Before Making More Housing Policy, Examine the Data
By James R. Follain
Over the last four years, an alarming number of property owners have faced the prospect of foreclosure or have already abandoned their homes or investments. Neighborhoods throughout the country are strewn with vacant houses that have declining values and need expensive renovations.
How can the number of such houses be reduced, and the harm they are doing to communities be minimized? That’s one of the most daunting questions facing policymakers today.
The size of the inventory of distressed real estate varies widely among housing markets and can have substantial negative impacts on property owners, their families, lenders seeking to make new loans and deal with bad ones, the neighborhoods in which these properties are located, and the speed of the recovery of the housing market.
One of the major federal programs to address this issue is the Home Affordable Modification Program (HAMP), which reduces mortgage payments for qualified owners of distressed properties. One version of HAMP entails principal reductions for borrowers. These policies are designed to provide lenders and borrowers a less expensive alternative to the foreclosure process and to keep owners and families in their homes. Ideally, they can provide a “win-win” for both parties and speed the reduction in the number of distressed properties and their negative impacts.
However, they face two great challenges:
- One is the inherent complexity of the circumstances that lead to foreclosure. Some property owners face this situation because of imprudent decisions or greatly miscalculating the future path of house prices. Many may be unemployed, victims of the Great Recession.
- The second is the need to promote a recovery of the ailing housing market, while acknowledging that the response to today’s crisis may foster expectations about how future crises will be handled. The result may be that future property buyers will take on more risk than they should, assuming they won’t actually incur the costs of the risk. Economists often call this trait the “moral hazard” problem.
These challenges have been highlighted in two recent exchanges between the U.S. Treasury and the Federal Housing Finance Agency (FHFA), the current conservator for the two major government-sponsored enterprises (GSEs) providing home loans — Fannie Mae and Freddie Mac. One involves a July 31, 2012 letter to the Senate Banking, Housing, and Urban Affairs Committee from the Director of FHFA, Edward DeMarco. DeMarco writes that “Fannie Mae and Freddie Mac’s adoption of HAMP Principal Reduction Alternative would not make a meaningful improvement in reducing foreclosures in a cost-effective way for taxpayers.” He specifically states such a program would give “borrowers who are current on their mortgage a message that the government endorses forgiving a portion of mortgage debt if hardship can be demonstrated,” which, in a nutshell, is the moral hazard problem.
The Secretary of the Treasury, Timothy Geithner, was quick to reply in a July 31, 2012 letter. Geithner, while recognizing DeMarco’s legal authority to make such a decision, said, “I do not believe that it is the best decision for the country, because, as we have discussed many times, the use of targeted principal reduction by the GSEs would provide much needed help to a significant number of troubled homeowners, help repair the nation’s housing market, and result in a net benefit to taxpayers.”
Both men offered research to buttress their views in supporting documents by members of their respective staffs but, obviously, they reach different conclusions from the empirical work or, perhaps, weigh components of the evidence differently.
I have studied some of the evidence offered by both men (though some of it is proprietary and not available to the public). I have also conducted my own research about the impact of the distressed housing inventory on the economic recovery and followed closely an example of a mortgage modification program being applied in New York State via the Settlement Conference, a judicial process that seeks to promote mortgage modifications as an alternative to foreclosure. I have three strong opinions on the issue.
First and foremost, we need more empirical evidence before moving more aggressively forward with principal reductions. There is simply too much uncertainty about how people will respond to such programs and how soon housing markets will recover with or without them.
Second, an effective program is surely going to involve taking serious account of widely varying differences among local housing markets. These include differences in the size and concentration of the distressed inventory as well as in the underlying market conditions that affect the ability of housing prices to recover. (Some of these are discussed in an Institute interview.) For example, the huge run-up in house prices in places like Stockton, California, was likely based upon unsustainable assumptions about the growth in future population and the demand for housing, and led to excessive housing supply. As a result, it will probably take much longer for house prices to recover in Stockton than in places where housing price decline was fueled by a cyclical drop in employment. This is just one of many local factors that affect expectations about future house price appreciation and impact the success or failure of mortgage modification programs. In short, there is no “one size fits all” principal reduction alternative for all housing markets in this country. An effective national initiative to promote modifications should take account of varying local conditions.
Third, and more optimistically, there is growing availability of geographically granular data that I believe can be of great help in the study of this issue. I have championed the use and potential of these data in a series of articles with two colleagues entitled Lessons from the Data. With more time and resources devoted to the analysis of the policies using these new data, I am confident that a better PRA program can be developed and it will likely be highly tailored to specific areas and to borrower circumstances. We need to accumulate more data and do more research about the impacts of this policy before launching still more aggressive and expensive principal reduction programs. Such work will not lead to a flawless plan void of uncertainty about the impacts of mortgage modifications. But, in my opinion, more research that exploits the extraordinary and emerging data available to study this topic can help policymakers build better remedies.
Of course, such opinions reflect my passion for empirically oriented research and impatience with those who champion moving ahead quickly, with little concern for potential secondary, negative outcomes. They also reflect a widely accepted fact about the house price boom and bust of the early 2000s: many market participants, regulators and economists failed to adequately appreciate and surely did not anticipate the enormous negative consequences of a sudden, dramatic decline in house prices within a finance system built upon relatively untested mortgage types and highly complex and interdependent mortgage-backed securities. In short, not nearly enough research was done prior to the crisis to forecast what might occur.
Which leads me to ask: Have we not already seen the limitations of policies in the housing market that rest upon relatively limited research?