A market is where people come to make exchanges. Every market has its own rules, and markets thrive or wither, in part, depending on the choice of those rules. Clear rules for payment; the penalties for nonpayment, fraud, and nonperformance; and the rules for resolving disputes, for example, usually induce growth of the market, increasing the expected net benefit to each party. Redistributive rules among or between buyers and sellers, however, usually lead one or more parties to leave the market. U.S. financial markets today face several major new policy threats. Most of the new threats have a common pattern: the government is using existing regulatory authority or proposing new authority to aid some parties in the market at the expense of others. Moreover, the government is no longer satisfied to affect the allocation of new financial flows; it is trying to reallocate the stock of assets — about $3 trillion of bank reserves, nearly $5 trillion of pension funds, and over $2 trillion of insurance assets. Those measures will reduce the affected financial markets and increase the demand for other types of credit or risk-spreading instruments. The major new policy threats to banks, pension funds, and insurance are addressed in that order.
Federal bank regulators and the Department of Justice have increasingly reinterpreted their authority under existing law to develop an extensive system of credit allocation. The four statutes under which bank regulations are issued are the Fair Housing Act of 1968, the Equal Credit Opportunity Act of 1974, the Home Mortgage Disclosure Act of 1975, and, most important, the Community Reinvestment Act of 1977. The common objective of those four laws was to reduce the alleged discrimination in bank lending to minorities. I say "alleged" because the premise that banks discriminate is both implausible and unsupported.
The premise is implausible because it implies that banks consistently forgo profits by denying loans to minority applicants with good credit records. For many people, evidence that blacks and Hispanics are denied bank credit at a substantially higher rate than are whites and Asians is evidence of discrimination. The record is not sufficient evidence of discrimination, however, if credit risk is correlated with race. A major study by the Federal Reserve Bank of Boston in 1992 found some evidence of discrimination by race, even after controlling for other correlates of credit risk; that study, however, has been subject to sharp analytic criticism, charging massive errors in the data base and that the results were due to two banks that specialized in lending to minority loan applicants. Two other types of evidence are probably more important:
1. The relative denial rates by race are about the same for minority- owned banks as for other banks.
2. If the denial rates were based on discrimination rather than credit risk, the default rate on approved loans to blacks and Hispanics would be lower than on loans to other groups; there is no evidence of that effect.
In summary, there is no consistent evidence that banks discriminate among loan applicants by race, either consciously or inadvertently. In Washington, however, no good deed goes unpunished. Two major banks with records of outreach to minority borrowers have been subjected by the Department of Justice to what is best described as extortion. In a major 1993 case, following actions against three small banks, the Federal Reserve held up approval of several proposed acquisitions by Shawmut National Corporation pending resolution of a discrimination suit brought by Justice against Shawmut's mortgage company subsidiary. The facts of the case are clear. During the period when the alleged discrimination occurred, Shawmut had an aggressive program to increase mortgage lending to minority applicants. Shawmut relaxed its normal lending criteria, substantially reduced the rejection rate on loan applications by minorities, and doubled the amount of new mortgage lending to minorities. Although no private person filed a discrimination complaint, the Department of Justice charged Shawmut with discrimination, based on findings that some of the loan officers had not been as aggressive as others in approving loans to minority applicants and that Shawmut had no internal review procedure to ensure that all the loan officers used the same lending criteria. In order to remove the barrier to approval of its proposed acquisitions, Shawmut agreed to settle that absurd case, set aside $1 million as a settlement fee, and worked with Justice to find some "victims" of the alleged discrimination to share the fee.
A later case is, if anything, even more outrageous. The Community Reinvestment Act had previously been interpreted to obli-gate banks to serve the communities from which they drew deposits. In 1994, however, the Department of Justice charged the Chevy Chase Federal Savings Bank of Maryland with a crime not covered by law, failure to open enough branches in predominantly black areas of Washington and a suburban Maryland county. In fact, Chevy Chase had made many loans and issued credit cards to residents of those areas and had an unusually low rate of denying loans to minority applicants. In this case the Department of Justice, again as the price of granting other regulatory permits, forced Chevy Chase to agree to open four branches in black areas, budget advertising in black publications, adopt more employment quotas, and make loans to blacks at below-market interest rates.
Last October the Department of Justice made a similar charge against Barnett Banks, the largest bank in Florida, and plans to bring suit unless Barnett agrees to a settlement. I do not fault the banks for consenting to this extortion; it is difficult for any one bank, even a large one, to take on the federal regulators and the Department of Justice. These are classic cases in which the individual businesses should have been defended by their trade association. Instead of challenging the government in these cases, however, the American Bankers Association took the position that all financial institutions should be subject to the same type of onerous regulation. With friends like that, American banks may need a new trade association. For banks, the bottom line is that what they do best — evaluating credit risk — has made them vulnerable to charges of racial discrimination. The government attempt to force banks to cross-subsidize credit will further reduce the bank share of the credit market with consequences that are quite different from the presumed objective of the government. The rich will still be able to borrow against their mutual funds, and the poor will be left with check-cashing outlets and pawn shops.
For years state and local government pension funds have been vulnerable to mandates to make politically favored "social investments." Several studies have found that the average yield on those investments is substantially lower than on other investments. As a rule, teachers, police, and other state and local employees bear the costs of those low-yield politically targeted investments. A higher share of total compensation must be set aside to fund the promised pension benefits, reducing the direct wages of public employees. Similarly, if the pension assets are not sufficient to fund the promised benefits, those public funds may go bankrupt with the consequent loss of pension benefits. The rationale for financing politically targeted investments at the expense of public employees has never been made quite clear.
For the most part, private pensions have not yet been required to make investments in politically favored groups. Private defined benefit plans are regulated because they are also insured, and the regulations authorized by the Employee Retirement Income Security Act are designed solely to ensure the safety and soundness of those plans, for the protection of both the plan participants and the federal pension insurance fund. A large and increasing share of private pensions, moreover, is in defined contribution plans for which each plan participant has the opportunity to choose his or her own fund manager and investment portfolio.
Until recently, ERISA has been interpreted to preclude investments that are not expected to yield the prevailing rate on other investments of the same risk and liquidity. In November 1992, however, a Department of Labor report discussed a procedure for valuing the "net externalities" of investments as a way of broadening the prevailing rate test to include economically targeted investments (ETIs). And in September 1993, Olena Berg, Assistant Secretary of Labor for Pensions and Welfare Benefits, announced a more expansive interpretation of the prevailing rate test that would "allow collateral benefits to be considered in making investment decisions where such investments are prudent and provide a competitive risk-adjusted return to the plan." She especially encouraged pension fund investment in firms that invest in their own workforce.
So there was ample warning of the next step. On June 22, 1994, the Department of Labor issued an interpretative bulletin stating that the fiduciary standards of ERISA "do not prevent plan fiduciaries from deciding to invest plan assets in an ETI if the ETI has an expected rate of return that is commensurate to rates of return of alternate investments with similar risk characteristics." The department claimed that this was the original interpretation of ERISA and backdated the new interpretative bulletin to the date that ERISA was first effective, July 1, 1975. At a congressional hearing, Labor Secretary Robert Reich and Housing and Urban Development Secretary Henry Cisneros claimed that the new interpretative bulletin would make it possible to tap into private pension funds to finance their favorite projects without increasing the risks to those funds, but the secretaries never quite explained how that was possible. At the same hearing, I testified that the new bulletin "was either meaningless or mischievous — meaningless if it does not weaken the strict ERISA standards, mischievous if it does."
So far, the threat to private defined benefit plans is only a cloud on the horizon, but it is a dark and ominous cloud. As is the case with the public employee plans, the direct losses from politically directed investments would be to the plan participants, in the form of either lower wages or a failure of the plan to pay the promised benefits. In this case, moreover, the losses would be shared by the federal pension insurance fund, and ultimately the taxpayer, if plans fail to pay the guaranteed benefits. Several commentators have suggested that ERISA be broadened to protect public employee pension plans against pressure to invest in politically favored groups. The first priority, I suggest, is to ensure that ERISA is administered as intended — to protect the safety and soundness of private defined benefit plans. Twenty-one years ago, Congress passed ERISA to protect private pensions against the consequences of irresponsible private pension sponsors. It would be a sad irony if the officials responsible for administering ERISA undermined pensions in the interests of politically targeted investments. For pension funds, the bottom line is that what they do best — choosing a portfolio that best serves plan participants — is increasingly vulnerable to political pressure. Public plans are already vulnerable, and private plans may soon be as well. The probable effects would be a continued erosion of real wages, defined benefit pension plans, and, possibly, total private saving for retirement.
For the most part, insurance markets are regulated by state governments, not the federal government. That has not protected insurance markets, however, from the types of redistributionist policies the federal government has already applied to banks and, potentially, to pension funds. Many state governments, for example, subsidize high-risk drivers by assigning them to broader auto insurance pools and then mandating (more accurately, trying to mandate) universal coverage. Some states have required insurance companies to offer the whole range of policies in that state in order to broaden the base for mandated cross-subsidies. In response, some low-premium companies have withdrawn entirely from those states.
For better or for worse, California is often the wave of the future, and a recent case is most disturbing. After some very bad underwriting and huge claims as a result of the Northridge earthquake, 20th Century Industries asked the California Department of Insurance for a minor premium increase on its auto insurance policies. Two intervenors withdrew their objections to the increase in exchange for a commitment by 20th Century to deposit $1 million in Oakland banks, contribute $50,000 to local organizations, and market its policies in Oakland and outer San Diego County. The lead intervenor, a vocal community activist, also expects $115,000 in fees from 20th Century for her intervention. This is another case where it does not pay the directly affected company to challenge such extortion; the company should have been defended by some association of insurance companies to avoid setting a precedent for more such outrages.
There are two major potential federal threats to insurance markets. In April 1992, Rep. John Dingell, then chairman of the House Energy and Commerce Committee, introduced a major bill that would impose federal regulations on the insurance industry. That bill would also establish a national reinsurance fund through which solvent insurers would cover the policy claims on failed insurers, but without a federal guarantee. Insurers that do not elect to participate in the federal system or do not meet the federal solvency standards would continue to be regulated by the state insurance commissions. All insurers would continue to be subject to continued state regulations of insurance rates.
The primary problem of the Dingell bill is the separation of solvency and rate regulation; state insurance commissions would have even less incentive to be concerned with insurer solvency in setting insurance rates. The other problem of the proposed bill is that there is probably no way to avoid an implicit federal guarantee of a national reinsurance fund, with all of the consequent problems of the deposit and pension insurance funds and the several federal credit programs.
The Dingell bill is seriously flawed but could provide the basis for useful legislation. One productive outcome would allow national insurers to elect federal solvency regulation in exchange for full immunity from federal or state rate regulation. In the absence of such a provision, the seriously flawed system of state insurance regulation is probably better than any federal regulation. The most innovative alternative, recently proposed by the majority leader of the Michigan senate, would be to create one or more interstate insurance compacts that would determine both the solvency rules and the reinsurance guarantee for those companies that elected to operate under their rules. Such voluntary compacts are more likely to lead to a spread of good rules than are either the separate actions of the states or the likely outcome of the process to establish a federal insurance code. The other major federal threat to insurance markets is health reform. Most of the major health reform proposals considered in 1994 had some form of "community rating," a warm and fuzzy term that disguises its meaning — a legislated cross-subsidy from low-risk people to high-risk people. Community rating would increase the premiums on low-risk people and reduce the premiums on high-risk people. Further, unless community rating were paired with an effective mandate on coverage, some low-risk people would drop health insurance, and the insurance companies would try to avoid making a market for high-risk people. The experience in New York State since the spring of 1993 has already confirmed those effects.
For insurance companies, the bottom line is that what they do best — offering the best possible policies for people in each risk pool — is increasingly disparaged as "cherry picking." Unless there is a major change in the rhetoric of the debate on these policy issues, the insurance companies will end up as paper pushers in a government-managed prepayment system, not as entrepreneurs who make the market for insurance.