Questions arise surrounding the default rate as the affordable housing industry fights to keep the program alive
On February 14th we posted a story on the Obama administration’s proposed elimination of the USDA Rural Development’s 538 multifamily guaranteed loan program. In the Terminations, Reductions and Savings report that accompanied the proposed budget the administration cited the rising cost of the program caused by a dramatic increase in the program’s default rate:
…the defaults in these programs have been much higher than initially projected, and the increase has happened quickly, making them more expensive than their direct loan counterparts. In addition, the direct loan programs have very low defaults, even though they tend to serve the much lower income residents/communities
We also noted that the program’s default troubles were not news to the affordable housing industry. A CARH email newsletter had noted the troubling rise in default rate several months earlier:
In the USDA’s budget submission, the score for FY2011 appropriations increased nearly ten-fold from FY2010… It appears that certain defaults in the Section 538 program, together with changes resulting from the lack of interest credit subsidy, have been cited as reasons for this scoring increase.
Looking at the 2011 budget assumptions, we found that the subsidy rate increased from 1.15 in 2010 to 9.69 in 2011, driven by an increase in the default rate from 1.49 to 11.73.
Affordable housing advocates have since rallied around the program in an effort not only to save the program for FY2012, but also to maintain it’s funding levels in FY2011. These efforts culminated in a recent joint letter to both the house and senate:
We would urge you as you complete consideration of the Fiscal Year 2011 budget to provide the necessary appropriations to allow for a program level of $129 million and then in Fiscal Year 2012, consider allowing fees to be charged, thus making the program revenue neutral.
While the efforts to maintain a funding level of $129 million in 2011 have proven unsuccessful (the recently released FY2011 CR, H.R. 1473 only provides $30 million), the fight to keep the program alive in FY2012 goes on. The letter, signed by numerous advocacy groups, strongly contests the default rate at the center of the program’s proposed elimination:
We refute both statements, particularly the default rate… The default rate and therefore the subsidy rate for the program are incorrect as relayed by the Administration in its FY 2011 and FY 2012 budgets. The Council for Affordable and Rural Housing (CARH) has numbers that can demonstrate the default rates to be less than the agency transmitted in their budget.
Further investigation revealed that the increase in default resulted from just five properties going into foreclosure. The table below shows the series of loans that went into default, grouped by repurchase date to indicate which loans belonged to each of the five properties:
Also included in the table is a calculation of the default rate. This calculation, taken by summing the default amounts, and dividing the sum by the total outstanding principle in the program, results in a default rate of 8% not 11.73%. “Somewhere there is a disconnect, said Rob Hall of Bonneville Multifamily Capital. While concerned with this disconnect, he was more troubled by the way in which the numbers do not paint an accurate picture of the program’s cost, “Historically [this default rate] is not indicative of the deals getting done in the past 7 years.”
A different program
Underlying this belief is the fact that four of the five loans were originated in the early years of the program, before updated underwriting practices went into effect: “Four of the five deals were not tax credit properties. All the deals closed in the last 5,6,7 years have been LIHTC properties. They are completely different than deals from the early days of the program, which had virtually no equity.”
The changes Mr. Hall is referring to stem from a 2004 Notice of Funding Availability that awarded points to 538 loan applicants based on Loan to Value ratios and a 2005 NOFA that awarded points for Loan to Cost ratios. Additionally, since 2005, Rural Development has pushed to have every 538 loan securitized by the Government National Mortgage Association (GNMA). Because GNMA requires a Loan to Cost Ratio of 50% or less, this move has been accompanied by subsequently tighter NOFA requirements. The combined effect of these policies is that the agency has been funding loans that have more equity, generally in the form of tax credits. Pictured left is a clip from the 2005 Notice of Funding Availability that details the debt to cost preference scheme.
Given the historically low rate of defaults in the LIHTC program, the correlation between 538 and tax credit deals has significant implications for the 538 program’s projected cost. A recent Ernst and Young report showed the default rate for the 23-year history of the program at .83%, and an annual rate of about .03%.
“Recent 538 deals should be nearly identical to that TC rate because they are all the same deals. The default rate is well below 1% for 538 loans originated in the last 5-7 years.” Said Mr. Hall. He also noted while the loans that went into default should never have been underwritten, “the lenders behind these loans are no longer involved with the program and the officers who originated the loans are no longer with the program.” He continued, “Members of the industry just want the default rate to reflect this, something reasonable like 2-3%. The higher default rate is leading the budget people and congress to believe the program is very expensive when it is not. “
Symptom of broader ills
A recently released GAO audit on the USDA’s 514/515 Farm Labor housing program suggests that the questions surrounding the 538 program could be indicative of broader problems for the agency. In particular, the GAO found that, “Rural Development (RD) overestimated its credit subsidy costs for the fiscal year 2010 FLH loan cohort”, attributing this overestimate to errors in calculating the default rate, “we found that the primary driver of the change from the fiscal year 2010 credit subsidy estimate to the re-estimate was the default cost component and, more specifically, how this cost component was calculated.”
The report also provides insight into the mysterious credit subsidy rate formula: “Four cost components comprise the credit subsidy estimate for the FLH program: defaults, net of recoveries; interest; fees; and a component labeled “all other,” which includes prepayments.” Going on to explain exactly how the agency erred in their default rate calculation:
Specifically, when the fiscal year 2010 budget formulation credit subsidy estimate was calculated, the estimated default cost component was inflated by a prepayment estimate. That is, RD overstated the estimated default cost component to reflect the effect of prepayment. RD, includes the impact of prepayment estimates in the all other cost component
It remains unclear whether RD uses the same credit subsidy formula for the 538 program and whether mistakes in the same prepayment calculations could have contributed to an inflated 538 default rate. However, some industry experts find it hard not to see similarities between the two issues, “My take is that if there is a problem with determining the subsidy rate for one program, there may be issues in determining rates for other programs.” , said Colleen Fisher, Executive Director of the Council for Affordable and Rural Housing (CARH), “I think that it shows that there are some issues that need to be worked out between OMB people and the budget people and the program people that actually know what the story is.”
She went on to reiterate that the program today is a vastly different one than produced four of the five problem-properties, “The program is too important now and we have really seen it improve since its early days. From an infancy period where it had some issues, we can now use this program to do some good”, she continued, “I think if we can get through FY2011, I think we might be seeing the agency reviewing the subsidy rate, that’s all premised on the crazy FY2011.”