As this article goes to print, our nation is nearing the 100-day mark until the presidential election of 2012. While the candidates debate, most polls rate the economy, jobs, and housing as differentiators. For the issue of housing to be included as top of mind speaks volumes to the challenges the housing market has brought to the policy makers on the one hand, and the confidence of the voters on the other. Policy makers want to f ind answers to the next gen-eration of housing finance and have sought to balance the fragility of the housing market with appropriate loan modification programs as an overall economic recovery takes hold. The questions of how to solve our homeown-ership conundrum are generational; and they are due additional quantitative analysis and scrutiny. Many factors affect the success of the various loan modification programs, and in this article, we review the different loan modification efforts that have been observed in the non-agency market and assess the performance of these different modification types. As the campaign rhetoric fades past November, we will continue to analyze the data surrounding the success and failure of the various loan modification programs and the impact these programs are having on our nation’s housing recovery.

With a growing number of modifica-tion programs in effect today and even more in debate, it is crucial that all housing analysts consider the performance drivers of modifica- tions. The standard method of reporting per-formance of modifications (re-default rates) has, to date, been largely driven by single-factor (bivariate) analysis. For example, the re-default rates calculated and published by the Office of the Comptroller of the Cur-rency (OCC) and similar ones calculated by the research departments of investment banks, while helpful, provide only a one- or two-dimensional view of re-default rate drivers. For instance, the OCC published a report indicating that modifications completed on loans serviced for “private” investors under-perform those serviced for the GSEs (Fannie Mae and Freddie Mac) and bank-owned port-folios.1 On the surface, this underperformance could ostensibly be the result of ser vicers’ pol-icies and practice on modifications to non-agency investors versus mortgages they hold on their own books or mortgages held by the GSEs. However, a deeper analysis based on regression analysis provides for an alternative theory: underperformance of non-agency loans could be a function of borrower quality rather than the relationship of the servicer to the risk-holder.