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

  Notes

  1 Introduction

  1 I use the term “Middle East” somewhat broadly throughout this book, comprising North Africa, the entire Arab world, Iran, Turkey, and Islamized Spain under Umayyad rule. Essentially, I use the term to represent the Islamic world west of South Asia. I apologize if this broad use of the term offends, but repeatedly using the term “Islamic world west of South Asia” would be an incredible nuisance, both for you and for me.

  2 See Lewis (2002).

  3 For more on the connection between urbanization and economic development, see de Vries (1984), Bairoch et al. (1988), and Bosker et al. (2013). Numerous works of economic history have employed city size as a proxy for premodern economic development; a few of these works include de Long and Shleifer (1993), Acemoglu, Johnson, and Robinson (2005), Dittmar (2011), and Cantoni (2015).

  4 In 800, there are twenty-two cities, because four cities tie for nineteenth place on the list. I am extremely grateful to Maarten Bosker, Eltjo Buringh, and Jan Luiten van Zanden for sharing their data from their 2013 paper with me. These data were used to create the maps in Figures 1.1–1.4.

  5 For more on these data, see Pamuk (2011).

  6 Wages were comparable in Istanbul with parts of southern and central Europe. For more on European wage data, see Allen (2001). For more on Ottoman wage data, see Özmucur and Pamuk (2002).

  7 For more on European and Asian wage data, see Allen et al. (2011). These wage data are broadly consistent with Angus Maddison’s GDP data, updated in Bolt and van Zanden (2014). Allen et al. (2011) provide evidence, consistent with Maddison’s evidence that real incomes of Chinese workers were well behind those of workers in northwest Europe in the eighteenth century.

  8 Examples of these views are found in Weber (1922), Cromer (1908), von Grunebaum (1966), and Lewis (1982, 2002). For an overview of this literature, see Kuran (1997, pp. 49–53). For a classic criticism of this approach, see Said (1978).

  9 This certainly does not mean that all economic arguments that depend on cultural explanations are wrong. Greif (1994a), Guiso, Sapienza, and Zingales (2006, 2009), and Tabellini (2010) are among a small set of good economic works that take culture seriously. Alston et al. (2016) provide an insightful example from Brazil in the late twentieth and early twenty-first centuries of how culture is endogenous to beliefs and institutions. Alesina and Giuliano (2015) provide a nice review of this literature.

  10 Numerous scholars have challenged Weber’s thesis since it was first proposed. The most damning criticism is that the “capitalist spirit” predated the Reformation – the Italian city-states of the late medieval period were highly capitalist, yet they did not spark modern economic growth. On this point, see in particular Sombart (1967 [1913]) and Tawney (1926 [1954]). For a more general overview of the literature from the last century on Weber’s hypothesis, see Iannaccone (1998), Delacroix and Nielsen (2001), and Iyer (2016).

  11 This book adds to works such as those of Robert Allen (2009), who argues that relatively high wages in London encouraged more investment in capital-saving technologies than in the rest of the world, and these technologies were at the heart of the Industrial Revolution. Allen’s hypothesis is convincing on many fronts, although it is less strong on the reasons for high wages in London in the first place. An institutional story consistent with the one presented in this book may help set the stage where Allen’s story begins.

  12 van Zanden (2009, pp. 59–60) has a nice discussion of some alternative arguments for the presence of relatively weak family ties in medieval Europe.

  13 Also see Greif, Iyigun, and Sasson (2012) and Greif and Iyigun (2013) for how differences in family structures encouraged different institutional responses for risk-sharing in England and China, and see Greif and Tabellini (2015) for how these differences affected inter- and intragroup cooperation in Europe and China.

  14 This is not to say that Kuran ignores the supply side or that I ignore the demand side. Both sides enter into both of our arguments. I simply place more weight on the supply side, and Kuran places more weight on the demand side.

  15 This theory gives a historical twist to endogenous growth theory – formulated by Paul Romer (1986) and Robert Lucas (1988) and extended by Oded Galor (2011) – which places human capital at the center of perpetual economic growth. Galor is the leader in the “unified growth theory” field, which argues in favor of the importance of prehistorical forces such as geography and endowments. Galor suggests that these forces affected demographic transitions, the evolution of technology, and the acquisition of human capital in ways that explain modern day development. Unlike the geography hypotheses explored in this section, unified growth theory can account for reversals of fortune, although it has a very hard time accounting for the timing of the reversals of fortune.

  16 Besley and Persson (2009, 2010) and Acemoglu (2005) extend this argument, noting that investments in fiscal capacity arise endogenously because of common interests in the provision of public goods. Gennaioli and Voth (2015) take this argument one step further, noting that once fiscal and state capacity becomes important for war-making, a divergence arises between internally cohesive states and those without cohesion, with the latter set of states dropping out of existence. Bates (2001) takes a slightly different interpretation of the “war made the state” argument, suggesting that since European states invested in war, they could ultimately use their increased capacity to coerce for economically beneficial activities, such as the protection of property rights and the termination of feuds. Other important works in this literature, especially those of Dincecco (2009, 2011), stress the role that representative institutions played in generating fiscal capacity through increased taxation and lower sovereign credit risk. Also see Karaman and Pamuk (2013), who argue that the connection between representative institutions, war, and fiscal capacity is dependent on the economic structure of the regime. They suggest that the interests of representative assemblies align with the ruler with respect to war in urban settings, where the ruler and elites jointly govern, but not in rural settings, where local control over coercive power dominates.

  17 These insights have spurred a large literature. For instance, Irigoin and Grafe (2013) argue that fiscal capacity is a function of coercive power and it follows an inverted-U shape; initial investments in coercion pay off well, but eventually diminishing returns kick in. If coercion is too great, the legitimacy of the ruler may be undermined. Dincecco, Fenske, and Onorato (2016) argue that the type of conflict matters for long-run fiscal capacity, as they find no correlation between historical warfare and per capita GDP in sub-Saharan Africa. Johnson and Koyama (2014) and Anderson, Johnson, and Koyama (2016) argue in a series of papers for an important – although nonobvious – effect of increased legal and fiscal capacity: it reduced persecution (of witches, minority religions, and so forth) in Europe in the early modern period.

  18 One hypothesis in this literature that deserves special attention, because it focuses on the Western Europe-Middle East comparison, is Blaydes and Chaney (2013). They argue that European feudalism arose because of the weak fiscal capacity of rulers following the fall of the Roman Empire, and in this system economic elites were able to negotiate with rulers through parliaments. They also argue that such negotiating organizations never emerged in the Middle East because rulers relied on slave armies to extend their power and collect taxes and therefore did not need to negotiate with the elite. Both Blaydes and Chaney and I argue that the lack of constraint by the elite on Middle Eastern rulers relative to their European counterparts was a crucial piece of the long-run economic divergence between Western Europe and the Middle East, but we differ on the reasons why the elite did not constrain Middle Eastern rulers. For example, it is not completely clear within the Blaydes and Chaney framework why the Ottoman military elites in the provinces, who maintained relations with the central government under the timar system, could not have come together in a manner similar to European parliaments to constrain the sultan.
My explanation is that European kings needed to negotiate with the elite to a greater degree than Ottoman sultans did because their position was weaker due to their weaker legitimacy. At the same time, my explanation does not address exactly why parliaments arose in Western Europe in the first place in the manner that they did, except to say that rulers were in a weaker position vis-à-vis the elite because of weaker legitimacy. Blaydes and Chaney shed light on this point, noting that parliaments arose after a long tradition of feudal relations slowly evolved into semi-organized bodies throughout Western Europe.

  19 For more on this argument, also see Hoffman (2011, 2012). For a comprehensive account of the role that military might had on world economic history, see Findlay and O’Rourke (2007).

  20 Numerous reasons have been given for the relative fractionalization of Europe, including the presence of outside threats (Alesina and Spolaore 2005; Ko, Koyama, and Sng 2016), trade patterns (Friedman 1977; Alesina and Spolaore 1997), and geography (Diamond 1997). Others have noted that the presence of numerous independent city states helped foster the rise of a merchant class (Pirenne 1925; Jones 1981; Stasavage [2014] takes an alternative view, arguing that independent cities initially had higher growth rates but ultimately failed, possibly due to the stifling of trade by guilds), and that fractionalization encouraged technological discovery (Lagerlöf 2014).

  21 In fairness to Jones and Mokyr, they are hesitant to make a causal claim connecting conservatism to bad outcomes. Goldstone (2000) presents an alternative take on Mokyr’s argument, suggesting that the key technological advances – along with the acceptance of Newtonian science and the Glorious Revolution settlement – were “accidents” of history and are not explainable causally. While the present book acknowledges the importance of individual events and in no way suggests that history is deterministic, it argues that institutional environments make certain outcomes, including the important ones Goldstone studies, more likely to arise in certain places at certain times.

  22 Cross-country regressions do suggest a possible connection between religion, especially Islam, and economic development. See Grier (1997), Barro and McCleary (2003), and Guiso, Sapienza, and Zingales (2003). For a different view, see Noland (2005). Yet, it is far from clear that these works are picking up anything more than a correlation – causation is a very different story. There are certainly aspects of religious belief that affect economic performance, however. In particular, religion may incentivize (or disincentivize) one to attain education (see, e.g., Berman 2000, Becker and Wößmann 2008, 2009; Botticini and Eckstein 2012; Chaudhary and Rubin 2011, 2016; Meyersson 2013). Such incentives have clearly played an important role in long-run economic outcomes, although they cannot account on their own for the “reversal of fortunes,” the long-run rise of Europe, and intra-European differences. They are important contributing influences nonetheless.

  23 Gregory Clark lists eighty reviews of his book on his website, a very large number for an academic book.

  24 Some of Clark’s comparisons rest on data from the revisionist literature spearheaded by Pomeranz (2000), who argues that China and Europe were economic equals as late as the eighteenth century. Allen et al. (2011) provide evidence that this was not the case, especially in the comparison between northwestern Europe and China.

  25 Blaming Sykes-Picot is a common trope in the popular press, who reinvigorated the thesis when Islamic State declared the goal of creating a new map in the Middle East. For a sample of articles on this topic, see Osman (2013), Sazak (2014), Ignatius (2014), and Howorth (2014). Danforth (2013) presents this argument with a twist, suggesting that the “divide-and-rule” policies of the British and French are responsible for persistent violence. Academic contributions to the debate generally focus on the trade capitulations the Ottoman government gave to the European powers. For instance, see Ahmad (2000).

  26 Of course, geography may affect the types of institutions a society has. For example, Michalopoulos, Naghavi, and Prarolo (2015) argue that numerous features of the Islamic economic system stem from the agricultural endowments and pre-Islamic trade routes of the Muslim world. Rodrik, Subramanian, and Trebbi (2004) test the role of geography versus the role of institutions and economic openness and find that institutions can account for most of the difference in worldwide economic outcomes. Kenneth Pomeranz (2000) also makes an argument in The Great Divergence that places importance on geography. Pomeranz is primarily concerned with the divergence between Europe and China. He makes the “revisionist” argument that Europe and China were on relatively equal economic footing in the eighteenth century, and that fortuitous circumstances allowed Europe to pull ahead (e.g., access to coal, discovery of the New World). There is convincing evidence suggesting that the timing of the divergence is not as Pomeranz suggests (see Allen et al. 2011), and Pomeranz’s theory has difficulty explaining intra-European differences.

  2 The Propagation of Rule

  1 An example that has received significant attention from economists is the degree to which the law protects corporate investors from expropriation by corporate insiders. Such protections are important because they encourage investment. La Porta et al. (1997, 1998) show that these protections are more prevalent in countries with a legal tradition based in common law than in civil law–based countries, and consequently financial development is much greater in common law countries. The legal origins literature has provided many nice insights (see, most prominently, La Porta et al. 1999, Glaeser and Shleifer 2002, Djankov et al. 2002, 2003, Glaeser et al. 2004, and La Porta et al. 2008). That literature seeks the specific effects of the transplantation of different strains of common and civil law. This is not the point of the present chapter, which seeks to understand how the content of specific laws emerges and persists.

  2 See van Bavel et al. (2015).

  3 Wintrobe (1998) provides a comprehensive overview of the constraints facing dictators and how these constraints affect the manner in which the dictator rules.

  4 This definition of elite is similar to the one proposed in Wallis and North (2014).

  5 My definition is similar to the definition of legitimacy proposed by Seymour Martin Lipset (1959, p. 86), who defined legitimacy as “the capacity of a political system to engender and maintain the belief that existing political institutions are the most appropriate or proper ones for the society.” The definition proposed in this book is also similar to those suggested by Greif (2010) and Greif and Tadelis (2010), who define legitimacy of a political authority as “the extent to which people feel morally obliged to follow the authority.” A more expansive definition would include beliefs about the beliefs of others. That is, people may view a rule-maker as legitimate if they believe that others believe the rule-maker has the right to make the rules. Since this aspect of legitimacy is not the focus of this book, I do not include it here.

  6 This motivation of course does not fit all types of rulers. For example, Ronald Wintrobe (1998) analyzes “tinpot” rulers, like many dictators of sub-Saharan Africa, who aim to minimize the cost of staying in power so they can pocket excess rents produced by society. Yet, most goals that one might ascribe to a ruler – even growing wealthy – are not possible if the ruler is not in power. Similar points are made in Gill (1998, ch. 3; 2008, ch. 2).

  7 This idea is similar to Wintrobe’s (1998) insights on dictators, who maintain their rule by investing in repression and loyalty.

  8 For an extended discussion on the concept of legitimizing agents, see Coşgel, Miceli, and Rubin (2012a, 2012b) and Greif and Rubin (2015).

  9 For a nice rational choice analysis of how and why religious authorities legitimize political rule, see Gill (1998, especially ch. 1, 3; 2008, ch. 2).

  10 These examples and many more are in Masud et al. (1996).

  11 Chaney is quoting Robert Irwin’s 1986 book The Middle East in the Middle Ages (p. 50).

  12 For more on the economic effects of the “Pax Mongolia”, see Needham (1954) and Findlay and O’Rourke (2007). For more on the
use of violence to promote economic good in general, see Bates (2001).

  13 Avner Greif (2006b, p. 30) also proposes a definition of institutions favored in this book: “[institutions are] … a system of rules, beliefs, norms, and organizations, that together generate a regularity of (social) behavior … Each component of this system is social in being a man-made, nonphysical factor that is exogenous to each individual whose behavior it influences.” Other important works discussing the nature and consequences of institutions include Greif (1993), North (1981), North, Wallis, and Weingast (2009), Williamson (1985, 2000), Ostrom (1990, 2005), Aoki (2001), Acemoglu, Johnson, and Robinson (2001, 2002, 2005), Acemoglu and Robinson (2006, 2012), David (1994), and Helpman (2004, ch. 7).