Progressives and conservatives both want private capital to assume more responsibility in the home mortgage market. But conservative proposals for reforms based on the unfounded assertion that private financial institutions are ready to enter segments of the market formerly covered by a government guarantee could be profoundly detrimental to our economic recovery.
Congress last month began a precarious test of the strength of the U.S. housing market by letting some federal backing for residential mortgages disappear. On October 1, the maximum loan amount eligible for government guarantees, or conforming loan limits, dropped in areas containing more than two-thirds of the U.S. population. The new limits reduced the size of mortgages eligible for Federal Housing Administration insurance and for Fannie Mae and Freddie Mac (the two mortgage giants currently under government conservatorship) to purchase or securitize. Congress has since voted to restore FHA’s higher limits, but left those of Fannie and Freddie at their new, lower levels.
The changes were expected to affect only between 1 percent and 2 percent of the mortgage market, according to analysis from the Federal Reserve and New York University’s Furman Center for Real Estate and Urban Policy. The critical question was whether the private sector would fill the gaps created by scaling back the federal government’s role.
At this point there is no reliable evidence on how the new loan limits will affect the housing market. But that hasn’t stopped some conservative analysts from trying to prove their point.
Edward Pinto, a resident fellow at the American Enterprise Institute, recently claimed that the housing market has actually been helped by the new loan limits. He points to a chart issued by the analytics firm Loan Processing Services that shows a recent surge in “other” (not government-guaranteed) lending in August as evidence that the private market is ready to fill in gaps left by Fannie and Freddie. But he blatantly misuses the data.
First, Pinto’s analysis relies on loan data that lags behind the mortgage market, which is important because it often takes weeks or even months for a loan to process. Indeed, even the organization that compiled the dataset highlighted the danger of using data on the most recent loans. During a conference call when LPS released its latest data, LPS Vice President of Applied Analytics Herb Blecher said that the “other” totals were misleading “because it takes time for loans to transfer to the appropriate investor.” That’s why we see a “larger concentration in this … ‘other’ category than will ultimately be the case when all the loans are fully boarded,” Blecher said.
In reality, many of the loans labeled as “other” in the LPS dataset are destined for Fannie or Freddie in the coming weeks, so Pinto overestimates the private sector’s role. What’s more, the LPS dataset also includes refinances, which were the driving force behind the boost in August lending. But increased refinancing does nothing to support a badly needed recovery in the home purchase market at a time when the FHA, Fannie, and Freddie backstop about 90 percent of all home mortgage loans. The appropriate statistic is not total loan activity but existing home sales, which actually decreased between August and September in the lead-up to the new loan limits.
Not only is Pinto’s analysis an invalid use of unreliable data, it is also an unfounded assertion of causality. The LPS data show that between July and August 2011 “other” entities started to take on more of the U.S. mortgage market. Pinto attributes this to changes in loan limit guarantees from the FHA, Fannie, and Freddie, which didn’t take effect until October 2011. That timeline simply doesn’t make sense for causality.
To be sure, many mortgage originators stopped accepting loan applications prior to the October 1 deadline to prepare for the upcoming changes. But most sales in August were likely based on mortgages applications filed earlier and under the old rules. Are we to believe that applications for the loans that closed in August were constrained by expectations that these conforming loan limits would drop months later? That’s a bit of a stretch.
Even if this “other” segment of the market is actually growing, Pinto’s analysis says nothing about what proportion of these new private mortgages would have been eligible for a government guarantee under the old loan limits. If his hypothesis is correct, then one would expect to see a high percentage of these new loans fall between the previous limit and the current one—say between $625,000 and $729,750 in New York City. But Pinto presents no such finding.
The upshot: We simply do not have the data to know whether the private sector filled any gaps in the mortgage market caused by the short-lived decrease in loan limits. Perhaps we will one day, but the window will likely be too small to glean any real conclusions. In any case, we must not be blind to the fact that the faith and confidence of the federal guarantee is the foundation keeping our weak housing market standing today.
In the coming months, policymakers can monitor the impact of lowering the conforming loan limits, specifically the extent to which either or both FHA and the private sector take over segments of the market formerly covered by Fannie and Freddie. As Congress continues its risky experiment with a very weak U.S. housing market, it would be irresponsible to declare success without real evidence that private firms alone can pick up the slack.
Mark Willis is a research fellow at the Furman Center for Real Estate and Urban Policy. Janneke Ratcliffe is a Senior Fellow at the Center for American Progress. John Griffith is a Research Associate with the Economic Policy team at the Center.