As Chris Whalen is a friend, I won’t dwell on his misrepresentations of facts such as stating I have criticized increased GSE fees (I have not) or I haven’t criticized the National Association of Realtors (I have) or his odd assertion that “the FHA is really doing a pretty decent job.”

The FHA’s own numbers indicate that over the last 37 years, some 3.25 million foreclosures will be the result of their decent job, with half a million the result after 2008.

Instead I will focus on the flaws in the central thesis as set for in the title of his February HousingWire opinion piece: “FHA critics light on viable solutions.” He correctly describes my position as calling for the FHA to only engage in responsible lending, which he notes I define as averaging a 5% claim rate.  He also correctly describes my position that FHA’s mission should be focused on helping higher-risk working class borrowers and first-time borrowers with less than 680 FICO scores.

We part company when he asserts that this would inevitably force up FHA claim rates to well above the desired 5% claim rate and goes on to claim: “FHA cannot achieve one Pinto principal (5% average claim) without violating another — namely not going after business that the private markets are willing to support”— the prime, 720-plus FICO risk bucket.  He concludes: “It is easy for conservatives to criticize the FHA and other government agencies. The hard part is coming up with workable solutions that will not hurt the US economy.”

Chris is just flat out wrong. Both of my goals are not only attainable, they represent a workable solution that will help, not hurt the U.S. economy.  On Feb. 6, I testified to that effect before the U.S. House Financial Services Committee. 

First, my approach for FHA would responsibly target insured lending to the 24% of all households (not home owners) with a FICO score of 580-675. This market segment provides plenty of potential borrowers for FHA to responsibly insure loans and help, not hurt, neighborhoods.

Second, my proposal would introduce risk-mitigating process changes to the way FHA does business today.  In this regard, I have borrowed liberally from the best practices of the Veterans Affairs’ loan guaranty program.   These include a return to appraisal panels (as currently used by the VA), the use of residual income in conjunction with debt-to-income (as currently used by the VA), limiting seller concessions to 3% (as proposed by then-FHA Commissioner David Stevens in 2010), and a gradual reduction in the FHA’s claim payment percentage from 100% today to 25% (the VA’s effective percentage today).

My research has found markedly different performance metrics for the FHA versus the VA, particularly over the last 15 years. View the VA and FHA serious delinquency rates graph and focus on the far right portion of the performance.

Third, my proposal would balance down payment, loan term, FICO score, and debt-to-income loan features so that higher risk FHA borrowers have an opportunity to achieve meaningful equity within four years of purchase without being exposed to the catastrophic foreclosure rates of 15, 20, or even 30% that its current policies invite. FHA’s historical average claim rate of 10.6% masks the concentrated pain inflicted on higher risk borrowers. The  different loans chart provides a FICO breakdown along with the respective projected foreclosure rate.

The grid below sets out how this balance might be achieved with respect to loans A-D above, without relying on cross-subsidies from Loans E-G above.  The bubble-and-bar graph below highlights the entirety of the results of FHA policy and approach to insuring mortgages.

This balanced approach uses one or more lower risk compensating factors (shorter term, lower DTI, or larger down payment) to offset one or more high risk factors (say a FICO of 580-619 or a 30-year term). This approach also maintains the same present value mortgage insurance premium per insured loan.

I estimate that loans originated to the above standards would have an average FICO of about 640 as compared to FHA’s average of around 695 today. This average FICO, by Chris’ own admission, would not compete with the private sector. And it would responsibly expand lending and help the economy, not hurt it as Chris alleges.

Finally, I estimate that the loans resulting from combining best practice risk-mitigating process changes along with balancing high risk factors with compensating factors would allow FHA to both concentrate on its mission to help working class and first time homebuyers and reduce its average claim rate to 5-6%.

I welcome comments and criticisms, but I trust not even Chris would deny that this is a workable solution that helps, not hurts, the U.S. economy.