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

  This chapter suggests that such arguments do not properly account for the path-dependent consequences that followed from the relaxing of interest restrictions in Western Europe and, conversely, the “path not taken” in the Middle East due to the persistence of formal interest restrictions. Path-dependent consequences can be difficult to trace because the results are by definition far removed from the initial causes. This chapter traces out one such consequence: the growth of branching of Western European financial institutions over long distances. Branching is a process by which a central holder of wealth transacts with numerous other offices (i.e., branches) that do business in varied locations; most major modern banks employ this form of organization. Branching emerged in late-medieval Europe precisely due to incentives associated with the relaxation of interest restrictions in Western Europe, albeit through nonobvious channels. This was a key development in the growth of Western finance, and it arose because merchants and lenders acted according to the incentives imposed on them by the surrounding institutions. Such an innovation might not have been imaginable in the Middle East, since nothing like the institutional buildup that facilitated branching in Western Europe ever emerged in the Middle East. To be clear, this is not to say that interest restrictions were burdensome because they prevented lending of any type: people in both Western Europe and the Middle East found ways around the restrictions and frequently lent to known relations. The claim here is that the manner in which people found ways around the restrictions precipitated institutional developments associated with larger-scale lending by long-lived organizations – what we now call banking. And as these organizations became more complex in Western Europe relative to the Middle East, organizational differences built on themselves in a path-dependent manner so that, over time, what were once small differences blossomed into very large ones.

  Another reason the connection between relaxations of interest restrictions and financial growth is complex is that any explanation must take into account why interest restrictions were relaxed in the first place in Western Europe but not the Middle East. If there were a region in the early medieval period where demand was high for a relaxation of interest restrictions, it was the Middle East, not Western Europe. This is where the framework proposed in Chapter 2 helps guide intuition as to why it was Western Europe, and not the Middle East, where interest restrictions were ultimately relaxed.

  Interest restrictions are useful to analyze for reasons beyond their long-run effect on the financial systems of Western Europe and the Middle East. They provide a nice point of comparison between the two sets of economies because they prevailed in both Islam and Christianity throughout most of the medieval period. Both sets of restrictions initially emerged for similar reasons, as rulers and religious authorities imposed restrictions in an economic context where most borrowing occurred primarily for consumption purposes. In that environment, borrowers were often poor or recently affected by some negative economic event. They borrowed to prevent starvation in the face of a bad harvest or to ensure enough seed was available for the following year’s crop. These borrowers generally faced extremely high interest rates, often leading to lifetime indebtedness or much worse, such as selling one’s children into slavery. Since social safety nets were weak and markets for insurance against disaster were missing, religious interest restrictions encouraged interest-free lending among neighbors and discouraged demand for high-interest loans.4

  Interest restrictions imposed a very different set of hurdles, however, when investment borrowing became feasible. Under such conditions, outright bans on interest prevented mutually agreed-on transactions involving moderate interest and thus inhibited economic growth. Yet, interest restrictions persisted for centuries in both religions in the face of commercial expansion. The persistence of interest restrictions in both religions is a puzzle with no obvious answer. Religious reinterpretations in favor of commerce occurred frequently in both religions, so the answer cannot be some simple appeal to the inviolability of religious dictates or religious hostility to commerce. The questions that need answering are: Why did interest restrictions persist for centuries longer in the Middle East than in Western Europe? Can the political economy framework proposed in the previous chapters shed light on these differences? What was the ultimate economic and institutional impact of these restrictions? This chapter answers these questions. First, however, some historical context is necessary.

  History of Islamic Interest Restrictions

  Taking interest on loans has always been a sin in Islam. The Qur’an has numerous passages detailing the sinfulness of riba, which was a pre-Islamic usurious process in which lenders doubled and redoubled the principal of a loan when the debtor was unable to pay at maturity.5 This effectively entailed enslavement for debtors, as debts mounted after a single default. Restrictions on taking interest were therefore an optimal response to a situation where extremely high-interest loans created massive social problems. Early Muslims quickly equated riba with interest of any kind. Some Qur’anic injunctions against riba include:

  Those who consume riba shall not rise except like the one who has been struck by the Devil’s touch. This is because they say that selling and riba-making are one and the same thing, whereas God has made selling lawful and has forbidden riba. Whoever receives an admonition from his Lord and desists, he shall have his past gains, and his affair is committed to God; but whosoever reverts – those are the inhabitants of the Fire, therein dwelling forever. God destroys riba but makes alms (zakat) prosper … O ye who believe! Protect yourselves from God and remit what is left of riba if ye be faithful. If ye do not, be prepared for war from God and His Prophet: but if ye desist, ye shall receive back your capital without doing injustice or suffering injustice. If, however, anyone is in difficulties, let there be a delay till he is able to pay, although it is better for ye to remit if ye only knew. (Surah Al-Baqarah ii.274–280)

  Believers, do not live on riba, doubling your wealth many times over. Have fear of God, that you may prosper.

  (Surah Al-Imran iii.130)

  That which you seek to increase by riba will not be blessed by God; but the alms (zakat) you give for His sake shall be repaid to you many times over.

  (Surah Al-Rum xxx.39)

  In theory, an absolute ban on interest should have detrimentally affected economic outcomes once investment lending became feasible during the economic expansion of the first four Islamic centuries. In practice, however, the ban hardly meant that capital flows from financier to merchant halted altogether. In such a profitable environment, lenders simply found ways to circumvent interest restrictions. One set of tactics frequently employed were straightforward ruses, known as hiyal (singular, hila), designed to facilitate evasion of the ban.6 A famous example of a hila is the previously discussed double sale, which worked as follows: Abdul buys a rug from Mahmud for 50 dinars, and Mahmud immediately buys the rug back from Abdul for 55 dinars, payable in a year. The upshot is that Mahmud holds onto his rug, receives 50 dinars from Abdul, and owes Abdul 55 dinars in one year. This is ostensibly a loan at 10 percent interest, with the interest being the difference between the two prices. Yet, because it is not officially a loan but is instead two sales, both of which are legal in Islamic law, the double sale does not violate the letter of the law as long as 50 dinars is a reasonable price for the rug. The double sale was commonplace by the ninth century, and Muslims employed it in Medina as early as the eighth century.7

  Other forms of circumventing interest restrictions were common. For instance, documents exist showing numerous loans and purchases on credit in twelfth-century Tunisia where lenders made profit by exchanging in different forms of currency.8 In the Ottoman period (1299–1923), a common form of transaction was istiğlal, which involved the debtor giving his creditor a piece of real estate, supposedly as a sale, but actually as a pawn. Such contractual forms were common well before the Ottomans, and hiyal became more complicated over time as business transactions became more complex.

>   Clerics went beyond merely permitting hiyal: they actively participated in creating hiyal, too. As early as the second Islamic century, clerics wrote treatises recognizing and formulating various forms of hiyal.9 Authorities brought customary commercial law into agreement with Islamic law (Shari’a) in this way. Although the legality of hiyal differed over time and place, especially among different legal schools, they usually received stamps of approval from leading jurists.

  It is no coincidence that the rise of Middle Eastern economies was also the period of massive reinterpretation of Islamic law via hiyal. The flurry of reinterpretation of interest law played a role in permitting commerce to flourish: more transactions occurred, and long-distance trade was common and profitable. In the first four Islamic centuries, legal flexibility was ubiquitous, and economic growth was the result.

  This all changed at some point toward the end of the fourth Islamic century. Evidence of further advancements in Islamic law on interest recedes, and as a result, transactions involving overt, guaranteed interest were not a common means of extending commercial credit in medieval Islam. Not coincidentally, this slowdown in reinterpretation of interest law coincided with the proverbial “closing of the gate of ijtihad,” whereby Islamic reinterpretation of all types of laws slowed immensely. By the mid-twelfth century, contracts stipulating interest existed, but the parties either derived interest from another type of contract or concealed interest in another way.10 If the parties did not conceal the interest payment, one party could bring the other to court where the transaction was voidable without further legal consequence.

  The Ottomans permitted more straightforward interest-bearing lending. For example, the Ottoman Grand Mufti Ebu’s-su’ud (c. 1490–1574) permitted lending at moderate interest under the euphemistic designations “transaction” or “legal transaction.” He considered charging greater than 15 percent a criminal offense, but he allowed exceptions liberally.11 Data collected by Timur Kuran (2013) and employed in Kuran and Rubin (2017) indicate that taking moderate interest was normal in Istanbul in the seventeenth and eighteenth centuries – the real interest rate on private loans was around 19 percent – although the contracts usually cite some ruse. Likewise, in a study of seventeenth-century judicial records in Anatolian Kayseri, Ronald C. Jennings shows that lenders regularly charged interest on credit in accordance with Islamic law and “secular” law (kanun) and with the consent and approval of the judge’s court, religious scholars, and the sultan. But almost all interest-bearing transactions Jennings observes involved some sort of ruse.12 The same was true in the seventeenth century in the important commercial city of Bursa: interest ranging between 10 to 15 percent was legal, but the parties primarily conducted such transactions via ruses.

  It is unlikely that most lenders actually resorted to such ruses. Numerous scholarly works indicate that lip service paid to Islamic law – rather than the actual undertaking of the ruse – prevailed throughout the Ottoman Empire. For example, waqf (pious trust) trustees required borrowers to deposit a pledge, suggesting that they lent at interest directly. Courts approved the instruments used by the cash waqfs in Bursa, but their relatively constant returns suggest that economic interest prevailed.13

  It was clearly possible to lend at interest without fear of punishment throughout much of Islamic history. As long as the parties employed a ruse, courts upheld the transaction should the borrower renege. Yet, the mere fact that lending via ruse occurred between two individuals who knew each other does not mean that this was conducive for the growth of larger-scale organizations such as banks. In order for a court to uphold a ruse, the parties actually had to go through with the ruse. It might have been possible for friends or neighbors to avoid doing this, since social sanctions provided a strong incentive to pay back loans between two parties that knew each other. The same is not true of bigger organizations like banks, where lenders know prospective buyers much less intimately. Having to actually go through with a ruse with unknown relations for large sums of money makes lending much more risky and costly. It meant a greater outlay of capital in order to conduct the ruse, increased monitoring so that the borrower would not renege before the ruse is complete, and much greater risk of default should the ruse not actually occur. It also made taking deposits and openly paying interest on those deposits next to impossible.

  This could not be more contradictory to the essential elements of modern banking. Modern banks engage in some monitoring – they check credit scores – but they have recourse should the borrower renege. Modern banks accept deposits, which are attractive to depositors because of the interest paid, and invest these deposits in more profitable pursuits – often loans at interest to other bank customers. These activities were too risky or costly for a medieval Middle Eastern financier to undertake. Hence, nothing akin to a bank ever arose in Middle East indigenously. Not only were banks capable of pooling resources nonexistent, but there is no evidence from the medieval period that there were individuals whose exclusive occupation was banking. Many of the elements commonly associated with Western banking were simply absent in the medieval and early modern Middle East. In its absence, lenders extended credit primarily via partnerships, which remained small both in terms of size and capital outlay.14

  The data cited above provide further evidence of the lack of banking. Jennings’s Ottoman data suggest that most loans were small, debtors were often poor, and lending was highly decentralized. Neither banks nor a class of big moneylenders existed; individuals incurred 97 percent of all debts and single individuals extended 99 percent of all credit.15 In Kuran’s (2013) Ottoman data set, most loans were small and made either by individuals or cash waqfs. The cash waqf was an Ottoman institutional innovation that could have substituted for banks – but did not – and it thus warrants further discussion.

  The institution of the waqf, or pious foundation, existed since the first Islamic century. Originally, the assets of waqfs had to be immovable, but this requirement was relaxed beginning in the eighth century and relaxed even further during the Ottoman period in order to permit lending using waqf funds.16 These lending waqfs, or cash waqfs, functioned primarily in the following way: the waqf manager distributed its endowed capital as credit to a number of borrowers and spent the return of the investment on social and religious purposes. The cash waqf founder earned returns without violating the Shari’a ban on interest by lending via sleeping partnership or legal ruse.17

  Leading jurists did not always grant approval of cash waqfs. Because they earned income primarily through interest-bearing loans, cash waqfs provoked controversy. Yet, as the cash waqf became customary and essential to financial dealings, jurists were more willing to accept their validity. An episode known as the “cash waqf controversy,” which ensued throughout much of the sixteenth century, secured their ultimate acceptance. The controversy reached a peak in 1545 after a jurist issued an opinion opposing the cash waqf, which by this time was well established. The Grand Mufti Ebu’s-su’ud countered this opinion, and his opinion carried the day. Ebu’s-su’ud’s ultimate concern was not with juristic texts, but with what was in popular usage and for the welfare of the people, which clearly entailed maintaining cash waqfs.18

  This episode suggests that Islamic law was flexible in responding to overwhelming demand for an “anti-Islamic” action. But this response – the cash waqf – is also indicative of the limitations set by the broader political economy equilibrium where religious authorities legitimized rulers. During the reign of Bayezid II (1481–1512), the number of cash waqfs in Istanbul increased steadily: in 1505, more cash waqfs than land waqfs were established, and by 1533, the cash waqf became the rule, not the exception.19 The popularity of cash waqfs was due not only to the lack of banks and other alternatives able to meet the demand for interest-bearing loans but also to idiosyncrasies of cash waqfs themselves. Where jurists approved them, they allowed moneylenders to operate within Islamic law.20 On top of the lip service paid to the Shari’a by the waqf manager, cash
waqfs were insulated from the charge of sinfulness though their inclusion in the waqf system, which imbued them with a certain level of sacredness. Yet, this feature also meant they were much more rigid than Western banks, since managers had to spend waqf returns on prespecified social and religious purposes. While these investments may have done a great deal of good for society, they forced waqf managers to invest funds in an inefficient manner, since they could not reroute funds to the most profitable opportunity. On top of this, waqf law prevented cash waqfs from expanding their business and branching out, since they were required to spend their proceeds on prespecified purposes rather than reinvesting them in the waqf. This meant that a process where growth begat growth was stifled, and cash waqfs remained small. Cash waqfs therefore never grew into anything remotely resembling modern banks. Such a process did occur in Western Europe, however, due to the different incentives associated with lending at interest that evolved during the Commercial Revolution of the tenth–fourteenth centuries.