Political Economy Lecture Notes Daron Acemoglu, Exams of Political Economy

Download Political Economy Lecture Notes Daron Acemoglu and more Exams Political Economy in PDF only on Docsity!

Daron Acemoglu

These are preliminary notes. They include typos and

mistakes.

CHAPTER 1

General Issues

The first lecture is a broad overview of the main issues and the evidence. The rest of the course will be about formal models. However, it is useful to begin with a relatively broad discussion, without formalizing every claim. Some of the statements here will be formalized in the models discussed later, and for those that are not formalized, you will have the tools to formalize them depending on your own interests.

1.1. What Are Institutions? 1.1.1. Different views of institutions. One of the first things you learn in economics as an undergraduate is that the way that societies are organized, or briefly their “institutions”, are an important determinant of their economic performance–though perhaps without being clear what “institutions” are. For instance, you learn that competitive markets, in conjunc- ture with well-defined and enforced property rights, allocate resources efficiently. An economy with such a set of institutions is then contrasted to other ways of organizing the economy. For example, the most common argument is that the Soviet Union did less well economically because it tried to allocate resources via central planning (though see Allen, 2003). This debate about the pros and cons of socialist versus capitalist institutions dates back at least to the 1920s. One argument (now obviously discredited) was that socialism would not face incentive problems because people’s preferences would change when society was changed (just another example that one has to be careful about claims regarding “manipulation of preferences”). In response, von Mises raised the ‘calculation problem.’ He argued that it would be simply impossible for a socialist central planner to compute efficiently the allocation of resources. In the 1930s Lange (1936, 1937) famously argued that since the central planner knew the Walrasian system of supply and demand equations, he would be able to mimic the mar- ket and solve the equations for the allocation of resources. The famous response to Lange came from Hayek (1945), who argued that the price system was able to communicate and aggregate information much more efficiently than an individual or a computer could do. Of

if institutions matter, they themselves are the outcome of some type of negotiation among individuals and social groups, so that dysfunctional and highly inefficient institutions can be avoided. Leaving these issues aside for a second, it is also useful to note that the institutional structure, while important in economic models, is often left implicit and almost always taken as exogenous. For example, in the Arrow-Debreu model people have endowments and there are competitive markets in which they can buy and sell. It is clear who has what and nobody gets to try to take someone else’s endowment and nobody refuses to honor a contract. However, even when they explicitly enter into the analysis (for example, the vast literature on missing or imperfect markets), most often these institutions are treated as exogenous, and there is little effort to understand why these institutions vary across countries. Usually market imperfection of markets are tied to some fundamental “technological” problem, such as an informational asymmetry, rather than institutional differences across societies. Perhaps more surprisingly, much of political economy also treats institutions as exogenous. There are many theoretical and empirical works on the effect of different electoral systems or governance structures on economic performance, but they typically start from an exogenously- given set of institutional rules, such as a given set of electoral rules, a dictatorship, a particular form of democracy etc.. This course is explicitly about understanding both the effects of institutions on economic outcomes and the emergence and persistence of institutions in equilibrium.

1.1.2. What are institutions? There is no correct answer to this question, and the appropriate answer most probably depends primarily on the use that we want to put the notion of institutions to. Douglass North, for example, emphasizes the role of institutions as “to reduce uncertainty by establishing a stable (but not necessarily efficient) structure to human interaction.” This is useful but somewhat too loose a definition. An alternative, but related, definition is that institutions refer to “the rules of the game” or more specifically to the extensive form of the exact game that the agents are playing (e.g., as suggested by David Kreps). This definition is also useful, but may be too encompassing. At least in the context of the models we look at in this course, what the institutions are will be clear. Social scientists are still struggling for the most useful general definition. Probably the first question we should address is whether, and how well, institutions adapt to the economic requirements of the society. For example, the classical Marxist approach views the “superstructure,” which roughly corresponds to the notion of institutions here,

as simply determined by the underlying economic forces and not having large independent effects. More pertinent to the discussion here, following Acemoglu (2003a) and Acemoglu, John- son and Robinson (2005), we may want to distinguish between several different views of institutions. In coming up with this list we abstract from other interesting approaches which are beyond the scope of this course, for instance the evolutionary models of Nelson and Winter (1982) and Young (1998).

(1) Efficient institutions view: As already hinted above, according to this view, institu- tions may matter for economic outcomes, but societies will choose the institutions that maximize their total surplus. How this surplus will be distributed among dif- ferent groups or agents does not affect the choice of institutions. An example of this view would be Demsetz’s theory of property rights, which is also very similar to the theory of property rights that is advanced in North and Thomas. According to Demsetz (1967), enforcing property rights has some costs, but also is beneficial eco- nomically. When the benefits exceed the costs, property rights will be enforced. As an example, he offers variations in property rights in land among American Indians. In many Indian societies, there were no property rights in land because land was abundant, and the inefficiencies from overhunting were relatively small. However, property rights developed following the commercialization of the fur trade, because with the possibility of selling fur, the overhunting problem (‘tragedy of the com- mons’) became more serious, so the benefits from establishing property in over land increased. This caused the emergence of property rights in land among the tribes engaged in fur trading. The underlying reasoning of this view again comes from the Coase Theorem. Therefore, even in a world where institutions matter, when different economic parties can negotiate costlessly, they will be able to bargain to internalize potential externalities. The farmer, who suffers from the pollution created by the nearby factory, can pay the factory owner to reduce pollution. The same reasoning can be applied to political situations. If the current laws or institutions benefit a certain group while creating a disproportionate cost for another, these two groups can negotiate to change the institutions. By doing so they will increase the size of the total surplus (“the pie” that they have to divide be- tween themselves), and they can then bargain over the distribution of this additional surplus.

Therefore, equilibrium institutions will not be those that maximize the size of the overall pie, but the slice of the pie taken by the powerful groups. Why doesn’t a Coase theorem type reasoning apply? This is what we will discuss in detail in many different parts of this course. But for now, it is useful to note that one possible and obvious reason for why the Coase theorem may not apply in politics is commitment problems. If a ruler has political power concentrated in his hands, he cannot commit not to expropriate assets or revenues in the future. The enforcement of property rights, which would encourage investment by the agents, requires that he credibly relinquishes political power to some extent. But according to the Coasian bargain, he has to be compensated for what he could have received using this power. Herein lies the problem. When he relinquishes his power, then he has no guarantees that he will receive the promised payments in the future. Therefore, by their very nature, institutions that regulate political and social power create commitment problems, and prevent Coasian bargains that are necessary to reach efficient outcomes. (3) The ideology/beliefs view: According to this view societies may have different insti- tutions because people have different beliefs about what is best for society. Some societies get it right and some get it wrong and those that get it right ex post are prosperous. It is clear that belief differences and ideology do matter in practice, the real question is whether this can systematically explain the massive variation in outcomes we observe in the world. It is also possible of course that people care about the organization of society for non-economic reasons and are prepared to sacrifice output in order to have a set of institutions that they feel are better. (4) The incidental institutions view: While the efficient institutions view is explicitly based on economic reasoning, a different approach, which downplays choices and emphasizes the development of institutions as a by-product of other social interac- tions, is more popular among many political scientists and sociologists. According to this view, the set of political and economic institutions emerge not as a choice of economic actors, but is an incidental consequence of other actions. An interesting example of this is the work by Tilly (1990). Building on the Weberian tradition, Tilly proposed a theory of the formation of modern states, which argues that mod- ern state institutions such as fiscal systems, bureaucracy and parliaments are closely related to the need to raise resources to fight wars and thus arose in places with incessant inter-state competition. Building on this work Herbst (2000) argued that

the absence of such processes led to a very different pattern of state formation his- torically in Africa, something which left African countries with states which were incapable of providing order or public goods. Another example would be the lit- erature on legal origins where the fact that Latin American countries have French legal origin stems from the coincidence that they were colonized by the Spanish who happened to have a legal system which was more compatible with the Civil Code than the Common law. What distinguishes the efficient institutions view from the other three is that according to the efficient institutions view, there should not be meaningful institutional differences translating into different economic outcomes– institutional differences should simply reflect differences in economic environments, rather than cause such differences. Therefore, as noted above, empirical evidence that shows that “exogenous” institutional differences matter for differences in economic outcomes will support one of the other three views. We will discuss this type of empirical evidence below. What distinguishes the first two approaches from the incidental institutions view is that according to the incidental institutions view, we cannot try to understand institutional differ- ences as emerging from different economic calculations. As a result, we cannot ask questions of the following form: “why aren’t the existing set of institutions being replaced by a new set of institutions that are more beneficial for the whole society or for certain groups?” These types of counterfactual questions are a very attractive feature of the economic approach. The first two approaches are therefore more in line with economic research in general, and will be in the starting point of the approach in this class. But there are important differences between these two views as well. In the social conflict view, conflict between social groups is an essential element of institutions and differences in the nature of this conflict will lead to different sets of institutions. In contrast, in the efficient institutions view, conflict between different groups or agents is not important, and institutional differences will mostly emerge from differences in the economic environment or the costs of designing institutions. The approach in this course will be very much based on the social conflict view of institu- tions. While the other views may also contribute to differences in institutions we observe in practice, the perspective in this course will be that the bulk of the differences in institutions emerges as an equilibrium outcome from well-specified games, in which there is conflict of interest between individuals or social groups. This is a natural perspective for an economic approach to institutions, since it emphasizes both economic interests and equilibrium. It also

easier to change than economic institutions. Again this distinction is not very tight. We refer to labor market policies, such as the extent of firing costs or whether or not trade unions have a closed shop agreement as ‘labor market institutions.’ Is it important to distinguish between policy and institutions? Consider the following example: currently in Zimbabwe the government of President Robert Mugabe is expropriating the property of the white farmers. In the 1960s and 1970s a succession of governments in Ghana used the monopoly purchasing power of the Cocoa Marketing Board to tax farmers at punitively high rates (Bates, 1981). Though governments in Ghana did not try to take the land of the cocoa farmers, they made the value of their assets almost zero. What is the difference? It is conventional to subdivide institutions into:

• Formal institutions, for example, whether the country in question has a Supreme Court, separation of power, parliamentary system etc.
• Informal institutions, which determine how a given set of formal rules and informal institutions function in practice. For example, many Latin American countries have a presidential system similar to the U.S., but in practice, they have very different “political institutions”. [...Currently, we have very little understanding of how informal institutions work, and this might be an interesting area for future research...] One terminology that we have used (Acemoglu, Robinson and Verdier, 2003) is to say that societies where behavior is strongly conditioned by the formal institutions are ‘strongly institutionalized’ while those where it is not are ‘weakly institutionalized.’ An interesting example of the difference comes from the history of the Supreme Court in the US and Argentina. Starting in 1935 the Supreme Court began to rule against President Rooselvelt’s New Deal Policies including ruling that the National Industrial Recovery Act was unconstitutional. In response after his landslide 1936 re-election, Roosevelt proposed judicial ‘reform’ which in effect would give him the right to nominate 6 new judges (there were 9 at the time) and ensure a majority. Despite huge majorities in Congress and Senate and a clear mandate, this created a massive outcry in the press and opposition in the Senate. In response the Supreme Court made some compromises, ruling that the Social Security Act and the National Labor Relations Act, two key pieces of New Deal legislation, were constitutional. Moreover, one of the most conservative judges resigned, allowing FDR to appoint a democratic. This made it impossible for FDR to get the change in the rules through the Senate and he had to drop it. This is an interesting example of how the formal rules which constitute the separation of powers and checks and balances have to be supported by informal norms.

Contrast this to the experience of Argentina under Perón. When Perón (who had been secretary of labor in the previous military Junta which had ruled since 1943) was first demo- cratically elected president in 1946 the Supreme Court had recently ruled unconstitutional an attempt to create a new national labor relations board. The labor movement was a key pillar of the Peronist movement and he sought the impeachment of 4 or the 5 members of the Court. In the end 3 were removed and the Chamber of Deputies and the Senate supported this. Following this the Court did not provide any checks on Perón’s policies. There then followed a sequence of transitions between civilian and military governments in Argentina (see Acemoglu and Robinson, 2006, Chapter 1). The 1946 impeachment established a new norm so that whenever a political transition too place, the incoming regime either replaced the entire existing Supreme Court or impeached most of its members (Helmke, 2005, Chapter 4). The attached figure from Iaryczower, Spiller and Tommasi (2002) dramatically shows the declining average tenure of Supreme Court judges in Argentina over this period. Fascinatingly in 1990 when the first transition between democratically elected governments occurred, the incoming Peronist President, Menem, complained that the existing Supreme Court, which had be appointed after the transition to democracy in 1983 by the Radical President Alfonsín, would not support him. He then proposed an expansion of the Court from 5 to 9 members which was duly passed and which allowed him to name 4 new judges. We will sometimes refer to the cluster of institutions, consisting of economic institutions, political power and formal and informal political institutions, simply as “institutions”. It may also be tempting to follow political scientists and sociologists and classify institu- tions into two groups (with the implicit understanding that most real world institutions will fall somewhere in between):

• Predatory (“bad”) institutions: as institutions that do not encourage investment and economic development.
• Developmental (“good”) institutions: institutions that permit or encourage invest- ment and growth.

Even though, this may be a useful device for discussion or even more formal thinking, there is a potential problem. Certain arrangements that are good for economic development may later become bad for in investment and development. We will see formal models illustrating this issue later in the course. For now, when convenient, we will sometimes talk of bad and good institutions.