State Foreclosure Prevention Working Group August 2010
Memorandum on Loan Modification Performance
Introduction and Summary of Key Findings For over two years, the State Foreclosure Prevention Working Group,
Our data indicate that some recent loan modifications are performing better than loan modifications made earlier in the mortgage crisis. Loans modified in 2009 are 40 to 50 percent (40% – 50%) less likely to be seriously delinquent six months after modification than loans modified at the same time in 2008. This improvement in loan modification performance suggests that dire predictions of high redefault rates may not come true. This positive trend suggests that increased use of modifications resulting in significant payment reduction has succeeded in creating more sustainable loan modifications.
In addition, recent modifications that significantly reduce the principal balance of the loan have a lower rate of redefault compared to loan modifications overall. The State Working Group believes that servicers should strategically increase their use of principal reduction modifications to maximize prospects for success. Only one in five loan modifications reduce the loan amount; in fact, the vast majority of loan modifications actually increase the loan amount by adding servicing charges and late payments to the loan balance.
Finally, while loan modifications have consistently increased over time, the numbers of foreclosures continue to outpace loan modifications. Nearly three years into the foreclosure crisis, we find that more than 60% of homeowners with serious delinquent loans are still not involved in any loss mitigation activity. Furthermore, with the significant overhang of seriously delinquent loans, the State Working Group anticipates hundreds of thousands of foreclosures will occur later this year absent additional improvements in foreclosure prevention efforts.
1 has collected delinquency and loss mitigation data from most of the largest servicers of subprime mortgages in the country. This memorandum looks at trends in loan modifications of nine non-bank mortgage companies servicing 4.6 million loans across the country as of March 2010.
This memorandum analyzes data submitted by nine servicers providing longitudinal data on loan modification performance. Since the inception of monthly data collection in October 2007, these nine servicers have completed over 2.3 million foreclosures as compared to 760,000 loan modifications. As of March 31, 2010, these servicers report 778,000 borrowers seriously delinquent (60+ days late on mortgage payments).
Impact of HAMP Program on Loss Mitigation Pipeline
As shown in Chart 1, permanent loan modifications dipped in the Spring and Summer of 2009 as servicers transitioned to the federal Home Affordable Modification Program (HAMP). The HAMP program requires a three month trial period. Accordingly, loans that would have been modified immediately in the middle of last year were instead placed into trial repayment plans, which should have become permanent after three months of successful payments from homeowners. For a variety of reasons, servicers have struggled to transition trial plans into permanent loan modifications. As shown in Chart 2 below, it appears that servicers have begun to work through the backlog of trial plans needing conversion to permanent modifications, but servicers’ conversion ratio is still far short of pre-HAMP levels.
Despite the increase in trial modifications, more than six out of ten (62.5%) seriously delinquent borrowers were not involved in any form of loss mitigation efforts. The biggest failure of foreclosure prevention efforts continues to be the inability to engage homeowners in meaningful loss mitigation efforts in the first instance. Beyond the usual factors driving borrower non-response, some reasons for the low involvement of struggling homeowners include mixed messages communicated to struggling homeowners regarding foreclosure and loss mitigation opportunities, a lack of transparency in loss mitigation options and process, inconsistent and confusing information provided to homeowners during the process, poor customer service delivery, and long delays in the modification process.
Type of Modification
The vast majority of loan modifications now involve some reduction in the homeowner’s monthly payment. Of loan modifications tracked by the State Foreclosure Prevention Working Group in the first quarter of 2010, 89.3% involved some reduction in payments, including 77.6% that significantly decreased payments (i.e. decreased by more than 10%). This data is consistent with data for the large national banks covered by the OCC and OTS mortgage metrics report.
While payment reduction is now commonplace, the State Working Group remains concerned over the absence of loan modifications significantly reducing outstanding loan balances. In the first quarter of 2010, only 13.7% of all modifications reported to the State Foreclosure Prevention Working Group involved principal reductions greater than 10%; in fact, 70.4% of loan modifications
increased the unpaid principal balance.3 With home price declines of 30% since 20064 and almost 25% of all homeowners with a mortgage owing more than their home is worth,5 the failure to meaningfully reduce principal limits the success of current foreclosure prevention efforts. The HAMP program has recently introduced a principal reduction alternative to its standard waterfall to give servicers the option of prioritizing the reduction of principal; however, we believe the optional nature of this alternative and its inapplicability to GSE loans will likely significantly limit its impact in the HAMP program.
A loan modification does not guarantee that a borrower will be able to remain current on the mortgage. Even the best-designed loan modification has some risk of redefault; however, a loan modification that fails to address the borrower’s repayment ability and the factors underlying the default may set the homeowner up for failure. Redefault expectations are incorporated into the servicer’s decision whether or not to even offer a loan modification to a struggling homeowner. Therefore, loan modification performance is very important both for the long-run efficacy of the program as well as a factor in determining the universe of eligible borrowers. Some analysts have predicted redefault rates of 65% to 75%.
6 The State Working Group is more optimistic. The reason for our optimism is that loans modified in 2009 are performing substantially better than those modified in 2008, as shown by Chart 4 on the next page.
7 For example, 30.8% of loans modified between August and September in 2008 were seriously delinquent after 6 months, but only 15.3% of loans modified in August and September of 2009 were seriously delinquent after 6 months.8 That amounts to a 50% reduction in the redefault rate.9 The OTS and OCC report a similar reduction. In recent mortgage metrics reports, the OCC and OTS report that 48.1% of loans modified in the third quarter of 2008 were 60 or more days delinquent 6 months after modification,10 but that redefault rate fell by more than 40 percent (to 27.7%) for loans modified in the third quarter of 2009.11
A comparison of five reporting servicers
12 demonstrates how the improvement in redefault rate is evident even when controlling for the type of loan modification. For instance, the redefault rate at six months for loans with significant payment reductions fell from almost 31.4% for loans modified in August to September of 2008 to just 11.8% for loans modified in August to September of 2009, a more than 62% reduction. Similarly, the redefault rate for loans with significant principal reductions fell from 35.4% to 12.9%, over a 63% reduction. While there is understandable fear that loan modification programs may be overused and that they may become less effective in the effort to reach the maximum number of borrowers, our research suggest that servicers’ loss mitigation offers are becoming more successful for those borrowers that are able to secure a loan modification.
While servicer performance is still short of what is needed and the HAMP program has not been a silver bullet, we find that there has been some improvement in foreclosure prevention efforts.
Loan modifications have increased, significant payment reduction is the norm, and loan modification performance is improving. The improved performance of recent vintages of loan modifications validates the policy of offering sustainable loan modifications. We encourage servicers and the Treasury Department to monitor this trend and to adjust redefault expectations in their models as evidence permits. If experience reflects lower redefaults than anticipated, revised adjustments will enable the HAMP and non-HAMP loan modification programs to reach more struggling homeowners.
Despite the progress noted in this memorandum, the number of seriously delinquent loans moving toward foreclosure remains at near all-time highs. As servicers pass through the initial wave of successful HAMP-eligible borrowers, the State Working Group is concerned that many of the currently delinquent loans will accelerate into foreclosure in the second half of the year. The State Working Group believes that unnecessary foreclosures will occur without further efforts and resources of servicers to reach homeowners, and, where appropriate, to offer loan modifications with significant principal reduction. These unnecessary foreclosures will be a needless drag on the recovery of the housing market and will continue to delay a broader economic recovery.
1. The State Working Group is more fully described in our first report from February 2008, available at: http://www.csbs.org/regulatory/Documents/SFPWG/DataReportFeb2008.pdf. The State Working Group currently consists of representatives of the Attorneys General of 12 states (Arizona, California, Colorado, Florida, Illinois, Iowa, Massachusetts, Nevada, North Carolina, Ohio, Texas, and Washington), three state bank regulators (Maryland, New York and North Carolina), and the Conference of State Bank Supervisors. Data analysis and graphs for this memorandum were prepared by Center for Community Capital, University of North Carolina at Chapel Hill.
2. For the first quarter of 2010, the OCC/OTS reports that over 87% of all loan modifications involve a payment reduction, with 72% reducing payment by more than 10%.
See OCC and OTS Mortgage Metrics Report, First Quarter 2010 (Jun 2010) at p. 33, available at: http://www.occ.treas.gov/ftp/release/2010-69a.pdf.
This is generally consistent with results from the OCC/OTS metrics report. The OCC and OTS report that only 2% of modifications in the fourth quarter of 2009 involved principal reduction, while 82% included the capitalization of missed payments and fees, thereby increasing the amount owed. See OCC and OTS Mortgage Metric Report, infra note 2, at p. 26. The State Working Group notes with some surprise the decline in the percentage of loan modifications with principal reduction for the large national banks and thrifts between 4th quarter 2009 and 1st quarter 2010 (from 7% in 4Q 2009 to 2% 1Q 2010).
The S&P/Case-Shiller National House Price Index fell 32% from its peak in the second quarter of 2006. See S&P/Case-Shiller Home Price Indices: 2009, A Year In Review (January 2010).
First American CoreLogic estimates that more than 11.3 million, or 24%, of all residential properties with mortgages, were underwater at the end of 2009. See Media Alert: Underwater Mortgages On the Rise According to First American CoreLogic Q4 2009 Negative Equity Data (February 2010), available at: http://www.loanperformance.com/infocenter/library/Q4_2009_Negative_Equity_Final.pdf
U.S. RMBS Servicers’ Loss Mitigation and Modification Efforts Update II, Fitch Ratings (Jun 16, 2010)
7 For purposes of this memorandum, redefault is defined as 60+ days late or foreclosed.
8 Due to limited data availability, August and September are the only months for which we have overlapping redefault rates specifically for 6 months after origination; however, a decline is evident with other cohorts. For example, the redefault rate at 9 months for loans modified in May and June fell from 37.4% in 2008 to 26.7% in 2009, a 29% reduction.
9 The decline in default rate has been broadly consistent across all nine servicers. The range of redefault rates six month after modification was 17-46% for loans modified in August and September of 2008 and only 10-25% for those modified at the same time in 2009.
OCC and OTS Mortgage Metrics Report, Fourth Quarter 2009 (Mar 2010) at p. 34, available at: http://www.occ.treas.gov/ftp/release/2010-36a.pdf.
OCC and OTS Mortgage Metrics Report, infra note 2 at p. 36. 12 Note that the redefault rates of loan modifications with payment and principal reductions are based on the 5 (out of the 9 total) servicers who provided performance data on all types of modifications. The overall redefault rate for loans modified in August and September by these 5 servicers was 32.3% in 2008 and 15.2% in 2009.
Conference of State Bank Supervisors