Research program
Institutions and Economic Growth
University Co-ordinator
Università Commerciale "Luigi Bocconi" MILANO -
Economia politica - MILANO(MI)
Research Unit Leader
Guido TABELLINI
Description
As explained above, the project wants to address three distinct but related questions: a) What are the economic consequences of alternative electoral rules in modern democracies? In particular, how does the electoral rule influence the relative political power of poor vs rich citizens, and through this channel, how does it contribute to determine the overall level of taxation? b) What happens to economic policy and to economic growth when a country becomes a democracy? Which features of democracies are most conducive to strong economic performance and which instead may retard economic development ? c) What is the role of past political institutions in shaping individual values and social norms, and how can this explain observed differences in the level of development among European regions?The first two parts of the project will be joint with Torsten Persson of the Universiy of Stockholm. The third part will be done by myself alone. Each part will result in one or more scientific papers. These papers will be presented and discussed in an international research group on institutions and economic growth, that is being formed and to which I will participate. The group is organized by the Canadian Institute for Advanced Research, under the coordination of Elhanan Helpman of Harvard University, and with the involvement of about 20 leading international scholars of various disciplines (economics, political science and economic history) including a recent Nobel prize winner.These three parts ask different questions, but they are linked by a common theme: the economic effects of institutions. The first part looks at the effects of formal institutions of modern democracies on fiscal policy. The second part studies the distinction between democracies and non-democracies, with reference to a variety of policy choices. The third part looks at the historical legacy of past political institutions through individual attitudes and social normsHere I describe the specific research project and methodology on each of these issues.a) The consequences of alternative electoral rules.In Persson, Roland and Tabellini (2003), we compared fiscal policy under proportional vs majoritarian electoral rules. Our central conclusion (based on theoretical and empirical analysis) was that proportional electoral rules lead to more fractionalized party systems, to more frequent coalition governments, and through this channel to larger overall government spending. In that paper, the mechanism whereby coalition governments lead to more government spending was the so called "common pool problem": each party in the coalition wants to target redistribution to benefit his own constituency, and the costs that fall on taxpayers at large are not fully internalized. In that formulation, the economic conflict underlying political disagreements in society was over targeted benefits. But this is only one (and perhaps not the main) source of economic conflict in modern societies. Another and more traditional economic conflict is between rich and poor, or more generally between "left and right". How does the conflict between rich and poor play out under different electoral rules? The theoretical framework recently developed in Persson, Roland and Tabellini (2003) can be fruitfully extended to answer this question. The general idea motivating this new research project is that the political influence of marginal groups of very poor voters can be much larger under proportional electoral rules than under majoritarian elections. The reason is that, with proportional elections, these marginal groups of voters are represented in Parliament, and thus have a serious chance of belonging to a coalition government. When this happens, they can exert considerable influence on policy formation. Under majoritarian elections, instead, these extreme groups of voters are much less likely to be represented in the legislature; the party system typically consists of two large parties, both of which fight for the support of the middle classes. Hence, extreme left wing voters (and more generally marginal groups of voters) are likely to be less influential both in the electoral competition and in government formation. In principle, this feature of majoritarian elections can penalize both marginal groups on the right and on the left, and hence need not create a systematic difference in the implied size of government between proportional and majoritarian elections. But given the asymmetry of income distribution, the effect on left wing influence could be predominant. This effect of the electoral rule can be reinforced by the endogeneity of voters' participation. Voters' turnout is known to be much larger under proportional than under majoritarian elections; this could reflect the perception that indeed, under majoritarian elections, marginal groups of left wing voters are not very influential, further reinforcing the impact of the electoral rule on policy formation.We plan to address this problem both in theory and with empirical analysis. First we will solve the theoretical model, extending the model in Persson, Roland and Tabellini (2003) to incorporate this new dimension of economic conflict, between rich and poor. We will then take the prediction of the theory against cross country and panel data. Much of the data has already been collected. But we will need to collect new data, on the ideological positions of parties in the legislature and in government, and on the distribution of ideological preferences of voters. Being able to address this issue is important in order to understand the precise mechanisms through which electoral rules influence policy decisions and in particular the overall level of taxation. Making progress on this question, in turn, is essential in order to characterize the normative tradeoff that a country faces when reforming or designing its electoral system, and how this tradeoff interacts with the specific economic and political situation. b) Democracy vs Non-democracies and economic development. The literature summarized in point 2.4 above, on the interaction between democracy and development, has mainly looked at cross country comparisons (implicitly or explicitly). Episodes of transition from autocracy into democracy that exploit only the within country variation have not been carefully studied, instead. Moreover, almost all of the literature has focused exclusively on the interaction between development and democracy, neglecting the question of how democracy influences specific policy choices. Finally, the vast heterogeneity between democracies and non-democracies has been overlooked. In our research, we plan to focus on these neglected aspects of the problem. A large data set on political institutions that dates back to 1800 now exists (the POLITY IV dataset). We plan to study episodes of transition from autocracy into democracy, and viceversa. We have already started collecting some data, and done some preliminary analysis. We have found a few robust empirical regularities. First, economic policy and economic performance tend to improve after episodes of trade reform and economic liberalizations. This finding helps us to identify instances of meaningful and effective economic reforms, confirming earlier results by Sachs and Werner (1995) and by Wacziarg and Welch (2003). Second, we found that episodes of economic liberalizations tend to be preceded by political liberalizations. The timing of events strongly suggests that causation runs from political to economic reforms and not viceversa. First, a country becomes a democracy. Then, it liberalizes its economy and improves its economic institutions. But despite the positive association with subsequent economic liberalizations, on average transitions to democracy do not seem to be followed by growth accelerations nor by improvements in economic policies. Why is that so and how can it be consistent with the positive association between democratization and economic liberalization? Several alternative conjectures might explain this puzzle. One possibility is that it is the combination of economic and political liberalization that promotes better policies and institutional improvements. Another possibility, suggested also by preliminary results in Persson (2004), is that it is the type of democracy (eg. parliamentary or presidential) also makes a big difference. In particular, transition into a weak democracy could be counterproductive, relative to remaining a non-democracy, perhaps because of the risk of reversal. In general, to address these questions, we plan to study in detail these episodes of economic and political reforms, to try and gain a robust understanding of which features of democratic institutions are helpful for economic development and which features instead might be counter-productive. Making progress towards these issues is essential, both to assist new emerging democracies, and to gain a better understanding of the deeper determinants of government incentives to enact good policies. c) Past political institutions, individual values and differences in development among the European regionsIn the third part of our research project, I plan to study the link between historical institutions and individual attitudes and values, as well as the transmission of these values through time. My goal is to try and explain at least part of the observed differences in economic development among European regions as the result of different attitudes and values of its inhabitants. To sort out obvious problems of reverse causation, I plan to try and trace back the origins of these different cultural attitudes to the different political history of different European regions. The empirical research strategy will exploit the World Value Survey data base, that contains individual opinions and values over a large range of issues on a European scale. These values and opinions seek to measure individual trust (social capital), but also confidence on the state and other institutions, propensity towards risk, willingness to innovate, propensity to cheat the government, confidence in the market systems, and so on. Observed regional differences in these values and opinions are very large. I will then relate summary measures of these values, on the one hand to the political history of the region, on the other hand to current regional economic performance. Given the political and economic history of Europe, carrying out this empirical analysis at the regional level (rather than at the national level) could make it possible to separately identify the effect of past institutions or of other historical differences between regions. The political history of Europe is very rich, and could teach us important lessons. Several European regions that are now part of the same country (in Italy but also in parts of Germany and Belgium) were ruled by very different political institutions in the 1700s and in the first half of the 1800s. This variety can be exploited to explain currently observed regional difference in individual values and opinions. The next step would then be to ask whether these historically determined differences in values and opinions in turn can explain currently observed differences in economic development, after taking into account other observable determinants of development. The analysis would thus resemble the early work by Putnam 1993), except that it would be carried out on a European scale, and with a larger set of measures of individual values and opinions and more systematic and rigorous quantitative methods.
|
