The voices of Urban Institute's researchers and staff
September 20, 2017

How HARP saved borrowers billions and improved the housing finance system

Before 2009, borrowers were unable to refinance if they had little or no equity in their homes, even if they were current on their mortgage payments. This left many stuck in loans with 6 percent interest rates, even as rates dropped below 4 percent, costing borrowers significant monthly savings and the struggling economy much-needed stimulus.

In 2009, the government-sponsored enterprises (GSEs), Fannie Mae and Freddie Mac, introduced the Home Affordable Refinance Program (HARP) to address this problem for the loans that they guarantee, allowing borrowers with little to no equity in their homes to refinance into new loans, often reducing their monthly payments significantly.

HARP struggled early on, as many of the rules that the GSEs, mortgage insurers (MIs), and lenders had put into place over the years to manage their traditional refinancing risk locked out the borrowers that this program was intended to help. So, in 2012, policymakers, regulators, and key industry participants worked together to overhaul the program.

The impact of HARP since has been huge, making it arguably the most successful housing policy initiative coming out of the crisis. The Federal Housing Finance Agency’s (FHFA) refinance report indicates that during an eight-year period from 2009 to 2017, about 3.5 million HARP refinances were completed.

Nearly 2 million refinances (57 percent of the total) were completed in the two years following the revamp alone. On each refinance in the program, the average borrower saves 1.66 percent on their interest rate and almost $200 in monthly payment, generating total savings thus far in excess of $35 billion.

Participation in the program is limited to borrowers who took out their original loan before the program was launched, a population that has dwindled in recent years. This has led to a fall-off in the program’s numbers: a total of 10,000 HARP refinances were completed in second quarter of 2017 (Q2 2017) in comparison with over 300,000 at its peak in Q3 2012.

Originally set to expire at the end of 2013, HARP was extended several times to allow as many eligible borrowers to refinance as possible, with the most recent extension taking the end date out to December 31, 2018. We expect volumes to continue to dwindle, however.

The FHFA’s estimate as of March 2017 indicates 143,000 additional borrowers can still benefit from HARP, though even this modest sum overstates the number of new borrowers who will ultimately participate. The borrowers who are left in the eligible pool have passed over numerous opportunities to refinance at much more favorable interest rates than we have today, making them unlikely candidates for participation going forward, particularly if rates continue to rise.

HARP's legacy

The impact of the program reaches beyond the 3.5 million borrowers who’ve participated, and even beyond the more than $35 billion in borrower savings. Overhauling the original HARP program required policymakers and industry participants to cut through prohibitive obstacles that not only held this program back, but slowed refinancing down more broadly.  

To see how, it helps to look more closely at how policymakers finally fixed HARP.

The key to making HARP work effectively was to reduce the costs and risks in refinancing borrowers with little to no equity in their homes, so that these loans would be offered to more borrowers and at a lower cost. But three significant impediments stood in the way:

  1. Lenders had to get a manual appraisal for the new loan, adding hundreds of dollars of cost per loan and risks arising from committing to the property’s value.
  2. Lenders had to secure new mortgage insurance on the new loan, adding more transaction costs and the risk that no mortgage insurer would want to take on the additional risk of a borrower with a high loan-to-value percentage (LTV).
  3. Lenders didn’t want to take on the risks associated with underwriting a new high-LTV borrower. This meant that borrowers could only get a HARP loan from their existing lender, and without competition over their loan, borrowers would see worse pricing and less savings.

The first step in addressing these problems was recognizing that they arose because of rules designed to mitigate risks in loans made to new borrowers, not those made to borrowers for which the GSEs or MIs already held the credit risk. In backing a loan made to a new borrower, the GSE or MI needs to be comfortable with a whole host of underwriting issues that clarify precisely what risk they are being asked to take.

But where they already hold the credit risk, as they do for borrowers looking to refinance through the HARP program, the GSE doesn’t need to know with precision the updated value of the property, because it already owns the risk; similarly, the MI doesn’t need to decide whether the risks posed are worth backing, because it already owns the risk; and the GSE doesn’t need to apply the same stringent underwriting process rules that it applies to filter out risks that it doesn’t want to take, because it already owns the risk.

Indeed, to the degree rules in place to address these concerns keep borrowers from participating in the program, they leave the GSEs and MIs exposed to greater risk of borrower default, because they are keeping borrowers from reducing their monthly mortgage payment.

With this in mind, policymakers adjusted the rules:

  1. The GSEs would produce appraisals through their automated valuation system, which provided accurate-enough valuations for the required mortgage-backed securities disclosures.
  2. Mortgage insurers would transfer their coverage from the old loan to the new one, avoiding all of the costs and frictions of running an entirely new approval process.
  3. And the GSEs would reduce the underwriting assurances they required of lenders making HARP loans, even if the borrowers were coming from other lenders.

Once these three steps were taken, lenders were able to automate their participation in HARP, not only for their own borrowers but for borrowers currently serviced by other lenders, leading to a dramatic increase in the number of borrowers who benefited and how much they saved in doing so. This has in turn benefited the GSEs as well, because in expanding and deepening borrower payment reductions, the GSEs are reducing the default rates for many of their higher-risk borrowers.

Of course, the logic behind the lessons learned here applies not only to HARP loans, but to high-LTV loans backed by Fannie and Freddie more broadly. So the FHFA and the GSEs have extended the steps taken to fix HARP to remove similar impediments in refinancing for all borrowers who take out a loan backed by the GSEs after October 1, 2017, have no more than 5 percent equity in their home, and have been paying on time for at least 15 months (for specifics of the programs, see here and here).

Unlike HARP, which was always intended to be temporary, the new programs will be permanent, making it easier for all borrowers who find themselves in this predicament going forward to refinance at competitive rates, putting more money in their pockets each month, lowering the risk to the GSEs and stimulating the economy.

And for that, we have HARP to thank.

Dickey and Arthur Cook saved $480 a month by refinancing their under water mortgage using the government's Home Affordable Refinance Program. The couple is shown in front of their home in Corona on April 17, 2013. Photo by Glenn Koenig/Los Angeles Times via Getty Images.


As an organization, the Urban Institute does not take positions on issues. Experts are independent and empowered to share their evidence-based views and recommendations shaped by research.