Journal of Islamic Monetary Economics and Finance. Vol. No. , pp. 471 - 484 p-ISSN: 2460-6146, e-ISSN: 2460-6618 INCENTIVES. SOCIAL NORMS. AND BUSINESS CYCLE: AN EXAMPLE OF BUSINESS LOANS PROVISION BY ISLAMIC BANKS Ishaq Bhatti1 and Suren Basov2 Department of Accounting. Economics. Finance and Data Analytics. La Trobe University. Australia, bhatti@latrobe. Department of Economics. University of Melbourne. Australia, surenbasov@gmail. ABSTRACT The interaction of social norms and incentives is a subject of growing interest in economic literature. Basov and Bhatti . pointed out that invoking a social norm is both a blessing, since it allows mitigating moral hazard problem, and a curse, since it restricts the class of admissible contractual arrangements. In this paper, we reiterate this point using particular example of the effects of restrictions imposed on contracts by Shariah law on the optimal risk-incentive trade-off. We show that extra rigidity imposed by Shariah law leads to a greater reluctance to invest into daring new ideas, which are profitable in expectation, but may also result in significant losses. A shared set of social norms between the lender and the entrepreneur allows mitigating adverse consequences of the excess rigidity through creation of good will and may even lead to an improved performance. The adverse consequences may vary according to the stages of business cycle. As a result, recessions can have negative long-term effects and longer booms may be followed by longer recessions. We also hypothesize that turning a social norm into a law will deprive it of the ability to generate good will, while leaving the negative aspects intact. We find a tentative support of this hypothesis by comparing relative performance of Islamic banks in three regions: South East Asia . Malaysi. Middle East, and the UK. Keywords: Social norms. Incentives. Trust. Islamic banks. Shariah law. JEL classification: D4. D8. Article history: Received : July 18, 2022 Revised : July 29, 2022 Accepted : August 11, 2022 Available online : August 30, 2022 https://doi. org/10. 21098/jimf. Incentives. Social Norms, and Business Cycle: An Example of Business Loans Provision By Islamic Banks INTRODUCTION The interaction of social norms and incentives is a subject of growing interest in economic literature (Abdelsalam et al. Gul and Ng, 2. A social norm is an understanding that governs the behavior of members of a society which can be used to promote social control and is self-enforced by feeling of guilt or shame in those who break it. The latter differentiates a social norm from a convention. kind of behavior that is expected and is optimal to conform, provided the others do so, though breaking it does not elicit any negative emotions. For example, driving at the left side of the road is a convention: given that everybody drives on the left it is the best response for a driver to follow this rule, no feeling of shame or guilt for driving on the right side is needed. Tipping in a restaurant, on the other hand, is a social norm. A patron who does not frequent a particular restaurant will not leave a tip, unless she feels shame or guilt for not doing so. Pioneering papers by Akerlof . 6, 1980, 1. mark the birth of literature that includes social norms in microeconomic analysis. It was followed by extensive literature, studying the effects of social norms on the optimal incentives. For example. Bernheim . explicitly studies the effects of the norm of conformity. Bohnet. Frey, and Huck . investigate the importance of trust, while the research by Fehr and Schmidt . concerns the norms of fairness and reciprocity. Kandel and Lazear . Barron and Gjerde . Che and Yoo . Huck. Kybler, and Weibull . , and Huck and Rey-Biel . extend this litera ture to incorporate peer effects. Sliwka . and Fischer and Haddart develop models where social norms arise endogenously. In early papers the agents were driven by universal ethical principles, such as reciprocity or trust. Such norms can govern bilateral relationships. Recent papers, in particular the ones that endogenize the social norms, require to have an entire population of agents. Festry . provides a review of this literature1. The main lesson to be learned from this literature is that relying on the social norms creates a nontrivial trade-off. On the one hand, social norms may be used to mitigate moral hazard. On the other, they restrict the set of allowable contracts that can be signed between the agents. In doing so, the social norms introduce rigidities in the relationship between economic actors that can lead to undesirable consequences, e. unemployment and suboptimal portfolio selection. Though this trade-off was in the background of the models since Akerlof . , it was first made explicit in Basov and Bhatti . , who combine norms of trust and honesty, assuming that honesty should be activated by showing trust. Basov and Bhatti follow earlier literature, modelling social norm as an abstract ethical principle. This paper follows closely the modelling strategy developed in Basov and Bhatti . , however the particular social norm under consideration here is motivated by Islamic law. Under the law, the principal-agent relationships are partnerships usually governed by MushArakah and MudArabah. Both types of contracts are quite similar with the difference that under MudArabah partyAos capital is entirely managed by another party, while under MushArakah a joint management is possible. Both types of contracts are like traditional incentives 1 For a book level treatment of interaction of social norms and incentives, see Basov . See Basov and Bhatti . and Uddin et al. for basic concepts of Islamic Finance products. Journal of Islamic Monetary Economics and Finance. Vol. Number 3, 2022 contracts, where pay ment of the agents depends on the profits she generates to the principal. However. MushArakah and MudArabah contracts are more rigid in several ways. First, the profits of the venture are shared between the two parties per a pre-agreed ratio, while the losses are shared per shares of capital invested by the parties in the venture2. This implies that contracts must be piece-wise linear with slope fixed for negative values of profits. Moreover, the above-mentioned contracts do not allow a fixed payment. Arbi. Basov, and Bhatti . considered optimal contracts under such a restriction and have shown under some additional assumptions on distribution of profits and preferences, that optimal contract will have a fixed slope irrespective of the riskiness of the venture. Therefore, risky projects will not be undertaken, even if they have highly positive NPV. In a recent paper. Ebrahim and Sheikh . suggested that a way out may be to combine a MudArabah and Ijarah contracts3, since the latter allows a fixed wage. We will take the latter point of view in this paper and allow a fixed payment in the incentive The paper is organized as follows. In Section 2 we develop the basic static In this model an Islamic bank finances an entrepreneur engaged in a business venture. The expected profit of the venture depends on the effort put by the entrepreneur. The effort is non-contractible therefore, the payment to the entrepreneur depends only on realized profit . In the case of loss, the sharing rule is exogenously fixed by the amount of the initial capital put in by the entrepreneur, while in the case of profit it reflects the optimal risk-sharing versus incentives provision trade-off. We show that the extra rigidity that comes from a predetermined way to share the losses is particularly handicapping during the It also leads to a greater reluctance to invest into daring new ideas, which are profitable in expectation, but may also result in significant losses. Though extra rigidity imposed by Shariah law is usually costly, there may be a silver lining that emerges from a shared cultural heritage between the bank and the entrepreneur. Let us assume that at the top of the offering a wage schedule, the bank may offer a suggested effort level. In accepting the contract the entrepreneur promises to exert the suggested effort, though this promise cannot be enforced in a court. The idea behind the promise is that Muslims have a moral obligation to be good to their word. Of course, the contract once offered will attract both: sincere Muslims and opportunists. The degree of opportunism is the entrepreneurAos private information. The bank offers a menu of contracts that consist of the share of the up-front cost the bank finances, her share of profit, and suggested effort levels, to maximize her profits in the presence of private information. To investigate the effects the shared social norms can have on different stages of the business cycle, in Section 3 we extend the basic static model in several ways. First, we allow the bank to choose whether to operate as a conventional or an 2 Some schools of Islamic jurisprudence are even more restrictive. For example. ShafiAoy school insists that both gains and losses must be shared proportional to investment (Udovitch, 1. This leaves no room to provide incentives to reward entrepreneurial talent and we will not consider such models in this paper. 3 For a definition of an Ijarah contract and a detailed discussion of other Islamic financial instruments, see Basov and Bhatti . Incentives. Social Norms, and Business Cycle: An Example of Business Loans Provision By Islamic Banks Islamic bank and to recommend an effort to the entrepreneur. Following Basov and Bhatti . we assume that an agent experiences a psychological cost of deviating from the recommended effort if she perceives the bank to be a bone fide Islamic bank, which she does if the contracts offered by the bank up to that date complied with the Shariah law. We also assume that the economy can be in one of two states: a boom or a recession, with the marginal product of effort higher during the boom. We show that the bank will have stronger incentives to break restrictions imposed by Sharia law during the recession. We conclude in Section 4. II. THE STATIC MODEL In this Section we start by proposing a general contracting framework in the presence of social norms and then adapt it to study business loans, provided by Islamic banks. A General Contracting Model in the Presence of Social Norms Suppose two parties enter a bilateral contracting relationship. The first party . he principa. has an access to set E of enforceable contracts. The society also possesses a social norm that specifies set of N c E of appropriate contracts. The second party . he agen. controls set A of available actions. The principal offers to the agent contract enforceable A OO E and recommends action yi* OO A. The gross payoff x to the principal is distributed according to F . es, y. The recommendation by the principal is assumed to be not enforceable. The agent decides whether to accept the contract and if yes then what action to take. The cost of action to the agent is . i, yi*) = c. i, yi*)I(A OO N) . where c(A) is the physical cost of action and cp(A. A) is the psychological cost of not following the principalAos recommendation. We assume that c. Ou 0 and cp. i, yi*) Ou 0 and cp. i*, yi*) = 0. One can interpret formula . in the following way. If the principal offers a contract that respects the social norm the agent feels a warm glow and feels bad about not following the principalAos recommendation. The warm glow, however, disappears if the contract chosen by the principal does not respect the social norm. Note that formulation . assumes that agents, while feeling a warm glow to principals who abide by the norm do not feel any spite towards the violators, neither does a violator suffer a cost from the society at large. We will say about a society where social norms operate in such a way that it possesses mild social If, on the other hand, one or both of the above assumptions are violated, we will say that a society possesses hard social norms Note that while social norms can never have negative effects on economic performance, the latter is not true if the social norms are hard. Journal of Islamic Monetary Economics and Finance. Vol. Number 3, 2022 Business Loan Provision by Islamic Banks Suppose an entrepreneur possesses a project that requires up-front cost K > 0 to finance. Given effort e > 0, the profits from the project, yes, are distributed per distribution function F . es, . Exerting effort is costly. For simplicity, we assume that the cost is quadratic, i. To see how this formulation fits in the general framework developed above, assume A=R and write e, for effort, to denote a generic element of A. The bank cannot observe the effort, so it can affect the level of effort only indirectly by offering a wage schedule. To keep matters simple, assume that the bank is restricted to offering a piece-wise affine wage scheme, i. where I(A) is the indicator function of set A, i. = 1 if yes OO A and I. = 0 We will also assume that . That is, in set E is the set of piece-wise linear contracts, satisfying . One can justify this set by assuming that the bank can use only three instruments. The bank issues some shares to finance the project run by entrepreneur and sells some of them at an initial public offering (IPO). It promises the entrepreneur a fixed wage, offers her some shares in the project, and some number of protective European put options, with the strike price equal to the price of a share, obtained at the IPO. Coefficient 1 . is the share of profits . the entrepreneur is entitled to due to the profit-loss (PLS) sharing agreement. Another way to understand constraint . is to assume that if the enterprise experiences a loss and the share price goes down the bank can commit to partially reimburse the entrepreneurAos losses, but cannot impose any extra penalties. From a technical point of view, restriction without this restriction . is important, since without it MirrleesAo contract (Mirrlees, 1. can be used to closely approximate the first best. According to Usmani . , a MudArabah contract cannot involve lump-sum payments to either party, which will force = 0. Arbi. Basov, and Bhatti . consider optimal contracts under such a restriction and have shown under some additional assumptions on distribution of profits and preferences, that optimal contract will have 1 = 2 = 0. 5 irrespective of the riskiness of the enterprise. As a result, sufficiently risky projects will not be undertaken, even if they have positive NPV. In a recent paper. Ebrahim and Sheikh . suggest that a way out may be to combine a MudArabah and an Ijarah contract, since the latter allows a fixed Here we will consider such a hybrid contract and allow for a non-zero value of . However, we will respect the constraint a MudArabah contract imposes on Incentives. Social Norms, and Business Cycle: An Example of Business Loans Provision By Islamic Banks the way the parties share losses. Such a contract requires that the share of losses endured by the entrepreneur should equal her fraction of capital investment. Formally, if the entrepreneur possesses own funds L OO . K], and chooses to invest I O L then . Combining . one obtains: Coefficient 1, on the other hand, is assumed to be unrestricted. To close the model, assume that entrepreneurAos utility is linked to the distribution of her income. O, and effort she chooses to exert by: where E stands for the expected value of the wage and Var for its variance. the case when the wage schedule is linear in profits and distribution of profits is normal with mean that depends on e, expression . corresponds to the certainty equivalent of an agent with a CARA Bernoulli utility: Under more general assumptions on wage schedule preferences described by . are not derivable from any expected utility model, however, we will still assume them for analytical tractability. Let us also introduce the following notation: where subscript v stands for variable. One can show that . Journal of Islamic Monetary Economics and Finance. Vol. Number 3, 2022 where (A) stands for the cumulative normal distribution function and . The Optimal Contract. When There are No Psychological Costs Let us start by considering the situation, when the psychological costs are zero. an entrepreneur accepts the contract she will choose an effort to solve: She will choose to participate if . The wage schedule is given by . , with coefficients i satisfy . The bank chooses , 1, 2 to maximise its expected profits, . subject to . , . , and . Let us assume that profits are distributed normally with a mean equal to e, where e is the exerted effort and variance E2. Proposition 1. Get us assume that the entrepreneur has su cient funds then constraint . does not bind at equilibrium, i. 2 < min. L/K). Proof. Let us assume that constraint . does not bind, i. the entrepreneur has sufficient funds. Then constraint . will bind if and only if 1 = 2, i. If constraint . The principal chooses 1, 2, e to maximize the total certainty equivalent: subject to the incentive compatibility constraint and constraint . The constant term is then chosen to ensure the participation constraint holds. The Lagrangian for the problem is: Incentives. Social Norms, and Business Cycle: An Example of Business Loans Provision By Islamic Banks where is the Lagrange multiplier for . and AA is the Lagrange multiplier on . Let us assume that constraint . is binding, write the first order conditions with respect to 1 and 2 and evaluate them at 1 = 2. Following this procedure, one obtains: which implies . By Kunh-Tucker Theorem. AA Ou 0, implying 2 = 0 and since we assumed that constraint . is binding, this implies 1 = 0, but such contract will implement zero effort and therefore should expect zero profits. The principal can do better by offering small positive 1. Example. Let us consider almost risk neutral entrepreneur operating in the environment of extreme uncertainty, i. consider a limit EIe O. IIe 0, and IE2 Ie I0>0. Then IH. IeI0/2,IHAo. Ie0 and the total certainty equivalent becomes: which should be maximized subject to: The solution is . In this example, this solution applies as long as: and the entrepreneur invests . Journal of Islamic Monetary Economics and Finance. Vol. Number 3, 2022 Note that the riskier the project or the more risk-averse the entrepreneur, the smaller is the fraction of funds she invests. If condition . is violated, then constraint . binds, i. Since fixing 2 partially protects entrepreneur against losses, the bank will choose to increase 1 to provide the entrepreneur with stronger incentives, provided the effort is sufficiently productive. Let us revisit the above example of extreme uncertainty and almost risk neutral manager, but assume that the latter has no funds, i. L = 0. Then the Islamic bank will be forced to set 2=0 and one can calculate that in this limit and the optimal contract has slope Entrepreneur with no funds faces stronger incentives than the one with sufficient funds, provided . In the above model Islamic banks maximize the same objective as the conventional ones, but face more stringent constraints. As a result they can perform at best at a par with the conventional ones, and usually will underperform in comparison. Basov and Bhatti . further pointed out that such underperformance can be exacerbated by positive assortative matching in the market for human capital. However. Islamic banks can also play on a shared cultural heritage with the entrepreneur to mitigate the moral hazard problem and boost the performance. We will enrich the model in the next Section to take this possibility into account. We will also consider a dynamic version of the model, where an economy can be in a boom or in a recession. We will show that recessions can lead to opportunistic behavior by the banks, eroding trust and having longlasting consequences. The Optimal Contract With Positive Psychological Costs In this Section, we enrich the model by assuming that on the top of providing financial incentives, the bank can recommend a level of effort to the entrepreneur. Though this level of effort is not contractible, the entrepreneur will endure psychological cost if she chooses to deviate from it. We assume that this cost is given by: Incentives. Social Norms, and Business Cycle: An Example of Business Loans Provision By Islamic Banks . where e* is the level of effort recommended by the bank. The problem of bank is like the one we solved in the previous Section, but with cost of effort given by the sum of physical cost . and psychological cost . Coefficient can be viewed as the degree of trust between the bank and the entrepreneur. It can be boosted by the common cultural heritage, for example, by common faith. If the latter is the case, an Islamic bank operating in a Muslim country can have a better ability to tap into this channel than a conventional one. Let us introduce . Assume that k <<1, i. the probability of a loss is negligible. In that case slope 2 is irrelevant and we can use the results obtained in Basov and Bhatti . to argue that at equilibrium the bank would recommend the efficient level of effort and the slope of the optimal compensation schedule will decrease, exposing the entrepreneur to less risk in the case of gains and countervailing effects of the loss-sharing restriction. Since without a loss sharing restriction, having an extra instrument will always be beneficial, one may argue that an Islamic bank may outperform a conventional one during a boom when loss-sharing is not important. Basov and Bhatti . argue that in the latter case the optimal contract will contain the following slope of the optimal incentive scheme and the recommended effort: Let AS denote the profit the bank will earn if it relies on the social norm and shares losses proportionally to the investment, while AF is the profits earned by the bank that fully on the financial contract. If E<< the probability of a loss is negligible and AS > AF, since relying on the social norm provides the principal with additional In the opposite case, <*. BUSINESS CYCLE AND PERSISTENCE OF SOCIAL NORMS In this Section we assume that productivity of effort can fluctuate due to random Such shocks in productivity are known to be the main course of business cycles . ee, for example. Long and Plosser, 1. Let us consider a dynamic model Journal of Islamic Monetary Economics and Finance. Vol. Number 3, 2022 where time is discrete and the bank and the entrepreneur enter a new contractual relation in every period. Assume that >* and OO{L,H} with H>L>0. We will interpret periods when =H as booms and the ones with =L as recessions and assume that Pr (=L)=p irrespective of the history, i. every period the economy has the same probability to be in a recession4. We also assume that . during boom the bank is better off to rely on a social norm, while during a recession it prefers to rely on a purely financial contract. We will also assume that that determines the psychological cost evolves in the following way: OO. O) if OAE