FHFA Acting Director Ed DeMarco made an uncharacteristic step of announcing the GSEs will be raising their guarantee-fee sometime next year to be more in line with what the private sector would charge for the same insurance. This is an idea we have long advocated for, as it would help lower Fannie and Freddie competition with the private sector. Back in February I testified before a House committee that:
One way to decrease the government's exposure to housing market risk would be to begin increasing the guarantee fee (g-fee) charged by Fannie and Freddie for ensuring payment on mortgage-backed securities to investors. This could be done slowly so as to not cause the GSEs to exit the mortgage market overnight. Over time this would increase the cost of doing business with the GSEs and create room for private capital to be more competitive with the government agencies. In the meantime, the GSEs would be collecting more revenue to put back towards the cost of bailing them out and taxpayers would be protected from the risks of bailing out Fannie and Freddie again in the future.
A story in MortgageServicingNews cited DeMarco as indicating the g-fee change would be gradual. Just how gradual is important though. While I recognize that it should not change overnight, my definition of "slowly so as to not cause the GSEs to exit the mortgage market overnight" should be placed in context of my separate proposal to lower the conforming loan limit by 20% each year from the previous year's cap so that after five years the conforming loan limit goes from $729,000 in 2011 to $239,000 in 2016—and then all remaining affordable housing subsidies could be shifted into FHA. So a gradual increase in the g-fee should follow the same logic at a minimum, meaning that by 2016, the g-fee should be so high that the private sector would not want to do business with Fannie and Freddie.
Frankly I think it would be best to get the g-fee up to private sector standards within 1-3 years, and the private sector really should welcome it as there will be more competition in the mortgage space and room for private businesses to win back market share from the GSEs.
The natural question is: how much should the fee go up? I looked at this question back in an April blog post and the reality is that even a three year increase to the private sector equivalent would see substantial jumps each year:
So how much should the fee go up? There are a range of potential answers to the question and it all depends on the approach to estimating the fee and what factors are considered. So for the sake of simplicity, I would point to the CBO.
In a paper last year, the Congressional Budget Office outlined it's budget treatment of Fannie and Freddie. Part of the report estimated the subsidy cost to the government of being the backstop of the GSEs. For example, CBO estimated that the cost of providing a guarantee to Fannie and Freddie cost $291 billion in 2009. This meant the GSEs would have to charge 440 basis points more than usual if they were to operate the same way but be a fully private company without the government backstop.
The report estimated that the GSEs should charge 150 basis points (or 1.5%) more in their g-fee for 2011 in order to account for the subsidy, 120 basis points for 2012, and 110 basis points for 2013 (see table 2 on page 8). Those are significant subsidies. FHFA will have to determine if it agrees with the CBO assessment and then be bold.
See the whole blog post "On Raising G-Fees" here.