Economic Origins of Dictatorship and Democracy

by Daron Acemoglu

  8. Democracy in a Picture

  The previous discussion illustrates the various empirical implications of our theory for the circumstances under which a society becomes and stays a democracy. To fix ideas, it is useful to use simple pictures to illustrate the circumstances under which different regimes arise. To map our comparative statics about inter-group inequality into the data, we do this in the context in which the elites are the rich and the citizens are the poor. Recall the four “paths” of political development that we sketched in Chapter 1. The first, the British path, was one of fully consolidated democracy. The second, the Argentine path, was that of unconsolidated democracy. The third path, that of Singapore, was persistent nondemocracy in which the political status quo can be sustained without serious repression. The fourth path, that of South Africa, was persistent nondemocracy with repression. The comparative statics of our theory allow us to depict these different outcomes in a picture.

  In essence, the different political outcomes occur because these societies differ fundamentally in their underlying economic structures, and it is this that motivates the title of our book. In addition, we also emphasize differences in political institutions, to some extent historically determined, to some extent consciously chosen with the nature of the regime in mind. To keep the pictures simple, we assume that conditions are such that a revolution never occurs in equilibrium, and we also abstract from the use of concessions (e.g., they are always insufficiently credible to stop revolt) so that if revolution is a threat, a nondemocratic regime must choose between repression or conceding democracy.

  Consider Figure 2.1, which captures the predictions of our theory for democratization. On the horizontal axis, we plot inequality, with moves from left to right corresponding to greater inequality. The origin represents a completely equal society. On the vertical axis, we plot the historically determined costs of repression that are exogenous, such as the extent to which repression destroys assets in society. We have divided the resulting square into different regions that represent how different structures lead to different paths of political development. When inequality is sufficiently low, the cost of repressing or mounting coups is irrelevant because the poor are sufficiently content under the political status quo not to rock the boat. This corresponds to Region A, where there is nondemocracy that remains unchallenged, and in it we place Singapore. In Region B, inequality is higher and revolution becomes a threat. However, the cost of repression is sufficiently high that democracy is created. In this region, we place both Britain and Argentina. Finally, in Region C, inequality is so high that revolution is a threat to nondemocracy but the cost of repression is sufficiently low that democracy can be avoided. This is the case of South Africa until 1994. The cost of repression in South Africa might have been lower because the disenfranchised groups were black Africans and Coloureds, and exclusion and repression were justified by an explicitly racial philosophy.

  Figure 2.1. Democratization.

  Figure 2.2. Democratic Consolidation.

  To study the consolidation of democracy, we must turn to Figure 2.2. Here, there are just two regions separated by an upward-sloping curve. When the cost of a coup is zero, the rich are always willing to undertake a coup. However, as the cost of a coup rises, inequality must be sufficiently high (i.e., democracy must be sufficiently costly to the rich) for it to be worthwhile. Figure 2.3 has just two regions. We have placed Britain in Region A. Once created, democracy will consolidate if it is not too redistributive and if coups are sufficiently costly. However, when inequality is very high, the costs of a coup may be sufficiently low that it is attractive. This is the case in Region B, where democracy is unconsolidated; here, we have placed Argentina. Singapore is, of course, not in this picture because it has yet to make the transition to democracy.

  Figure 2.3. Democratization in South Africa.

  These simple pictures also allow us to trace out the paths of political development of different countries. For example, we look at the history and future of democracy in South Africa in Figure 2.3. Why did South Africa finally move so belatedly to a democratic regime? The arrows in Figure 2.3 capture part of the story. From the mid-1970s onward, inequality fell in South Africa, making democracy less threatening for the white elite. At the same time, the industrial sector rose at the expense of the agricultural sector, and human and physical capital became more important. In terms of the picture, this means that at a given level of inequality, the elites are less willing to repress. This moves the boundary between Region C and Region B downward. Changes in the global environment, particularly globalization, also have the effect of moving the same boundary downward, implying that for fixed levels of inequality, the cost of repression had to be lower to justify the persistence of dictatorship. Thus, some time between 1970 and 1994, South Africa moved out of Region C into Region B, and democracy was created.

  What does the future hold for South Africa? To see this, we must turn to Figure 2.4, which asks whether democracy will consolidate after apartheid. Because inequality is still very high, one might conjecture that South Africa would be in Region B and, therefore, an unconsolidated democracy. Nevertheless, the impact of development of the South African economy, increasing importance of physical and human capital, and increased globalization has the effect of moving the boundaries between Regions A and B down. Now, for a given cost of a coup, inequality must be higher to justify mounting a coup against democracy. Moreover, as discussed in Chapter 1, the structure of political institutions after apartheid was designed specifically to protect the interests of the whites, a factor that again moves this line down. Thus, although one cannot be certain of the future (witness the evolution of democracy in Zimbabwe since 1980), one might hope that South Africa had transitioned into Region A rather than Region B.

  Prediction in the case of Singapore seems much easier. Figures 2.1 and 2.2 suggest that if and when Singapore becomes a democracy, it is very likely to consolidate.

  Figure 2.4. Democratic Consolidation in South Africa?

  9. Overview of the Book

  The remainder of our book develops the arguments outlined in this chapter. The remainder of this part continues to lay the scene. In Chapter 3, we survey the empirical evidence about cross-country patterns of democracy. We show that richer countries are more likely to be democratic, more educated countries are more likely to be democratic, and more unequal countries are generally less democratic. We emphasize the basic correlations in the data and do not take a strong view on causal relationships. Chapter 3 also discusses the large literature in political science and sociology on the creation and consolidation of democracy, and we explain how our research contributes to this work.

  Part 2 surveys existing models of collective decision making in democracies and nondemocracies. In Chapter 4, we focus on democracies and provide a simple analysis of basic issues in the study of collective choice, electoral politics, and competition, which is useful in later parts of the book. We also introduce some basic models of two-group distributional conflict, paying special attention to the relationship between inequality and redistribution, the implications of different political identities, and the factors that determine the distribution of power in democracy. In Chapter 4, we also propose a reduced-form model of the distribution of power in a democracy. The appendix at the end of the book develops a series of models that provide microfoundations for this reduced form. In Chapter 5, we analyze nondemocracy with particular attention to the collective-action problem and the issue of commitment.

  Part 3 provides our approach to democratization. In Chapter 6, we introduce our basic model of democratization. This chapter formalizes many of the issues already mentioned in this introductory chapter, giving us ways to think about the role of political power and the role of political institutions in allocating future political power. It illustrates how democratization creates a credible commitment to future redistribution by transferring political power to the majority in society. It also shows how democratizat
ion may be a response by the elite in the face of a credible threat of revolution by the majority. We see the possibility of an inverted U-shaped relationship between inter-group inequality and democracy in this chapter. Chapter 7 then develops our basic model of coups against democracy and studies the circumstances under which democracy, once created, consolidates.

  Part 4 discusses a number of important extensions to this basic framework and some applications. In Chapter 8, we analyze how the presence of a large and affluent middle class affects the balance of the distributional conflict between the elites and the citizens in ways that can help create and consolidate democracy. Chapter 9 introduces factor endowments and markets to endogenize the distribution of income and discusses the impact of the structure of the economy on the creation and consolidation of democracy. In this chapter, we also conjecture about mechanisms that might account for political development - that is, the question of why and whether countries transition to democracy as they become richer, and the potential reasons for the relationship between income and democracy. Chapter 10 extends our model to allow for international trade and mobility of factors of production among countries and studies how globalization alters and adds to the results we have derived until this point.

  Part 5 discusses the future of democracy and concludes the book.

  3

  What Do We Know about Democracy?

  In this chapter, we provide an overview of previous research on democracy. We depict some of the most salient “stylized facts” about democracy that have been emphasized and much debated in the literature. Because a reevaluation of the existing empirical evidence is not our main focus, we present these patterns diagrammatically without using formal econometric techniques. Although these patterns may not correspond to causal relationships, they are still informative about the correlates of democracy in the data, thus potentially informative about the type of models we should develop. In the final two sections we discuss the existing approaches to democracy and explain how our approach differs from and contributes to the existing literature.

  1. Measuring Democracy

  The first challenge facing a quantitative analysis of the patterns of democracy is to develop reliable and informative measures. There has been much controversy over this issue in political science, mostly because there is disagreement about what actually constitutes a democracy. Many scholars, however, accept the definition proposed by Schumpeter (1942), who argued that democracy was

  ... the institutional arrangement for arriving at political decisions in which individuals acquire the power to decide by means of a competitive struggle for the people’s vote. (p. 250)

  In practice, then, democracy is associated with a particular set of institutions, such as free and fair elections, the accountability of politicians to the electorate, and free entry into politics. Even accepting a Schumpeterian definition, countries differ as to the extent to which any of the institutional conditions are satisfied. This suggests to most scholars the need to make a finer distinction than simply between a democracy and a nondemocracy.

  Our first and main measure of democracy is the Freedom House political rights index, which has been used by many other scholars in quantitative work on democracy (e.g., Barro 1997, 1999). This index ranges from 1 to 7, with 7 representing the least political freedom and 1 the most freedom. A country gets a score of 1 if political rights come closest to the ideals suggested by a checklist of questions, beginning with whether there are free and fair elections, whether those who are elected rule, whether there are competitive parties or other political groupings, whether the opposition plays an important role and has actual power, and whether minority groups have reasonable self-government or can participate in the government through informal consensus. The main checklist includes three questions on the electoral process, four questions on the extent of political pluralism and participation, and three questions on the functioning of government. For each checklist question, 0 to 4 points are added, depending on the comparative rights and liberties present (0 represents the least, 4 represents the most). These scores are totaled and used to determine where a country resides on the 1 to 7 scale.7 Following Barro (1999), we supplement this index with the related variable from Bollen (1990, 2001) for 1960 and 1965, and we transform both indexes so that they lie between 0 and 1, with 1 corresponding to the most democratic set of institutions.

  The Freedom House index, even when augmented with Bollen’s data, enables us to look only at the postwar era. The Polity IV dataset, on the other hand, provides information for all countries since independence starting in 1800. To look at pre-1960 events and to check on our main measure, therefore we also look at the other widely used measure of democracy: the composite Polity index, which is the difference between the Polity’s democracy and autocracy indexes.8 The Polity democracy index ranges from 0 to 10 and is derived from coding the competitiveness of political participation, the openness and competitiveness of executive recruitment, and the constraints on the chief executive. For instance, constraints on the executive is coded on a 7-point scale running from “unlimited authority” where “there are no regular limitations on the executive’s actions (as distinct from irregular actions such as the threat or actuality of coups and assassinations),” to “executive parity or subordination” where “accountability groups have effective authority equal to or greater than the executive in most areas of activity.” A country would receive the lowest score for constraint on the executive if “constitutional restrictions on executive action are ignored” or “there is no legislative assembly or there is one but it is called or dismissed at the executive’s pleasure.” A country would receive the highest score, on the other hand, if “a legislature, ruling party or council of nobles initiates much or most important legislation” or “the executive is chosen by the accountability group and is dependent on its continued support to remain in office.” The Polity autocracy index also ranges from 0 to 10 and is constructed in a similar way to the democracy score based on scoring countries according to the competitiveness of political participation, the regulation of participation, the openness and competitiveness of executive recruitment, and the constraints on the chief executive. To facilitate comparison with the Freedom House score, we also normalize the composite Polity index to lie between 0 and 1.

  Figure 3.1. Evolution of Democracy 1960-2000.

  Figure 3.2. Evolution of Democracy 1960-2000.

  Both of these measures enable us to distinguish between different shades of democracy. An alternative empirical approach has been defended and used by Przeworski and his coauthors (Przeworski, Alvarez, Cheibub, and Limongi 2000, Chapter 1), who argue that a simple dichotomy between democracy and nondemocracy is the most useful empirical definition. In addition to the dichotomous classification, these authors add other provisos to the definition of a democracy, most important that a country cannot be democratic unless a political party has been observed to lose power. Hence, according to Przeworski et al. (2000), Botswana has never been a democracy because, even though all agree that elections are free and fair, that there is free entry into politics, and that the government is accountable to the people, the Botswana Democratic Party has won every election since independence in 1966. Japan would not have been a democracy for most of the post-Second World War period until the Liberal Democratic Party lost power, and South Africa today is not counted as a democracy because the ANC has formed the government since the end of apartheid.

  Although there is a lively debate among political-science scholars about the virtues of continuous versus dichotomous measures, none of the patterns discussed here depend on this choice. Our preference is for more fine grained measures, although dichotomous measures also have advantages. For example, they enable a clearer discussion of transitions from and to democracy; in the following section, we use the dichotomous measures developed by Przeworski et al. (2000) and augmented by Boix and Rosato (2001) to discuss transitions to and from democracy.

  2. Patterns
of Democracy

  Figures 3.1, 3.2 and 3.3 plot the values of the normalized Freedom House, Polity scores, and the augmented Przeworski et al. (2000) index, respectively, for our basic 1960-2000 sample. These figures show that Organization for Economic Cooperation and Development (OECD) countries entered the period almost fully democratic and stayed there. In contrast, democracy declined in other parts of the world, particularly in Latin America and Africa, although from the mid- 1970s onward, we can detect what Huntington (1991) calls the “third wave” of democratization. Figure 3.4 uses the Polity data back to 1840 for all the countries that were independent during this period. This picture vividly displays the onward march of democracy in the OECD in the period leading up to the First World War and shows evidence of the “first and second waves of democracy,” the first before the First World War and the second after the Second World War.

  Figures 3.5 and 3.6 show the most famous correlation about democracy, first investigated by Lipset (1959): rich countries tend to be more democratic. Figure 3.5 shows this by plotting the average Freedom House index during the 1990s versus the average log gross domestic product (GDP) (income) per capita during the 1990s (in purchasing-power party terms, calculated from the latest version of the Summers-Heston data set; Heston, Summers, and Atten 2002). Figure 3.6 does the same using the average Polity score during the 1990s. Both figures show a strong positive relationship between income and democracy. The richer countries, such as the United States, Canada, Australia, New Zealand, and those in the European Union, are all democratic, whereas the poor countries in sub-Saharan Africa, South Asia, and Central America are less democratic.