CoreLogic recently awarded its Academic Research Council Excellence Award to a study conducted by professors from three universities who looked at reasons why so many homeowners fail to refinance their homes even when it is financially advantageous to do so.  The award for scholarly research in the real estate and mortgage fields was given to Benjamin J. Keys, Harris School of Public Policy, University of Chicago; Devin G. Pope, Booth School of Business, University of Chicago; and Jaren C. Pope, Department of Economics, Brigham Young University and their paper "Failure to Refinance."

According to the study, housing decisions can have substantial long-term consequences for household wealth accumulation because almost two thirds of the median household's total wealth is from housing.  Public policies have been crafted to encourage homeownership but their effectiveness hinges on homeowners' wise housing decisions.

One decision is the choice to refinance a home mortgage.  Failure to do so can cost a household tens of thousands of dollars in savings.  When mortgage rates reached all-time lows in the vicinity of 3.35 percent in late 2012 a household with a $200,000 mortgage and an interest rate of 6.5 percent would save roughly $130,000 over the life of the loan by refinancing.

Yet many did not, and while this appears puzzling the authors say it is consistent with recent work in behavioral economics. Homeowners may have many reasons for failing to refinance; bad credit or a lack of equity or a change of moving in the near future.  Often the benefits of refinancing are not immediate but accrue over time and there can be a number of costs, both financial and non-financial that households must pay in order to complete refinancing.    Further many households have limited experience with calculating the financial benefits they might receive.  

The authors attempt to provide empirical evidence about how many households appear to be suffering from a failure to refinance and the magnitude of their mistakes by analyzing a nationally representative sample of 1.5 million single family mortgages from the CoreLogic database that were active in December 2010.  They merged this with 2010 zip-code level census information on median income, education levels and other variables.  Given the information available the authors could calculate how many households would save money over the life of the loan were they to refinance at the prevailing interest rate.  The data also allowed them to reasonably separate out homeowners who could not refinance from those who sub-optimally fail to do so.

They concluded, based on what they called conservative assumptions, that about 20 percent of households in December 2010 had not refinanced when it appeared profitable to do so.  They also calculated that the median household holding on to that too-high rate mortgage would have saved approximately $45,000 over the life of the loan by refinancing or approximately $11,500 when adjusting for discounting for time and tax incentives. 

By December 2012, when interest rates reached historic lows approximately 40 percent of those non-refinanced 2010 households were still in their homes and still had not refinanced.   These results suggest that the size and scope of the problem of failing to refinance is large.   

A typical active loan in December of 2010 was paying 5.52% interest, had 23 years remaining and an unpaid balance of just over $200,000. The average loan-to-value ratio at origination was approximately 70% and in 2010 was 74%. While the average interest rate being paid is 5.52%, there is substantial variation with many households paying interest rates near the market rate in December 2010 (~4.3%) and other households paying interest rates well over 6%.   The distribution of rates was narrower in the sample restricted to those households that appeared able to refinance.

Assuming that all households could refinance in December 2010 at the prevailing rate of 4.3 percent the authors estimated the savings after adjusting by upfront costs and estimated that 91.4 percent could save money over the life of the loan by refinancing.  Taking into consideration the mortgage interest rate tax deduction, the probability of moving and discounting of money over time that was reduced to 41.2% of households in the full sample.  They also put the present discounted value of refinancing after all considerations at $13,000.

The 41.2 percent was further reduced to 31.1 percent by screening out households with low credit scores and/or high loan to value ratios at their loan's origination.  Further reductions were made by estimating damaged credit or diminished equity after origination.  The final estimate is that 20% of households in December of 2010 were sub optimally in a state of not refinancing.  The unadjusted savings available to this 20 percent of households averaged $45,473 although there was a great deal of variation in that number.  The present-discounted value of forgone savings was approximately $11,500.



 If interest rates had increased sharply starting in December 2010 the authors estimate that approximately 20% of households would have lost their chance to refinance even though it would have been optimal for them to do so. Interest rates, however, continued to decline through the end of 2012, providing an opportunity for the 20% of households that failed to refinance in December 2010 to do so and realize even greater savings from the lower rates.  However they found that 40 percent of households that should have refinanced in 2010 were still living in their house by December 2012, continued to make full and on-time monthly payments, yet had not refinanced despite the further decline in interest rates.

The available data does not allow the authors to provide detailed information about these households that fail to refinance despite the large financial stakes however it was possible to make some broad assumptions.  First, the failure to refinance is more prevalent among households that have worse credit, and slightly more prevalent in neighborhoods with lower education and income levels but the differences were small.

In an attempt to get more detailed information the authors partnered with a non-profit company called Neighborhood Housing Services of Chicago (NHS) which provides housing related services primarily to lower-income communities including homeowner education and foreclosure prevention.  NHS is also a mortgage lender and servicer and actively encourages clients to refinance when interest rates decrease.

In July of 2011, NHS sent a letter to 446 client households with an offer to refinance their current mortgage loan at a 4.7% interest rate with no up-front money required. The letters went only to households pre-determined to be eligible and who would benefit from refinancing.  Eighty-four percent of recipients did not respond to the offer; the 16 percent who did would go on to pay $24,500 less in total interest payments over the life of the loan while the 84 percent that did not respond saw forgone savings of $17,700.

NHS sent out a second letter in July 2012 offering 140 of primarily the same client households the opportunity to refinance at 3.99 percent and 75 percent did not respond, forgoing unadjusted savings of $24,700 while the 24.3 percent who did respond had savings of $29,900.

After a third salvo of 193 letters was sent in May of 2013 with only a 13 percent response rate the authors worked with NHS to conduct a survey by phone of non-refinancing households.  While the responses were thin, up to ¼ of households said they did not open the letter.  Of those that did 1/3 said they intended to pursue the issue but didn't get around to it.  Another third did not find the savings significant enough to warrant a call and about a third said they would be willing to have a loan officer call them to discuss a loan.

 The authors also cite the parallel with their research of the current government effort to encourage refinancing.  In 2009 the Treasury Department introduced the Home Affordable Refinance Program (HARP) designed to help current borrowers under federally guaranteed programs refinance even with high loan-to-value ratios and estimated that 4 to 5 million borrowers could take advantage of it.  By September 2011, however, less than a million borrowers had actually done so - remarkably similar to the results of this analysis.  Modifications to the program have encouraged greater participation but the total take up rate remains low.

The writers say the magnitude of the financial mistakes that households make suggest that psychological factors such as procrastination and the inability to understand complex decisions are likely barriers to refinancing. One policy that has been suggested to overcome the need for active household participation would require mortgages to have fixed interest rates that adjust downward automatically when rates decline. To the extent that it is undesirable to reward only those households that are able to overcome the computational and behavioral barriers of the refinance process, policies such as an automatically-refinancing mortgage may be beneficial.