Comment Detail

Comment

FHFA is seeking comment specifically regarding “whether it is appropriate to continue to link upfront guarantee fees to the [Enterprise Regulatory Capital Framework] ERCF, set risk-based upfront guarantee fees for both Enterprises, and set a minimum threshold for an Enterprise’s return on capital.”

Regarding upfront guarantee fee (G-fee)…

As described in the RFI, the loan level price adjustments (LLPA) is one factor of many that sets a portion of the upfront portion of overall G-fee. Although the LLPA is necessarily subsidizing lower credit score borrowers at the expense of higher credit score borrowers, the difference in the overall G-fee is minimal (and convoluted by many other factors). However, in the current rising home price AND rising interest rate environment, every unfair fee component will be called into question. It was a political hot button. This is the low hanging political fruit of unfair fees to be charged on middle class borrowers that has ripened. The emphasis on this one fee though is far less problematic than the overall problem that the ERCF has created.

This problem was avoidable had the ERCF actually been created to be a “risk based” capital framework. As stated in the RFI, “In general, the ERCF capital requirements are substantially higher than the requirements under which the Enterprises previously operated. Specifically, the aggregate increases in the Enterprises’ capital requirements reflect the following components of the ERCF: (1) a risk weight floor minimum requirement for mortgage exposures, (2) a countercyclical adjustment to single-family credit risk capital requirements that increases single-family requirements when real house prices are significantly above their long-run trend (and vice versa), and (3) risk-insensitive capital buffers for stability, stress, and leverage.” The RFI states the problem. All the additional buffers were created specifically to achieve an arbitrary “bank-like capital” target of at least 4.0% regardless of the actual risk that Fannie’s or Freddie’s business actually faces.

Former Director Calabria started with his arbitrary 4 percent target and wrote a rule to achieve it. He did not actually look at risk, which is evident in the fact that shortly after the rule was published, stress test results (that were delayed by Calabria at the time) confirmed that the GSEs would NOT NEED ANY CAPITAL in a severe stress situation. They remained cash-flow positive throughout the stress period. The cash-flow positive stress tests have continued to date. A true risk-based rule would be consistent with such stress test results, which include parameters that have been tested in the real world. There is already an arbitrary 2.5% capital standard in HERA that the GSEs would need to maintain if they have capital in excess of the risk based standard that FHFA was supposed to create. There was absolutely no need for the ERCF to create a higher arbitrary standard. That higher standard is what now causes the current problem. The higher arbitrary ERCF requires a similarly higher arbitrary G-fee requirement. That is not only bad for low credit score borrowers, but every person that is seeking a mortgage. Even if homeowners aren’t getting a GSE mortgage, the bank held loans are priced based on the G-fees of the GSEs. The only people winning because of this ERCF rule are “Too Big to Fail” banks and other large private financial institutions, getting improved margins on their own loans.

The solution to this problem is to re-write the ERCF to be a truly ‘risk-based’ standard. Remove all the extra buffers as one idea, although the pro-cyclicality of the overall rule should probably also be addressed, so we might do well to start over. You probably should rewrite the entire ERCF from scratch. It should clearly reflect the stress tests and how they relate to capital requirements. It should have some adjustment to address pro-cyclicality. The sooner this is accomplished, the better for all Americans.

As far as “minimum threshold for an Enterprise’s return on capital”, 10-12% always seems like a good target return. It seems pretty standard for large businesses like the GSEs.