Senior Fellow Peter J. Wallison

Nor am I implacably hostile to Government Sponsored Enterprises in general. I believe they can and do have a role in cases of market failure. The Federal Home Loan Bank system was created in 1932 as part of a Hoover administration effort to provide a steady source of funding for long term residential mortgage lending. At the time, S&Ls were unable to lend for housing because they were dependent on deposits, and in the midst of the Depression deposits were not a stable funding source. Hence, the establishment of the FHLBs can be defended as a legitimate way to address what seemed to be a market failure. The system made it possible for member S&Ls to use their mortgage portfolios as a source of liquidity.

Fannie Mae was established in 1938 to address another aspect of market failure--the inability of banks to sell off their mortgage portfolios so they could make more housing loans. Fannie Mae and, eventually, Freddie Mac performed well in helping banks become more liquid and in creating a national market for home mortgages.

As Ronald Reagan once said, however, the closest we'll ever come to immortality is a government agency. And that captures the major problem I have with GSEs. Once they have addressed the market failure that was the reason for their establishment, they bend all their efforts to stay alive and relevant--and regrettably, usually succeed. For that reason, my colleague Alex Pollock has suggested that a GSE should never be created with perpetual life; it should always be subject to a sunset provision that will bring its continued relevance back to Congress for consideration. Of course, this doesn't guarantee the right result. In my view, neither Fannie and Freddie nor the FHLBs now serve any purpose for which it is necessary to be a GSE, but no one in Congress is seriously suggesting their privatization or other elimination, despite the fact that far-reaching regulatory legislation is currently under consideration.

But not all GSEs are the same, or raise the same issues of public policy. Fannie and Freddie, in particular, are different from other GSEs. As partially privatized entities, with shareholders and a board of directors, they are subject to incentives and the beneficiaries of advantages that make them truly dangerous. As private shareholder-owned corporations, the fiduciary obligations of their directors virtually require that they exploit the companies' government subsidies to the maximum extent possible, and expand the companies' activities as far as their mission description can be stretched. The truly stunning pecuniary rewards to their managements--considering they were operating in part on the taxpayers' dime--are another engine of rapacity.

The funds generated by Fannie and Freddie's extraordinary profitability were poured into a powerful lobbying network, and they developed ways to shower benefits on the politicians in Congress who alone had the authority to rein them in. They also mastered the fundraising game, not only providing financial support to those in Congress who were able to protect them, but also marshalling other industries--the realtors, the homebuilders and the securities industry--to make contributions as Fannie and Freddie directed. By the year 2000 they had developed such an effective network to control their political risk that Fannie was regarded as the most powerful company in the United States.

But for their arrogance--born of their sense of invulnerability--there might never have been a hope of controlling Fannie and Freddie. Thank heaven for human foibles. Freddie's admission in 2003 that it had manipulated its financial statements, and the subsequent finding by OFHEO that Fannie had done the same, broke the spell. Legislative action, that had initially seemed wildly improbable when I first started criticizing them in 1999, actually became plausible, and at the moment there is tough legislation in the House of Representatives--controlled by the Democrats, no less--that would have been unimaginable only a few years ago. My view is that there is more than a 50-50 chance that the Frank bill will pass both the House and Senate in close to its current form, and will give the new regulator of Fannie and Freddie the power for the first time to control them.

I go through all of this detail to explain why I have always seen Fannie and Freddie as different from--and posing fewer policy problems than--the Federal Home Loan Banks. Although in both cases the organizations have outlived their usefulness and thus their reasons for continued existence, the incentives of the Fannie and Freddie managements were to exploit their subsidies to the maximum degree possible and to take risks for the possibility of increasing their profits. This is what made them dangerous--a word I don't think is an overstatement--a danger to the taxpayers and through systemic risk to the economy as a whole.

I don't see the FHLBs in the same light. Like Fannie and Freddie, their subsidy gets passed through to their shareholders, but their shareholders are competing with one another, and thus I think the subsidy ultimately gets passed along to borrowers in lower interest rates. It's not an efficient or sensible system--and it may result in lower returns to savers--but it doesn't reward a relatively small group of shareholders and managers the way Fannie and Freddie do. Even more important, while the FHLBs have a strong incentive for self-preservation, they don't seem to me to have the incentives for high levels of growth and profitability that promote and encourage risk-taking. The dangers of systemic risk, it seems to me, are negligible.

Having said this, I can see some incentives for risk-taking in the FHLB system, although not in the same league as Fannie and Freddie. In theory, an organization like the FHLB system is particularly prone to risk-taking. As GSEs, the Banks are not subject to market discipline, and because the system is a cooperative enterprise with joint and several liability, any Bank is encouraged to take risks because all the others will bear some of the possible losses. But what are the spurs to risk-taking? One of them could be retaining member banks and their capital. It is important to recall that the capital provided to the system by member banks is withdrawable, and that banks won't join the system, or stay in it, unless they see significant benefits for themselves. Lower interest rates and higher dividend rates will attract and retain members, and both of these can be improved by greater leverage, asset growth and risk-taking. Another source of risk-taking is management compensation. If management compensation is tied to profitability, that could turn out to be a problem. Still, while these incentives are there, they are modest compared to the incentives that play on Fannie and Freddie.

Although I don't see any urgent reason to worry about the FHLB system, I am not predicting immortality. The system was initially established to assist mortgage lenders--first the S&Ls and now any depository institution that maintains at least 10 percent of its assets in mortgages or mortgage-related instruments. There are now over 8,000 members of the 12 Banks in the system, which is over 92 percent all insured banks and S&Ls in the country. Although the original purpose of the system was to provide liquidity to the S&L industry, and the 10 percent requirement is a vestige of this original purpose, it doesn't take long to realize that although the system makes loans on the collateral of mortgages, the money it lends does not have to be used for mortgages. In other words, the system is providing a subsidy to most banks in the country, and that subsidy can be used for any purpose. Indeed, many of its members are ordinary and quite large commercial banks that have realized that the FHLBs are an easy way to get inexpensive funds.

This, combined with the likelihood that Congress will adopt a new regulatory structure for the system, has significant implications for the future.

First, not only is the FHLB system no longer needed to assist housing, it can't even credibly argue that it is doing so. Fannie and Freddie are certainly no longer needed to assist housing either, but at least they can make a reasonably plausible argument that requiring them to pay state and local taxes, or imposing a fee for the benefits they receive from their GSE status, is, as they put it, a tax on housing. There is a direct connection between what they do and the housing market. This argument was and still is a powerful source of political support, but it is not in my view an argument that the FHLBs can credibly make.

As Congress looks around in the years ahead for sources of revenue, the FHLBs look to me like easy pickin's. Fannie and Freddie can always call on the homebuilders and realtors to vouch for their importance to the housing business, but the FHLBs can only call on their member banks. To be sure, the banks, and especially small banks, are a powerful constituency, but their advocacy in this case would be more self-serving than the realtors and the homebuilders, who are not direct beneficiaries of Fannie and Freddie's largesse.

Moreover, even if we think about the member banks as advocates, the numbers are not good. The largest five members of each of the 12 FHLBs own substantial amounts of the Banks' outstanding stock, ranging from 30 percent of the Chicago Bank to 80 percent of San Francisco. The average ownership is about 57 percent on a weighted average basis. The actual advances to member banks are even more concentrated, with these ranging from 29 percent in Des Moines to over 80 percent in San Francisco. The average is over 61 percent on a weighted average basis. That means the major borrowers from these Banks are very large financial institutions, not the local community institutions that it would be good to show to Congress.

One of the best arguments in favor of the system is that it provides access for small banks to the international capital markets. However, this argument I undercut by the fact that most of the access is provided to very large institutions that could and do access the international capital markets on their own. The same problem arises in connection with the FHLBs' effort to get into the business of offering letters of credit in support of the tax exempt revenue bonds of state and local governments. The argument there is that the FHLBs are providing financial assistance to underserved communities. But this effort will be dwarfed by the financing the system is providing to members that are lending to private equity firms, hedge funds and other substantial borrowers. When the data is published, it won't look much like the system is focused on underserved communities. In other words, the FHLBs are in danger of looking like a huge subsidy program for Citibank, Chase, WAMU and the others. That will be a damaging fact for an organization that is trying to convince Congress that it has some major public policy purpose that should not be taxed.

Second, the system now holds, in the aggregate, over $1 trillion in assets, up from only $175 billion in 1989. This is about 1/6th of the deposits of the entire banking system. Given the fact that the FHLB system offers subsidized funding, the only natural limit on its size is the size of the banking system iself. As Americans become more sophisticated, and begin to move more of their savings out of bank deposits and into tax-favored savings vehicles such as 401(k)s invested in mutual funds, deposits may cease to be the least expensive source of bank funding. Mutual funds now hold over $10 trillion in assets and continue to grow.

How will the system be perceived when it is the source of one-third or one-half the funding of the banking industry. At that size, it could start to look like the United States had set up a system to subsidize its banks. This could become a trade issue, and the U.S. could be subject to challenge in the WTO grievance process. In the 1990s, Germany was forced to restructure its Landesbank system because these wholly owned government depository institutions were deemed to be subsidized by the state governments that owned them and thus in violation of EU rules. Having adopted this reform, the EU is not likely to look sympathetically at a U.S. government-backed structure that seems to be funneling subsidized funds to U.S. commercial banks.

Even within the United States, non-depository institutions that compete with banks in offering lending services will resent the fact that members of the FHLB system--which is limited almost entirely to depository institutions--have a source of funding that is rated AAA and can gather funds at a rate about 35-40 basis points lower than any fully private organization. This will bring pressure on the FHLB system to explain why it should continue to exist--or at least why it should be permitted to provide subsidized funding to institutions that don't need it.

In any event, any structure that grows beyond $1 trillion is bound to attract attention, and some in the administration and Congress are bound to start asking whether it makes any sense to continue to subsidize Citibank and other large internationally active banks.

Finally, and perhaps most important in the short run, as I indicated earlier I believe this Congress will enact legislation that changes the regulatory structure for the FHLBs, providing for the regulation of the system jointly with Fannie Mae and Freddie Mac. It is difficult to overstate the profound impact that this will have on the FHLB system. The most significant fact about the new regulator is that it will have a broader mandate than the FHLBs alone. When an agency is responsible for a single and well-defined group of institutions it almost always comes to see the regulated group as clients. Its treatment is sympathetic, and occasionally promotional.

The Federal Housing Finance Board's decision, in 1998--despite the Treasury's opposition--to approve Bank issuance of letters of credit to back tax exempt state and local revenue bonds, is a good example of this sympathetic approach. This is understandable, since the agency itself depends for its continued existence on the financial and political health of the group it is regulating.

This, I think, will change when the FHLBs are regulated and supervised by an agency that is responsible for both the Banks and Fannie and Freddie. The approach of the new regulator will be focused much more clearly on a precise and narrow definition of mission and on control of risk-taking.

There will be considerable political pressure on the new agency to control the activities of Fannie and Freddie--a substantial part of the new legislation is concerned with assuring that the enterprises do not expand their activities beyond their housing finance mission--and it will be difficult for an agency that is focused on a careful definition of Fannie and Freddie's mission to switch gears when considering new activities for the FHLBs.

The same is probably true for safety and soundness and risk-taking. The new agency will be charged with keeping tabs on the adequacy of Fannie and Freddie's capital and on the interest rate risk associated with their portfolios of mortgages and mortgage-backed securities. This sensitivity to safety and soundness and interest rate risk will inevitably affect the agency's approach to the supervision of the FHLBs, causing a much less sympathetic view of portfolio growth and leverage than the FHLBs have experienced thus far.

And finally, the agency's attention to Fannie and Freddie's capital position will have a profound effect on the FHLBs, whose capital is withdrawable by members if they are no longer getting the benefits they want or expect from membership. Without any reason to promote the growth of the FHLB system, the new agency is likely to tighten its regulation of the FHLBs, requiring more capital and less risk-taking. In this case, the benefits for member banks could decline, and this could produce a reduction in capital, triggering further restrictions on asset growth in order to maintain capital adequacy levels.

Altogether, then, while I don't see any reason for Congress or anyone else to seek urgent changes in the FHLB system, over the long term I think that the absence of a good justification for the continued existence of the system--together with what is likely to be a significant change in the approach and attitude of its regulator--could result at least in substantial restraints on the system's growth, and possibly its eventual decline.

Peter J. Wallison is a senior fellow at AEI.