The Big Lie about the financial crisis is perpetuated by a series of phony factoids. This one is the false connection between government housing policy and low down payment loans.
Michael Cembalest, the Chief Investment Officer of JPMorgan, has taken a shine to the discredited research of Edward Pinto of The American Enterprise Institute. Pinto points out that Cembalest, “wrote an article that makes use of research from my Government Housing Policies in the Lead-up to the Financial Crisis: A Forensic Study relating to diminishing down payments and Fannie’s competition with FHA.”
Cembalest, like Pinto, asserts that Fannie Mae and the FHA, pursuant to government policy, extended an extraordinary amount of loans to borrowers who put virtually nothing down. Therefore, the government enterprises inflated the housing bubble. There were just two problems with Pinto's "research," however. First, Pinto’s numbers on Fannie’s diminishing down payments were little more than wildass guesses; they weren't based on actual data. Second, his assumptions about diminishing down payments had no connection whatsoever to government housing policies or Fannie’s underwriting guidelines. It was yet another case where a right wing "think tank" fabricates a phony factoid to deflect away Wall Street's culpability.
Pinto refers to chart titled, “A look back at the origins of the housing crisis,” on page 4 of the November 1, 2011 edition of J.P. Morgan’s Eye on The Market. It’s one of those hockey stick charts, where the line suddenly shoots up to show wild exponential growth. It shows the percentage of home purchase mortgage loans a minimum combined loan-to-value of 97%, i.e. borrower home equity at 3% or less, that were originated by FHA and Fannie Mae. The percentage of 3%-down mortgages rises sharply, from about 10% in 2000 to almost 40% in 2007. That’s right; by 2007 almost 40% of the mortgages were extended to homebuyers who put almost nothing down.
Cembalest lifts the numbers from page 26 of Pinto’s analysis of “Government Housing Policies in the Lead-up to the Financial Crisis,” where it shows that Fannie Mae’s originations of 3%-down mortgages skyrocketing after 2000. It also shows the percentage of high LTV loans extended by FHA, which we’ll get to in a second. FHA had a tiny share of the market during this period, so its numbers are far less meaningful.
Pinto had no idea what Fannie’s real numbers were. That is, he had no idea about the extent to which other lenders, not Fannie, offered 2nd lien piggyback loans at the time of the purchase. Although Fannie requires that information as part of its underwriting process, and it adjusts its pricing accordingly, it does not disclose the extent to which private market 2nd lien loans are attached to its first lien mortgages.
Fannie itself was subject to lending limits that were very well known. By law, Fannie could not take the primary credit risk on any mortgage loan in excess of 80% of the home’s appraised value. If a loan had an LTV higher than 80%, then the first loss was covered by private mortgage insurance. Also, Fannie’s policy was that if any loan had an LTV higher than 80%, then the insurance coverage must bring Fannie’s credit exposure well below 80%. This made sense, because higher LTV loans are more likely to default, and therefore demand a stronger comfort margin.
For example, if a loan had an LTV of 85%, the minimum insurance coverage was 12%, so that Fannie’s net risk exposure would be no more than 73% of the total. If a loan had an LTV of 95%, then the minimum insurance coverage was 30%, leaving Fannie’s net risk exposure at 65% of the total. It was the private insurance market, not government policy, which determined Fannie’s appetite for high LTV loans.
In addition, Fannie imposed a minimum down payment policy for home purchases. Its underwriting guidelines stated: “The borrower must make a 5% minimum down payment from his or her own funds on a purchase transaction…The lender must verify that the borrower is making the required minimum contribution; otherwise the case is not eligible for sale to Fannie Mae.” (So, for example, if a purchase transaction had 2% upfront closing costs, a 5% down payment might result in a net LTV of 97%). It’s also worth remembering that during the 2001-2007 period, the large majority of loans originated by Fannie were not for home purchases; they were refinancings of existing mortgages.
In 2007, about 10% of Fannie’s mortgage originations (both home purchase and refinance loans) had an LTV>90%, and within this segment, the average LTV of these loans was 97.2%. The number applied to all of Fannie’s 2007 originations, which were evenly divided between home purchases and refinancings. Pinto also referred to a well known 2004 study by SMR Consulting, which showed that piggyback loans for home purchase transactions rose to 42% in 2004. For all types of loans, the 2004 percentage was 22%.
So, based on what little he had, Pinto assumed that private lenders, like J.P.Morgan Chase, were willing to issue 2nd lien piggyback loans as a substitute for higher LTV loans covered by private mortgage insurance. Perhaps. And, perhaps that phenomenon turned out to benefit Fannie in the long run, since the once-AAA mortgage insurers have gone bankrupt. What remains indisputable is that private financiers accommodated demand for high LTV loans, not Fannie and not government housing policy.
As for FHA, it was a tiny part of the mortgage market during the height of the real estate bubble. It’s market share fell from about 8% in 2001 to about 2% during 2005-2007.
Although the FCIC eviscerated Pinto's "research," Cembalest has been a big fan of his work nonetheless. Pinto's Fannie/Freddie mythology is endlessly touted by FCIC Commissioner Peter Wallison, who was singled out by Barry Ritholtz as one of the chief proponents of The Big Lie. Wallison's penchant for lying has been well documented.