AEI’s FHA Disinformation Campaign Ignores Basic Finance

How one discredited “mortgage expert” from the American Enterprise Institute launched an ongoing disinformation campaign to destroy a successful government program that helped stabilize the mortgage markets.


Much of the brouhaha concerning the fate of the Federal Housing Administration can be traced to the actions of one dishonest man, a crackpot who is treated with utmost deference by the current Chair of the House Financial Services Committee and by friends in the media.

Genesis of the Disinformation Campaign

As recently as last September, the House was capable of passing a piece of legislation, known as the FHA Emergency Fiscal Solvency Act of 2012, with a lopsided bipartisan vote of 402 to 7.  The Senate version of the bill, sponsored by Pat Toomey, was co-sponsored by Richard Burr, Kay Hagan and Mark Warner. It seemed likely to pass in the Senate as well, until December 13, 2012, one day after The New York Times  published a favorable story on the crackpot research of Edward Pinto of the American Enterprise Institute. On that date, the bill was sent back to Committee to die.

According to two knowledgeable sources, neither of whom are Democrats, Pinto was lobbying in the Senate to kill the bill, and he persuaded Sen. David Vitter and two other GOP senators to do just that.

Whether or not he had a hand in killing the legislation, one fact is indisputable. The only reason why anyone pays attention to Pinto’s disinformation campaign is because last year’s bipartisan bill died. And the only reason why anyone would take Pinto’s work seriously is if that person were ignorant of the subject matter, or shared Pinto’s contempt for the truth. His campaign went into overdrive on November 16, 2012, the day HUD released the latest annual actuarial study of FHA’s mortgage insurance portfolio. Right away, Pinto reveals his duplicity.

A Profound Misunderstanding of an NPV

FHA’s annual actuarial study is primarily a net present value calculation, prepared by an outside consultant, of the insurance in place as of July of the current year. The NPV, the 2012 portfolio had an NPV of negative $13.5 billion, a considerable downward slide from the 2011 portfolio, which was valued at $1.19 billion. The primary reasons for the slide were traceable to revised assumptions, which pertain to cash flows extending more than 30 years, and to a different methodology for calculating the NPV, used by a newly hired  consultant.

The NPV number refers to a static portfolio in liquidation, as opposed to the economic value of an ongoing enterprise. Pinto demonstrates an inability to differentiate between the two concepts, which is a pretty big deal.  Because, notwithstanding all the controversy surrounding the mortgage crisis, there remains one universally accepted truism: If you don’t understand the difference between a static loan portfolio and an ongoing lending enterprise, you don’t know WTF you are talking about.

So, upfront, excuse me for belaboring some obvious points, which seem to escape Pinto. An NPV is a way to count your chickens before they hatch; it involves a number of forward-looking assumptions. As any kid who ever took a course in finance knows, if you tweak the assumptions, you can change the NPV dramatically.  Similarly, there are different methods for calculating an NPV. Change the method, and you change the outcome.

(Many, including myself, would take issue with the consultant’s change from a stochastic analysis to a Monte Carlo simulation for 30-year projections, but that’s beyond the scope of this piece.)

Also, given the way that discounted cash flows work, changes that occur sooner in time have a bigger impact than changes that occur further out in the future. So, when loans perform better or worse than expected in the first year, the updated NPV of the same portfolio can change dramatically.

I know, you’re thinking, “Duh.”   But lot’s of people miss this point, which is the dirty little secret of credit ratings for private label mortgage-backed securitizations. A synonym for NPV is “credit enhancement,” which, for the rating agencies, represents the margin for error, or safety cushion. Credit ratings are supposed to be stable over time, whereas NPVs for RMBS can be wiped out very quickly.

Static Loan Portfolios Versus Ongoing Enterprises

The NPV of any static loan portfolio is an estimate the projected future income from performing  loans, used to offset the projected future credit losses from defaulting loans.  In FHA’s case, the static pool consists of insurance policies on mortgages, but the same concept still applies. If loans in a static portfolio prepay at a faster-than-expected rate, that shortfall in income is lost forever. The NPV goes down and the balance between good loans and bad loans irrevocably shifts for the worse. This is one of the key risks assumed by investors in any private label residential mortgage backed securitization.

And early prepayments are exactly what triggered the collapse of the subprime mortgage market in the late 1990s. A multitude of subprime RMBS portfolios were weakened when creditworthy borrowers refinanced at faster-than-expected rates beginning in 1998.  Back in the 1990s, people had this quaint notion that an originator should have skin in the game.  Many now-defunct subprime originators, like ContiFinancial and Southern Pacific Funding, acquired the equity tranches of deals securitizing the mortgages they originated. And with faster-than-expected prepayments, the NPVs of their bond holdings plummeted in value, which wiped out their capital, prompting banks to cut their credit lines and, viola, they went out of business rapidly.

This is really basic and really important: Rapid prepayment impacts the risk of principal recovery for private label deals in a way it does not for mortgage securities sold by Fannie Mae and Freddie Mac. Forget about any implicit or explicit government support, we’re talking about structure in structured finance deals.  Private label RMBS are non-recourse, whereas the government sponsored enterprises guarantee their deals. Anyone who equates the credit risk diversification of private label deals with that of GSE securitizations is simply too ignorant or dishonest to be taken seriously.

Historically, the biggest trigger for rapid prepayments is falling interest rates. This is bad news for private label deals, but not for buy-and-hold lenders, which can easily replace early prepayments with new or refinanced loans.  Similarly, if an investor in one GSE securitization sees the mortgage pool declining more rapidly than expected, he knows that the corporate guarantor is still generating new business, which enhances its ability to honor that guarantee.  And since the GSEs extend credit during all stages of the real estate cycle, an individual investor is less concerned about credit losses from a mortgage pool that was booked at the wrong time.

The FHA, which insures mortgages, is like a balance sheet lender; it books new insurance policies on new mortgages when other loans prepay faster than expected. But the annual NPV of FHA’s portfolio, calculated by an outside consultant, Integrated Financial Engineering, Inc., assumes that FHA will never book a new insurance policy ever. Prepaid mortgages represent a permanent shortfall in cash flows. This fiction may be useful for analyzing the extant portfolio, but, again, the volatility in the different annual numbers is mostly traceable to revised assumptions and methodologies. And no one would confuse that valuation with GAAP accounting, which is based on the idea that you don’t book income or losses until the period when they actually occur.

The consultant calculated the annual NPV  based on forecasts by Moody’s Analytics as of July 2012. The revised assumptions of lower interest rates, triggering faster prepayments, lowered the NPV by $8 billion. The revised assumptions of reduced home price appreciation lowered the NPV by another $10.5 billion.

As the consultant wrote, “We project that there is approximately a 5 percent chance that the Fund’s capital resources could turn negative during the next 7 years.” Such concerns were addressed in the FHA Emergency Fiscal Solvency Act of 2012.

Ed Pinto and His Cherry Picked Factoids 

The same day that the actuarial study was released, Pinto rushed out his crackpot analysis to frame the media narrative. He said the FHA was masking its financial problems, because the latest interest rate forecasts, which were lower than those in July 2012, meant that the company was $31 billion in the hole. If you have no idea what he’s referring to, his words sound confusing, but not ridiculous.

Today the FHA released its FY 2012 actuarial study and as documented by FHA Watch, the FHA’s financial condition continues to deteriorate.  This report should be cause for significant concern for Congress and taxpayers. As expected, the report shows that the FHA main single-family insurance program has a negative economic value of negative $13.5 billion. Even under generous accounting rules that no other financial entity gets to use, it is insolvent.

To make matters worse, this report is already obsolete and outlines a conservative estimate of the true losses incurred by the FHA. The projection of negative $13.5 billion is based on Moody’s July 2012 forecast projecting 10 Year Treasuries in CY Q3:12 to be over about 2.2% and climbing to 4.59% by 2014.   Today the 10-year is at 1.57%.  Under that same forecast, mortgage rates are projected to double to 6.58% by CY Q3:14.

The base case scenario ignores the Fed’s September QE 3 announcement.  FHA has once again ignored intervening events that dramatically change the base case findings in their annual report.  If the current low interest rate scenario were substituted, the FHAs FY 2012 is a negative $31 billion. Yet, FHA chose cherry pick a piece of “good news”–the study projects that FHA will generate $11 billion in new economic value in FY 2013 and seize on it as evidence the 2012 deficit will be largely wiped out and all will be fine.  This ignores the of the real negative $31 billion hole in 2012.   No matter how bad things get today, FHA continually paints a rosy picture.   The SEC would be all over a public company that played by FHA’s rules.

“Yet FHA chose to cherry pick…”? Hey Ed, projecting much?

Let’s get to the big stuff first. Pinto says that the “economic value” of the ” FHA main single-family insurance program,” is not negative $13.5 billion, but negative $31 billion, based on an interest rate outlook that was even lower than that forecast in July.

Pinto conflates a static portfolio at a point in time with FHA’s ” single-family insurance program” which operates as a business.  He says that FHA is dishonest in its presentation because the NPV, would change if it were based on updated projections from Moody’s Analytics, which show that interest rates are expected to be lower than previously forecast.

Again, under the methodology used by Integrated Financial Engineering, lower interest rates translate into faster prepayments, thereby reducing future income which is assumed to be lost forever.  It’s sort of like assuming that every wage earner in his 30s who changes jobs or is temporarily laid off will never get another payroll check for the rest of his life.

Historically, about 60% of those loans that prepay because of lower rates end up refinancing with FHA.

And since interest rates have risen over the past few months, the outside consultant will need to reverse that year-old assumption that hammered the NPV. So, we can expect, at minimum, an $8 billion increase in the net amount.

Of course, Pinto ignores what he wants to ignore.

The same holds true for home price appreciation. The consultant assumed that housing prices would increase by 1% during 2012, which is why the NPV fell by $10.5 billion. As it wrote:

Moody’s July 2012 house price index forecast is very similar to the alternative scenario called “mild second recession in July 2011.  Compared to its July 2011 forecast, Moody’s Analytics’ July 2012 local house price growth rate forecast is more pessimistic in the short run. In fact, The difference is that the 2011 “mild second recession” has a deeper short-term HPA drop in 2012, but rebounded back to exceed the July 2012 forecast by 2014 and stayed higher thereafter.

And once again, home price appreciation in the near term has a much bigger impact on the NPV than price appreciation further out in time. And the impact of price increases and decreases in huge, especially in terms of loss severity on defaulting loans. For instance, in California during in 2005, loss severity subprime defaults was under 2%; in 2008 it was 70%.

As we know, Moody’s Analytics was way off in its short term home price forecasts. Home price appreciation, especially in the bubble states, has been quite robust over the past year. As the firm stated in June, “Housing has gone from a major weight on the economy  to an important source of growth.”

Consequently, if the NPV were calculated today, using the current FHA portfolio and the current projected prices increases from Moody’s Analytics, FHA’s NPV would be positive.

This is but one part of Pinto’s multi-layered smear campaign against a program that has never relied on government support for 78 years.  Pinto also cherry picks to pervert history and malign the dead, by deceitfully conflating decades-old examples of poor FHA management and oversight with the agency’s underlying business model. More on this later.

In 2011, the head of Moody’s Analytics  reminded us how the FHA stepped in to help the entire economy:

U.S. home prices have fallen by more than a third; without the FHA, the decline would have been substantially worse. Many more homes would have been foreclosed, and private financial institutions would have faced measurably greater losses. Aggressive intervention by the FHA saved the housing market and the economy from a much darker fate.

Why are Pinto and his AEI cohorts so vitriolic in their attacks on FHA? Because, over the past 18 years, we have seen how private financing of higher-risk mortgages, especially in private label deals, has proved to be an unmitigated failure, whereas FHA’s long-term track record, which shows a foreclosure rate less than one-half that of subprime securitizations, has proved to be a self-sustaining success.

David Fiderer

What's been said:

Discussions found on the web:

Posted Under