FROM THE OFFICE OF PUBLIC AFFAIRS

During the first half of this year, the Federal Home Loan Bank System issued $1.2 trillion in debt securities and supplanted the U.S. Treasury as the world's largest issuer of debt. That's quite a distinction, considering that most Americans have no awareness of the System. One can criticize the $1.2 trillion figure as misleading because the System does much short-term, even overnight, borrowing and uses the proceeds to fund much short-term, even overnight, lending. But I'm curious: has anyone here ever wanted, or known someone who wanted, an overnight home mortgage? And so our story begins.

The Home Loan Bank System is a so-called "government-sponsored enterprise." It's a privately owned company, or set of twelve companies, chartered by the federal government. It exists to further a public purpose centered on housing finance. And, in return, the government gives it benefits not available to fully private businesses.

Let's take a quick look at some of those benefits. The Home Loan Bank System has its own line of credit at the Treasury. It is exempt from federal corporate income tax. It is exempt from state and local corporate income taxes, and so is interest on its debt securities. It is exempt from registering its securities with the Securities and Exchange Commission. Public funds can be invested in those securities. Those securities can serve as collateral for government deposits. Those securities are issued and transferred through the Federal Reserve's electronic book-entry system, just like Treasury bonds.

All that brings us to the most important benefit of all. Capital market participants, looking at these and other specific benefits, evidently believe that the government implicitly stands behind the System. These market participants accordingly lend the System hundreds of billions of dollars at rates only slightly above those on Treasury securities rates below those available to even the highest-rated private borrowers.

The Home Loan Banks have played an important role in developing the residential mortgage market as we know it today. They continue to provide some valuable services to their member institutions. They offer their members a reliable source of funds, and assist members in managing interest-rate risk and remaining competitive in housing finance. Their Affordable Housing Program has won excellent reviews for helping lower-income people become homeowners. Yet much has changed since the System was created in 1932, near the depth of the Great Depression.

Today I'll talk about the meaning of those changes. My remarks have three main parts. First, I'll identify the logical foundation of the Home Loan Bank System and describe how that foundation has eroded. Second, I'll explain how that erosion raises questions about the System's reason for being. And third, I'll outline the sorts of reforms we at the Treasury believe are necessary to assure that the System furthers a meaningful public purpose.

The Logical Foundation of the Home Loan Bank System Has Eroded

Let's begin by looking at why Congress created the Home Loan Bank System. Anyone who has seen the holiday classic It's a Wonderful Life has a good sense of the problems besetting local housing finance during the early 1930s. Jimmy Stewart's character, George Bailey, faced almost insurmountable obstacles in keeping his building and loan association in business. When times got tough, depositors ran for their money. Yet, as George pointed out, that money wasn't sitting in the vault; he had used it to make loans to the depositors' friends and neighbors. To keep his institution afloat, George had to pay depositors out of his own pocket and lock horns with the town's sinister banker. George could have avoided many of his problems if he had just had access to a reliable outside source of funds.

Enter the Home Loan Bank System. The financial problems of the Bailey Building and Loan Association correspond closely to the reasons why Congress created the System. "[M]ortgage credit ha[d] dried up," according to the House Banking Committee's report on the Federal Home Loan Bank Act, and Congress sought "to place long-term funds in the hands of local institutions" and to counteract any Depression-related "drift of money away from . . . home financing activities."

The Federal Home Loan Bank Act sought to encourage the use of long-term, self- liquidating mortgages, to overcome geographic impediments to the flow of mortgage credit, and to give thrift institutions access to a lender of last resort. The Act's implicit premise its basic operating assumption, its logical foundation was that by providing low-cost funding to a depository institution that had made home loans in the past, the System could induce that institution to make more home loans in the future. In 1932, a combination of three circumstances rendered this basic premise logical. First, the System made advances only to thrift institutions. Second, thrifts generally had narrow charters that permitted them to invest in little more than residential mortgage loans. And third, thrifts generally lacked reliable outside funding sources and suffered from severe liquidity problems. Given those combined circumstances, Home Loan Bank advances necessarily supported housing finance.

Over time, each of those three circumstances has dramatically changed thereby eroding the basic premise of the System.

First, thrift institutions have expanded far beyond home mortgage lending. They can and do engage in the full range of retail financial services. They can also deal in derivative instruments and make commercial loans and commercial real-estate loans.

Second, Home Loan Bank membership is no longer restricted to thrifts, much less housing-focused thrifts. Commercial banks and credit unions have been free to join since 1989. In fact, a depository institution can become a member even if it has never made a home mortgage and can remain a member even if it never makes such a mortgage. If you're a depository institution and you want to join, you need only wear a small fig leaf for an instant of time. You can just put 10 percent of your assets into mortgage-backed securities, which you buy in the multi-trillion-dollar mortgage-related securities market. Once you sign up for membership, you can immediately sell those securities. And then, under current law, you'll have all the connection to housing that you'll ever need to remain a Home Loan Bank member in good standing. You can be morbidly allergic to home loans, but as long as you have eligible collateral on your books and that includes any sort of U.S. government or agency securities you can take out advances, use the proceeds for any lawful purpose, and rejoice in the generosity of Congress and the fungibility of money.

Third, residential mortgage lenders no longer suffer from a general lack of liquidity. Capital markets have grown deeper, wider, and more efficient; and they are now truly national markets. An enormous secondary market for mortgages has arisen, in which mortgage lenders can readily convert mortgages into cash or other liquid assets. Karen Shaw Petrou has summarized the changes as follows: "In the 60 years since the system was created, capital markets have become so efficient and mortgage securitization so effective that even the smallest bank or thrift can fund itself with a flick of a computer key."

A Crisis of Legitimacy

What, then, is the System's reason for being? Making secured loans to depository institutions with eligible collateral isn't much of a public purpose. Lots of private firms gladly do that every day, without a government subsidy. So what exactly does the System do that would not otherwise get done? And how exactly does it earn its valuable government- conferred privileges?

Let's look now at key activities of the System: making advances; running the affordable housing and community investment programs; holding a large investment portfolio; making the so-called REFCorp payments; and conducting the new programs that have begun to proliferate.

Advances

The erosion of the System's basic premise has, as just discussed, taken a conspicuous toll on the System's core function of making advances to member institutions. Institutions of any size with eligible collateral can get advances and use them for any purpose. And advances far from remaining a vital source of liquidity for member institutions have become one of many available funding options.

The vast majority of advances involve short-term, even overnight, funding which is unlikely to be used to make mortgage loans. For example, of the new advances made during the twelve months ending in October 1998, over 70 percent had maturities of less than one month. Such short-term funding is readily available from fully private sources.

Not only are advances predominantly short-term but they go predominantly to large institutions that generally have ready access to the capital markets. Small banks and thrifts although presumably having less direct access to those markets, and correspondingly greater need for advances receive only a small proportion of advances. As of the second quarter of 1998, institutions with $500 million or more in assets had 85 percent of all outstanding advances, and institutions with $1 billion or more in assets had 77 percent. The top five users of advances account for less than 0.1 percent of System membership but almost 21 percent of all advances. The top 50 users account for less than 1 percent of the System's membership but almost 57 percent of all advances.

Affordable Housing Program and Community Investment Program

The System's Affordable Housing Program subsidizes both rental and owner-occupied housing for low-income households. Lenders often combine an AHP subsidy with assistance from other governmental and private programs. By all accounts, the program is a winner. But it amounts to only the greater of $100 million a year or 10 percent of the System's net income.

The System's Community Investment Program makes loans at cost to finance the purchase or rehabilitation of homes, and commercial and economic development projects, that assist low-income households. Last year the System made $3.2 billion in CIP advances, yet had $202 billion in total advances outstanding at year-end.

So the AHP and CIP are good, but represent only a tiny fraction of what this $420 billion System does.

Investment Portfolio

Over the course of this decade, the System has developed an enormous investment portfolio. As of October 31, 1998, this portfolio stood at $150 billion equal to 36 percent of the System's total assets and 41 percent of the System's outstanding debt. One way to think about it is that the System never loans to its member institutions 41 cents of every dollar that it borrows and borrows at low rates because of its perceived government backing. Instead, the System invests that money in Fed funds, mortgage-backed securities, commercial paper, reverse repurchase agreements, and the like. In so doing, the System conducts a massive arbitrage between the government-sponsored enterprise debt market and the private debt market. The System then pockets the difference between its own cost of funds and the returns on its investments.

The System's arbitrage investments further no public purpose. The markets for those investments are deep, liquid, and extremely efficient. They don't need the System; they'd work perfectly well without it. Even the System's holdings of mortgage-backed securities do nothing appreciable to expand homeownership. Although those holdings averaged $47 billion in 1997, they represent less than 3 percent of the $1.7 trillion in outstanding government- related mortgage-backed securities. The System adds no value to the mortgage-backed securities market and was not intended to do so.

REFCorp Payments

Then what about the System's role in making payments on the so-called REFCorp bonds? In 1989, the Bush Administration persuaded Congress to finance part of the thrift clean-up with these off-budget bonds and have the Home Loan Banks pay $300 million a year toward the interest on those bonds. This was intended as a sort of continuing tax on the thrift industry. Little did policymakers suspect that commercial banks would so quickly come to dominate the System's membership. And little did they suspect that the System would soon develop a massive arbitrage portfolio not only to pay for the REFCorp tax but to help itself to an extra serving of government subsidy.

Some System insiders refer to the REFCorp payments as the System's "fiscal mission." Quite a mission. The System, exempt from all corporate income taxes, uses its relationship with the government to reap arbitrage profits and then share some of those profits with the government. Not bad so far, but there's more. Some people believe that if Congress were to abolish or privatize the System, any lost REFCorp payments would trigger the pay-as-you-go requirement in the Congressional Budget Act and thus require offsetting tax increases or spending cuts (which could go beyond applying the corporate income tax to a privatized System). Yet one may doubt whether the REFCorp payments actually make the government better off, since it stands to reason that the System's arbitrage borrowing at rates close to those on Treasury securities may increase the cost of financing the public debt by increasing the supply of competing securities. In any event, the REFCorp obligation does not justify the System's investment arbitrage portfolio any more than Congressional cost-cutting would justify a federal agency in sponsoring a mutual fund and using the income to replace money that Congress had chosen not to appropriate.

New Programs

In seeking to sum up this review of the System's activities, we might say: The System is big; the System is busy; but most of what the System does would get done anyway. The status quo hardly makes a ringing case for the System's government-conferred privileges.

The System has accordingly sought to expand into new activities (and to promote existing activities as though the System sought to become the lender of first resort). New activities help perpetuate the System in several ways. By expanding the System's business lines, they give depository institutions additional reasons to become members. These activities can also help extend the System's political network. And in some cases the activities may help meet some significant unmet needs.

In a recent example of dubious expansion, the Federal Housing Finance Board broadened the Home Loan Banks' authority to issue financial guarantees in the form of standby letters of credit. The Home Loan Banks could use this authority for a wide array of purposes that do little or nothing to expand homeownership, such as credit-enhancing municipal bonds and asset backed securities. The market for such guarantees is already highly competitive. But expanded credit-enhancement offers member institutions another carrot and may also help cultivate additional constituencies for the System.

More broadly, just because a government-sponsored enterprise has some capability to conduct a given activity doesn't mean it should do so. From the Treasury's standpoint, proposals to expand the Home Loan Bank System raise questions about whether a demonstrable market failure exists and, if so, whether the proposal is the best way to correct it.

Real Home Loan Bank Reform

If the Home Loan Bank System did not exist today, no one would seriously propose to create a government-sponsored enterprise with anything like the System's current mix of activities. Still, the System does some good and could do more. We at the Treasury would support legislation that preserves a Home Loan Bank System genuinely reformed and refocused on a meaningful public purpose. In that context, I'd like to suggest three process principles and three policy principles.

The process principles are easily stated and deceptively simple. First, do no harm. Second, get the job done right. And third, do not preempt needed reforms.

Mindful of these considerations, we oppose piecemeal changes that would dissipate pressure for real reform and, in some cases, create perverse incentives never to undertake such reform. For example, if Congress in piecemeal fashion gives the System's insiders the relatively few things they seem to want from Congress (e.g., devolving management authority from the Finance Board to the Home Loan Banks, or liberalizing borrowing rights and membership terms for depository institutions with less than $500 million in assets), it removes the incentive to go along with other reforms. Thus piecemeal change can preempt real reform.

But piecemeal change could do worse than that; it could actually obstruct real reform. Proposals to reallocate the REFCorp obligation among the 12 Home Loan Banks provide a case in point. Current law requires the System to contribute a fixed $300 million a year toward REFCorp interest payments, and arbitrarily allocates that obligation among the 12 Banks. Reform proposals would commonly replace the fixed dollar obligation with a requirement that each Home Loan Bank contribute a specified percentage of its net income toward REFCorp payments. Such a change would make eminent sense in the context of broader reform. But if made piecemeal in particular, without curtailing the System's swollen investment portfolio such a change could conceivably impede real reform. Because once restated as a percentage of income, the REFCorp obligation might be construed to create powerful budget incentives to expand the System. Specifically, pay-as-you-go rules might treat legislation curtailing the System's arbitrage portfolio (and thus shrinking the System's net income) as revenue-losing, even if it would actually protect the taxpayers by reducing the liabilities covered by the System's perceived government guarantee. Those rules might also treat legislation expanding the System (and thus increasing the System's net income) as revenue-raising, even if the System had no good policy reason to conduct the expanded activity.

In addition to the three process principles I've just outlined, I would also propose three substantive policy principles. First, any legislation should tightly connect advances (and other activities) with the System's public purpose.

Second, any legislation should include objective, enforceable limits on the System's investment portfolio. We have endorsed generally limiting the total dollar amount of bonds a Home Loan Bank can have outstanding (the so-called consolidated obligations) to the total dollar amount of the Home Loan Bank's outstanding advances. We would, in effect, limit a Home Loan Bank's investments to its capital plus its member deposits, which would still let the System hold a sizeable investment portfolio ($45 billion as of October 31, 1998). Thus the System could raise funds at government-subsidized rates only if it advanced those funds to its members.

Third, legislation should embody the principle of equal rights and responsibilities for all Home Loan Bank members and preserve the desirable aspects of the System's current capital structure. We believe that the current structure a single class of redeemable stock would, with only modest changes, be adequate even in the context of voluntary membership for all institutions (which we support). The current structure lets capital expand and contract along with demand for advances. It fits the System's cooperative character. And it would also facilitate additional changes if Home Loan Banks decided to undertake them.

Conclusion

The Home Loan Bank System's original reason for being has eroded and left the System seeking ways to make itself useful and justify its continued existence. Yet the System is not a fully private business; its various privileges amount to a significant government subsidy. The System should not have free rein to expand into lines of business for which reasonably effective private markets already exist. Congress not the System should decide whether to expand the System's activities. And Congress should permit such expansion only after a rigorous process in which it ascertains that a market failure exists and concludes that the System is the best way to correct that failure.