EVENTS

Date:	Tuesday, February 15, 2005
Time:	2:00 PM -- 4:00 PM
Location:	Wohlstetter Conference Center, Twelfth Floor, AEI 1150 Seventeenth Street, N.W., Washington, D.C. 20036

February 2005

Do Fannie and Freddie Charge Too Much for Guaranteeing Mortgage-Backed Securities?

At a February 15 AEI conference, experts considered the question of whether guaranty fees (G-fees) are properly priced or whether institutional barriers are keeping them unnecessarily high. The accounting problems of Fannie Mae and Freddie Mac have focused attention on the two ways they make their profits: a substantial portion of their earnings comes from buying and holding mortgages and mortgage-backed securities (MBS), but they also earn substantial amounts by charging fees for guaranteeing that investors will receive full payment on their MBS. Many observers have noted that Fannie and Freddie’s G-fees are far higher than their actual losses, and recent commentary suggests that these fees add significant costs to what a homeowner might pay on a mortgage--raising questions about whether Fannie and Freddie are really reducing the costs of home ownership. The Office of Federal Housing Enterprise Oversight (OFHEO), the regulator of Fannie and Freddie, has begun an investigation into how G-fees are priced, and why--if the two companies are competing--G-fee rates remain so high.

Jay Brinkmann
Mortgage Bankers Association

A G-fee is the amount paid to a government-sponsored enterprise (GSE) each month to compensate the GSE for collecting cash flows from servicers, paying the holders of the mortgage-backed securities (MBS), providing information to investors on the underlying loans, and guaranteeing the timely payment of all principal and interest to investors. Recent question about the economics of G-fees have addressed credit losses, appropriate levels at which G-fees are determined, and alternative guaranty structures that might be more efficient.
 
Some complicating issues surround the G-fees calculation. For instance, risk-based capital generally declines over time. G-fee levels that generate low returns on equity (ROEs) initially may generate very high ROEs in later years. The G-fee remains at a constant rate. Earnings decline with principal balances but operational costs remain constant. There are factors that cannot be controlled: home prices, the economy, and thus credit losses; interest rates and prepayments; required capital and taxes. Factors that can be controlled, however, include administrative expenses, required return on equity, required capital, and dividend payouts.
 
There are potential market impacts of having a wide range of profitable G-fees. The reported average is a little over 20 basis points, with a range on either side that is not clearly related to the risk of the loans delivered. Many small lenders have sought shelter under the low negotiated fees of the giant lenders, and that has driven some of the consolidation in servicing. Other lenders have combined to form negotiating groups that often include concessions in other areas. The lack of market competition, resulting in an opaque G-fee setting process, required nondisclosure agreements, and lack of an avenue of appeal, make many lenders reluctant to speak up on issues where they might differ with the GSEs.
 
That said, there are ways to reduce costs to borrowers. Encourage more competition, particularly in the long-term fixed rate mortgage market, more innovative fee structures to allow greater credit for risk sharing in the early years of a mortgage, greater disclosure to lenders, and, absent greater competition, involvement of a regulator in pricing practices and creating an avenue of appeal.

Lawrence White
New York University Stern School of Business

Fannie Mae and Freddie Mac are engaged in two lines of business: investing in residential mortgages and MBS, portfolio lending (highly leveraged) and credit risk and interest-rate risk; and swapping (issuing) MBS for mortgages (guarantying timely payment of principal and interest to investors and credit risk). Both lines of business are very large today. Fannie has $1 trillion in assets and $1.3 trillion in net (outstanding) MBS and Freddie has $800 billion in assets and $770 billion in net (outstanding) MBS.
 
Fannie and Freddie are both very special. They are publicly traded companies; but they have congressionally legislated charters. The president can appoint five of their eighteen board members. They pay no state or local income taxes, and they are not required to register their securities with the SEC. G-fees on the MBS that they issue appear to be relatively high (relative to losses) and quite profitable. Their G-fees are not uniform and apparently are not risk-based. Although losses are greatest in early years, fees are constant as a percentage of UPB. Guaranty fees appear to be the result of oligopolistic market structure. Change in this system could come from regulatory mandates, Basel II, the Federal Home Loan Banks, or privatization. Price regulation of the guaranty fees would create a highly inflexible structure and more regulation in this dimension is not needed. Change may be forthcoming.

Alex J. Pollock
AEI

This is a very informative presentation of how G-fees work and how they produce extremely high sustained returns on equity for Fannie and Freddie. The single most important question answers itself if properly rewritten to change one word and add one phrase: “Absent competition, are the fee levels set competitively with only two principal providers of this type of insurance, with their government charters protecting them from other entrants? The obvious answer is: of course not. I believe one correction is required in the return on equity calculations. In all financial companies, we must add to the return on equity the income earned by investing the capital in financial assets. This is not reflected in Brinkmann’s basis point analysis. There are changes that can be made. More competition, one way or another, is in fact the only good answer.
 
Privatization is one way of ensuring more competition. Bert Ely’s “mortgage holding subsidiary” proposal is another idea. FHLBs are immediately available competitors, being GSEs themselves. Brinkmann also suggests “innovative fee structures to allow greater credit sharing,” which is the fundamental idea of the FHLBs’ Mortgage Partnership Finance program, the “partnership” reflecting credit risk structures. An excellent pro-competitive action would be for the Congress to enact the proposed amendment to the Federal Home Loan Bank Act suggested by the Mortgage Bankers Association and the National Association of Home Builders. This would make explicit the power of FHLBs to securitize mortgages in competition with Fannie and Freddie. There is little doubt that in this case lenders and home buyers would rapidly benefit from competitively priced G-fees, which would more closely reflect the actual risks.

AEI research assistant Jessica Browning prepared this summary.