Central Banking after Bear Stearns By Allan H. Meltzer Carnegie Mellon University and the American Enterprise Institute
Prepared for the Euro 50 Group Paris, October 3, 2008
My comment on current central banking is mainly about the Federal Reserve. I refer to the ECB only for comparison. A critical difference between the two is their relative independence. The ECB’s mandate is price stability. The Maastricht treaty protects it. The Federal Reserve is the agent of Congress and is often reluctant to ignore Congressional wishes.
Recent Federal Reserve actions to respond to financial problems are unprecedented. Comparisons to the Great Depression miss a main point. The Federal Reserve did not “rescue” banks during the depression. It did not consider failures as its responsibility. The federal government acted through agencies like the Reconstruction Finance Corporation, the Home Loan Banks and the Homeowners Loan Corporation. Rescues and assistance were fiscal not monetary actions. Calling recent Federal Reserve action unprecedented is not an overstatement. In the deep recession of 1920-21, Congress urged the Federal Reserve to help distressed farmers who bought land during the wartime commodity boom and inflation. By 1920-21, the United States had a severe deflation. The Federal Reserve was less than ten years old, but it was sufficiently mature to tell the Congress that it was not a rescue agency. Congress established the Federal Land Banks, fiscal not monetary assistance. In the early 1930s, Congress leaned on the Federal Reserve to help the housing industry, then in a much deeper and more widespread crisis then now. The Federal Reserve responded negatively again. Congress created the Home Loan Banks and later other agencies. Current large-scale lending to Bear Stearns, other security dealers and investment banks and an offer to lend to the failing government housing agencies is not only unprecedented in scale, it is mistaken. A central bank is not an all purpose facility. It should follow its own earlier
precedents by telling Congress to vote its appropriations instead of avoiding a vote by pushing them off to the Federal Reserve’s balance sheet. As I write this, there are some signs of a welcome change in policy, refusal to bailout Lehman and Merrill, followed by the inconsistency of bailing out AIG. I doubt that the financial system can survive in its present form if the bankers continue to make the profits and the taxpayers assume the losses. One or the other is likely to change. The choice will have major consequences for the future of the U.S. financial system.
What Were the Mistakes? The financial crisis did not just happen. Mistakes by regulators and regulated contributed. To reduce future risk of crisis, the origins become important. And the system must change. First, the regulators made some mistakes. My first law of regulation says that lawyers and bureaucrats make regulations, but markets decide to circumvent the most costly rules. The regulators act as if incentives do not matter. In the Basel Agreement, regulators told banks to hold more capital if they took more risk. Banks responded by putting many of the risk s off the balance sheet. We changed from a system that was not well monitored to a system that was not monitored at all. No one knew where the risks and losses were. The banks later learned that they had not rid themselves of the risks they put off their balance sheet. Second, the compensation systems in financial markets create perverse incentives. Why did MBAs who graduated in the past from the world’s leading business schools buy and sell paper that they know was nearly worthless? They received high rewards for doing so, and often lost their jobs if they didn’t. Their managers and the managers’ management shared in the profits. There were a few exceptions, but only a few. Third, rating agencies profited also. They enabled buyers and sellers to ignore the actual quality of the paper. Fourth, unlike the ECB, the Federal Reserve responded aggressively to what they thought would be a serious recession. This was a mistake. The economy suffered a wealth loss, a permanent reduction in wealth imposed by the wealth transfer to oil exporters. Real incomes in the oil importing countries must fall. Monetary expansion can postpone the adjustment, but it cannot prevent it. At the same time, housing prices fell, lowering wealth also. We do not know
2
the degree to which homeowners believed earlier housing price increases were permanent additions to wealth and how much they were treated as transitory. Concern about recession caused a near panic response by the Federal Reserve in January. The Board seems to have yielded to pressures from Congress and exaggerated claims about a coming depression by large lenders especially in New York. The most vociferous of the lenders mistook their portfolio losses for economic collapse. The economy muddled along. Output may decline as the economy adjusts to its wealth losses. The 3.25 percentage point reduction in interest rates did not prevent the economic slowdown. Current problems of slow growth are not primarily monetary. This is the third large oil price increase since 1973. The Federal Reserve should have learned to distinguish between a one-time permanent reduction in wealth and income that lowers the economy’s growth path and a recession. In the latter, the economic growth path remains unchanged. Output declines below the path and later returns to it. The Federal Reserve is in the money business. It can compensate for a decline in aggregate demand in a recession, but it is not in the oil business. It cannot compensate for the rise in oil prices. A central banks proper response to an oil price increase is to prevent the rise in the relative price of oil from becoming an increase in the sustained rate of change in the general price level. Keeping short-term real interest rates at negative values is counterproductive. The Federal Reserve should slowly raise the real value of the federal funds rate to zero. Journalists often ask whether I approved of the assistance the Federal Reserve gave to Bear Stearns and promised to Fannie Mae and Freddy Mac. They seem to accept that by answering “yes,” I accept U.S. government policy. I try to explain that they are misled by their question. If the regulators wait until most options are foreclosed, they reduce choice to either taxpayer assistance or risk of major financial panic. Writing more than 130 years ago, Walter Bagehot urged the Bank of England to pre-announce its lender-of-last-resort policy. By following its pre-announced policy, the regulator avoids having to make its decision in the midst of a crisis. In its 95 year history, the Federal Reserve has never announced a lender-of-last-resort strategy. Sometimes it lends; sometimes it doesn’t. The absence of a clear policy increases uncertainty and encourages failing institutions to push for assistance.
3
Recently Treasury Secretary Henry Paulson proposed to broaden the Federal Reserve’s responsibility and authority to regulate non-bank financial institutions. His proposal ignores the Federal Reserve’s record as a regulator. With help from the IMF it prolonged the Latin American debt problem for most of a decade to protect the New York banks. A solution to the problem came from the Treasury, after Citicorp wrote down the value of its holdings and other banks followed. The Federal Reserve ignored the saving and loan problem in the 1980s. It was slow to act against the growing housing problem in 2005 and 2006. In 1991, Congress passed deposit insurance changes, FDICIA, to prevent the Federal Reserve from lending too much to failing banks that the losses shifted to the deposit insurance fund. All investment banks and broker-dealers balance their accounts every night. Many borrow in the overnight loan market to settle their accounts. Instead of regulating portfolio decisions, all financial firms should be subject to FDICIA. If they cannot balance their accounts, the regulator can require them to stop paying dividends to stockholders. If problems continue, the regulator takes control, eliminates management and stockholders while the firm has positive net worth. Later, the firm is merged or sold. The Federal Reserve should announce a version of Bagehot’s rule. It should lend at a penalty rate to all distressed financial institutions that present good collateral. It should not prevent failures. Counterparties with acceptable collateral should be permitted to borrow. A rule of this kind worked well in Great Britain for about a century. And it worked well in the 1987 stock market decline. As Bagehot explained, it can end a panic quickly without shifting losses from failing firms to the taxpayers. The Bagehot rule recognizes that regulators are supposed to protect the public’s interest, not the interests of private firms. Pre-commitment is essential. The regulator is besieged from many sides warning that the country will suffer a depression if he fails to rescue the failed firm. Some will tell him or her that history will attach his or her name to the depression. I have watched well-intentional Treasury secretaries yield to this pressure, often against their own beliefs. Pre-announcing a policy does not guarantee that it will be followed. It only increases resistance to salvaging failed firms. Next, a word about the failed mortgage companies Fannie and Freddy. Many have pointed out that the risks to taxpayers were well-known and widely discussed. I would add that the agencies were never needed if their purpose was to subsidize home ownership. Many years
4
ago, I published a paper showing that providing mortgages does not increase housing. The base idea is simple: Mortgages are pieces of paper; houses are real assets. At first, this view was met with scorn. But I believe it is now well-accepted. The subsidy that Fannie and Freddy received from the implicit, now explicit, guarantee subsidized housing. Treasury guaranties were the vehicle carrying the subsidy. I can think of two reasons why the subsidy took this form. First, it avoided the often contentious budget battles. Key members of Congress could increase the subsidy by this indirect means. Second, it became a source of funds or contributions to friendly Congressmen. Neither reason is attractive or compelling. The Treasury should close Fannie and Freddy and urge Congress to put housing subsidies in the budget. We do not need and will not benefit from new regulations described as reform. We must change the incentives and clearly announce that credit policy will follow Bagehot’s rule so that losses are borne by those who incurred them and the regulators will restrict their efforts to supporting the market by discounting negotiable paper at a penalty rate to avoid abuses. Finally, let me suggest an answer to the critical question: How and when will the mortgage and housing debacle end? My answer is that it cannot end until markets have a rather firm anticipation of where housing prices will settle. The reason is that no one can correctly value many mortgages until housing prices are expected to stabilize. Uncertainty about mortgage values and underlying housing values restricts trading. Lenders are reluctant to commit for a month or more until they have more certainty about the quality of a borrowers’ portfolio. There is not much that public policy can properly do. Buying mortgages can transfer the risk to the taxpayers, but it cannot eliminate the risk. And it increases the probability that the crisis will return, perhaps in a new form. The decline in housing prices continues, but the rate of decline may be falling. Once prices are expected to reach a sustainable level, markets will gradually improve.
5