Engineering the Financial Crisis: Systemic Risk and the Failure of Regulation

Conclusion

CONCLUSION Because our argument has covered a lot of ground, a summary is in order. We present it with its vulnerabilities (as we see them) exposed; after the summary, we briefly consider counterarguments directed at those vulnerabilities . ‘‘Moral Hazard’’ and the Financial Crisis We began in Chapter 1 with the fact that the mortgage-backed bonds on the balance sheets of U.S. commercial banks appear to have been either guaranteed by the GSEs or rated triple-A. This well-known fact poses serious problems for the two most prominent hypotheses about the cause of the crisis, both of which are ‘‘moral-hazard’’ stories. According to one of the moral-hazard stories, bankers made bets on mortgage-backed bonds that they knew were risky because they also knew that they would be bailed out if disaster struck. According to the other story, they knowingly made bets on risky mortgage-backed bonds because they were paid to pursue short-term profit at the expense of long-term safety. In Chapter 1, we pointed out that there is literally no evidence that either of these hypotheses is true (see also Appendix I); and we noted that if either or both of them were true, it would be difficult to explain why banks bought low-yielding GSE-guaranteed and triple-A privately issued mortgage bonds, including triple-A CDOs, instead of much-higher-yielding double-A, single-A, triple-B, or double-B bonds or CDOs.1 Moreover, contrary to conventional wisdom, neither commercial banks in general nor the biggest of them were, on the whole, increasing their leverage to the legal limit in the years before the crisis (although some were indeed levering up, while others were deleveraging). For the moral-hazard stories to be true, bankers should have been taking risks that would pay off, and one way to increase the payoff on any bet is to use as much borrowed Conclusion 145 money as possible. In reality, bankers seem to have been trying to play it safe by betting on government-guaranteed or privately issued but lowyielding , tranched, overcollateralized mortgage bonds rated AAA, and by falling well short of using all the leverage they could to make these bets. The Basel Thesis The relatively low yields of the mortgage bonds and the relatively low leverage of the commercial banks form, respectively, the conceptual beginning and end of Chapter 2. First, we explained U.S. commercial banks’ and savings and loans’ $1.324 trillion in GSE-guaranteed and triple-A mortgage bond purchases (Table 1.1) as due to the 80 percent capital relief that these types of bonds received under the Basel I accords and the Recourse Rule, the latter of which applied only to U.S. banks. What is of particular interest is the $472 billion portion of this amount that was invested in AAA-rated ‘‘private-label’’ mortgage bonds (PLMBS). In 2001, the year in which the final version of the Recourse Rule was issued, PLMBS issuance increased by nearly 100 percent after having actually declined in 2000 (Figure 2.1). By the eve of the crisis, U.S. commercial banks had accumulated $383 billion of AAA-rated PLMBS and $90 billion of AAA-rated CDO bonds (Table 1.1). This amounted to 4.3 percent of the total assets of U.S. commercial banks and S&Ls, or three times the proportion of PLMBS and CDOs to be found in other U.S. investors’ portfolios (Table 2.3). The Mystery of Banks’ Steady Leverage Levels However, if banks were accumulating all of these relatively low-yielding PLMBS/CDOs primarily because of the capital relief they offered under Basel I and the Recourse Rule, why weren’t they using the capital relief by increasing their leverage, which remained steady (in the aggregate) and low during the years before the crisis? What is the point of accumulating lowyielding bonds that, under the Recourse Rule, allowed banks to borrow more if they did not then borrow more? Our answer, presented at the end of Chapter 2, turned on the distinction between usable and legally mandated capital cushions. A prudent unregulated bank would maintain a capital cushion because leverage can 146 Conclusion be dangerous: unpredictable defaults on its loans and losses on its other investments may prevent a bank from being able to pay back the funds (borrowed from depositors and other creditors) with which the bank has financed these loans...