Why do institutions differ?

We saw that economic institutions matter, indeed are central in determining relative prosperity. In terms of the different fundamental theories that we discussed, there is overwhelming support for the emphasis of North and Thomas on institutions, as op­posed to alternative candidate explanations which emphasize geography or culture.

In the Introduction (Section 1.2), we began to develop such a theory based on social conflict over economic institutions. We have now substantiated the first point we made there, that economic institutions determine prosperity. We must now move to substan­tiate our second point, that economic institutions must be treated as endogenous and what which economic institutions emerge depends on the distribution of political power in society. This is a key step towards our theory of economic institutions. In the process of substantiating this point however it is useful to step back and discuss other alter­native approaches to developing a theory of economic institutions. Broadly speaking, there are four main approaches to the question of why institutions differ across coun­tries, one of which coincides with the approach we are proposing, the social conflict view.

5.1. The efficient institutions view - the Political Coase Theorem

According to this view, societies will choose the economic institutions that are socially efficient. How this surplus will be distributed among different groups or agents does not affect the choice of economic institutions. We stress here that the concept of efficiency is stronger than simply Pareto Optimality; it is associated with surplus, wealth or output maximization.

The underlying reasoning of this view comes from the Coase Theorem. Coase (1960) argued that when different economic parties could negotiate costlessly, they will be able to bargain to internalize potential externalities. A farmer, who suffers from the pollution created by a nearby factory, can pay the factory owner to reduce pollution. Similarly, if the current economic 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 that they can divide between themselves, and they can then bargain over the distribution of this additional surplus.

Many different versions of the efficient economic institutions view have been pro­posed. Indeed, assuming that existing economic institutions are efficient is a standard methodological approach of economists, i.e., observing an institution, one tries to un­derstand what are the circumstances that lead it to be efficient. For instance, Demsetz (1967) argued that private property emerged from common property when land be­came sufficiently scarce and valuable that it was efficient to privatize it.

Williamson and North and Thomas do not specify how different parties will reach agreement to achieve efficient economic institutions, and this may be problematical in the sense that many economic institutions relevant for development are collective choices not individual bargains. There may therefore be free riding problems inher­ent in the creation of efficient economic institutions. Nevertheless, the underlying idea, articulated by Becker (1958) and Wittman (1989), is that, at least in democracies, com­petition among pressure groups and political parties will lead to efficient policies and collective choices. In their view, an inefficient economic institution cannot be stable be­cause a political entrepreneur has an incentive to propose a better economic institution and with the extra surplus generated will be able to make himself more attractive to vot­ers. The efficient institutions view regards the structure of political institutions or power as irrelevant.

The notion that a Coasian logic applies in political life as well as in economics is re­ferred to by Acemoglu (2003a) as the Political Coase Theorem. Although the intuition that individuals and groups will strive towards efficient economic outcomes is appeal­ing, there are both theoretical and empirical limits to the Political Coase Theorem. First, as argued by Acemoglu (2003a) and further discussed below, in politics there is an in­herent commitment problem, often making the Political Coase Theorem inapplicable.

Second, the Political Coase Theorem does not take us very far in understanding the effect of economic (or indeed political) institutions on economic outcomes - in this view, economic institutions are chosen efficiently, and all societies have the best possi­ble economic institutions given their needs and underlying structures; hence, with the Political Coase Theorem, economic institutions cannot be the fundamental cause of in­come differences. However, the empirical results we discussed above suggest a major role for such institutional differences.

The only way to understand these patterns is to think of economic institutions vary­ing for reasons other than the underlying needs of societies. In fact, the instrumental variables and natural experiment strategies we exploited above make use precisely of a source of variation unrelated to the underlying needs of societies. For example, South and North Korea did not adopt very different economic systems because they had dif­ferent needs, but because different systems were imposed on them for other exogenous reasons. In sum, we need a framework for understanding why certain societies consis­tently end up with economic institutions that are not, from a social point of view, in their best interests. We need a framework other than the Political Coase Theorem.

5.2. Theideologyview

A second view is that economic institutions differ across countries because of ide­ological differences - because of the similarity between this and the previous view, Acemoglu (2003a) calls this the Modified Political Coase Theorem. According to this view, societies may choose different economic institutions, with very different impli­cations, because they - or their leaders - disagree about what would be good for the society. According to this approach, there is sufficient uncertainty about the right eco­nomic institutions that well-meaning political actors differ about what’s good for their own people. Societies where the leaders or the electorate turn out to be right ex post are those that prosper. The important point is that, just as with the efficient institutions view, there are strong forces preventing the implementation of policies that are known to be bad for the society at large.

Several theoretical models have developed related ideas. For example, Piketty (1995) examined a model where different people have different beliefs about how much effort is rewarded in society. If effort is not rewarded then taxation generates few distortions and agents with such beliefs prefer a high tax rate. On the other hand if one believes that effort is rewarded then low taxes are preferable. Piketty showed that dispersion of beliefs could create dispersion of preferences over tax rates, even if all agents had the same objective. Moreover, incorrect beliefs could be self-fulfilling and persist over time because different beliefs tend to generate information consistent with those beliefs. Romer (2003) also presents a model where voters have different beliefs and showed that if mistakes are correlated, then society can choose a socially inefficient outcome. These models show that if different societies have different beliefs about what is socially efficient they can rationally choose different economic institutions.

Belief differences clearly do play a role in shaping policies and institutions.

Nevertheless, the scope of a theory of institutional divergence and comparative de­velopment based on ideology seems highly limited. Can we interpret the differences in institutional development across the European colonies or the divergence in the eco­nomic institutions and policies between the North and South of Korea as resulting from differences in beliefs? For example, could it be the case that while Rhee, Park, and other South Korean leaders believed in the superiority of capitalist institutions and pri­vate property rights enforcement, Kim Il Sung and Communist Party members in the North believed that communist policies would be better for the country?

In the case of South versus North Korea, this is certainly a possibility. However, even if differences in beliefs could explain the divergence in economic institutions in the im­mediate aftermath of separation, by the 1980s it was clear that the communist economic

policies in the North were not working. The continued effort of the leadership to cling to these policies, and to power, can only be explained by leaders looking after their own interests at the expense of the population at large. Most likely, North Korean leaders, the Communist Party, and bureaucratic elites are prolonging the current system, which gives them greater economic and political returns than the alternative, even though they fully understand the costs that the system imposes on the North Korean people.

Differences in colonial policies are even harder to explain on the basis of differences in ideology. British colonists established different economic institutions in very differ­ent parts of the world: in the Caribbean they set up plantation societies based on slavery, supported by highly oppressive economic institutions. In contrast, the economic insti­tutions that developed in areas where the British settled, and where there was no large population of indigenous people to be captured and put to work, and where slavery could not be profitably used, such as northeastern United States, Canada, Australia and New Zealand, were very different. Moreover, differences in the incentives of the colonists in various colonies are easy to understand: when they did not settle, they were choosing economic institutions simply to extract resources from the native population. When they settled in large numbers, economic institutions and policies emerged in order to protect them in the future and encourage investment and prosperity.

These considerations make us tend towards a view which emphasizes the actions of key economic and political agents that are taken rationally and in recognition of their consequences, not simply differences in beliefs. We do not deny that belief differences and ideology often play important roles but we do not believe that a satisfactory theory of institutional differences can be founded on differences in ideology.

5.3. The incidental institutions view

The efficient institutions view is explicitly based on economic reasoning: the social costs and benefits of different economic institutions are weighed against each other to determine which economic institutions should prevail. Efficiency arises because in­dividuals ultimately calculate according to social costs and benefits. Institutions are therefore choices. A different approach, popular among many political scientists and sociologists, but also some economists, is to downplay choices and to think of insti­tutions, both economic and political, as the by-product or unintended consequence of other social interactions or historical accidents. In other words, historical accidents at critical junctures determine institutions, and these institutions persist for a long time, with significant consequences.

Here, we discuss two such theories. The first is the theory of political institutions developed by Moore (1966) in his Social Origins of Dictatorship and Democracy, the second is the recent emphasis in the economics literature on legal origins, for example as in the work of Shleiferandhis co-authors [LaPortaetal. (1998,1999), Djankovetal. (2002, 2003), Glaeser and Shleifer (2002)].

Moore attempted to explain the different paths of political development in Britain, Germany and Russia. In particular, he investigated why Britain evolved into a democ­racy, while Germany succumbed to fascism and Russia had a communist revolution. Moore stressed the extent of commercialization of agriculture and resulting labor re­lations in the countryside, the strength of the ‘bourgeoisie’, and the nature of class coalitions. In his theory, democracy emerged when there was a strong, politically as­sertive, commercial middle class, and when agriculture had commercialized so that there were no feudal labor relations in the countryside. Fascism arose when the mid­dle classes were weak and entered into a political coalition with landowners. Finally, a communist revolution resulted when the middle classes were non-existent, agriculture was not commercialized, and rural labor was repressed through feudal regulations. In Moore’s theory, therefore, class coalitions and the way agriculture is organized deter­mine which political institutions will emerge. However, the organization of agriculture is not chosen with an eye to its effects on political institutions, so these institutions are an unintended consequence. Although Moore was not explicitly concerned with eco­nomic development, it is a direct implication of his analysis that societies may end up with institutions that do not maximize income or growth, for example, when they take the path to communist revolution.

Beginning with the work on shareholder rights [La Porta et al. (1998)], continuing to the efficiency of government [La Porta et al. (1999)] and more recently the efficiency of the legal system [Djankov et al. (2003)], Shleifer and his co-authors have argued that a central source of variation in many critical economic institutions is the origin of the legal system. For example, “Civil laws give investors weaker legal rights than common laws do, independent of the level of per-capita income. Common-law countries give both shareholders and creditors - relatively speaking - the strongest, and French-civil-law countries the weakest, protection” [LaPortaetal. (1998, p. 1116)].

These differences have important implications for resource allocation. For example, when shareholders have poor protection of their rights, ownership of shares tends to be more highly concentrated. Djankov et al. (2003) collected a cross-national dataset on how different countries legal systems dealt with the issue of evicting a tenant for non­payment of rent and collecting on a bounced check. They used these data to construct an index of procedural formalism of dispute resolution for each country and showed that such formalism was systematically greater in civil than in common law countries, and is associated with higher expected duration of judicial proceedings, less consistency, less honesty, less fairness in judicial decisions, and more corruption. Legal origins therefore seems to matter for important institutional outcomes.

Where do legal origins come from? The main argument is that they are historical accidents, mostly related to the incidence of European colonialism. For example, Latin American countries adopted the Napoleonic codes in the nineteenth century because these were more compatible with their Spanish legal heritage. Importantly, the fact that Latin American countries therefore have ‘French legal origin’ is due to a historical ac­cident and can be treated as exogenous with respect to current institutional outcomes. What about the difference between common law and civil law? Glaeser and Shleifer (2002) argue that the divergence between these systems stems from the medieval period and reflects the balance of power between the lords and the king in England and France. Once these systems established, they persisted long after the initial rationale vanished.

Although we believe that historical accidents and persistence are important, in reality the aspect of choice over institutions seems too important to be denied. Even if insti­tutions have a tendency to persist, their persistence is still a choice, in the sense that if the agents decided to change institutions, change would be possible. There are im­portant examples from history of countries radically changing their legal systems such as in Japan after the Meiji restoration, Russia after the Crimean War, and Turkey un­der Mustafa Kemal in the 1920’s. Another example might be central planning of the economy. Though many countries adopted this way of organizing the economy some abandoned it while others, such as North Korea and Cuba, still maintain it. The point here is that though institutions may in some circumstances be the incidental outcome of history, at some point people will start to ask why society has the institutions that it does and to consider other alternatives. At this point we are back in the realm of choice.

5.4. The social conflict view

According to this view, economic (and political) institutions are not always chosen by the whole society (and not for the benefit of the whole society), but by the groups that control political power at the time (perhaps as a result of conflict with other groups). These groups will choose the economic institutions that maximize their own rents, and the economic institutions that result may not coincide with those that maximize total surplus, wealth or income. For example, economic institutions that enforce property rights by restricting state predation may not be in the interest of a ruler who wants to appropriate assets in the future. By establishing property rights, this ruler would be reducing his own future rents, so may well prefer economic institutions other than en­forced private property. Therefore, equilibrium economic institutions will not be those that maximize the size of the overall pie, but the slice of the pie taken by the powerful groups.

The first systematic development of this point of view in the economics literature is North (1981), who argued in the chapter on “A Neoclassical Theory of the State” that agents who controlled the state should be modeled as self-interested. He then argued that the set of property rights that they would choose for society would be those that maximized their payoff and because of ‘transactions costs’, these would not necessarily be the set that maximized social welfare. One problem with North’s analysis is that he does not clarify what the transactions costs creating a divergence between the interests of the state and the citizens are. Here, we will argue that commitment problems are at the root of this divergence.

The notion that elites, i.e., the politically powerful, may opt for economic institutions which increase their incomes, often at the expense of society, is of course also present in much of the Marxist and dependency theory literature. For example, Dobb (1948), Brenner (1976,1982) and Hilton (1981) saw feudalism, contrary to North and Thomas’s (1973) model, as a set of institutions designed to extract rents from the peasants at the expense of social welfare.^[264] Dependency theorists such as Williams (1944), Wallerstein (1974-1980), Rodney (1972), Frank (1978) and Cardosso and Faletto (1979) argued that the international trading system was designed to extract rents from developing countries to the benefit of developed countries.

The social conflict view includes situations where economic institutions may initially be efficient for a set of circumstances but are no longer efficient once the environment changes. For example, Acemoglu, Aghion and Zilibotti (2002) show that though certain sorts of organizations may be useful for countries a long way from the technological frontier, it may be socially efficient to change them subsequently. This may not happen however because it is not privately rational. An interesting example may be the large business enterprises (the chaebol) of SouthKorea. Inthe context of political institutions, one might then develop a similar thesis. Certain sets of institutions are efficient for very poor countries but they continue to exist even after they cease to be the efficient institutional arrangement.

In stark contrast to the efficient institutions view, political institutions play a crucial role in the social conflict view. Which economic institutions arise depends on who has political power to create or block different economic institutions. Since political insti­tutions play a central role in the allocation of such power they will be an intimate part of a social conflict theory of economic institutions.

What distinguishes the social conflict view from the ideological view is that social conflict can lead to choices of economic institutions which cause underdevelopment even when all agents have common knowledge that this is so. What distinguishes it from the incidental view is that it emphasizes that institutional choices which cause underdevelopment are conscious choices, rather than the result of some historical acci­dent. The aspect that distinguishes the social conflict view from the efficient institutions view is that it does not assume that institutions are always efficient. This is one possible outcome but it is not the only one or indeed the most likely. Why is this? Why cannot efficiency be separated from distribution? We discuss this issue in the next section.

6.