A 2000 United States Department of the Treasury study of lending trends for 305 cities from 1993 to 1998 showed that $467 billion of mortgage lending was made by Community Reinvestment Act (CRA)-covered lenders into low and mid level income (LMI) borrowers and neighborhoods, representing 10% of all U.S. mortgage lending during the period. The majority of these were prime loans. Sub-prime loans made by CRA-covered institutions constituted a 3% market share of LMI loans in 1998, but in the run-up to the crisis, fully 25% of all sub-prime lending occurred at CRA-covered institutions and another 25% of sub-prime loans had some connection with CRA. In addition, an analysis by the Federal Reserve Bank of Dallas in 2009, however, concluded that the CRA was not responsible for the mortgage loan crisis, pointing out that CRA rules have been in place since 1995 whereas the poor lending emerged only a decade later. Furthermore, most sub-prime loans were not made to the LMI borrowers targeted by the CRA, especially in the years 2005–2006 leading up to the crisis. Nor did it find any evidence that lending under the CRA rules increased delinquency rates or that the CRA indirectly influenced independent mortgage lenders to ramp up sub-prime lending.
To other analysts the delay between CRA rule changes (in 1995) and the explosion of subprime lending is not surprising, and does not exonerate the CRA. They contend that there were two, connected causes to the crisis: the relaxation of underwriting standards in 1995 and the ultra-low interest rates initiated by the Federal Reserve after the terrorist attack on September 11, 2001. Both causes had to be in place before the crisis could take place. Critics also point out that publicly announced CRA loan commitments were massive, totaling $4.5 trillion in the years between 1994 and 2007. They also argue that the Federal Reserve’s classification of CRA loans as “prime” is based on the faulty and self-serving assumption: that high-interest-rate loans (3 percentage points over average) equal “subprime” loans.
Others have pointed out that there were not enough of these loans made to cause a crisis of this magnitude. In an article in Portfolio Magazine, Michael Lewis spoke with one trader who noted that “There weren’t enough Americans with [bad] credit taking out [bad loans] to satisfy investors’ appetite for the end product.” Essentially, investment banks and hedge funds used financial innovation to enable large wagers to be made, far beyond the actual value of the underlying mortgage loans, using derivatives called credit default swaps, collateralized debt obligations and synthetic CDOs.
As of March 2011 the FDIC has had to pay out $9 billion to cover losses on bad loans at 165 failed financial institutions. The Congressional Budget Office estimated, in June 2011, that the bailout to Fannie Mae and Freddie Mac exceeds $300 billion (calculated by adding the fair value deficits of the entities to the direct bailout funds at the time).
Economist Paul Krugman argued in January 2010 that the simultaneous growth of the residential and commercial real estate pricing bubbles and the global nature of the crisis undermines the case made by those who argue that Fannie Mae, Freddie Mac, CRA, or predatory lending were primary causes of the crisis. In other words, bubbles in both markets developed even though only the residential market was affected by these potential causes.
Countering Krugman, Peter J. Wallison wrote: “It is not true that every bubble—even a large bubble—has the potential to cause a financial crisis when it deflates.” Wallison notes that other developed countries had “large bubbles during the 1997–2007 period” but “the losses associated with mortgage delinquencies and defaults when these bubbles deflated were far lower than the losses suffered in the United States when the 1997–2007 [bubble] deflated.” According to Wallison, the reason the U.S. residential housing bubble (as opposed to other types of bubbles) led to financial crisis was that it was supported by a huge number of substandard loans – generally with low or no downpayments.