تجزیه و تحلیل مخاطرات اخلاقی ناشی از صندوق بین المللی پول

Three channels through which the IMF rescue package may affect international lending can be distinguished: debtor-side moral hazard, creditor-side moral hazard, and debtor and creditor-side moral hazard. We show that if the rescue package fully benefits the debtor, no credit contract between him and the creditor arises. The other two kinds of moral hazard, where the creditor receives the rescue package either fully or in part, increase the scale of international lending relative to the case where no rescue package is forthcoming. The increase is larger if the creditor receives the whole rescue package than if it is shared between the creditor and the debtor. These results are based on the analysis of two sequential credit relationships, the first one between a bank and a government and the following one between the IMF and the government. Each of these credit relationships is characterized by asymmetric information and modeled by a moral hazard model. The two moral hazard models are linked by considering the different channels of the IMF rescue package.

The International Monetary Fund (IMF) has played a prominent role as a provider of emergency liquidity to troubled countries in the past decade. The number and size of the rescue packages have given rise to criticism of the IMF. Friedman (1998), Meltzer (1998) and Schwartz (1998) argue that the provision of emergency liquidity by the IMF affects the efficiency of international financial markets. In other words, the provision of an IMF rescue package has the effect that creditors lend and debtors borrow too much compared with the case when no IMF rescue package is forthcoming. In this paper we focus on this distortion, i.e. excessive lending and borrowing due to the rescue package relative to the level that would prevail without the IMF support. We show that the existence and the extent of this distortion mainly depends on two factors; the probability that the IMF will lend money in a financial crisis; and, if this lending occurs, the question of who receives the money. The first point is to be expected, so we do not go into detail here. Rather, we concentrate on the second point. This is a special feature of the IMF rescue package because three recipient scenarios can be distinguished. First, the rescue package may be fully beneficial for the debtor. An example is where the government of a troubled country receiving financial support from the IMF uses the whole amount to finance its deficit. Second, the rescue package may fully benefit the creditor, i.e. the bank. In this case, the government of the troubled country spends the whole amount of the IMF support on meeting its obligations to a foreign bank. The third scenario is where the rescue package is shared between both parties, the debtor and the creditor. The first and the third scenario are special features of IMF lending. The second one, where the creditor receives the rescue package in full, corresponds to the situation in traditional insurance. We will take the example of deposit insurance. Here, the bank (which corresponds to the government in the IMF case) always gives the money received from the deposit insurance to its depositor (who corresponds to the bank in the IMF case). Why does not the bank always receive the whole amount of the rescue package from the government in the IMF case? The reason is that a foreign creditor cannot enforce a repayment from a sovereign debtor. Therefore, it is up to the government whether or not to spend the rescue package for servicing its liabilities towards the bank abroad. This in turn implies that the scenarios whereby the government keeps either all or part of the rescue package are special features of the IMF's “safety net”. These two scenarios also reflect the view that creditors should be involved in bearing the cost of the crisis. In these scenarios, the creditor either bears the whole cost or a part of it. The three scenarios of recipients are three different channels through which the rescue package has an effect on international lending. All recipients of the rescue package are assumed to behave rationally. Since the various channels represent different circumstances, the behavior of the recipients is expected to be different for the three channels and compared to the one without IMF support. The rescue package implies that the government pursues a worse policy whereas the bank lends more money. For the behavior change of the recipients in the three scenarios, terms associated with moral hazard exist in the literature (e.g., Lane and Phillips, 2000). “Debtor-side moral hazard” means that the debtor is benefiting fully from the rescue package and consequently changes its behavior. The terms “creditor-side moral hazard” and “debtor and creditor-side moral hazard” describe the other two scenarios and the resulting behavior changes of the recipients. The impact of each of these transmission channels of the rescue package on the level of international lending depends on the preferences and information structure of the involved parties. As described in more detail below the relationships between a debtor, a creditor and the IMF are two bilateral, sequential credit relationships under asymmetric information. Therefore, we analyze the question whether the rescue package leads to excessive lending and borrowing using microeconomics. More precisely, based on a contract theory approach, we investigate how each of these transmission channels of the IMF rescue package influences the level of international lending compared to a situation where a rescue package is absent. We can demonstrate that no credit contract exists between the creditor and the debtor, if the rescue package fully benefits the debtor. If however, the creditor receives either the whole or at least part of the rescue package, a credit contract arises. The resulting level of international lending in both of these cases is higher than when no rescue package is forthcoming. This deviation represents a distortion in the financial markets. The distortion is larger if the creditor receives the whole rescue package than if it receives only part of it. Hence, the model implies that the case of the creditor being the beneficiary of the rescue package is both good and bad in the following sense; on the one hand, the creditor must benefit from the rescue package in order for credit transactions to take place. These volumes, however, are inefficient. Since this inefficiency is reduced by involving the creditor in bearing the costs of the crisis, the burden should be shared between the creditor and the debtor. The lending process underlying all three transmission channels consists of two credit relationships. First, the bank lends money to the foreign country's government and second, in the event of a financial crisis in the borrowing country, the IMF possibly lends money to the government in trouble. This second credit relationship reflects a further distinction between the IMF and a traditional insurer in that the IMF's financial support is not simply a payout in cash such as an insurer would make but a loan to be repaid with interest. Asymmetric information is inherent in credit markets. In cross-border credit relationships it is in fact particularly prevalent because of the geographical and cultural distance between lender and borrower. Both credit relationships are therefore assumed to be characterized by asymmetric information in the form of hidden action. This is equivalent to the classical notion of moral hazard, which is formalized by the standard moral hazard model for any principal–agent relationship (e.g., Ross, 1973). We model this lending process as a game with two sequential moral hazard models, i.e. it is a sequential three-player moral hazard model. In doing so, the moral hazard model of the Bank–Government relationship differs from the standard two-player moral hazard model applied to a credit relationship in that the subsequent IMF–Government relationship is considered. The two relationships are linked by the government or the bank, or both, depending on the respective transmission channel of the IMF rescue package under consideration. This three-player moral hazard model helps us to answer questions such as how the willingness of the creditor to lend and the simultaneous willingness of the debtor to borrow is compared to the traditional two-player approach. We explain financial crises by incentive problems, and our approach is novel in distinguishing between asymmetric information and the IMF rescue package causing the incentive problems. Dooley (1997), Krugman (1998) and Mc Kinnon and Pill (1996) emphasize only the latter cause; in fact, they consider the incentive problem occurring in the presence of government guarantees to private borrowing. We apply the asymmetric information framework to model the occurrence of a financial crisis. More precisely, we focus on the problem that a creditor in a cross-border financial relationship cannot observe the debtor's investment policy. Choosing a moral hazard model of the credit relationship between the bank and the government is therefore natural. In this framework, loan default by the government defines a financial crisis. The loan default in turn is caused by bad government policy and external factors. This means that the worse the policy pursued by the government, the more likely a financial crisis is. Due to the external factors, however, an unsound policy does not necessarily cause a financial crisis. Conversely, sound government policy may also result in a financial crisis. Whereas we use the asymmetric information framework to explain why a crisis takes place, we apply the incentive problem due to the IMF rescue package against the background of a financial crisis. In this situation we investigate whether the IMF rescue package leads to an inefficient functioning of the financial system, which in turn makes a financial crisis more likely to happen. The sequential three-player moral hazard model forms the basis of this analysis. This, of course, is not the only possible way to study a crisis. Other approaches focus on self-fulfilling debt runs (e.g., Rodrik and Velasco, 1999) and bank runs (e.g., Chang and Velasco, 2000 and Chang and Velasco, 1999). In these models the illiquidity is at the center of the crisis. Since multiple equilibria follow, the crisis itself occurs as a switch to an equilibrium run. Based on these approaches similar results to ours are reached. Chang and Velasco (2000) for instance extend the classic Diamond–Dybvig model (Diamond and Dybvig, 1983) to a monetary, open economy to link financial fragility and exchange rate crises. Among other things, they show that a central bank providing emergency credit to commercial banks in case of trouble cannot prevent runs under fixed exchange rates whereas it can under flexible ones. In our model, independent of the exchange rate regimes but dependent on the transmission channels of the rescue package, the latter prevents the current crisis, but only at the cost of making a financial crisis more likely to happen in the future. The paper is organized as follows. Section 2 describes the sequential three-player moral hazard model considering debtor-side moral hazard. Section 3 presents the bank's optimal loan scheme under uncertain IMF lending which causes debtor-side moral hazard. Section 4 describes the model of debtor and creditor-side moral hazard and the model of creditor-side moral hazard. Both models are characterized by ensuring a contract between the three parties. Section 5 presents these contracts, and Section 6 summarizes.

This paper investigates the three transmission channels of the IMF rescue package, namely: debtor-side moral hazard, creditor-side moral hazard, and debtor and creditor-side moral hazard. To address these channels, we solve a sequential three-player moral hazard model. This model consists of two standard moral hazard models which are linked by considering the different channels of the IMF rescue package. Each of the two moral hazard models considers the incentive problem due to asymmetric information and the link between the two models takes into account the incentive problem due to the rescue package. If the rescue package fully benefits the government, an enlarged conflict between the bank and the government over the issue of sound policy arises. Asymmetric information in the form of hidden policy enables the government to pursue its own interests. The government pursues an unsound policy in its own interest because it is less costly, and due to IMF lending. The former reason is standard. It is at the root of the standard two-player moral hazard model. We show the loan scheme solving the enlarged conflict. To address the standard conflict, the loan contingent on the good state of nature has to be larger than the loan contingent on the bad state of nature. To additionally address debtor-side moral hazard, the difference between the two contingent loans has to be larger. Since only the loan contingent on a financial crisis is affected by IMF lending, this loan is reduced. Consequently, the bank's loan scheme addressing the enlarged conflict deviates by a lower loan contingent on the bad state of nature from the loan scheme addressing only the standard conflict. The government's expected utility from the loan scheme addressing the enlarged conflict is, therefore, smaller than the government's expected utility from the loan scheme addressing the standard conflict. The latter is precisely equal to the government's reservation level of utility. Hence, no credit contract between the bank and the government arises if the rescue package fully benefits the latter. To ensure borrowing by the government from the bank under IMF lending, the model is extended. This extension considers the IMF rescue package also benefiting the bank and, consequently, changing its behavior. In other words, by making a financial crisis less severe for the government and the bank, the government pursues an unsound instead of a sound policy and the bank lends more money. Then, the loan scheme solving the enlarged conflict is influenced by the bank being more willing to lend money. Besides debtor and creditor-side moral hazard, creditor-side moral hazard is addressed. If the bank fully receives the rescue package, only the standard conflict over a sound policy between the bank and the government arises. This conflict is solved by the bank's optimal loan scheme, which is influenced by the bank being willing to lend more money due to the IMF rescue package. The resulting volume of international lending deviates considerably from the one when no IMF rescue package is forthcoming. This deviation or inefficiency in financial market is smaller if the rescue package is shared between the bank and the government. Hence, the model shows that from an efficiency standpoint, the bank should be involved in bearing costs of the crisis. If the bank, however, bears the whole cost, which means that the government receives the rescue package in full, no credit contract arises. Hence, the rescue package should be shared between creditor and debtor.