It is well known that subprime mortgages in the U.S. housing market were a key contributor to the Great Recession. What is less clear is who should be blamed for the proliferation of these mortgages and their increasingly important role in the banking industry. Was it the irresponsible consumer, spending beyond his or her means and taking on too much debt? Was it the loan officer, handing out mortgages to unsuspecting low-income borrowers on usurious terms? Was it the banker, securitizing subprime mortgages into asset-backed securities (ABS) that could be traded and used to fuel risky activities? Was it the rating agency, slapping AAA-ratings on assets backed by subprime loans with high default rates (for which they are now being investigated by federal prosecutors: http://online.wsj.com/article/SB10001424052970203721704577156963813900028.html)? Was it the insurer, providing unlimited amounts of insurance against loan default and lulling owners of mortgage debt into a false sense of security? Or perhaps it was it the regulator, letting bankers run awry with new and complicated assets without proper safeguards and controls?
A new research paper authored by the Federal Reserve claims that at least one group can likely be absolved of responsibility for the subprime mortgage crisis: government housing agencies. The paper (found here: http://www.federalreserve.gov/pubs/feds/2011/201136/201136pap.pdf) finds that federal government policies aimed at increasing homeownership did not lead to more risky lending strategies and therefore did not contribute to the subprime mortgage bubble. The paper examines areas where more loans were made by government housing agencies and finds that these areas actually experienced lower delinquency rates and less risky lending practices than those areas with less government agency presence (the paper also talks about the Community Reinvestment Act of 1977, but I will focus on the government agencies in this post since they seem to have received more public attention). The paper attempts to prove not that agency housing goals led to more loans to low-income borrowers (which they inevitably did), but instead that these loans were not any risker than those that would have been made regardless of agency goals – in fact, they were in many ways less risky.
For background, the government sponsored entities (GSEs) Fannie Mae and Freddie Mac have received a lot of flack for their business practices and have been blamed as key contributors to the crisis. Fannie and Freddie are responsible for buying mortgage loans from banks and securitizing pools of these loans into asset-backed securities, which they then sell to various institutions. They do not issue mortgages off of their own balance sheet. By purchasing loans from banks and spreading out the risks of mortgage delinquency and default (if many loans are packaged into one product and distributed throughout the economy, risk is spread among more people and it is less likely that a few concentrated defaults will bring down an entire bank), Fannie and Freddie provide liquidity to the mortgage market and enable banks to make more loans at lower rates.
The two GSEs were encouraged by the federal government to increase mortgage lending to low-income families, namely through the 1992 GSE Act. The Act set quotas requiring a certain percentage of the mortgages purchased by the GSEs to be loans made to low-income borrowers. The percentage was chosen every year by the federal government and steadily rose over time. Some claim that this policy required Fannie and Freddie to encourage banks to relax their underwriting standards and issue mortgages to those less able to make their payments on time, mortgages that would eventually become subprime. A downward spiral of more delinquencies, defaults, and subprime mortgages began to take hold as homeowners were increasingly unable to make payments. Lax underwriting standards may have fueled the housing bubble by increasing demand for home ownership and driving up home prices. When the bubble burst, many low-income borrowers were forced to default or become delinquent on their mortgages. If Fannie and Freddie hadn’t encouraged weaker lending practices, the story goes, much of the concurrent downward spiral in defaults and upward spiral in home prices could have been avoided.
The Fed paper shows that this claim is tenuous at best. There are many ways that Fannie and Freddie could have achieved their goal of buying more loans made to low-income borrowers without forcing banks to lower their underwriting standards. Just proving that loans to low-income borrowers (many of which would eventually be subprime) increased does not prove that GSE policies caused the increase in subprime mortgages. A causal link must be established. To see if such a link exists, the Fed looked at areas of the population that would be most affected by GSE policy and measured the percentage of borrowers late on their mortgage payments by at least 90 days between 2004 and 2006 (the run-up to the crisis) as the independent variable, with delinquency rates in 2008 as the dependent variable. If GSE policy induced banks to make loans to less qualified borrowers than they otherwise would have, then areas with more loan sales to the GSEs should see more bad outcomes down the line – aka more subprime mortgages in 2008. The Fed also measures house price changes to determine if GSE policy contributed to the housing bubble by increasing demand for housing in areas affected by GSE policy.
The results of this analysis show that mortgage lending by banks in areas most affected by GSE policy was actually associated with lower delinquency rates in 2008, or less subprime mortgages, than areas with more lending by independent mortgage banks. GSE policy may actually be associated with better, not worse, loan policies (although not at very statistically significant levels). The analysis also shows that GSE policy is not strongly associated with an increase in housing prices in the run up to the crisis.
It is easy to understand why GSE policy has been blamed as a contributor to the subprime mortgage crisis. The policy encouraged lending to more low-income borrowers so that the GSEs could meet their quotas for mortgage purchases, and low-income borrowers were more likely to become delinquent or default on their mortgages in 2008. Since more loans were made to low-income borrowers leading up to 2008, subprime mortgages began to build up until the bubble burst and the crisis hit with full force. Showing that GSE policy directly affected the increase in subprime mortgages, however, is more difficult, and the results suggest that they did not. Granted, Fannie and Freddie had to be bailed out by taxpayers and are still a drain on government resources, so this isn’t to suggest they have not been part of the problem at all. They just didn’t directly or materially increase the amount of subprime mortgages or contribute to the housing price bubble.
As we discussed in the previous post, economics is largely value-free on its own but is often used by policymakers and others as a vehicle to understand and promote favorable economic actions, however they may be defined. No matter who is rightly to blame (or not to blame) for the housing crisis, it is certain that regulation was insufficient and perverse incentive structures led lenders and borrowers to inefficient outcomes. Although Fannie and Freddie may be off the hook according to the Fed paper, policymakers in general are hardly off scot-free. Lack of proper regulation of the banking industry undoubtedly contributed to the recession. Ultimately, all of the parties named at the beginning of this post are likely blameworthy in some way.