The Impact of Institutions on Long-Run Development

22.1.1. Institutions and Growth. As already emphasized in Chapter 4 “institutions” matter—at least when we look at clusters ofeconomic and political institutions over long hori­zons.

We summarize these variations across societies as “institutional differences” (or differ­ences in institutions and policies). The term is slightly unfortunate, but is one that is widely used and accepted in the literature. Institutions mean many different things in different contexts, and none of these exactly correspond to the meaning intended here. As already emphasized in Chapter 4, by institutional differences, we are referring to differences in a broad cluster of social arrangements including the security of property rights for businesses as well as for regular citizens and the ability of firms and individuals to write contracts to facilitate their transactions (contracting institutions), the entry barriers faced by new firms, the socially-imposed costs and barriers facing individual decisions in human capital accu­mulation, and incentives of politicians and individuals in providing or contributing to the provision of public goods. This definition of institutions is quite encompassing. To make theoretical and empirical progress, one typically needs a narrower definition of institutions. Towards this goal, I have already distinguished between economic institutions, which corre­spond to the security of property rights, contracting institutions, entry barriers and other economic arrangements, and political institutions, which correspond to the rules and regu­lations affecting political decision-making, including checks and balances against presidents, prime ministers or dictators as well as methods of aggregating the different opinions of indi­viduals in the society (for example, electoral laws). In terms of the notation introduced in the introduction on to this part, the effect of economic institutions is summarized by the mapping ρ (∙), while the implications of political institutions for the types of economic institutions and policies that will arise is captured by the mapping π (∙).

It is also useful to note at this point that the difference between economic institutions and policies is not always clear, so it is often the combination of economic institutions and policies that matter not simply the “institutions”. For example, we can refer to security of property rights or to the quality of contracting institutions as economic institutions but we would not typically refer to tax rates as institutions. Yet insecure property rights and 100% taxation of all income have much in common. One difference might be that institutions are more durable than policies. Motivated by this, in Section 22.4, I will make a distinction between economic institutions and policies whereby economic institutions provide a framework in which policies are set. The role played by the durability of political institutions will be further studied in the next chapter. Finally, in Section 22.8, I will discuss another potential reason why taxation and security of property rights might be different. Nevertheless, the contrast of insecure property rights and 100% taxation illustrates the large overlap between economic institutions and policies, and when the distinction between the two is unclear or unimportant, I will typically refer to “institutions and policies”.

The evidence presented in Chapter 4 suggests that institutional differences do matter for economic growth. The focus of this section is not to review this evidence but build on it and ask the next question: if economic institutions matter so much for economic growth, why do some societies chooses institutions that do not encourage growth? In fact, based on available historical evidence we can go further: why do some societies choose institutions and policies that specifically block technological and economic progress? The rest of this chapter and much of the next chapter will try to provide a framework for answering these questions. I start with an informal discussion of the main building blocks towards an answer.

The first important element of the political economy approach is social conflict. There are few (if any) economic changes that would benefit all agents in the society. Thus every change in institutions and policies will create winners and losers relative to the status quo. Take the simplest example: removing entry barriers so that a previously monopolized market becomes competitive. Economic theory tells us that this is desirable in the sense that it removes distortions and creates a “potential Pareto improvement”. In the context of growth, we often focus on the implications of changes in institutions and policies on the level of income or the rate of economic growth. In this respect as well, removing entry barriers is likely to be a beneficial reform, since the removal of monopoly power will increase the quantity transacted in the market and raise real incomes. Nevertheless, not all parties in the economy will be winners from the removal of entry barriers. While consumers will benefit because of lower prices, the monopolist, who was previously enjoying a privileged position and high profits, will be a “loser”. The effect on workers depends on the exact market structure. If the labor market is competitive, workers will also benefit, since the demand for labor will increase with the entry of new firms. But if there are labor market imperfections, so that the employees of the monopolist were previously sharing some of the rents accruing to this firm, they will also be potential losers from the reform. Thus if we start with the status quo of a monopoly and consider the reform of liberalizing markets (removing entry barriers), there will not be unanimous support for this proposal. Put differently, there will be social conflict over the policy of “market liberalization”.

The presence of social conflict over institutions and policies is not specific to reform. If instead of starting with the status quo we were deciding how markets should be organized without reference to any past arrangements, the same conflicts would be present.

Therefore, because of the different allocations that they will induce, individuals will have different, conflicting preferences over economic institutions. So if there are conflicting pref­erences over collective choices in general (and over institutions and policies in particular), how do societies make decisions? Political economy is the formal analysis of this process of collective decision-making. If there is social conflict between a monopolist that wishes to retain entry barriers and consumers that wish to dismantle them, it will be the equilib­rium of a political process that decides the outcome. This process may be “orderly” as in democracies, or disorderly or even chaotic as in other political regimes as illustrated by the all too frequent civil wars throughout human history. Whether it is a democratic or a non- democratic process that will lead to the equilibrium policy, the political power of the parties with conflicting interests will play a central role. Put simply, if two individuals disagree over a particular choice (for example, about how to divide a dollar), how powerful each is will play an important role on the ultimate choice. In the political arena, this corresponds to the political power of different individuals and groups. For example, in the monopoly example, we may expect the monopolist to have political power because it has already amassed income and wealth and may be able to lobby politicians. In a non-democratic society where the rule of law is tenuous, we can even imagine the monopolist utilizing thugs and paramilitaries to quash the opposition. On the other hand, in a democracy, consumers may have sufficient political power to overcome the interests and wishes of the monopolist through the ballot box or by forming their own lobbying groups. Whatever the outcome, political power will play an important role.

The second key element of the political economy approach is commitment problems, which will act both as a source of inefficiency and also augment the distortions created by social conflict. Political decisions at each date are made by the political process at that date (for example, by those holding political power at that point); commitment to future sequences of political and economic decisions are not possible unless they happen to be “equilibrium commitments” arising as part of the equilibrium (here, we will see that whether we use the concept of subgame perfect equilibrium or Markov perfect equilibrium will play a role in shaping the extent of available commitments).

At this point, it is important to distinguish between non-growth-enhancing policies (or distortionary policies) and Pareto inefficiency. Many political economy models will not lead to Pareto inefficiency (though some will). This is because in some reduced-form way their equilibrium can be represented as a solution to a weighted social welfare function (see Section 22.6). The resulting allocation, by virtue of maximizing this weighted social welfare function given the set of available instruments, will be a point along the constrained Pareto frontier of the economy. Nevertheless, many such allocations will involve distortionary and non-growth­enhancing policies (think, for example, of an allocation in which a dictator such as Mobutu in Zaire expropriates all the investors in the country; it is possible to change policies to increase investment and growth, but this will typically imply taking resources and power away from Mobutu). Interestingly, when commitment problems are present, and especially when we focus on Markovian equilibria, the political equilibrium will typically lead to a constrained Pareto inefficient allocation, because there will often exist future policies that can make all parties better-off, but those policies will not arise as part of the equilibrium.

Consider a situation in which political power is in the hands of a specific group or an individual—the political elite. To simplify the thought experiment, let us ignore for now any constraints on the exercise of this political power (this is essentially where we will begin in the next section). Then the elite can set policies in order to induce allocations that are most beneficial for themselves, and thus the political equilibrium can be thought of as the solution to the maximization of a social welfare function giving all the weight to the elite. Even though the resulting equilibrium will not necessarily be Pareto inefficient, it will typically involve non-growth-enhancing policies. Why and when will the exercise of political power by the elite lead to such distortionary policies?

I will argue that there are two broad reasons for why those with political power will choose distortionary policies. The first is revenue extraction, that is, the attempt by the elite to extract resources from other members of the society using a limited menu of fiscal instruments. Central to this source of distortionary policies is two aspects of the society: (1) a decoupling between political power (which is here in the hands of the the elite) and economic power (which lies with the entrepreneurs and the workers); (2) a limited set of fiscal instruments. These two aspects combined imply that the elite will use the available fiscal instruments to transfer resources from the rest of the society to themselves, which is the first potential reason for distortionary policies. We will also see that the same type of distortionary policies emerge even when there is no political elite, but decisions are made democratically (Section 22.7). The restriction to a limited set of fiscal instruments, such as linear taxes that discourage investment or work effort is important here. Had there been non-distortionary taxes, such as lump-sum taxes, the elite could extract resources from the rest of the society without discouraging economic growth. But lump-sum taxes are often not possible, and more generally, most forms of redistribution do create distortions by reducing incentives for work effort or by discouraging investment.

I will argue, however, that the second reason for the use of distortionary policies by the political elite is potentially more damaging to economic growth. The elite will also choose distortionary policies because it will often be in competition with other social groups in society. This competition may be economic. For example, the elite may also engage in production and understand that by taxing and creating distortions on other entrepreneurs, they will be able to reduce their demand for factors (for example labor) and thus increase their profits. I will refer to this as the factor price manipulation motive for distortionary policies. The competition between the elite and other social groups may also be political. The elite might foresee that enrichment by other groups will pose a threat to their political power and to their ability to use and benefit from their power in the future. When this is the case, they will use distortionary policies to impoverish their political competitors. I will refer to this as the political replacement motive for distortionary policies. The rest of the chapter will illustrate how distortionary policies can be adopted for extracting resources from different social groups and for factor price manipulation and political replacement motives. An interesting implication of the models I will present will be that factor price manipulation and political replacement motives will often lead to greater distortions and will be more damaging to the growth potential of a society than the revenue extraction motive.

Next, this basic framework also enables us to illustrate the additional inefficiencies created by commitment problems. In particular because the elite cannot commit to future policies, there will be a holdup problem, whereby investments, once undertaken, may be taxed at prohibitively high rates or expropriate. Holdup problems are likely to be important in a wide range of circumstances, for example, when the relevant investments are in long-term projects and assets, so that a range of policies will be decided after these investments are undertaken.

Much of the current chapter is devoted to understanding how the potential conflict over different economic allocations leads to different preferences over economic institutions and policies. The next two sections focus on distributional conflict in a simple society, consisting of different social groups. Throughout this chapter, I will take the distribution of political power as given and in the next few sections, political power—and thus the authority to decide policy than economic institutions—will be in the hands of that group of individuals to whom I will refer to as “the elite”. I will investigate how the desire of this group of individuals to influence the allocation of resources in their favor may lead to distortionary policies that reduce investment and output. I will also highlight how these problems can become more severe in the presence of commitment/holdup problems.

Section 22.4 starts the investigation of how inefficiencies in policies might translate into inefficient (economic) institutions. In particular, in this section I will show how the same forces leading to distortionary policies will affect two aspects of economic institutions, whether effective constitutional limits on taxation and expropriation arise endogenously, and whether there will be regulation (blocking) of new technologies. Economic institutions preventing future high taxes may emerge if holdup problems are important and the main source of distortionary policies is revenue extraction. In contrast, when factor price manipulation or political replacement motives are important, economic institutions limiting distortionary policies are unlikely to emerge. On the contrary, in this case, economic institutions that explicitly block the adoption of more efficient technologies may emerge. These results underlie the claim above that factor price manipulation and political replacement motives are typically more damaging to economic growth than the revenue extraction motive.

Throughout this chapter, I will focus on comparative statics that illustrate which types of societies are more likely to adopt growth-enhancing policies, and which others are likely to try

to block economic growth. The major comparative static exercises will look at the effects of the nature of production technology, the distribution of resources within the society, whether the politically powerful compete with economically productive agents in factor markets, the extent of holdup problems, the importance of natural resources and whether or not political power is contested. While the field of the political economy of growth is still in its infancy, it is only by developing such comparative statics that it can contribute to a systematic analysis of why some societies grow and become rich, and others stagnate.

22.2.