By Kat Aaron and Mary Kane
October 8, 2010
With more than 1 million total foreclosures predicted this year, the government is finally taking steps toward resolving a vexing dilemma: It has very few details about this very big problem.
Since the subprime mortgage market collapse in 2007, regulators, Congress, and consumer advocates have relied on what one housing expert describes as "a lack of even marginally accurate or complete data" on the level and nature of foreclosure activity. Incredibly, almost three years into the collapse, there is no nationwide, government- collected data on foreclosures.
"We can tell you how many grapefruit are grown in every state in the country. But we can't tell you how many mortgages are being made, how many individual loans, and even the characteristics of those loans," says Guy Cecala, publisher of Inside Mortgage Finance, an industry publication. "We've got much better agricultural data in this country than financial data."
In the absence of a government database, corporate providers serve as the main source for numbers on foreclosures. This information relies on sampling and is proprietary, expensive, and not standardized. The firms piece together their data from industry and public sources, including county assessors' offices, courthouse filings, newspaper listings, auction notices, and loan reporting from mortgage companies. Some firms compile their numbers from actual completed foreclosure sales, while others track notices of default, which occur much earlier in the foreclosure process.
But the foreclosure information gap is about to change. At least, it's supposed to.
Tucked into the financial-reform bill passed in July is a requirement for a new national foreclosure database. Buried in the 1,447th section and offering few specifics, the requirement calls for the creation of a database aggregating delinquency, defaults, properties owned by banks, and houses that are underwater.
Consumer and neighborhood activists say a national foreclosure database is a welcome development. It could be used to detect foreclosure trends and allow for quicker responses from policy-makers. It could track the behavior of individual lenders and servicers, including what percentage of a lender's loans go into foreclosure. It also could offer details on the buyers of foreclosed homes. Just understanding where foreclosures are happening isn't enough. To craft effective solutions, regulators need to understand details of the mortgage, like the total debt owed to various lien holders and whether a homeowner is underwater and has a second or third mortgage on top of the original loan.
However, some remain skeptical. Whether the government finally compiles a usable foreclosure database, critics say, will depend on a long list of conditions, including whether it ends up fully funded, requires lenders and servicers to participate, and is accessible. None of which, of course, is a given. And the agency tasked with implementing the database, the Department of Housing and Urban Development, has not inspired confidence among consumer advocates.
"I think potentially, this could be fabulous," says Alan Mallach, a senior fellow at the National Housing Institute and a nationally known housing and community-development expert. "What you need to make this happen is an entity that is really committed both to making it happen and to making sure the data is put out there in a way that is really usable by people and not just academics. And to do that costs a lot of money."
If the government wants to try to put together a useful database, it should begin collecting loan-servicing and origination information like escrow payments, payment due dates, and late-payment fees, Cecala says. When a database finally gets built three or four years from now, servicing mortgages will be the challenge, not new foreclosures. "This is the chance to get all this stuff," he says.
And, to be particularly useful, the information would have to be made available in as close to real time as possible -- at least quarterly. Right now, the home-mortgage data the government collects has a lag time of nearly a year.
The smartest thing Washington can do is consult with local neighborhood and community-development groups about what shape the database should take, says Geoff Smith, senior vice president at the Woodstock Institute, a Chicago-based community-development organization. His organization, and others like it, has a long track record of collecting on-the-ground information on foreclosures and bank-owned homes.
"Sometimes things in Washington get done in a bubble," Smith says. Congress' decision to put the database in the hands of the Department of Housing and Urban Development has raised eyebrows in the consumer-advocate community. HUD will develop the database in consultation with the new Consumer Financial Protection Bureau. Even groups that worked closely on the bill can't explain that decision. Because the Consumer Financial Protection Bureau now will be collecting loan data under the Home Mortgage Disclosure Act (HMDA), the foreclosures database would have been a natural fit at the new bureau, some consumer advocates say.
"I am not sure why it is HUD," says Josh Silver, a policy researcher at the National Community Reinvestment Coalition. "I would have made it the same agency, the consumer bureau, that has control of HMDA. But the data is now a legal requirement, and we will work to make it the most robust data possible."
A scathing 2009 report from the Government Accountability Office criticized HUD's information-technology systems as outdated and inefficient and noted a troubling turnover rate of IT employees. At the neighborhood level, HUD has come under fire for letting its own foreclosed properties fall into disrepair or be sold off to speculators.
"The notion of putting HUD in charge of this has all sorts of implications," says Cleveland housing law professor Kermit Lind. "I'm still trying to think of a positive one."
Adds the National Housing Institute's Mallach: "Notwithstanding my respect for the people in the top jobs there, [HUD] is still a dysfunctional agency with only limited capacity and resources for something like this."
However, mortgage-industry consultant and former Fair Housing Administration (FHA) official Brian Chappelle is less concerned.
"I do think HUD can manage the foreclosure database. For all of their antiquated systems, FHA has not had one system glitch over the last several years even though their business has quintupled. If they are given the funding, HUD will be able to implement the system," Chappelle says.
HUD spokesperson Brian Sullivan says the agency itself has had to rely on a range of public sources in crafting its programs and has struggled with missing and incomplete foreclosure data. He has no details yet on how the database would be designed or funded, which is also not mentioned in the legislation. Given the challenges in collecting the data, costs could be high. Mallach, for example, offers a conservative estimate of between $10 million and $30 million.
"It will be fascinating to see how expensive it becomes to set up the government infrastructure necessary to compile, authenticate, manage, and maintain this database," says Rick Sharga, senior vice president at RealtyTrac, one of the nation's leading sources for foreclosure data. "I'm sure there will be some heavy-duty lobbying going on as vendors fight it out to see who gets what will probably be a pretty big contract."
Merging numbers between the foreclosure database and other related financial databases is yet another challenge. Banks and regulators collect data in a range of formats, making it hard to compare one set of information to another. Rep. Darrell Issa of California tried to change that by introducing an amendment to the financial-reform bill requiring all agencies to report financial data in one standardized format known as XBRL. That amendment was stripped out of the final legislation.
Servicers, loan originators, and Fannie Mae and Freddie Mac all have loan details that would be useful for a future database. But those entities, and others, likely will contend that standardizing the databases would be too costly and difficult, Cecala says, voicing the same argument that has for the past few decades made mortgage details harder to collect than counting grapefruit.