We do enjoy a good calling-out here on FT Alphaville, and on Wednesday, a pair of credit strategists at BarCap provided a neat example of the genre.
In a report headlined “Misleading Reporting of Mod Performance in the June HAMP Scorecard”, analysts Sandeep Bordia and Jasraj Vaidya deplored the reporting of the latest Hamp loan performance data by the US Department of Housing and Urban Development.
As they put it:
Starting this month, a new section was added to the report focusing on the performance of permanent modifications…Although we still believe that overall redefaults from HAMP will be better than from prior mods, we find that the data as reported in this table are misleading and fail to capture the full magnitude of redefaults from these modifications.
The analysts quibble is with the data presented in figure one, as below:
Here’s their take on the table (emphasis in the original):
Figure 1 shows loans that are 60+ /90+ days delinquent at the end of 3,6,9 months by quarter of modification. However, in the report, a footnote to the table states that “a HAMP permanent modification is canceled for non payment if it is more than 90 days delinquent”.We interpret this to mean that these loans are removed from the percentage delinquent numbers reported in this table. The report also tells us that 8,628 loans have been cancelled from the permanent HAMP modification stage till date (8823 permanent mods cancelled – 195 paid off). From Figure 1, we can approximate the number of 60+ borrowers today as (Q110 60+ at 3 months) + (Q409 60+ at 6 months) + (Q309 60+ at 9 months). Similarly, we can estimate the 90+ loans as of the latest date. These numbers turn out to be 8,205 60+ loans and 2,489 90+ loans. We believe that the total number of loans that have gone bad after the permanent mod stage is probably closer to the 60+ loans estimated above, plus the cancelled permanent modifications, which more than doubles the absolute number from 8,205 to 16,833 bad loans. The report does not contain enough information to allow us to calculate true redefault rates by quarter of modification, but we would expect them to exceed the level reported in Figure 1.
The tl:dr version: the data presented by HUD significantly understated the number of loans that went bad after receiving a permanent modification, according to the BarCap strategists.
Here’s more of their critique (emphasis ours):
Removing 90+ permanent mods from the delinquency calculation and basing the calculation only on successful modifications makes the redefault rates look too low. In our opinion, the data shown in Figure 1 continue down the same path by taking deeply delinquent borrowers out of the performance calculation and showing lower delinquency rates as a result. Given the nature of reporting available for most HAMP mods in Loan Performance, where only permanent mods are reported, we find that a more consistent approach is to use mod rates based on permanent mods and redefaults from permanent mods. On that definition, we believe that our base case expectation of about 60% lifetime redefaults on HAMP are still adequate.
The BarCap critique comes at a sensitive time for the HAMP-sters and the Treasury.
As the FT reported on Wednesday:
The Obama administration’s signature programme to help troubled homeowners is struggling to meet its objectives and suffers from a lack of “transparency and accountability”, according to a scathing assessment by the special inspector-general for the financial bail-out.
The report charges that more than a year after Hamp began, the programme has not made an “appreciable dent” in foreclosure filings. The quarterly report also slammed the Treasury department for not providing realistic benchmarks for success of the programme, accusing it of “clinging” to a “meaningless” goal of helping 3m-4m Americans in making trial modifications to their mortgages.
June 2010 Hamp scorecard – PDF
More than you probably ever wanted to know about the Hamp – FT Alphaville
The slow death of Hamp, the summer of delinquencies – FT Alphaville
More on that Hamp-lified moral hazard – FT Alphaville