Ron J. Feldman

Estimating and managing the federal subsidy of fannie mae and freddie mac: is either task possible?

Fannie Mae and Freddie Mac receive explicit and implicit off-budget subsidies from the federal government. This paper reviews the methods to estimate the dollar amount of the subsidies. None of the three techniques to estimate the indirect subsidy yield accurate point estimates. They do suggest that Fannie and Freddie could receive billions of dollars in subsidies in some years and much smaller amounts in other years. However, assessing the size of the implied subsidies is most valuable in demonstrating that Fannie and Freddie, not the federal government, control their size. Efforts to improve federal control face significant difficulties including informational asymmetries and the political incentives that have led to the status quo. These drawbacks bolster the rationale for eliminating federal support for Fannie and Freddie.

Comments on Collender and Frizell

Shifts in the liability structure of community banks—defined as those with assets less than

Too Big to Fail: The Hazards of Bank Bailouts

1 billion—occur at a glacial pace. The rapid growth of community bank membership in and borrowing from the Federal Home Loan Bank (FHLB) system is thus noteworthy. Sixty percent of these banks are now members, up from 20 percent just six years ago. Six years ago, a little more than half of community bank members did not borrow from the FHLB and only 11 percent had FHLB advances exceeding 5 percent of assets. Currently, one third of community banks have an advance-to-asset ratio of more than 5 percent whereas less than one third do not borrow. Over the same time period, there have been mounting concerns about the nearly 300 percent growth of government-sponsored enterprises like the FHLB (Stojanovic, Vaughan, and Yeager 2000). Despite the confluence of these two trends, there has been little analysis of the use of FHLB funds by community banks beyond descriptive statistics. In this context, “Small Commercial Banks and the Federal Home Loan Bank System,” by Robert N. Collender and Julie A. Frizell, fills an important gap by asking and answering interesting questions about these important trends. This analysis is particularly timely given changes made in law by the Gramm-Leach-Bliley (GLB) Act of 1999. Part of GLB aimed to increase community bank membership in and borrowing from the FHLB system. (For more details on this legislative change see Feldman and Schmidt 2000.) Collender and Frizell’s analysis suggests that this expansion has some risk. At the same time, Collender and Frizell are modest or narrow in their goals, methods, and interpretations. As such, most of my comments relate to expansions the authors’ should consider. First, given their applied emphasis and concentration on the riskiness of FHLB members and borrowers, Collender and Frizell should consider adding a regression explaining the amount borrowed from the FHLB in conjunction with their logit analysis of the decision to borrow. The factors that influence the borrowing decision may not be significant in explaining the amount borrowed, and thus reporting INTERNATIONAL REGIONAL SCIENCE REVIEW 25, 3: 304–306 (July 2002)