With all the subsidies and assistance, expansion of mortgages and housing should not surprise anyone. Between 1980 and 2007, the volume of mortgages backed or supported by the three government-chartered agencies rose from $200 million to $4 trillion, an unsustainable compound growth rate of 36 percent a year. As the volume rose, the quality of mortgages declined. Government encouraged this development; in 1999 the FHA introduced a zero down payment loan, as noted above. Lenders expanded subprime mortgages, mortgages to buyers with relatively poor credit histories. Soon afterward, mortgage lenders began to offer mortgages that did not require a down payment. Then they eliminated credit checks on some mortgages. Such mortgages are called Alt-A.
Purchases and support for these subprime and Alt-A mortgages put Fannie Mae and Freddie Mac at much greater risk than in the past. In December 2008 congressional testimony, the heads of three agencies explained that they were aware of the increased risk but believed it necessary to compete with the private market. They did not add that the Federal Home Loan Banks supplied almost half the funding for two large private lenders, Countrywide and Indy Mac, that later failed. Nor did they add that Fannie Mae and Freddie Mac owned $1.5 billion in assets related to subprime and Alt-A mortgages, about half the total outstanding. Prodded by members of Congress and the Clinton and Bush administrations, they lowered the quality of their portfolios to promote home ownership. With the failure of Fannie Mae, Freddie Mac, Countrywide, and Indy Mac, taxpayers will bear a considerable loss.
Edmund Gramlich, a member of the Federal Reserve’s Board of Governors, reported on the problem but did not formally warn the Board about the deterioration of loan quality. William Poole, president of the Federal Reserve Bank of St. Louis, spoke publicly about the taxpayer’s risk and urged remedial action. Alan Greenspan warned Congress about growth of Fannie Mae and Freddie Mac. There were many other warnings including from Senator Shelby, a senior member of the Banking Committee. Congress declined to act and several members denied that there was a problem. Congressional inaction increased the incentive for Fannie Mae and Freddie Mac to accept very risky loans.
There are homebuilders, mortgage lenders, and real estate agents in every congressional district. This alone encourages support for mortgage and housing subsidies and delays corrective action. It is very likely that the government will continue to subsidize homeownership. Reform should seek to put the subsidy on the budget and subject it to the appropriation process. Government mortgage market operations were a means of hiding the subsidy and often denying it. The subsidy took the form of a reduced interest rate on Fannie Mae and Freddie Mac borrowing, and did not require off-budget finance.
Fannie Mae and Freddie Mac are in receivership and under government control. They should be liquidated and terminated. Congress should vote the subsidy directly.
After much hesitation and policy change, the Treasury used most of the money in the first half of the Troubled Asset Relief Program (TARP) to supply capital to banks and other financial institutions. No large bank was allowed to fail. Once the banks received this assistance, many in Congress wanted to influence the banks’ lending. Congress urged them to lend even if it meant acquiring risky loans with substandard repayment prospects.
A better alternative would have required bankers to borrow part, perhaps one-half, of the additional capital in the market. That would have increased a bank’s cost, but it would deter some banks from borrowing from TARP and identify banks that the market considered insolvent. Those banks should fail. Failure means that shareholders lose their investment and management loses its job. The reorganized bank should be sold or merged.
The government and Federal Reserve treat all large banks as “too big to fail.” That encourages gigantism. Instead, policy should impose a different standard: if a bank is too big to fail, it is too big. The social cost of losses to taxpayers exceeds the social benefit of large banks. The new standard would increase the incentive for bankers to be prudent.
Role of the Federal Reserve
Many politicians, bankers, and journalists blamed the housing and mortgage crisis on the Federal Reserve. The basis of their complaint was that from 2003 to 2005 the Federal Reserve held the federal funds rate at one percent. This permitted credit expansion, much of which concentrated in the mortgage market.
During these years, Chairman Alan Greenspan believed and said that the country faced risk of deflation. That was a mistake. Deflation is very unlikely to occur in a country with a relatively large budget deficit, a longterm depreciating currency, and positive money growth. Critics are correct about this part of their criticism. Federal Reserve policy was too expansive as judged by the Taylor rule or the 1 percent Federal funds rate that held the real short-term interest rate negative in an expanding economy.
The next part is wrong. The Federal Reserve did not force or urge bankers and others to buy mortgage debt. That was the bankers’ decision. Prudent bankers avoided excessive accumulation of low-quality mortgages. Bankers could have purchased Treasury bills or other assets with lower risk. They decided to overinvest in very risky assets and to lower quality standards. They share responsibility.
One plausible explanation of the errors that many made was the socalled “Greenspan put.” Whether such a put was available, the belief was widespread that the Federal Reserve would prevent large losses, especially for large banks. Several bankers and investment bankers raised the leverage they accepted and invested in risky assets. Whether or not there was a Greenspan put, prior actions that prevented financial failures, for example protecting Long-Term Capital Management (LTCM), created moral hazard and reduced concerns for risk. Arranging the rescue of LTCM is the most recent example in a long history of preventing failures. Notable examples include First Pennsylvania Bank, Continental Illinois, and most of the New York money market banks during the Latin American debt crisis. Bankers had reason to believe that the Federal Reserve would prevent failures.
One of the criticisms earlier in this history of the Federal Reserve is that the Federal Reserve has not announced its lender-of-last-resort strategy in its ninety-five-year history. Sometimes institutions fail, sometimes the Federal Reserve supports them, and sometimes it arranges a takeover by others. There is no clear policy, no policy that one can discern. But there was a firm belief that failure was unlikely at large banks.
The absence of a policy has three unfortunate consequences. First, uncertainty increases. No one can know what will be done. Second, troubled firms have a stronger incentive to seek a political solution. They ask Congress or the administration for support or to pressure the Federal Reserve or other agencies to save them from failure. Third, repeated rescues encourage banks to take greater risk and increase leverage. This is the wellknown moral hazard problem.
As financial problems spread in 2008, pressure built on Bear Stearns. The Treasury and the Federal Reserve arranged a takeover. The Federal Reserve contributed by buying—not lending—$29 billion of risky assets. Markets improved. Many bankers claimed the worst was over. A few months later Lehman Brothers failed. Without prior warning, the Federal Reserve and the Treasury announced that they would not prevent the failure, a major policy change. Next the Federal Reserve prevented the bankruptcy of American International Group by replacing management and providing up to $80 billion in credit.
What conclusion could a portfolio manager draw? There was no clear pattern, no consistency in the decisions. Uncertainty increased. Portfolio managers all over the world rushed for the safety of Treasury bills. A classic panic of the kind described by Walter Bagehot followed. Officials did not announce or follow a clear strategy, as Bagehot urged. Regulators reacted to each subsequent rush for safety by guaranteeing in turn bank deposits, money market funds, commercial paper, and other instruments.
Influenced by Bagehot’s (1873) criticism, the Bank of England announced the lender-of-last-resort policy that it had followed in past crises and successfully followed the policy into the twentieth century. Panics and fail
ures occurred, but they did not spread or accumulate. The policy called for lending without hesitation in a crisis at a penalty rate against acceptable, marketable collateral. That policy induced prudent bankers to hold collateral, and it reduced uncertainty.
By guaranteeing deposits, money market liabilities, and other instruments, the Federal Reserve prevented bank runs and further breakdown of the payments system. Unlike response during the Great Depression, depositors could not demand gold from banks, but they could demand currency and use deposits to buy gold or Treasury bills with the same effect. Because banks and other financial firms were unwilling to lend to other firms, they too bought Treasury bills and held idle reserves. The Treasury and the Federal Reserve supported these demands by paying interest on idle reserves and by exchanging Treasury bills for less liquid assets.
The Federal Reserve acted creatively to establish new lending facilities to accommodate market demands. They put off to the future any consideration of how and when they can reverse these overly expansive actions.
One lesson from the current crisis is that the Federal Reserve should announce a lender-of-last-resort strategy and follow it without exception. A second lesson is that Congress should dispense with “too big to fail.” Banks and financial firms should not have incentives to become so large that they cannot fail. “Too big to fail” encourages excessive risk taking and imposes costs on the taxpayers. If banks considered too big to fail are not reduced in size, they should have substantially higher capital requirements including subordinated debt. The very high leverage ratios at large financial institutions responded to the incentives created by earlier rescues and belief in a Greenspan put.
One of the Treasury’s proposed reforms gives the Federal Reserve responsibility for maintaining financial stability. This is a poor choice. The Federal Reserve did nothing about growing savings and loan failures in the 1980s. Ending that crisis cost the taxpayers about $150 billion. The Federal Reserve worked with the International Monetary Fund to protect lending banks during the Latin American debt crisis. The crisis began to end when Citicorp’s chairman decided to recognize the losses by writing down the debt’s value. Others followed. Soon thereafter, the Treasury began a systematic program to write down the debt. The Federal Reserve did nothing.
Although Alan Greenspan warned publicly in 1996 about irrational exuberance in the equities market, neither the Federal Reserve or the Securities and Exchange Commission tried to prevent rampant stock market speculation. And it followed by doing nothing to prevent the large expansion of subprime, Alt-A, and other mortgage loans and the rise in housing prices. This error will cost taxpayers much more than the savings and loan failures.
Reading transcripts of Federal Open Market Committee meetings, one finds very little discussion of regulatory and supervisory credit problems. The Federal Reserve’s record does not support a proposal to increase its responsibility for financial stability. More important, regulation of this kind can succeed only if the regulator makes better judgments about risk than those whose wealth is at risk. A better change would make risk takers bear the risks they take. Failure should remove management and cost stockholders as in the FDICIA rule. Companies would not disappear. They would get new management and stockholders.
FDICIA
In 1991, Congress passed the Federal Deposit Insurance Corporation Improvement Act (FDICIA). A main reason for the act was to reduce Federal Reserve lending to failing banks, thereby reducing losses paid by the FDIC. FDICIA gave regulators authority to intervene in solvent banks when losses reduced capital below required limits and to assume control before a bank’s capital was entirely gone. The bank could then be sold or merged. Stockholders would take the loss and managers would be replaced. The regulators did not apply FDICIA standards to failing financial firms in this crisis. FDICIA should be extended to apply to all financial institutions. It is an explicit rule that, if enforced, is known to all interested parties. Prudent bankers will act to avoid failure and the loss of their jobs.
Regulation
The financial crisis brought many demands for increased regulation. Few recognize that regulation works best if it takes account of the incentives it fosters. The Basel Accords agreed to by developed countries are a timely example. The accords required banks to hold more capital if they acquired more risk. The rationale seems clear and unassailable. The practice was very different.
Instead of increasing capital, banks chartered new entities to hold the risky assets. The intent was to keep the risk off their balance sheets. When the mortgage crisis occurred, the banks had to assume the risk and responsibility for losses. Regulation failed, and so did circumvention. The cost to the public is very large. This experience shows again that lawyers and bureaucrats choose regulations, but markets circumvent costly regulations.
Successful regulation recognizes that it creates incentives for avoidance or circumvention. Successful regulation aligns the interests of the regulated with socially desirable outcomes. Successful regulation induces market action to eliminate externalities. Successful regulation recognizes that market participants respond to regulation by changing their actions to find a new optimum.
Regulators rarely respond to this dynamic process by adopting regulations in response to market outcomes. Because all countries have some type of deposit insurance, either de jure or de facto, regulation must limit risk taking. FDICIA provides an incentive to avoid excessive risk. Capital requirements also help to align incentives and avoid excessive risk taking. Regulations such as the Basel Accords do not meet this standard.
After the Treasury supported General Motors and Chrysler with what will be a growing bailout of automobile companies, the Federal Reserve accepted General Motors Acceptance Corporation (GMAC) as a bank, enabling GMAC to borrow at the discount window. GMAC at once began to offer zero interest rate loans for up to five years to borrowers with below median credit ratings. This appears to be a response to pressure from prominent members of Congress, a further sacrifice of independence. Many members of Congress want the Federal Reserve to allocate credit to borrowers that they favor. This avoids the legislative and budget process just as Fannie Mae and Freddy Mac did. It subverts the principles of an independent central bank.
Independence is not just important. It is a critical part of the institutionalization of a low-inflation policy. It prevents Congress and the administration from financing deficits by printing money. And it avoids pressures for credit allocation to politically favored groups.
Compensation and Incentives
MBAs who graduated from the world’s leading business schools purchased and sold mortgages that carried a high degree of risk. In many cases they accepted the credit ratings supplied by others without investigating accuracy. At many banks, traders were well rewarded for doing the transactions and likely fired if they failed to do so. Compensation systems at many firms rewarded short-term increases in revenue without regard for longterm losses. Compensation systems of this kind encourage excessive risk taking.
Not all firms behaved alike. We know now that J.P. Morgan Chase, Bank of America, and some others limited risk taking much more than Citigroup, Merrill Lynch, Bear Stearns, and other failures.
Setting compensation schedules is management’s responsibility. Congress cannot establish rules that managements cannot circumvent, if they choose to do so. An improved compensation system would spread rewards over time to permit losses to be recognized. This can be done in many ways. Regulators should encourage and monitor the actions that managements take, but should leave the choice of compensation schedule to management.
Rating Agencies
The mix of incentives facing rating agencies are well-known as a contributor to the credit crisis. The agencies applied a rating system that had worked for decades in rating corporate bonds. This may have misled users. More seriously, rating agencies at times adjusted their ratings to satisfy client demands.
Not all of the fault falls on the rating agencies, but they share
the blame. The clients did not look at the underlying securities or question the ratings except to ask for more favorable ratings. They too share the blame.
Rating agencies must develop compensation and incentive programs that reward accuracy of rating achieved over time. The aim is to give the agency and its personnel incentives for diligence and accuracy.
Transparency and Risk
More information improves decisions and reduces risk. But transparency and increased information are most useful when interpretation is clear. Better reporting of asset and liability positions is most useful when risk models permit users to interpret the information correctly.
Risk models contributed to the credit crisis. These models use standard distributions. They make no distinction between permanent or persistent and transitory changes. Deciding whether risk spreads had permanently fallen before the crash or would return toward historic averages played a role in the crisis. Similarly, risk models were not useful for deciding whether the increase in house prices, or the decline in 2007, would persist. Improving ability to judge persistence can improve judgments and economic performance. Using rating agencies’ judgments without due diligence is a mistake.
A History of the Federal Reserve, Volume 2 Page 80
