Page 494 - untitled

Peter J. Wallison 465

of the NTM problem or recognized its signifi cance before the bubble defl ated.
Th e Commission majority’s report notes that “there were warning signs.” Th ere
always are if one searches for them; they are most visible in hindsight, in which the
Commission majority, and many of the opinions it cites for this proposition, happily
engaged. However, as Michael Lewis’s acclaimed book, Th e Big Short, showed so
vividly, very few people in the fi nancial world were actually willing to bet money—
even at enormously favorable odds—that the bubble would burst with huge losses.
Most seem to have assumed that NTMs were present in the fi nancial system, but not
in unusually large numbers.
Even today, there are few references in the media to the number of NTMs that
had accumulated in the U.S. fi nancial system before the meltdown began. Yet this is
by far the most important fact about the fi nancial crisis. None of the other factors
off ered by the Commission majority to explain the crisis—lack of regulation, poor
regulatory and risk management foresight, Wall Street greed and compensation
policies, systemic risk caused by credit default swaps, excessive liquidity and easy
credit—do so as plausibly as the failure of a large percentage of the 27 million NTMs
that existed in the fi nancial system in 2007.
It appears that market participants were unprepared for the destructiveness of
this bubble’s collapse because of a chronic lack of information about the composition
of the mortgage market. In September 2007, for example, aft er the defl ation of the
bubble had begun, and various fi nancial fi rms were beginning to encounter capital
and liquidity diffi culties, two Lehman Brothers analysts issued a highly detailed
34
report entitled “Who Owns Residential Credit Risk?” In the tables associated with
the report, they estimated the total unpaid principal balance of subprime and Alt-A
mortgages outstanding at $2.4 trillion, about half the actual number at the time.
Based on this assessment, when they applied a stress scenario in which housing
prices declined about 30 percent, they still found that “[t]he aggregate losses in the
residential mortgage market under the ‘stressed’ housing conditions could be about
$240 billion, which is manageable, assuming it materializes over a fi ve-to six-year
horizon.” In the end, of course, the losses were much larger, and were recognized
under mark-to-market accounting almost immediately, rather than over a fi ve to six
year period. But the failure of these two analysts to recognize the sheer size of the
subprime and Alt-A market, even as late as 2007, is the important point.
Along with most other observers, the Lehman analysts were not aware of
the true composition of the mortgage market in 2007. Under the “stressed” housing
conditions they applied, they projected that the GSEs would suff er aggregate losses
of $9.5 billion (net of mortgage insurance coverage) and that their guarantee fee
income would be more than suffi cient to cover these losses. Based on known losses
and projections recently made by the Federal Housing Finance Agency (FHFA), the
GSEs’ credit losses alone could total $350 billion—more than 35 times the Lehman
analysts’ September 2007 estimate. Th e analysts could only make such a colossal
error if they did not realize that 37 percent—or $1.65 trillion—of the GSEs’ credit
risk portfolio consisted of subprime and Alt-A loans (see Table 1, supra) or that
these weak loans would account for about 75% of the GSEs’ default losses over 2007-

34 Vikas Shilpiekandula and Olga Gorodetski, “Who Owns Resident al Credit Risk?” Lehman Brothers
i
Fixed Income U.S. Securitized Products Research, September 7, 2007.