We examine trends in redefault rates of loans selected for modification during the period 2008-2011, using propensity score matching with self-cure loans serving as the control group. It appears that early on in the crisis, many servicers had limited experience selecting the best borrowers for modification. Over the course of the downturn, as modification activity increased, it appears that lenders engage in “learning-by-doing” that resulted in more successful loan modifications. That is, they became more adept at screening borrowers for modification eligibility and in selecting appropriate modification terms. This enabled servicers to decrease the redefault rate of newly modified loans each year from 2008-2011.
Moreover, we find that over the course of the downturn, the survival curves of modified loans became less concave and more closely aligned with those of self-cures. Early on, most borrowers receiving a loan modification quickly redefaulted, suggesting that the modification terms only provided a short-lived benefit and did not fundamentally affect the borrower's ability or willingness to repay the loan. Over time, we find that the behavior of borrowers receiving modifications more closely resembled that of self-curing borrowers, suggesting that for many borrowers, the loan modification terms facilitated their own efforts to get back on track.
To effectively separate the effect of the improving economy on declining redefault rates of loans receiving modification from the effect of servicers learning, we perform additional analysis. Controling for matched characteristics of the loans, we find that redefault rates for the matched modified loans fell much more quickly than the matched self-cure redefault rates. We attribute this phenomenon to the learning-by-doing process. Our study overall provides evidence supporting mortgage modification activities as a mean to protect homeowners from losing their home and to reduce credit risk in the banking system.