouraged to submit papers to the Working Paper Series. Manuscripts are assessed on the basis of their scholarly qualities—the extent of original research, the rigor of the analysis, the significance of the conclusions—as well as their relevance to contemporary issues in international affairs. Manuscripts should range between 25 and 80 double-spaced pages and must include an abstract of no more than 150 words. Authors should submit their paper as an e-mail attachment in a standard word processing application (Microsoft Word or Word Perfect) to the Publications Department at [email protected]. Orders: Working Papers are available for $7.00 each, plus $1.00 for shipping and handling, from the Publications Office, 1737 Cambridge Street, Cambridge, MA 02138. WEATHERHEAD CENTER FOR INTERNATIONAL AFFAIRS HARVARD UNIVERSITY 1737 CAMBRIDGE STREET CAMBRIDGE, MA 02138 TEL: 617-495-4420 FAX: 617-495-8292 www.wcfia.harvard.edu Abstract The globalization debate has largely been fought between those who prophesize a “race to the bottom” in government expenditure and those who foresee continued divergence, with some states better shielded from global economic volatility. However, over the past few decades there has, in fact, been “upward convergence” in the percentage of national income governments devote to public education, albeit amidst considerable cross- national variance. This phenomenon has an enormous distributional impact, yet it has been largely neglected by political scientists. What explains this tremendous shift? This paper argues that two forces in particular shape the aggregate pattern of human capital expenditure: the level of democracy and the level of openness of any given state. By developing a model of the political economy of education investment, and testing its implications over a dataset of 115 countries from 1960 to 2002, this paper provides a first cut at explaining this critical issue. Ben Ansell [1] Introduction The globalization debate has largely been fought between those who prophesize a “race to the bottom” in government expenditure and those who foresee a continued divergence among those states who can shield themselves from global economic forces and those who cannot.1 This dialog has, however, largely ignored a significant display of “racing to the top”, apparent in government investment in education or “human capital”.2 Since 1960 there has been a momentous increase in the proportion of national income that governments devote to educating their citizens. This is a phenomenon with an enormous distributional impact. Numerous econometric studies have shown a robust and substantively significant relationship between the provision of education and economic growth.3 Politicians have, perhaps unsurprisingly, been even more hyperbolic: the British Chancellor of the Exchequer and the governor of the Bank England declared that “education is the best anti-poverty and social and economic development strategy”.4 Figure One demonstrates the magnitude of the average increase in educational 1 On downward convergence see Rodrik 1997 and Kurzer 1992, on continued divergence see Kitschelt, et al. 1999, Swank 2002, and Hall and Soskice 2000. Burgoon 2001 provides an excellent summary of the debate over globalization’s impact on the welfare state. 2 The major exception is the literature on how globalization leads to the expansion of compensatory welfare states, e.g. Katzenstein 1985, Garrett 1998, and Rodrik 1998. However, this literature largely focuses on globalization as increasing macroeconomic variance and thus as a potential negative shock rather than as a positive opportunity. Moreover, most of these works focus solely on the OECD countries. 3 Well-known analyses of education’s positive impact on growth include Mankiw et al 1992, Barro 1997, Hall and Jones 1999, and Bils and Klenow 2000. 4 Gordon Brown and Eddie George in 2002, as cited in Wolf 2004. [2] Weatherhead Center for International Affairs expenditure: whereas in 1960 the international average expenditure on education as a percentage of GDP was 2.5%, by the mid 1990s this had almost doubled to around 4.7%. Figure One –Growth in Public Investment in Education 1960 to 1995 5 n o ti a c u 5 d . E4 c bli u P n4 o t n e p S P 5 . D3 G f o e g a3 t n e c r e P 5 . 2 1960 1970 1980 1990 2000 Year However, despite this overall upward trend in education, there remains considerable cross-national variance. The cross-national standard deviation of expenditure on education as a percentage of GDP has actually doubled since 1960. This implies that whereas in 1960, 95% of states spent between 0.5% and 4.5% of their national income on education (e.g., ranging from Nepal to the United States in 1960), by 1995 this range had increased to between 0.5% and 8.5%, (e.g., ranging from Equatorial Guinea to Denmark in 2000). Figure Two, which presents kernel density estimates of 2 Ben Ansell [3] public education spending across a sample of 115 states, demonstrates this significant increase in variation. Figure Two Kernel Density Function of Public Education Spending 1960 to 1995 5 2 . 2 . 5 y1 t si. n e D 1 . 5 0 . 0 0 2 4 6 8 10 Percentage of GDP Spent on Public Education 1960 1995 What then explains both this tremendous upward shift and the contemporary cross-sectional variation? This paper argues that two forces in particular shape the aggregate pattern of human capital expenditure: the level of democracy and the level of openness of a given state. Human capital investment greatly affects the future distribution of resources in society by using education to reshape the distribution of income. Thus we should expect that institutions like democracy that effectively represent [4] Weatherhead Center for International Affairs the interests of the poorer members of society will respond to demands for such future redistribution through the provision of public education. However, regime type alone cannot explain all of the observed variation. In a closed economy the increased supply of education will, all else equal, reduce the rate of return to educated workers. This creates strong incentives to restrain education spending. Thus, autarkic states, even democracies, may still fail to fully expand education provision. In an open economy, however, factor returns are determined by global, rather than, domestic market forces. An increased supply of education is thus no longer met by a serious decline in the returns to education. We should, hence, expect states that become more open to the international economy to increase their provision of education. Thus, the expansion of a country’s political sphere, through democratization, and their economic sphere, through globalization, should be met by a consequent increase in the sphere of education. By and large, political economists have focused on the impact of democracy and globalization on forms of immediate redistribution, like unemployment benefit.5 However, the study of their impact on future redistribution through education is a less trodden path. This paper attempts to remedy the general neglect of the cross-national determinants of education investment by developing a coherent formal framework and testing it using a global dataset spanning over forty years. The assertion that education policy and democracy are linked is not unique within the literature. A small literature on the relationship between education and democracy has developed recently, including important works by Lake and Baum (2001), Brown and Hunter (2004), Lindert (2004), 5 Burgoon 2001. Rodrik 1998. Moreover there is a large literature on the effects of political institutions on immediate redistribution, e.g. Persson and Tabellini 2000, Iversen and Soskice 2001, Stephens, et al 2003. 4 Ben Ansell [5] and Stasavage (2004). However, there has not been a coherent attempt to both formally model the connection between democracy and investment in education and test these findings over a global dataset. Lake and Baum, for example, examine the political determinants of secondary enrollments rather than educational spending, and although Brown and Hunter, Lindert, and Stasavage examine the latter variable they do so within limited regions. Moreover, with the exception of Lindert, none of these studies have fully addressed the problem of reverse causation that bedevils the link between democracy and education. Finally, and most critically, this paper also contributes an analysis derived from the same formal framework, of how openness to the international economy affects education expenditure. The structure of this argument is developed as follows: • In Section Two, the paper develops a model demonstrating how states with elite control will systematically under-provide human capital as compared to the preferences of a “true” democratic median voter. • Section Three tests this proposition using the well-known Polity IV dataset on democracy with a sample of 115 states from 1960 to 2002. This statistical analysis is followed by a brief case comparison of the relationship between regime change and educational expenditure in the Iberian peninsula and in the Philippines. • Section Four extends the formal model to incorporate open economy analysis and follows this formalization with further empirical analysis. Various measures of openness, including imports and exports, duties, and deviations from predicted levels of trade, are used to operationalize the openness mechanism and incorporated into the statistical analysis. A brief case comparison follows of