Bankers often extend credit to firms owned by their close business associates, members of their own families or clans, or businesses that they themselves own. On the one hand, this allows banks to overcome information asymmetries and creates mechanisms for bankers to monitor borrowers. But on the other hand, related lending makes it possible for insiders—bank directors—to expropriate value from outsiders, be they minority shareholders, depositors, or taxpayers (when there is under-funded deposit insurance). The evidence suggests that during an economic crisis insiders have strong incentives to loot the resources of the bank to rescue their other enterprises, thereby expropriating value from outsiders. In a crisis, loan repayment by unrelated parties worsens, and banks thus find it more difficult to reimburse depositors and continue operations. Consequently, insiders perform a bit of self-interested triage: they make loans to themselves, and then default on those loans in order to save their non-bank enterprises. Outsiders, of course, know that they may be expropriated, and therefore behave accordingly: they refrain from investing their wealth in banks, either as shareholders or depositors. The combination of tunneling by directors, the resulting instability of the banking system, and the reluctance of outsiders to entrust their wealth in banks results in a small banking system.
And yet, the economic histories of many developed countries (the United States, Germany, and Japan) indicate strongly that related lending had a positive effect on the development of banking systems. If related lending is pernicious, why then did it characterize the banking systems of advanced industrial countries during their periods of rapid growth? In fact, related lending is still widespread in those same countries.
In a recent paper, my coauthors and I seek to reconcile these two competing views about the effects of related lending (Cull, Haber, and Imai, 2011). One reason researchers have come to these two divergent views is sample selection bias. We only get to directly observe related lending using ex-post measures—and that ex-post evidence is not randomly distributed across countries or time. Consider, for example, the literature on related lending as looting: the loan books on which these studies are based are available precisely because the banks were intervened by governments in the aftermath of banking crises characterized by tunneling and fraud. Countries in which related lending was positive for the development of the banking system do not figure in these studies: because there was no looting, there was no crisis; because there was no crisis, there was no government intervention; and because there was no intervention, there are no loan books in the public domain.
A similar problem affects the literature that views related lending positively—as a mechanism to overcome information asymmetries. It is not an accident that studies that advance this view are all based on historical evidence. The loan books that inform these studies are in the public domain precisely because of the antiquity and durability of the banks that kept these books. Banks that did not survive for very long, because they were looted by their own directors, were less likely to produce loan books that could one day find their way into an archive or library.
In the core of the paper we analyze a unique cross-country data set covering 74 countries from 1990 to 2007 to investigate the effects of related lending on private credit growth. As noted above, it is difficult to observe related lending in practice. Regulatory authorities in many countries do not require banks to report the percentage of loans made to related parties, and even those that do have such requirements do not employ uniform definitions of a related party, so measuring the extent of related lending across countries is challenging. We therefore construct a proxy measure—an index of the permissiveness of related lending that summarizes whether regulators are tolerant of cross-ownership between banks and non-financial firms and whether restrictions on the ownership of bank capital by related parties or a single owner are less binding. We find that our index of related lending, on average, does not have any effect on the growth of credit. We do find, however, that there are conditional relationships: related lending tends to retard the growth of banking systems when the rule of law is weak, while it tends to promote the growth of banking systems when the rule of law is strong. And these conditional relationships are economically important—a country that ranks relatively low on the rule of law (the 25th percentile) that goes from zero to 100 on our index of related lending decreases private credit growth by 6.8 percentage points (Figure 1). By contrast, for a country at the 75th percentile on rule of law, private credit growth accelerates by 4.8 percentage points when the related lending index moves from zero to 100.
This leaves policy makers in developing countries in a quandary. The choice of appropriate policy is made complex by the facts that (1) developing the rule of law requires many years, and (2) depositors cannot detect abuses associated with insider lending. A series of additional empirical tests offers clues about situations in which related lending is most likely to lead to abuses by insiders, and those clues could offer some guidance to policy makers. For example, related lending appears most likely to lead to abuses in ethnically fractionalized societies. It also appears that ownership of banks by non-financial firms poses a greater threat of looting than the reverse—the ownership of non-financial firms by banks—and that this threat is most pronounced in times of systemic crisis. Finally, expanding the official powers of bank supervisors does not appear to root out related lending abuses, especially in less developed countries.
Taken as a group, our results indicate that there is no single “best policy” regarding related lending. Whether or not policy makers should deter bankers from extending credit to themselves and their business associates crucially depends on how well they can adapt those policies to the particular country in question.
Figure 1
Further Reading:
Cull, Robert, Stephen Haber, and Masami Imai, 2011. “Related lending and banking development.” World Bank policy research working paper 5570.
Join the Conversation