November 2008, Asli Demirguc-Kunt (DECRG)
On October 2-3, the World Bank and IMF research departments sponsored a conference on, “Risk Analysis and Management,” in Washington, DC which brought together staff, academics, policymakers and industry participants for a lively discussion. A wide range of topics was covered including securitization, contagion, and implications for bank regulation, all discussed within the context of the current financial crisis. Below, we provide some highlights.
Summarizing recessions, credit crunch episodes, and asset price busts for a sample of 21 OECD countries from 1960 to 2007, Stijn Claessens, M. Ayhan Kose, and Marco Terrones suggest that the current crisis is a rare one. Only three of the 122 recessions that they tracked achieved the triple crown: busts in both housing and equity prices coinciding with a severe credit crunch. If the past is a reliable guide about such ‘perfect storms,’ their analysis indicates that the depth and duration of the current crisis is likely to be severe. The reduction in residential investment and housing prices was already more severe than in past recessions in both the U.S. and other OECD countries (Figure 1). The same is also true for the reduction in credit. Moreover, their evidence indicates strongly that financial market difficulties persist over long periods, typically much longer than the recession itself.
In his discussion, Vincent Rhinehart pointed out that the depth and severity of the crises in their database depended on the policy choices that were made, and thus it would be useful to provide some rough cuts of the data based on the policy paths that were chosen. He also questioned whether the past would be a perfect indicator of what we can expect. Globalization of the financial services industry and coordinated actions across countries by financial regulators might mean that the crisis charts a different course. As to whether it will be a more benign course, opinions differed.
Papers also discussed how the causes of the crisis were rooted in incentive problems and potential reforms that would improve incentives by increasing transparency and accountability in government and industry. For example, Jerry Caprio, Asli Demirguc-Kunt, and Ed Kane pointed out that Fannie Mae and Freddie Mac were pressured by the U.S. Government (namely by HUD) to assume greater and greater quantities of mortgage loans and securitizations of dubious quality. In part, they were egged on by the competition among credit ratings organizations that produced unrealistically high ratings for many of these instruments.
Atif Mian, Amir Sufi and Francesco Trebbi also point to political factors at the root of the crisis. During the subprime mortgage boom of 2001-2006, representatives from districts with a higher share of subprime borrowers experienced sharp increases in campaign contributions from the mortgage industry, which in turn resulted in the passage of legislation that further contributed to the boom. As the crisis erupted, representatives from the districts with high default rates were much more likely to vote favorably for the various rescue plans. Politicians did therefore respond to the demands of their constituencies for mortgage relief, after controlling for their political party and ideology.
Part of the problem rests with securitization itself, because as Jerry, Asli, and Ed noted, outsourcing the funding side of an originator’s balance sheet undermines its incentives to monitor the quality of the loans it originates. However, since the decision to securitize a specific group of loans is non-random, providing empirical support for that proposition is difficult. Benjamin Keys, Tanmoy Mukherjee, Amit Seru, and Vikrant Vig solve this problem by examining default rates on loans with FICO scores near 620. A market rule of thumb meant that loans that fell just above the cut-off were securitized while those below were not. Securitization does appear to have led to lax screening by lenders. Default rates for the securitized loans were 20% higher than on the non-securitized (but very similar) loans.
A distinguished panel of economists reflected on the implications of the crisis for developing countries, and global lessons from the crisis for crisis management, structure of financial systems and financial regulation and supervision. Justin Lin emphasized that the effects are likely to vary widely across developing countries. The most vulnerable are those that came into the crisis in a weak fiscal position, with a large current account deficit, and heavily reliance on exports for growth. Vincent Rhinehart pointed out that the decline in equity prices is much larger than for housing prices, which he attributed in part to inconsistent policies on the part of the U.S. government in dealing with insolvent institutions. Those inconsistencies led to the credit crunch as the increase in counter-party risk made actors reluctant to deal, effectively deterring private capital from entering to restore viable but damaged institutions. He also cautioned that it is best for the government to deal with problems on a firm by firm basis, since a targeted program for weak institutions is likely to subsidize all. Claudio Borno (BIS) stressed the need to shift to a macro prudential regulation framework, focusing on the health and stability of the system as a whole, rather than only inspecting individual institutions. As part of that framework, one proposal is for regulation of financial institutions to be counter-cyclical with institutions building buffers during good times to draw down during bad. Finally, Mark Zandi (Moody’s) was perhaps the most upbeat of the panelists, pointing out that a number of hedge funds were emerging out of the ashes of the crisis and already moving in to purchase foreclosed properties, and thus some sources of private capital might yet play a role in containing the crisis.