Rulers, Religion, and Riches: Why the West Got Rich and the Middle East Did Not (Cambridge Studies in Economics, Choice, and Society)

by Jared Rubin

  The “Game” Played between Rulers and Their Propagating Agents

  The first task of any game theoretic framework is to identify the relevant players.15 In the present case, those players are a ruler and the set of potential propagating agents. This includes propagating agents that the ruler does not actually end up choosing; they may still play an important role in the game because they represent the outside option available for the ruler. A final set of players that have received little attention so far are the “citizenry,” or the “nonelite.” For the most part, they are in the background of the framework, since most policies result from the bargain between the elite. However, the citizenry place an important constraint on the actions of all parties; a ruler cannot stay in power if he does not have the support of the citizens, and a propagating agent loses its ability to support political power if it holds no influence over the citizens.

  The next step is to consider the motivations of the players. Building on the previous analysis, the framework assumes that the ruler’s primary goal is to stay in power, which he can achieve through some combination of legitimacy and coercion. The motivations of propagating agents depend on their identity. The military elite generally desires policies such as greater military spending, campaigns of conquest, or increased capacity to collect taxes. Economic elites generally support policies that increase their own wealth. Religious authorities often desire tax exemption, policies consistent with religious precepts, and suppression of rival religions.16

  The next task is to determine the nature of the relationship between the players. All players must get something out of interacting with each other, otherwise they would not do so. The benefit to the ruler of playing the game is clear: propagating agents bolster his claim to rule, which in turn helps keep him in power. The benefit to the agents of playing the game is whatever they receive in return for their support. For example, a ruler may enforce a religious law or subjugate a rival religious movement in order to maintain good standing with the religious establishment. Numerous pre-Reformation European kings did just this, assisting the Church in suppressing heretical movements. Or a ruler may provide public goods to satisfy the economic elite. Early Islamic leaders commonly endowed waqfs (pious trusts) that provided schools, hospitals, mosques, and other public goods.

  The enacted policies ultimately result from the ruler’s choice of propagating agents and the bargaining strength of those agents. In other words, the keys to understanding which policies result from the bargain are found in the answers to the following questions: How much do rulers need the agents to propagate their rule? Are the agents effective at propagating rule at relatively low cost? Are there good alternative sources of propagation for the ruler? What determines the magnitude of the costs and benefits of different laws and policies? This is where the society’s institutions enter the game. By providing the “rules of the game” institutions dictate the costs and benefits that a ruler faces of having rule propagated in different manners and thus the relative costs and benefits of enacting different policies.

  Propagating agents are effective when they are able to strengthen the ruler’s likelihood of staying in power. They are able to do so only when the rules of the game permit it. In some societies, religious authorities are highly effective at legitimizing rule. They are only effective, however, when they are influential with the population and when there is historical or theological content that permits them to effectively legitimize rule.17 When such doctrine exists, one of the rules of the game is that religious leaders have the capacity to propagate rule. And indeed, such doctrine exists in both Christianity and Islam. Equally important, the costs associated with religious legitimation are generally trivial relative to the costs of other types of propagation. As Anthony Gill (1998, p. 51) notes, “Ideology … is a relatively cost-effective form of control since people obey out of the belief that what the government does is right. By creating a system of values and norms, a strong ideology regulates citizens’ behavior by providing an internal guide to acceptable and unacceptable activity.”

  The costs and benefits of using specific propagating agents therefore depend on the society’s institutions. For instance, Eliana Balla and Noel Johnson (2009) show that both Ottoman and French rulers provided elites with tax farms to collect revenue. But due to differences in incorporation laws, French tax farmers were able to pool their capital and thus coordinate to constrain the king if he chose fiscal policies undesirable to them. There was no mechanism for Ottoman tax farmers to do this, and they were never able to credibly constrain the sultan. In other words, as French tax farmers became more coordinated, it became more costly for kings to use them to collect taxes, since they had more bargaining power and could thus extract more from him. For the Ottomans, using tax farmers was a rather inexpensive manner of collecting taxes, as the sultan had to give relatively little to tax farmers in return for the right to collect taxes.18

  In sum, the cost to the ruler of receiving propagation from an agent is how much the ruler has to give up in the bargain over propagation. These costs can take many forms, depending on whether the propagating agent(s) are members of the economic, military, or religious elite, or some combination of the three. The magnitude of these costs is dependent on the society’s institutions.

  Taking these institutionalized rules as given for the moment, let’s turn to the equilibrium actions of all the players. To solve for the equilibrium, consider what each player is thinking when they come to the bargaining table. First off, the ruler wants to stay in power. He knows that to do so he needs propagation from his agents. He also knows there are a host of policies that otherwise weaken his hold on power but are consistent with the interests of the agents. Meanwhile, the agents have some objective they wish to obtain in the bargain. The agents can withhold or provide less propagation as a bargaining tactic. If these tactics are credible, the agent can force the ruler’s hand.

  The parties’ relative bargaining power determines the equilibrium outcome of the bargain. The ruler considers how much propagation he needs to stay in power and rule effectively, how much propagation each agent can provide him, and the cost of receiving propagation from each of the agents. He then weighs these costs and benefits and chooses some combination of agents to propagate his rule in the most effective way possible.

  There are two possible outcomes of this bargain. First, if the chosen propagating agent(s) is highly effective or inexpensive, it is in a good bargaining position. In this case, the agent can threaten to withhold legitimacy, leaving the ruler without an important source of propagation. If this threat is credible, the ruler will make significant policy concessions to the agent even if those policies otherwise reduce his likelihood of staying in power.

  The game becomes more complex if the agents’ desires do not align with those of the citizens. In this case, the ruler must choose between supporting his propagating agents and enacting policies that benefit the citizenry. If the ruler ends up supporting a policy favored by the agents, any citizen who violates the policy faces a “double penalty,” since they are subject to sanctions from both the ruler and the agent. For instance, anyone attempting to break Saudi laws that subjugate women may face both spiritual sanctions from the religious establishment as well as jail time or fines. This creates an extra disincentive for citizens to push for changes to such a rule in the future. A vicious cycle is thereby created where, over time, citizens have little incentive to push for change, and the ruler and propagating agents do not need to bargain over these laws and policies because they are not of concern of the public.19 It is even possible that society forgets arguments in favor of change once they fall out of the mainstream, meaning that over time a push for change in these rules is not even a preference of the citizens.20

  The contrary logic holds when the agent(s) chosen in equilibrium is costly or ineffective. In this case, if the agent’s demands are too great, the ruler can credibly threaten to choose another form of propagation. Hence, the laws and p
olicies emanating from the bargain are more reflective of the ruler’s desires. This outcome also has dynamic consequences when the actions of the citizenry are considered. If economic conditions incentivize some citizens to push to change rules that are contrary to the agents’ desires, the ruler is more likely to enact these changes, since he does not fear losing the propagation the agent provides. Hence, the citizens are encouraged to push for even more rule changes in the future. Although they face costs imposed by the agent from doing so, they do not face penalties from the ruler. This can lead to a cycle moving in the other direction, where the ruler permits more and more as the citizens ask for more, to the point where the rules of the society are radically different than before the initial push.

  Institutional Change (and Stagnation) in the Long Run

  When laws or policies change dramatically over time, institutions may change to accommodate the new reality. In the long run, it is possible that the ruler’s cost-benefit calculation changes over time precisely because people act in their own interest. When this occurs, institutions change endogenously. Figure 2.3 represents this process in simplified form.

  Figure 2.3 Endogenous Institutional Change

  This book focuses on why institutions change – or do not change – when political rule is legitimized by religion. Religious propagation is effective when religious authorities exert influence over the population and when religious doctrine is conducive to legitimizing rule. This is even truer of the monotheistic faiths such as Islam and Christianity. Murat Iyigun (2015, ch. 2) argues, with support from an extensive data analysis, that monotheisms are good for propagating rule because they imbue both political and religious authorities with monopoly powers. But religious authorities cannot simply claim that rulers are legitimate if this is inconsistent with doctrine. Religions are somewhat unique in this way – consistency matters, and a primary reason that religious authorities are able to propagate rule in the first place is that they hold the keys to the “eternal” word of God. These “keys” are highly valued in most civilizations, historic and present, perhaps for evolutionary reasons: humans desire answers to the unknown and the unknowable, and religion offers these answers. This gives religious authorities power over the citizens – it is what makes them “elite” – but only to the extent that their answers are consistent. Inconsistent proclamations indicate that religious authorities do not have access to “eternal” laws, and such proclamations therefore undermine the ability of the authority to legitimize political rule.21 Religious authorities may attempt to rationalize big changes as being consistent with doctrine, but they are still constrained to choose from the set of rationalizations that are believable. In other words, the “rules of the game” are such that the efficacy of religious statements that legitimize rule are constrained by being consistent with doctrine. This does not mean that religious doctrine remains stagnant under all conditions; it simply indicates that the cost to religious authorities of changing their proclamations is greater than it is for other types of propagating agents.

  One upshot is that religious authorities are in a no-win situation when people flagrantly violate their dictates. Religious authorities have a full menu of policy objectives, with their views on the issue at hand being one of them. They stand to lose credibility – and hence the ability to propagate rule – if large swaths of the population openly transgress their position on one issue. For instance, the contemporary Catholic Church stands to lose members as it continues to cling to its positions on contraception, divorce, and homosexuality, although these issues represent a miniscule fraction of the Church’s overall doctrinal concerns. Hence, on this one issue, religious authorities have a short-term incentive to update their views. Yet, religious reinterpretation carries an important long-run cost: it undermines the very nature of the religious authority’s power. This is the main feature that distinguishes religious authorities from other types of propagating agents: intertemporal consistency matters.

  This religion-specific institutional feature means that the legitimizing relationship between the religious and political elite can change over time. If religious authorities reinterpret doctrine to maintain their relevance with the citizenry, they undermine their ability to legitimize rule in the long run, thus reducing the benefits of religious propagation. On the other hand, if they permit individuals to flagrantly violate their dictates without a response, they stand to lose their ability to influence actions in both the short run and the long run, as transgressing religious doctrine will become a norm. The upshot is that religious authorities with weak ability to propagate rule become even weaker over time. This process may continue to the point where rulers remove religious authorities from the set of agents who propagate their rule.

  Such long-run change does not occur, however, when religious authorities are highly effective at legitimizing political rule. In this case, it is unlikely that citizens transgress religious dictates in the first place, since the ruler will not support their transgressions. The religious authority therefore never has to make the “no-win” decision because it never faces any pressure to reinterpret doctrine. This means that the legitimizing relationship between religious authorities and the ruler reinforces itself over time.

  One consequence of this logic is that what are initially small institutional differences between societies can blossom into very large differences over time. Once a society begins moving toward an equilibrium where open violation of religious dictates is the norm, a cycle begins where these actions feed into institutional change, exacerbating the degree to which religious dictates are transgressed, which itself feeds into even more institutional change. Such a cycle is unlikely to occur when religious authorities are highly effective in legitimizing rule, since there is no impetus to initiate the cycle. An institutional divergence can occur even if the economic benefits from institutional change are the same in both societies: the mechanism described above rests on the narrow self-interest of the ruler and his agents, not the efficiency of their bargaining outcome.

  These insights are related to the second mechanism through which institutions change in the long run: path dependence. A path-dependent series of events is one in which the initial impetus that triggers a series of events is far, far removed from the eventual outcome, and the initial raison d’être that sparked movement down a certain pathway has no direct influence on long-run behavior. This occurs when each step along the path depends on what happened at the last step. The first chapter described these steps as “forks” along a society’s economic or institutional path. Once the society chooses one side of the fork, a new path emerges, far removed from the reasons the society chose the original path in the first place.22

  In the context of economic history, path dependence entails that institutions can originate in situations in which humans found a solution to a problem, but they can also persist long after that problem is relevant. In such a situation, institutions still provide the rules of the game, even though important parts of the game were unforeseen at the time of the institution’s creation. The historical setting on which this book focuses is case in point: rulers in the medieval Middle East and Western Europe legitimized their rule in part via religion because the rules of the game indicated that this was the most efficient solution to their problem of propagating rule given the conditions that they faced. Once these rules of the game were established, the desires of religious agents affected the laws and policies that arose in response to unanticipated and unforeseen events. An important historical example of such a process is the subject of Chapter 5, which analyzes why the Ottoman Empire failed to adopt Gutenberg’s movable type printing press. In short, the printing press threatened the sultan’s ability to propagate his rule, since it threatened the status of the religious establishment. The invention of the printing press was unforeseeable at the time that Ottoman propagating arrangements emerged. The suppression of printing in the Ottoman Empire was therefore a result of a path-dependent series of events and is thus
only explainable by digging into history and searching for the events that caused these events to arise in the first place.

  Testable Predictions

  Any good economic framework should provide falsifiable, testable predictions. Otherwise, there is no way to know whether it is correct or if it tells us anything about the real world. To this end, this chapter concludes by summarizing the framework’s major implications as they relate to the main question posed in this book: Why did Western European economies catch up with and eventually far surpass those of the Middle East?

  The first, and most straightforward, testable prediction relates to the outcome of the short-run bargain between rulers and their agents. If the religious establishment legitimizes the ruler, we should expect to see laws favoring religious authorities. The same is true if the economic or military elite propagate the ruler. In other words, the following prediction should hold:

  Testable Prediction #1: The laws and policies of a society should in part reflect the desires of the ruler’s propagating agents.

  Testable Prediction #1 implies that the motivations of propagating agents matter. If they primarily desire laws and policies that benefit themselves at the expense of the rest of the population, economic success is unlikely to result: even though their desires are harmful, they still get their voice heard because they play a valuable role in propagating rule. In an ideal world, the motivations of the propagating agent are consistent with economic success, meaning that agents benefit from the provision of public goods, secure property rights, impartial law and order, investment in public education, protection for long-distance traders, and so on. This ideal world has never come to fruition, but some societies have come closer than others to realizing it. These societies are ones where the economic elite have a nontrivial seat at the bargaining table. Unlike religious or military elites, economic elites have interests that align with the types of laws and policies listed above. Testable Prediction #2 summarizes this insight.