Religious Prohibitions Against Usury*
January 21, 1999
Word counts: text only = 6017; text plus footnotes and references = 7326.
JEL classification codes: B15 D23 D64 G22 K20 N13 N23 N33 N43.
Key words: usury, tie-in, consumption smoothing, reciprocal exchange, pooling, insurance, capital market, religion.
CLYDE G. REED
Department of Economics
Simon Fraser University
CLIFF T. BEKAR
Department of Economics
Lewis and Clark College
Abstract (88 words)
Religious prohibitions against usury constitute an explicit tie-in between religious reward and lending/borrowing at interest. We argue that prohibitions will be observed when the following conditions apply: consumption smoothing is required to maintain population levels; charity and reciprocal exchange (informal pooling) are necessary to ensure the adequacy of smoothing; solving the consumption smoothing problem is the responsibility of a monopoly religion; alternative smoothing via the capital market poses a threat to charity and informal pooling. We show consistency between these conditions and the chronology of Roman Catholic prohibitions.
Why do religions prohibit usury? From a non-historical perspective, interest rate policy might seem an unlikely candidate for inclusion as a central tenet of religious doctrine. Yet history is replete with examples: Vedic India, Judaism (from 500 BC to the twentieth century), Roman Catholicism (from the twelfth century to the nineteenth century), and Islam (from its origins circa 570-632 AD to the present). Our focus is on the Roman Catholic case for which we provide a detailed analysis. We then sketch a generalization of the analysis to other cases.
We propose an economic explanation of the Roman Catholic usury prohibitions that combines traditional theory of the firm, industrial organization, and recent research on reciprocal exchange . The analysis views usury prohibitions as the solution to a consumption smoothing problem. It contains three essential elements: the role of charity and reciprocal exchange (informal pooling) in providing economy-wide consumption smoothing and averting social/economic crisis; the substitutability of capital market transactions for charity and informal pooling in smoothing consumption; and the Church’s use of usury prohibitions to create a negative tie-in between salvation and consumption lending at interest. We find that when pooling and charity are the principle consumption smoothing devices available to a significant portion of the population, and when the effectiveness of pooling and charity is threatened by the alternative of the capital market, the Church rails against usury. When pooling and charity are not essential for consumption smoothing, and/or when pooling and charity are not threatened by the capital market, the Church pays little or no attention to usury.
1. Roman Catholic Prohibitions
Period 1: 500 to 1050. The prohibitions are primarily applied to clerics. Usury itself is poorly defined. ‘At no time was it said that usury was a sin against justice, nor was restitution of usuries prescribed as an obligation of justice. . . . while the taking of usury was treated as a serious sin, it was denounced as a form of avarice or uncharitableness.’
Period 2: 1050 to 1175. Usury is declared a sin prohibited by the Old and New Testaments. All interest rates greater than zero are considered usurious. Even the desire for a return beyond the good itself is declared sinful. Usuries are required to be restored in full before salvation is possible. Higher prices for credit sales are declared implicit usury.
Period 3: 1175 to 1350. Usury becomes a dominant concern for the Church. The peak of the ecclesiastical attack on usury is reached at the Council of Lyon in 1274 and the Council of Vienna in 1312. The punishments for usury include the following: usurers are refused confession, absolution, and Christian burial; the wills of usurers are declared invalid; rulers and magistrates of states or communities which permit usury face excommunication.
The prohibitions target only consumption (distress) loans. Although Church scholars may find this assertion controversial, since social policy was not a primary argument in Scholastic writings, it is a reasonable inference from two observations. First, the only individuals or groups singled out by the Church for public denunciation and punishment were lenders dealing in consumption loans . Second, usury prohibitions restricted the flow of loanable funds into consumption lending, but not into other uses . With specific reference to banking, Noonan (1957, p. 192) speculates
It seems exceedingly difficult to believe that virtually the whole capitalistic class of every medieval city would have habitually committed sins of usury by participating in either exchange or deposit banking, if they had truly believed the contracts to have been usurious. There may have been a kind of inculpable ignorance on the part of many which led them to regard the change in the form of the contract, and the fact that the profit was taken not from the poor but from successful businessmen or from States, as sufficient grounds for thinking that banking was not usury.
Our point is that the Church allowed this ‘inculpable ignorance’ to persist by persecuting pawnbrokers but not bankers.
Period 4: 1350 to 1500. Professional usurers are allowed to partake in Church services and to be buried in Church graveyards . Numerous types of loan contracts are explicitly declared non-usurious. The sin of usury is increasingly applied only to excessive interest charges. In the late fifteenth century the Church helps to bring into existence monti di pietà—public pawnshops financed by charitable donations and run for the benefit of the urban poor. Interest charges are explicitly sanctioned in order to cover the cost of operation (6-15 percent).
Period 5: 1500 to 1600. In the sixteenth century, ‘The Church reaffirmed its traditional doctrine on the matter of usury and reverted to the uncompromising attitude which had prevailed prior to the fifteenth century. The secular authorities, however reluctantly, continued to issue licences, but the Church henceforth refused to grant dispensation to the Lombards [professional pawnbrokers]. They were, and remained, excommunicated. According to Charles V's ordinance of January 30, 1546 (n.s.), licenced usurers were forbidden to attend mass or to enter any church under the penalty of forfeiting their licences. The same prohibition applied to anyone who was in partnership with them . . . or who participated in their management.’ 
Period 6: 1600 to 1830. Usury prohibitions are under constant theological attack within the Church. In 1830, ‘the Sacred Penitentiary issued instructions to confessors not to disturb penitents who lend money at the legal rate of interest.’  This is the effective end to usury prohibitions by the Catholic Church.
Consistency with the above chronology constitutes a minimum test for theories of the Roman Catholic prohibitions: explanations must provide answers to the following questions. (1) Why was usury not an important issue for the Church until the twelfth century? (2) Why did the prohibition peak in the late thirteenth and early fourteenth centuries? (3) Why did the Church relax the prohibition in the fifteenth century? (4) Why did the Church revert to pre-plague sanctions in the sixteenth century? (5) Why did the Church end the prohibition in 1830? (6) Why did the Church choose zero as the only acceptable interest rate during the peak period of prohibition? (7) Why did the prohibition target consumption loans but not investment loans?
2. Historical Context
2.1 The need for consumption smoothing.
The primary historical fact driving our analysis is the high variance of household incomes over time, around a mean close to subsistence. The fear of starvation in pre-industrial Europe was omnipresent . The thirteenth and early fourteenth centuries appear especially perilous. Phelps Brown and Hopkins (1956) document average real wages at their lowest level over seven centuries in the period 1270 to 1350. The European-wide famine of 1315 to 1318 resulted in a ten to fifteen percent reduction of population in many villages . Using thirteenth century data, McCloskey (1976) calculates that even relatively wealthy English peasants could expect to experience ‘disaster’ every thirteen years.
Yet starvation in the medical sense was relatively uncommon. Thus the Wrigley and Schofield paradox: for the early modern period in England, crisis years in grain production are poorly correlated with crisis years in mortality . Although studies of the thirteenth and early fourteenth centuries find some evidence of a positive relationship between poor harvests and mortality, the mystery of aggregate population increase and the absence of catastrophic depopulation remains .
These findings suggest the existence of powerful mechanisms for consumption smoothing in pre-industrial Europe. Without such mechanisms, it is difficult to see how the population could have survived even a single generation.
2.2 Mechanisms for consumption smoothing
The following mechanisms for consumption smoothing were available to households in pre-industrial Europe:
As the above list makes clear, most of the devices for consumption smoothing were available only to households with significant landholdings. Smallholders and wage labourers were forced to rely primarily on pooling, charity, and stealing to stay alive.
2.3 The Church and consumption smoothing
The Church in the Middle Ages shared attributes with the modern welfare state . ‘Almshouses, leper-houses . . . pilgrim centres . . . special provisions for education . . . the establishing of monastic hospitals . . . are signposts to a vast system of medieval poor-relief.’  On the other hand, Dyer (1989, p. 252) finds that ‘Gifts were deployed inconsistently and unevenly, and the recipients were either not selected by need at all, or were chosen by rules that excluded many categories of poverty. The overall quantity of charity was low, whether measured as a proportion of the donors’ wealth, or in relation to the needs of the poor.’ He concludes (p. 257), ‘The survival of the medieval poor still remains something of a mystery. Given the inadequacy of charitable institutions, the networks of relatives and neighbours must be assumed to have worked with some effect.’ Emphasis added. 
We define informal pooling as the process of giving and receiving through networks of relatives and neighbours, where reciprocity in the future is a reasonable expectation. Formal pooling refers to the process of contributing to a Church, Guild, government, or other insurance organization with the assurance of receiving payments in bad years. Both types of pooling can be analyzed in terms of insurance against consumption risk . Charity refers to situations of gift giving in which reciprocity is not expected. Charity also exists in formal and informal pooling whenever the implicit insurance premiums imply a cross-subsidy from rich (low risk agents) to poor (high risk agents).
The Church was the dominant instrument of formal pooling, via tithing and poor relief, throughout the Middle Ages. Because the tithe was a flat tax, and the population heterogeneous with respect to risk, Church based formal pooling always involved an element of charity mixed with insurance. Informal pooling and charity were supported by Church doctrine which maintained that giving to the poor directly was just as pleasing to God as giving indirectly through the Church. The poor ‘were valued by the rich, because alms-giving, an act of justice and mercy, wiped away sin.’ 
Church revenues derived principally from two sources: spiritual income (which included the tithe), and income from land. Savin (1974) calculates that the spiritual income of monasteries accounted for only one-quarter of their gross incomes in the early sixteenth century. The main source of their wealth was land . But Church landholdings were not independent of its pooling efforts. A social contract has been hypothesized for the Middle Ages in which feudal lords granted land to the Church with the expectation that the Church would maintain social order by providing sustenance and social services to the poor. Some empirical support for the existence of the contract is provided in the sixteenth century when Henry VIII was able to take back Church land because of the perceived failure of the Church to fulfill its obligations .
2.4 The capital market and consumption smoothing
Prior to the twelfth century the private capital market was extremely limited. With the commercial revolution of the twelfth and thirteenth centuries, however, borrowing and lending increased significantly. The introduction of new business techniques and economic institutions caused transaction costs to fall dramatically. This was most visibly apparent in the decline in interest rates on commercial loans .
Consumption lending in urban areas was dominated by pawnshops. The loans were for small amounts with very few outstanding for more than a year. The rural population tended to borrow from each other rather than from professional moneylenders .
In our analysis, the critical aspect of the capital market as a consumption smoothing device is that the poor were severely limited in their ability to participate. Clark’s (1981, p. 267) study of rural debt for the years 1382 to 1490 finds that ‘. . . none of Writtle’s borrowers appears to have stood among the poorer litigants identified in court rolls. Of these borrowers, 30 percent practiced a craft or a trade; 10 percent were substantial cultivators in possession of plows and draught animals. All participated actively in village life; 42 percent were creditors in their own right and 33 percent were involved in the land market. . . . Lenders trusted people like themselves. . . .’ 
The difficulty for the poor in securing consumption loans can be explained by two observations. The first is that the poor had little in the way of collateral . The second is that their expected earnings (relative to subsistence) were insufficient to allow repayment of loans at existing interest rates. This last point can be illustrated with real wage data. Assume an excessively optimistic scenario in which the average wage of unskilled labour was ten percent above starvation, and that workers never suffered from illness, injury or unemployment . Thus the only source of crisis could be a fall in the real wage more than ten percent below its average level. Using Phelps Brown and Hopkins (1956) data for the period 1300 to 1347, starvation wages are observed in 17 of the 48 years. In every instance the period of starvation wages lasted longer than one year, exceeding the duration of all but a few consumption loans. If the restriction on the term of the loans is relaxed, only one period (1331/32) allowed the wage earner to both avert starvation and pay back the loan. In all other periods the loan could not have been repaid before the next period of starvation wages began, even at a zero rate of interest.
3. Usury Prohibitions as a Tie-in Contract
The primary religious outputs produced by the medieval Church were salvation and insurance against consumption risk via pooling and charity. The core observation in our analysis is that, while Church doctrine linked pooling and charity with the purchase of salvation (as a joint purchase), usury prohibitions amounted to an explicit negative tie-in between salvation and participation in the private capital market. If parishioners chose to smooth consumption through lending/borrowing at interest, they were refused the right to purchase the services necessary to attain salvation—e.g. last rights, confession, Christian burial.
Our analysis begins with the motivation of the Church. Standard tie-in analysis would have the Church use its monopoly in salvation to generate tied sales of insurance. But with formal pooling accounting for only a small percentage of total Church revenues, this argument seems implausible. We hypothesize that the purpose of the tie-in was to generate consumption smoothing for the poor, smoothing that was necessary to sustain and increase population levels and to enhance worker productivity through positive morale and nutrition. The Church benefited from increases in population and worker productivity for at least three reasons. First, the mission of the Church was to save souls: starving parishioners were limited in their ability to attain salvation, and deceased parishioners were difficult (but not impossible) to save. Second, depopulation and declining worker productivity would lower the value of Church land. Third, mass starvation increased the probability that the state or peasant uprisings would usurp Church wealth. Population growth and worker productivity are connected to the usury tie-in through the impact of the capital market on consumption smoothing.
Consider a world without a capital market. Let peasants hold enough land such that they can engage in land-related consumption smoothing. Let them purchase insurance through the Church (i.e. tithing in good years and receiving alms in bad years) and through informal pooling. Finally, let them contribute to and receive charity. Suppose that the consumption smoothing problem is solved in this environment.
Now introduce the possibility of consumption lending/borrowing at interest. This constitutes an important relative price change leading to two substitution effects: the substitution of lending for charity, and the substitution of the capital market for informal pooling. Both may adversely affect community-wide consumption smoothing.
For a peasant in a surplus year, lending at interest to a credit worthy borrower is an alternative to giving to the poor. While charity increases rewards in Heaven, lending at interest increases material wealth in this lifetime. Thus the emergence of a local credit market for consumption loans could imply an expenditure shift out of charity into consumption lending. The usury tie-in reduces the size of this substitution effect.
The choice between the capital market and informal pooling for consumption smoothing depends on their relative costs . The implicit insurance premium in the capital market is the transaction cost involved in contract negotiation and enforcement. In the case of the informal pool, the premium is the networking cost associated with guaranteeing reciprocity. For low risk agents, the cost of informal pooling also includes cross-subsidies to higher risk agents—costs that are avoided in the capital market. The defection of the rich to the capital market, and the loss of their cross subsidy to the poor, threatens the survivability of those remaining in the pool. Recent theoretical results imply additional problems for the poor: an informal pool made up of poor agents is more difficult to maintain than one comprised of wealthier agents ; path dependencies can exist such that, if the scale of market transactions becomes large relative to reciprocal exchange, informal pooling could be driven out of existence altogether, even if it is more efficient . Thus if only the poor are left in the informal pool, whose very existence is now fragile, they could not survive without substantial support from the Church.
Rather than sustaining the poor from general revenues or allowing the poor to die, the Church chose instead to lessen the role of the private capital market as an instrument of consumption smoothing. This was accomplished by preaching that the practice of usury was the ‘death of the soul’.
But if low risk agents are able to consumption smooth with each other in the capital market, why not also in a separate informal pool? Also, why would the permanently poor be allowed in any insurance pool when they would frequently require pay-outs but never make contributions?
Our answer to the first question is that the nature of networking costs limits the size of the informal pool. The more impersonal nature of transactions in the capital market allows the size of this pool to be larger, perhaps large enough to justify a separate pool.
Our answer to the second question is that the permanently poor were not part of the informal pool. Nevertheless, the Church had an economic incentive to include them within a social safety net in order to maintain social order, and many of the permanently poor survived only through charity provided directly by the Church. In order to reduce these expenditures, the Church had an incentive to adopt policies that increased private charity. We have already argued that usury prohibitions led directly to increased charity by increasing the cost of an alternative expenditure—consumption lending at interest. Below we argue that private charity was also indirectly enhanced by usury prohibitions through their impact on informal pooling.
A fundamental problem for agents wishing to engage in reciprocal exchange is convincing others that reciprocity will be forthcoming in the future. This problem is solved by investing in a reputation as a person who helps those in need. Giving to charity is one way to establish such a reputation. In this context, charity can be viewed as posting a bond. Failing to reciprocate in the informal pool would cause an agent to lose the value of his/her reputation (brand name) . The corresponding bond required in the capital market is collateral. While reputation is also important in the capital market, it is not enhanced by giving to the poor.
Let us summarize the entire argument. The Church benefits from keeping the population alive and productive, which in turn depends on economy-wide consumption smoothing. Consumption smoothing for the poor requires extensive charity and informal pooling. The existence of a private capital market in consumption loans can threaten both. Usury prohibitions re-direct funds from interest earning consumption loans (to the relatively well-off) into charity and informal pooling. Informal pooling benefits the poor in several ways. It includes them in an insurance pool. They benefit from cross-subsidized insurance premiums. The reputation building aspects of informal pooling promote private charity.
Our analysis sheds light on another usury puzzle: given the widespread practice of usury, why did the Church persist in costly, but seemingly ineffective, prohibitions? From McCloskey and Nash (1985), ‘The prohibition of usury was irrelevant: that the sin of taking interest should be committed frequently is no more surprising than that the sin of adultery was.’ Our answer to the puzzle is that the transfer of even a limited number of households into the informal pool was valuable to the Church since it increased the probability of the pool’s survival. Moreover, any induced increase in pooling and charity, however small, reduced Church expenditures on the poor.
4. Explaining the Chronology of Roman Catholic Prohibitions
The above analysis establishes a rationale for the existence of usury prohibitions. Let the Church’s objective be to maximize the net benefits from the prohibitions by equating their marginal benefit with their marginal cost. To generate comparative static results, the effect of historical parameters on these benefits and costs must be explored.
We view six parameters as historically determining: transaction costs in the capital market, the distribution of land holdings, the means and variances of incomes, income inequality, the size of the urban population, and the role of government and other secular institutions in reducing consumption risk.
The effect of transaction costs on usury prohibitions depends on the level of these costs. Very high transaction costs rule out the capital market as a consumption smoothing alternative. The benefit to the Church of increasing the costs even further is negligible and we expect the absence of usury prohibitions. Moderate transaction costs imply that consumption loans would be characterized by relatively high interest rates and significant collateral. This is the worst case for the poor and we expect the Church to respond with usury prohibitions. Low transaction costs yield low interest rates and thereby raise the possibility of loans to the poor. The incentive for the Church to impose prohibitions is diminished.
Landholding allows for a high degree of consumption smoothing. When most of the population holds land, consumption smoothing through pooling and charity is not as critical for survival. We expect the Church to lessen usury prohibitions when a significant portion of the population holds land.
Falling mean income and a constant or increasing variability of income implies an increase in the risk of starvation. Increasing the number of people in the informal pool lowers this risk. Thus we expect the Church to intensify usury prohibitions in an effort to shift households out of the capital market and into the informal pool.
The greater income inequality, the larger the subsidy from rich to poor that is required in order to sustain economy-wide consumption smoothing. Charity and informal pooling are primary sources of income redistribution. But since increased transfers make the informal pool less attractive to the wealthy, they are motivated to defect into the capital market where cross subsidies are absent. We predict that the Church will increase usury prohibitions to halt this defection.
The rural/urban distribution of the population affects the networking costs of informal pooling . It is less expensive to establish a reputation for reciprocity in close knit rural environments. Networks in the countryside are more likely to be long lasting, perhaps intergenerational. The longer the expected life span of the pool, the larger are the future benefits of membership, the less likely is defection. Finally, informal pooling requires participants to be able to observe each other’s incomes and work effort in order to reduce the cost of detecting fraudulent demands for pay-outs. These costs are lower in rural areas. By implication, the larger the urban population, the more the Church must support informal pooling through usury doctrine.
When the state provides welfare to the poor, or when private insurance against consumption risk becomes generally available, the Church no longer has an incentive to provide insurance against consumption risk and the rationale for usury policies disappear.
Below we show consistency between these predictions and the chronology of the Roman Catholic prohibitions.
4.1 Why was usury not an important issue for the Church until the twelfth century?
The inefficiency of the private capital market made it a high cost alternative to informal pooling; most of the population in the tenth and eleventh centuries held land ; the relative homogeneity of incomes minimized cross subsidization problems in pooling.
4.2 Why did the prohibition peak in the late thirteenth and early fourteenth centuries?
The commercial revolution of the twelfth and thirteenth centuries increased the efficiency of the private capital market; the percentage of the population in the small holder/low wage earner categories increased dramatically ; increasing population levels gave rise to the cultivation of marginal lands causing lower means and higher variances in agricultural output ; with the general increase in population, cities became larger.
4.3 Why did the Church relax the prohibition in the fifteenth century?
Average landholdings increased ; the variance of output decreased as grain production was concentrated on better land ; urban populations declined with the general contraction of population associated with the plague.
4.4 Why did the Church revert to pre-plague sanctions in the sixteenth century?
Increased efficiencies in the capital market lowered interest rates ; a growing proportion of the population became small holders or landless peasants; increasing poverty and urbanization followed in the wake of renewed population growth .
4.5 Why did the Church end the prohibition in 1830?
Interest rates had fallen to very low levels allowing a larger proportion of the population to qualify for consumption loans ; new technology and increased investment in agriculture resulted in increasing outputs with lower variance ; the state took on more of the responsibility for poor relief in the form of poor law legislation ; innovations occurred in private insurance that reduced risk .
4.6 Why did the Church choose zero as the only acceptable interest rate during the period of peak prohibitions?
In the thirteenth and early fourteenth centuries, the Church was particularly intent on expanding the number of people in the formal and informal insurance pools and increasing charitable donations. A zero interest rate minimized the attraction of the capital market alternative.
Of course, a zero interest rate was often not optimal from a narrow perspective of equating the marginal benefits of prohibitions to their marginal costs. We might expect an optimal usurious interest rate lower than the market rate, but seldom zero. This optimal usury rate would also change from year to year along with the benefits and costs of usury. The Church, however, was required to justify usury prohibitions by reference to natural law. Under these circumstances, it would have been difficult to maintain an argument for a non-zero interest rate that fluctuates through time. It would also have been more difficult to detect usury violations if the target interest rate was constantly changing. The demand for zero interest on consumption loans is consistent with a broader calculus that includes the costs of both selling the argument for usury prohibitions and enforcing the prohibitions.
4.7 Why did the prohibition target consumption loans but not investment loans?
Although Church doctrine stressed the spiritual dangers of excessive wealth, the Church stopped short of severe sanctions on investment lending. Compared to the case of consumption loans, vigorous persecution of investment loans involved lower benefits and higher costs. At the village level, the potential for consumption loans was larger than for investment loans: large scale investment loans were made by merchant bankers in capital markets inaccessible to villagers. Consumption lending/borrowing is clearly a substitute for lending/borrowing in the informal pool—they are alternative methods of consumption smoothing. The relationship between investment loans and informal pooling is less clear. Finally, the Church would have found it extremely costly to persecute merchant bankers—the Church’s ‘most trusted emissaries and . . . highest dignitaries’ as well as sources of financial and political support.’ 
5. Other Historical Examples
We propose the following sufficient conditions for the existence of religious prohibitions against usury: (1) consumption smoothing is required to maintain population levels; (2) charity and informal pooling are essential for solving the consumption smoothing problem; (3) solving the problem is the responsibility of a monopoly religion; (4) consumption smoothing via the capital market poses a threat to charity and informal pooling. While the first condition is perhaps common to most societies, the rest are not. In modern industrialized economies, for example, charity and informal pooling may be extensive, but they are not necessary for solving the consumption smoothing problem. Both social welfare and rent seeking motivations may result in restrictions on market interest rates, but the instrument of enforcement is the state rather than a religious organization.
The detailed application of our hypothesis to other cases of religious prohibitions against usury is a topic for further research . Below we briefly consider a non-European example: usury in western India from the fourteenth to nineteenth centuries.
Hardiman (1996) reports that in India money lending (for both consumption and investment expenditures) was a professional occupation condoned by the state with minimal regulation of the interest rate charged. The usurer was considered ‘a person who provided "help" to the needy in the form of loans.’  This attitude is reflected in language. The word for usurer, sahukar, translates as ‘a doer of good and righteous deeds’ . This is in stark contrast to medieval Europe. From LeGoff (1988, p. 50), ‘With virtually no excuses available, the usurer remained, during the thirteenth century, one of the few men whose trade was condemned secundum se, "in itself", de natura, "by its very nature." He shared this unhappy fate with prostitutes and acrobats.’
Hardiman (pp. 126-7) describes the relationship between peasant borrower and sahukar as ‘a very personal one, commonly stretching back for generations on both sides.’ From our perspective, the key to understanding the contrast with Europe is the guarantee from the sahukar that client households would not starve when harvests were poor. This is a case in which lending/borrowing at interest, far from being a threat to community-wide consumption smoothing, was the mechanism through which it was realized. As such we would predict the observed absence of religious sanctions.
* We are grateful to Doug Allen, Yoram Barzel, Ken Carlaw, Nancy Clegg, Greg Dow, Steve Easton, Curtis Eaton, Henry Ferris, Brookes Hull, Michael Lebowitz, Richard Lipsey, Joel Mokyr, Krishna Pendakur, Mark Perlman, and Angela Redish for comments on earlier drafts of this paper.
 Alternative economic explanations applicable to the Roman Catholic prohibitions include Ekelund, Hébert, and Tollison (1989), Posner (1995), and Glaeser and Scheinkman (1998). In our view, these explanations fail to fully exploit the variance in the Roman Catholic data and, as a result, cannot account for the chronology of the Catholic case. It should be noted, however, that the primary empirical focus of the papers by Posner and Glaeser and Scheinkman is on modern examples of government imposed usury restrictions.
 Noonan (1957, pp. 16-17).
 Noonan (1957, p. 191).
 Ibid., p. 195. De Roover (1948, pp. 48-83) comments extensively on the wide-spread use of bills of exchange by merchant bankers to circumvent usury prohibitions on investment loans. In contrast, there were no methods of circumvention permitted by the Church for consumption loans.
 De Roover (1948, pp. 104-5, 151).
 Ibid., p. 151.
 De Roover (1974, p. 321).
 See Geremek (1997, pp. 56-58, 122-24). From data on English wheat prices, Hoskins (1964, 1968) infers a major grain shortage once every six years (on average) between 1480 and 1759. Similar findings from real wage data are reported in Wrigley and Schofield (1981). But note that the survival problem for the wage earner was even more difficult than the one portrayed by real wage figures, which only reflect aggregate output shocks. In addition, the individual worker faced idiosyncratic shocks due to illness, injury, local crop failure and unemployment.
 Jordan (1996).
 Wrigley and Schofield (1981).
 Dyer (1989, pp. 178-187, 270-1).
 Razi (1980) and Schofield (1997). Also see the discussion in Dyer (1989, pp. 123-127); Jordan (1997, pp. 102-6); Duby (1968, pp. 254-57).
 McCloskey (1976) estimates that in the thirteenth century, scattering of holdings decreased the ‘frequency of disaster’ by 44 per cent compared to consolidated holdings.
 For a list of public goods provided by the Church see Ekelund et. al. (1996, pp. 25-29). For an analysis of the Church’s comparative advantage in income redistribution see Hull (1989).
 Gilchrist (1969, pp.78-9).
 See also Geremek (1997, pp. 60-61). For specific examples of networking for the purpose of consumption smoothing see Bennett (1992).
 Our use of the word pooling, while illustrative, may also be confusing. In standard insurance analysis of homogeneous agents, pooling is shown to be effective in smoothing individual consumption only in an environment of idiosyncratic shocks. In the context of aggregate shocks with heterogeneous agents, however, pooling can still be effective in increasing community survivability by transferring income from those above subsistence to those below. Our analysis includes both aspects of pooling.
 Dyer (1989, p. 235).
 Estimates of the size of Church landholdings are found in Herlihy (1961) and Swanson (1989).
 For a description and analysis see Clegg and Reed (1994).
 Homer and Sylla (1996, pp. 136-143).
 De Roover, (1948, p. 164).
 See also Duby (1968, p. 253), who notes that in the thirteenth century, loans to ‘those in the humblest positions . . . do not appear in any of the surviving documents.’
 Hudson (1982, p. 27) lists the goods favored as collateral in pawnshops: ‘jewelry, gold and silver ware, pewter, saddlery, manuscripts and the more expensive kinds of textiles and furs.’ The collateral actually used by the poor consisted of used clothing, tools, cutlery, and other items worth small sums of money. See Dyer (1989, pp. 178-82) for a discussion of the possessions of the poor.
 Here unemployment includes limited work capacity constrained by nutrition. Fogel (1991, p. 46) reports that in eighteenth century France, for example, ‘the bottom 10 percent of the labour force lacked the energy for regular work, and the next 10 percent had enough energy for less than three hours of light work daily.’
 See Kranton (1996) for a general analysis of reciprocal exchange (also called gift exchange) and its interaction with market exchange. Kimball (1988) explores the specific application of gift exchange to insurance against consumption risk by medieval peasants. Assuming homogeneous agents and costless pooling, Kimball concludes that ‘informal insurance’ was a more efficient form of risk reduction than scattering in open fields. The alternative of smoothing through the capital market is not explored. Coate and Ravallion (1993) apply reciprocal exchange to consumption smoothing in developing countries, but not in comparison to other forms of smoothing. These formal models of reciprocal exchange have been applied exclusively to theoretical environments of homogeneous agents who do not face a subsistence constraint.
 This is a common result in the gift exchange literature. It follows from diminishing marginal utility of consumption. For low income agents the marginal utility of consumption is quite high. As a result they are more likely to defect from reciprocal exchange.
 Kranton (1996).
 This argument is an application of analysis found in Klein and Lefler (1981) who investigate the interaction between bond posting and reputation building in an environment in which agents must purchase goods before they can ascertain product quality. An implication of the Klein and Lefler analysis is that agents will post bonds in the form of ‘hostage capital’. Charity meets this requirement because the permanently poor, by definition, can never reciprocate.
 Coate and Ravallion (1991).
 For example, at the end of the eleventh century in England, only ten percent of the rural population were landless peasants, and they were full-time workers on the land of their manorial lord. Thirty two percent were smallholders (holding 5 percent of the land). The rest (58 percent) were sufficiently large landholders to allow full time work on their own land. Twenty eight percent of the rural population held enough land (30 acres or above) to require seasonal wage labour to supplement household labour. See Miller and Hatcher (1978, p. 22).
 Postan (1966, p. 619), finds that in England, by the thirteenth century, one half of the adult population had no land at all; 45 percent of holdings averaged less than three acres; 22 percent of the tenants held an average of 30 acres. Over the thirteenth century the percentage of small holders continued to increase—see Dyer (1989, pp. 124-5). For similar findings on the Continent see Geremek (1997, pp. 56-58) and Duby (1968, pp. 282-285).
 Because of the paucity of data on grain outputs, price data have traditionally been used as a proxy measure for the variability of harvests. In England ‘grain prices fluctuated by an average of 26.6 per cent from one year to the next’ for the period 1280 to 1350, but only 11.5 per cent between 1440 and 1500—see Baily (1988, p. 235).
 For England, Dyer (1989, pp. 141-2) cites the examples of two representative villages. In Holywell in 1252, 64 percent of holdings were less than ten acres. In 1451, the percentage had fallen to 41 percent. In Stoughton in 1341, 52 percent of the holdings were 11 acres or less. By 1477 the percentage had fallen to 16 percent, with 58 percent of the holdings larger than 30 acres. For Continental examples see Duby (1968, p. 339).
 See footnote 33.
 Homer and Sylla (1996, p. 120).
 In general, the economies of western Europe returned to conditions similar to those in the thirteenth and early fourteenth centuries. See Geremek (1997, pp. 92-141).
 Homer and Sylla (1996, pp. 297-318).
 McCloskey (1975) discusses the implications of innovations in agriculture (originating on the Continent and diffused to England) for risk reduction.
 King (1997).
 Pearson (1997).
 Noonan (1957, p. 14).
 We note, however, that a critical requirement in our analysis—usury prohibitions are strictly applied only to consumption lending—is common to both the Islamic and Jewish cases. For Islam see Kuran (1992a, pp. 27-31) and (1992b, pp. 53-4). For a specific explanation of the Islamic preference for risk sharing over straight interest (riba) in investment lending see Presley and Sessions (1994). Also note that for consumption loans to incorporate risk sharing, some form of pooling is required. See Kirshenbaum (1985, pp. 284-85) for a discussion of the Jewish case.
 Hardiman (1996, p. 123).
 Ibid., p. 123.
Baily, Mark (1988). ‘Peasant welfare in England, 1290-1348.’ Economic History Review, vol. 51, pp. 223-251.
Bennett, Judith M. (1992). ‘Conviviality and Charity in Medieval and Early Modern England.’ Past and Present, no. 134, pp. 3-18.
Clark, Elaine (1981). ‘Debt Litigation in a Late Medieval Village." In Pathways to Medieval Peasants, Papers in Mediaeval Studies 2 (ed. J. A. Raftis), Toronto: Pontifical Institute of Mediaeval Studies.
Clegg, Nancy W. and Reed, Clyde G. (1994). ‘The Economic Decline of the Church in Medieval England.’ Explorations in Economic History, vol. 31, pp. 261-280.
Coate, Stephen and Ravallion, Martin (1993). ‘Reciprocity Without Commitment.’ Journal of Development Economics, vol. 40, pp. 1-24.
De Roover, Raymond (1948), Money, Banking and Credit in Mediaeval Bruges. Cambridge: Mediaeval Academy of America.
_________________ (1974), Business, Banking, and Economic Thought in Late Medieval and Early Modern Europe: Selected Studies of Raymond de Roover (ed. Julius Kirshner), Chicago: University of Chicago Press.
Dyer, Christopher (1989), Standards of Living in the Later Middle Ages. Cambridge: Cambridge University Press.
Ekelund, Robert B. Jr., Hébert, Robert and Tollison, Robert D. (1989). ‘An Economic Model of the Medieval Church: Usury as a Form of Rent Seeking.’ Journal of Law, Economics, and Organization, vol. 5, pp. 307-331.
Ekelund, Robert B. Jr., Hébert, Robert F., Tollison, Robert D., Anderson, Gary M. and Davidson, Audrey B. (1996), Sacred Trust. New York: Oxford University Press.
Fogel, Robert W. (1991). ‘The Conquest of High Mortality and Hunger in Europe and America: Timing and Mechanisms.’ In Favorites of Fortune (ed. Patrice Higonnet, David Landes, and Henry Rosovsky), Cambridge: Harvard University Press.
Geremek, Bronislaw (1997), Poverty: A History. Oxford: Blackwell.
Gilchrist, John Thomas (1969), The Church and economic activity in the Middle Ages. New York: St. Martin's Press.
Glaeser, Edward L. and Scheinkman, José A. (1998). ‘Neither a Borrower Nor a Lender Be: An Economic Analysis of Interest Restrictions and Usury Laws.’ Journal of Law and Economics, vol. 41, pp. 1-36.
Hardiman, David (1996). ‘Usury, Dearth and Famine in Western India.’ Past and Present, no. 152, pp. 113-156.
Hirlihy, David (1961). ‘Church Property on the European Continent, 701-1200.’ Speculum, vol. 36, pp. 81-105.
Homer, Sidney and Sylla, Richard (1996), A History of Interest Rates. New Brunswick: Rutgers University Press.
Hoskins, W.G. (1964). ‘Harvest Fluctuations and English Economic History, 1480-1619.’ Agricultural History Review, vol. 12, pp. 28-46.
____________. (1968). ‘Harvest Fluctuations and English Economic History, 1620-1759.’ Agricultural History Review, vol. 16, pp. 15-31.
Hudson, Kenneth (1982), Pawnbroking. London: The Bodley Head.
Hull, Brooks B (1989). ‘Religion, Afterlife, and Property Rights in the High Middle Ages.’ Studies in Economic Analysis, 12, pp. 3-21.
Jordan, William C (1988). ‘Women and Credit in the Middle Ages: Problems and Directions.’ Journal of European Economic History, vol. 17, pp. 33-62.
______________(1996), The Great Famine. Princeton: Princeton University Press.
Kimball, Miles S. (1988). ‘Farmers' Cooperatives as Behavior Toward Risk.’ American Economic Review, vol. 78, pp. 224-32.
King, Steve (1997). ‘Poor Relief and English Economic Development Reappraised.’ Economic History Review, vol. 50, pp. 360-368.
Klein, Benjamin, and Lefler, Keith (1981). ‘The Role of Market Forces in Assuring Contractual Performance.’ Journal of Political Economy, vol. 89, pp. 615-641.
Kirshenbaum, Aaron (1985). ‘Jewish and Christian Theories of Usury in the Middle Ages.’ The Jewish Quarterly Review, vol. 75, pp. 270-89.
Kranton, Rachel E. (1996). ‘Reciprocal Exchange: A Self-Sustaining System.’ American Economic Review, vol. 86, pp. 830-851
Kuran, Timur (1992a). ‘The Economic System in Contemporary Islamic Thought.’ In Islamic Economic Alternatives (ed. Kwame Sundaram Jomo), London: Macmillan Academic and Professional Ltd.
___________ (1992b). ‘Economic Justice in Contemporary Islamic Thought.’ In Islamic Economic Alternatives (ed. Kwame Sundaram Jomo), London: Macmillan Academic and Professional Ltd.
LeGoff, Jacques (1988), Your Money or Your Life. New York: Zone Books.
McCloskey, Donald N. (1975). ‘The Persistence of English Common Fields.’ In European Peasants and Their Markets (ed. William Parker and Eric Jones), Princeton: Princeton University Press.
_____________(1976). ‘English Open Fields as Behavior Towards Risk.’ In Research in Economic History, vol. 1 (ed. Paul Uselding), Greenwich: JAI Press.
McCloskey, Donald N. and Nash, John (1984). ‘Corn at Interest: The Extent and Cost of Grain Storage in Medieval England.’ American Economic Review, vol. 74, pp. 174-87.
Miller, Edward and Hatcher, John (1978), Medieval England: Rural Society and Economic Change, 1086-1348. London: Longman.
Noonan, John (1957), The Scholastic Analysis of Usury. Cambridge: Harvard University Press.
Pearson, Robin (1997). ‘Towards an Historical Model of Services Innovation: the Case of the Insurance Industry, 1700-1914.’ Economic History Review, vol. 50, pp. 235-256.
Phelps Brown, Henry and Hopkins, Sheila V. (1956). ‘Seven Centuries of the Prices of Consumables, Compared with Builders’ Wage-Rates.’ Economica, vol. 23, pp. 296-314.
Posner, Eric A. (1995). ‘Contract Law in the Welfare State: A Defense of the Unconscionability Doctrine, Usury Laws, and Related Limitations on the Freedom to Contract.’ Journal of Legal Studies, vol. 24, pp. 283-319.
Postan, M. M. (1966). ‘Medieval agrarian society in its prime. England.’ In Cambridge Economic History of Europe I, 2nd edition (ed. M. M. Postan), Cambridge: Cambridge University Press.
Presley, John R. and Sessions, John G. (1994). ‘Islamic Economics: The Emergence of a New Paradigm.’ The Economic Journal, vol. 104, pp. 584-596.
Razi, Zvi (1980), Life, Marriage and Death in a Medieval Parish. Cambridge: Cambridge University Press.
Savin, Aleksandr N. (1974), English Monasteries on the Eve of Dissolution. New York: Octagon Books.
Schofield, Phillipp R. (1997). ‘Dearth, Debt and the Local Land Market in a Late Thirteenth-Century Village Community.’ The Agricultural History Review, vol. 45, pp. 1-17.
Swanson, Robert. N. (1989), Church and Society in Late Medieval England. Oxford: Basil Blackwell.
Wrigley, Edward A. and Schofield, Roger S. (1981), The Population History of England, 1541-1871. London: Harvard University Press.