Fannie Mae and Freddie Mac (the government-sponsored enterprises or GSEs) keep capital flowing for mortgage lending by guaranteeing that investors in mortgage-backed securities get what they pay for: namely, continued timely payment of interest and repayment of principal, even when borrowers miss payments or default. It’s a simple, high-value proposition. So why is it so hard to determine what fee they should charge in exchange for this guarantee? Our new commentary released today, and outlined in this post, explains why it’s hard and offers thoughts on how to overcome the challenge.

A new regulator pauses to reconsider. At his January, 2014 swearing in Mel Watt, the new head of the Federal Housing Finance Agency (FHFA) which oversees the GSEs, delayed a scheduled increase in these fees, know as guarantee or g-fees, until he could fully understand the impact the increase would have on the mortgage market. Interested parties have until September 8, 2014 to share their input on this issue.

We recently examined FHFA information released in a request for public input and conducted our own modeling. Our conclusion is that guarantee fee determination is an art, not a science-- and more like a Jackson Pollock than a da Vinci. A few decisions can change the whole picture.

Three underlying assumptions. There are three critical assumptions that must be made when calculating the appropriate fee:

1. Whether to count future g-fee premiums as capital;
2. What return on equity (ROE) to assume;
3. What, if any, capital buffer to require of the GSEs beyond that needed to cover expected losses under a stress scenario.

In our commentary, we develop a simple model, which shows how every one of these assumptions is open to interpretation;  and every interpretation makes a huge difference in the results.

Our commentary discusses several conclusions from our research:

There is no room to increase fees for the safest loans. Under any reasonable set of assumptions, raising the fees for the least risky borrowers will result in adverse selection, with banks keeping the highest quality loans on their own balance sheets and selling only higher-risk loans to the GSEs. The highest quality loans are those made to the least risky borrowers -- those with a high credit score [FICO] and a low loan-to-value ratio [LTV].

For other loans, the most appropriate fee depends on the chosen assumptions. For more risky borrowers, whether g-fees are currently too low or too high depends on the assumptions used in determining the “appropriate” fee. For example, if we assume a 10 percent after-tax ROE, and do not give credit for g-fees in setting the capital requirements, loans with a FICO of 620 to 700 and an LTV between 80 and 95 would command a g-fee of 138 basis points (bps). If, on the other hand, capital requirements include future premium income, the required g-fee drops to 100 bps. Dropping the after-tax return on capital to 5 percent and not giving credit for future premium income generates a g-fee of 83 bps; with credit for g-fee income, the g-fee would be 72 bps. The g-fee the GSEs charge for loans in this bucket is 80 bps, so the loans are either earning less than the required return on capital, or are being charged too much, depending on these two critical assumptions.

The GSEs’ mission should guide the chosen assumptions. Given that g-fees are heavily dependent on assumptions, it is our view that once total g-fees are set to cover expected losses under stress, expenses, and the 10 bps surcharge required by the Payroll Tax Act of 2011 is added, the GSEs’ mission, including the duty to serve established by the Housing and Economic Recovery Act of 2008, should be taken into account in determining the required amount of capital, the division of that capital among risk buckets, and any required return on capital.  In this context, the results of our modeling indicate that the FHFA could justify a modest reduction in pricing for the more risky loans, depending on the results of the GSEs’ internal modeling, which will be superior to ours.

No matter what decisions are made, transparency about the goals and the assumptions used in setting the fees will be critical to ensuring that lenders and investors alike understand the choices made and the rationale for any future g-fee adjustments.

Photo: Mel Watts, director of the Federal Housing Finance Agency.