Taking Stock
This chapter made a first attempt at investigating why institutions and policies differ across societies. Even though this question may be viewed as part of the study of political economy rather than economic growth per se, in the absence of satisfactory answers to this question, our understanding of the process of economic growth will be limited.
The evidence provided in Chapter 4 suggested that societies often choose different institutions and policies, with very different implications for economic growth. Thus to understand why some countries 1007are poor and some are rich, we need to understand why some countries choose growthenhancing policies while others choose policies that block economic development.
This chapter emphasized a number of key themes in developing answers to these questions. First, the sources of different institutions (and non-growth-enhancing institutions) must be sought in social conflict among different individuals and groups in the society. Social conflict implies that there is no guarantee for the society to adopt economic institutions and policies that will encourage economic growth. Such social arrangements will benefit many individuals in the society, but they will also create losers, groups whose rents will be destroyed or eroded by the introduction of new technologies or by the process of economic growth. When individuals in the society have conflicting preferences over institutions and policies, the distribution of political power in the society plays an important role in determining which institutions and policies will be chosen (and whether non-growth-enhancing institutions will be reformed).
I emphasized that non-growth-enhancing policies can emerge even without any significant Pareto inefficiencies. In particular, a range of political economy models lead to equilibrium allocations that would also result from the maximization of a weighted social welfare function.
The resulting equilibrium allocation will naturally be constrained Pareto efficient. However, Pareto efficiency does not guarantee high output or growth. I illustrated this first by focusing on a “simple society,” where individuals belong to a social group, the conflict of interest is among social groups, and all political power rests in the hands of a political elite. I showed that this environment, combined with linear preferences, implies that even the more restrictive Markov Perfect Equilibrium (MPE) concept leads to constrained Pareto efficient allocations. Despite the Pareto efficiency of the equilibrium allocations, we also saw that there are two distinct sets of reasons for distortionary policies, which will discourage investment and economic growth (suggesting as a byproduct that Pareto efficiency may not be the right concept to focus in the analysis of the political economy of growth). The first is revenue extraction, while the second results from competition between the elite and other social groups in the marketplace or in the political arena, and can take the form of factor price manipulation effects or political replacement effects. Revenue extraction involves the use of distortionary taxes by the elite (because those are the only fiscal instruments available to them) in order to extract revenues from entrepreneurs and workers in the society. Factor price manipulation results when the elite use policies in order to reduce the labor demand (or more generally the factor demand) coming from other social groups, so that they face more favorable factor prices. Finally, the political replacement effect emerges when the elite tries to impoverish social groups that might politically compete with themselves. The analysis demonstrated that revenue extraction, though distortionary, is typically much less harmful to economic growth than factor price manipulation and political replacement effects, because ultimately the elite can only raise revenues if the groups that are being taxed have the incentives (and the ability) to undertake investments and produce. In contrast, the factor price manipulation and the political replacement effects encourage the elite to pursue policies that harm groups that they perceive as their competitors. This typically leads to higher taxes and also to explicit actions to block technology adoption or other productivity-enhancing investments by competing entrepreneurs. The consequences of these types of policies for economic growth can be disastrous. Remarkably, however, all of this can happen while the equilibrium is still constrained Pareto efficient—it is so, because all weight is given to the elite without any regard to the welfare of the other individuals and groups in the society.In addition to providing a simple and useful framework for the analysis of policy, the framework with political power vested in the elite also leads to a range of comparative static results that shed light on what types of societies will adopt policies that encourage growth and which societies are likely to pursue non-growth-enhancing policies or even try to block economic development. The following are some of the main comparative static results: (1) taxes are likely to be higher when the demand for capital by entrepreneurs is inelastic, because in this case the revenue-maximizing tax rate for the elite is higher; (2) taxes are likely to be higher when the factor price manipulation effect is more important relative to the revenue extraction effect; (3) taxes are higher when the political power of the elite is contested and reducing the income level of the competing groups will lead to political consolidation for the elite; (4) in the absence of the political replacement effect, greater state capacity leads to lower taxes; (5) when the political replacement effect is important, both greater state capacity and greater rents from natural resources lead to more distortionary policies because they increase the political stakes, i.e., the value of holding on to political power.
Pareto efficiency is not a general property of political economy models, however.
In particular, once the timing of policies is changed so that taxes by the politically powerful are set after entrepreneurial investments or when long-term investments are important, serious holdup problems emerge. At the simple level, the holdup problem corresponds to a situation in which the politically powerful can not commit to not taxing investments once they have been undertaken. Anticipating these high taxes, entrepreneurs refrain from investment. In such an environment, MPE leads to constrained Pareto inefficient equilibria. Subgame Perfect Equilibria (SPE) then sometimes significantly improves over the MPE. Whether the MPE or the SPE is the appropriate equilibrium concept depends on institutional and historical details, which affect whether individuals and groups can coordinate their actions in equilibrium. In any case, we also saw that even with SPE the equilibrium might involve a lack of commitment to a desirable tax sequence by the elite. In this case, if feasible, the elite may introduce economic institutions providing greater security of property rights to entrepreneurs so as to encourage investments (and thus increase tax revenues). Thus the possibility of commitment problems provides us with one perspective for thinking about the emergence of secure property rights.To move beyond models in which political power rests with a pre-specified group, here the political elite, we need systematic ways of aggregating heterogeneous political preferences. After reviewing some basic political economy theory, I used the well-known Median Voter Theorem (MVT), which applies in certain economic environments, to investigate the determination of equilibrium policies in an economy with heterogeneous entrepreneurs. One of the most interesting results of this analysis is that the revenue extraction mechanism emphasized in the context of elite-dominated politics is also present in more complex societies with heterogeneity among entrepreneurs. In particular, if the median voter is poorer than the average individual (entrepreneur) in the society, he may want to use distortionary policies to transfer resources to himself.
This type of distortionary revenue extraction by the median voter is qualitatively similar to the use of distortionary policies by the elite to extract revenues from middle-class entrepreneurs. Nevertheless, this effect exhibits itself in a more general environment with heterogeneity among the entrepreneurs and also leads to a new comparative static result; when the gap between the mean and the median of the productivity distribution is greater, the incentives to extract revenues are stronger and policies are more likely to be distortionary.Finally, I emphasized that taxation is not the only relevant policy affecting economic growth. The provision of public goods, in the form of securing law and order, investments in infrastructure or even appropriate subsidies, might also be important for inducing a high rate of economic growth. Will the state provide the appropriate amount and type of public goods? In the context of a political economy model, the answer to this question depends on whether the politically powerful groups controlling the state have the incentives to do so. As already discussed above, the elite may want to block economic development in order to affect the factor prices that it faces or to secure their political position. Beyond this, the elite would only invest in public goods if they expect to reap the benefits of these investments in the future. This raises the issue of weak versus strong states. While an emphasis on taxes suggests that checks on the economic or political power of the state should be conducive to more growth-enhancing policies, weak states will be unwilling to invest in public goods because those controlling the state realize that they will not be able to tax future revenues created by these public good investments. Consequently, an intermediate strength of the state might be most conducive to growth-enhancing policies. The more important point here is that an analysis of the effect of economic institutions and policies on growth should take into account both the incentives to private agents and also the incentives to the government for providing the appropriate amount and type of public goods.
The material in this chapter is no more than an introduction to the exciting and important field of political economy of growth. Many issues have not been addressed. Among those the following appear most important: first, in addition to taxes, expropriation and public goods, whether the society provides a level playing field to a broad cross-section of society is important. For example, broad-based human capital investments, which may be quite important for modern economic growth, require the provision of appropriate incentives not only to a few businesses, but to the entire population. Similarly, security of property rights for existing businesses have to be balanced against the ease of entry for new firms. Second, the entire analysis here took the distribution of political power in the society, and the political institutions that generated this distribution of political power, as given. It is clear, however, that different distributions of power in the society will lead to different policies and thus to different growth trajectories. Consequently, it seems important to understand how the distribution of political power and equilibrium political institutions might evolve endogenously and whether it might interact with the economic equilibrium. Some of these issues will be discussed in the next chapter.
22.10.