Economic Origins of Dictatorship and Democracy

by Daron Acemoglu

  We have less to say on the extension of suffrage to women. In almost all European countries, voting rights were first given to adult men and subsequently extended to women. This reflected the then-accepted gender roles; when the roles began to change as women entered the workforce, women also obtained voting rights. It is likely, therefore, that the mechanisms that we propose better describe the creation of male suffrage than the extension of voting rights to women.

  Our dichotomous distinction between democracy and nondemocracy makes sense and is useful only to the extent that there are some important elements central to our theory and common to all democracies but generally not shared by nondemocracies. This is indeed the case. We argue that democracy, which is generally a situation of political equality, looks after the interests of the majority more than nondemocracy, which is generally dominated by an elite and is more likely to look after its interests. Stated simply and extremely, nondemocracy is generally a regime for the elite and the privileged; comparatively, democracy is a regime more beneficial to the majority of the populace, resulting in policies relatively more favorable to the majority.

  We claim that nondemocracy represents political inequality relative to democracy. In democracy, everybody has a vote and, at least potentially, can participate in one way or another in the political process. In nondemocracy, an elite, a junta, an oligarchy, or - in the extreme case - just one person, the dictator, is making the decisions. Hence, the contrast in terms of political equality makes sense. This, of course, does not mean that democracy corresponds to some ideal of political equality. In many successful democracies, there is one-person-one-vote, but this is far from perfect political equality. The voices of some citizens are louder, and those with economic resources might influence policies through nonvoting channels, such as lobbying, bribery, or other types of persuasion. Throughout the book, when we discuss political equality in democracy, it is always a relative statement.

  Overall, the outlines of our basic approach are taking shape. We think of regimes falling into one of two broad categories: democracy and nondemocracy. Democracy is thought of as a situation of political equality and characterized by its relatively more pro-majority policies. Often pro-majority policies coincide with pro-poor policies, especially a greater tendency to redistribute income away from the rich toward the poor. In contrast, nondemocracy gives a greater say to an elite and generally opts for policies that are less majoritarian than in a democracy.

  2. Building Blocks of Our Approach

  We have now determined the basic focus of our investigation: to understand why some societies are democratic, why some societies switch from nondemocracy to democracy, and why some democracies revert back to dictatorships. We have already mentioned some of the building blocks of our approach; it is now time to develop them more systematically.

  The first overarching building block for our approach is that it is economic.2 By this term, we do not mean that individuals always act rationally according to some simple postulates. Nor do we mean that there are only individuals, and no social groups, in society. Instead, we mean that individuals have well-defined preferences over outcomes or the consequences of their actions; for example, they prefer more income to less and they may prefer peace, security, fairness, and many other things. Sometimes masses of individuals have interests in common or even act collectively. However what matters is that individuals do have well-defined preferences that they understand. They evaluate various different options, including democracy versus nondemocracy, according to their assessments of their (economic and social) consequences. In such situations, the economic approach suggests that people often behave strategically and that their behavior should be modeled as a game. Game theory is the study of situations with multiple decision makers, interacting strategically. The basic tenet of game theory is that individuals choose between various strategies according to their consequences. Our economic focus and the presence of important interactions between various political actors render all the situations analyzed herein essentially “game theoretic.” We, therefore, make heavy use of game theory in modeling preferences over different regimes and transitions between these regimes.

  To see the implications of these assumptions, consider a group of individuals for whom democracy and nondemocracy have the same consequences in all spheres, except that democracy generates more income for them; they naturally prefer more income to less. Therefore, we expect these individuals to prefer democracy to nondemocracy. At some level, this postulate is very weak; but, at another level, we are buying a lot with our economic focus. Most important, we are getting a license to focus on the consequences of the regimes, and preferences over regimes are derived from their consequences. Such an approach is consistent with many historical accounts of the motivations of different actors. For example, in 1839, the Chartist J. R. Stephens argued:

  The question of universal suffrage ... is a knife and fork question, a bread and cheese question ... by universal suffrage I mean to say that every working man in the land has a right to a good coat on his back, a good hat on his head, a good roof for the shelter of his household, a good dinner upon his table. (quoted in Briggs 1959, p. 34)

  The alternative would have been simply to assume that one group dislikes democracy whereas another group likes democracy- for example, because of certain ideological preferences or biases (Diamond 1999). Indeed, Diamond (1992, p. 455) argues that

  democracy becomes truly stable only when people come to value it widely not solely for its economic and social performance but intrinsically for its political attributes.

  We are not denying that such ideological preferences exist, but we believe that individuals’ and groups’ preferences over regimes derived from the economic and social consequences of these regimes are more important. Later in the book, we discuss how introducing ideological preferences affects our results, and the general message is that - as long as these do not become the overriding factors - they do not affect our conclusions.

  Our second building block is that politics is inherently conflictual. Most policy choices create distributional conflict; one policy benefits one group whereas another benefits different individuals. This is a situation of political conflict - conflict over the policies that society should adopt. These groups - for example, the rich and the poor - have conflicting preferences over policies, and every policy choice creates winners and losers. For instance, with high taxes, the rich are the losers and the poor are the winners, whereas when low taxes are adopted, the roles are reversed. In the absence of such conflict, aggregating the preferences of individuals to arrive at social preferences would be easy; we would simply have to choose the policy that makes everybody better off. Much of political philosophy exists because we do not live in such a simple world, and situations of conflict are ubiquitous. Every time society (or the government) makes a decision or adopts a policy, it is implicitly siding with one group, implicitly resolving the underlying political conflict in one way or another, and implicitly or explicitly creating winners and losers.

  Although the economic approach emphasizes individual preferences and motivations, many individuals often have the same interests and sometimes make the same decisions. Moreover, groups of individuals may be able to act collectively if there are no collective-action problems or if they can solve any that exist. If this is the case, then we can usefully discuss conflict and who is in conflict with whom in terms of groups of individuals. These groups may be social classes, somewhat similar to Marxist accounts of history and politics, or they may be urban agents, ethnic or religious groups, or the military. Our focus on social groups as key political actors is motivated by our sense that the most important forces in political conflict and change are groups of individuals.

  Leaving aside issues of political philosophy related to how a just or fair society should reconcile these conflicting preferences, how does society resolve political conflict in practice? Let us make this question somewhat more concrete: suppose there ar
e two policies, one favoring the citizens and the other favoring the elites. Which one will the society adopt? Because there is no way of making both groups happy simultaneously, the policy choice has to favor one group or the other. We can think that which group is favored is determined by which group has political power. In other words, political power is the capacity of a group to obtain its favorite policies against the resistance of other groups. Because there are always conflicting interests, we are always in the realm of political conflict. And, because we are always in the realm of political conflict, we are always under the shadow of political power. The more political power a group has, the more it will benefit from government policies and actions.

  What is political power? Where does it come from? In thinking of the answers to these questions, it is useful to distinguish between two different types of political power: de jure political power and defacto political power. Imagine Thomas Hobbes’s (1996) state of nature, where there is no law and man is indistinguishable from beast. Hobbes considered such a situation to argue that this type of anarchy was highly undesirable, and the state, as a leviathan, was necessary to monopolize force and enforce rules among citizens. But, how are allocations determined in Hobbes’s state of nature? If there is a fruit that can be consumed by one of two individuals, which one will get to eat it? The answer is clear: because there is no law, whoever is more powerful, whoever has more brute force, will get to eat the fruit. The same type of brute force matters in the political arena as well. A particular group will have considerable political power when it has armies and guns to kill other groups when policies do not go its way. Therefore, the first source of political power is simply what a group can do to other groups and the society at large by using force. We refer to this as de facto political power. Yet, and fortunately so, this is not the only type of political power. Today, key decisions in the United Kingdom are made by the Labour Party, not because it can use brute force or because it has acquired de facto power through some other means but rather because political power has been allocated to it by the political system (i.e., it was voted into office in the last general election). As a result, among policies with conflicting consequences, the Labour Party can choose those that are more beneficial to its constituency or to its leaders. We call this type of political power, allocated by political institutions, de jure political power. Actual political power is a combination of de jure and de facto political power, and which component matters more depends on various factors - a topic that we discuss later.

  Finally, we refer to the social and political arrangements that allocate de jure political power as political institutions. For example, an electoral rule that gives the right to decide fiscal policies to the party that obtains 51 percent of the vote is a particular political institution. For our purposes, the most important political institutions are those that determine which individuals take part in the political decision-making process (i.e., democracy versus nondemocracy). Therefore, a major role of democracy is its ability to allocate de jure political power. In democracy, the majority has relatively more de jure political power than it does in nondemocracy. That democracies look after the interests of the majority of citizens more than nondemocracies is simply a consequence, then, of the greater de jure political power of the majority in democracy than in nondemocracy.

  3. Toward Our Basic Story

  Armed with the first two basic building blocks of our approach, we can now start discussing preferences over different regimes. Typically, there is political conflict between the elites and the citizens, and democracies look after the interests of the citizens more than nondemocracies. It is, therefore, natural to think that the citizens have a stronger preference for democracy than the elites. So, if there is going to be conflict about which types of political institutions a society should have, the majority of citizens will be on the side of democracy and the elites will be on the side of nondemocracy. This is a good starting point.

  We could add more empirical content to this structure by assuming that the elites were the relatively rich and the majority the relatively poor. Indeed, in many instances, the transition from nondemocracy to democracy was accompanied by significant conflict between poorer elements of society, who were hitherto excluded and wanted to be included in the political decision-making process, and the rich elite, who wanted to exclude them. This was most clearly the case in nineteenth-century Europe, particularly Britain, as we saw in Chapter 1, when initially the middle classes and subsequently the working classes demanded voting rights. Their demands were first opposed by the rich elite, who then had to concede and include them in the political system.

  In line with this account of political developments in nineteenth-century Europe, Aminzade (1993, p. 35) describes the arrival of universal male suffrage to French politics as follows:

  French workers, mainly artisans, constituted the revolutionary force that put the Republican party in power in February 1848 ... and working class pressure from the streets of Paris forced liberal Republican leaders ... to reluctantly concede universal male suffrage.

  Perhaps, more tellingly, the key players in the process of democratization saw it as a fight between the rich and the poor. Viscount Cranborne, a leading nineteenth-century British Conservative, described the reform struggle as

  ... a battle not of parties, but of classes and a portion of the great political struggle of our century - the struggle between property ... and mere numbers. (quoted in Smith, 1966, pp. 27-8)

  The conflict between the poorer and richer factions of society was also a defining characteristic of most instances of the introduction of universal suffrage in Latin America in the first half of the twentieth century - including the experiences in Argentina in 1912, as we saw in Chapter 1, but also in Uruguay in 1919, in Colombia in 1936, and in Venezuela in 1945. The arrival of democracy in South Africa and Zimbabwe similarly followed a conflict between the rich whites and poor blacks.

  This discussion, therefore, highlights how the majority of citizens want democratic institutions because they benefit from them and, therefore, will strive to obtain them. Given our definition of political power, we can say that the citizens are more likely to secure a transition to democracy when they have more de facto political power. Thus, we have already constructed a simple theory of democratization : the citizens want democracy and the elites want nondemocracy, and the balance of political power between the two groups determines whether the society transits from nondemocracy to democracy (and perhaps also whether democracy, once created, becomes consolidated or reverts back to nondemocracy later).

  This could be viewed as a simplified version of our theory of democratization. But, in fact, it is so simplified that some of the essential features of our theory are absent. Most important, the role that democracy or, more generally, political institutions play is trivialized.

  The theory says that democracy leads to social choices more favored by the majority of citizens; hence, the citizens prefer democracy to nondemocracy, and democracy results when the citizens have sufficient political power. However, if the citizens have sufficient political power, why don’t they use this power to simply obtain the social choices and policies that they prefer rather than first fight for democracy and then wait for it to deliver those policies to them? Is democracy simply a not-so-necessary intermediate step here? One could argue so.

  This is only a feature of the simple story we have told so far, and it is a characteristic of neither real-world political institutions nor of our theory. In practice, political institutions play a much more fundamental role than being a simple inter-mediating variable: they regulate the future allocation of political power between various social groups. They play this role because we do not live in a static world like the one described in the previous narrative but rather in a dynamic world, where individuals care not only about policies today but also about policies tomorrow. We can capture this important role of political institutions and obtain a more satisfactory understanding of
democracy and democratization by incorporating these dynamic strategic elements, which is what our theory of democratization attempts to do.

  4. Our Theory of Democratization

  Consider the simplest dynamic world we can imagine: there is a “today” and a “tomorrow,” and the elites and the citizens care about policies both today and tomorrow. There is nothing that prevents society from adopting a different policy tomorrow from the one it chose today. Thus, it is not sufficient for the citizens to ensure policies they prefer today; they would also like similar policies to be adopted tomorrow. Suppose we are in a nondemocratic society, which generally looks after the interests of the elites. Citizens have de facto political power today, so they can obtain the policies they like, but they are unsure whether they will have the same political power tomorrow. Given that we are in a nondemocratic society, tomorrow the elites may become more powerful and assertive and the citizens may no longer have the same political power. Can they ensure the implementation of the policies they like both today and tomorrow?

  This is where political institutions may be important relative to the static world described previously. Institutions, by their nature, are durable - that is, the institutions of today are likely to persist until tomorrow. A democratic society is not only one where there is one-person-one-vote today but also one that is expected to remain democratic at least in the near future. This durability was already implicit in our definition of political institutions as a means of allocating political power: they regulate the future allocation of political power. For example, democracy means that tomorrow there will be a vote to determine policies or to decide which party will rule and the whole population will participate. Nondemocracy means that much of the population will be excluded from collective decision-making processes.