Good evening and welcome to the Federal Reserve Bank of Richmond. For more than a decade now, we have worked with the Powell Center for Economic Literacy to put on this biennial conference. Our purpose has been to help enhance your ability to teach students of economics and broaden your perspectives on current economic issues. As you know, the past few years have been a time of great challenge for the Fed, and the challenges have abated less rapidly than I had hoped as the economy enters the sixth quarter of the recovery. Tonight I'd like to discuss our monetary policy challenges, but in light of the scholarly nature of this audience, I will draw on an illuminating piece of history from half a century ago. As usual, the views I express are my own and do not necessarily reflect the thinking of my colleagues on the Federal Open Market Committee¹.

Economists use a wide range of economic statistics to define and study the business cycle. The official dating of the recessions by the National Bureau of Economic Research, for instance, is based on looking for a consistent pattern of weakening (and eventually strengthening) in a broad set of indicators. But for most people, the single most salient measure of the state of the economy is the unemployment rate. This is a measure that people can relate directly to how they feel about the security of their job and their income. So in addition to its direct economic implications, the unemployment rate can have a powerful effect on the economic and political mood of a nation.

The unemployment rate is currently very high. Although it has come down a bit from its peak of more than 10 percent during the recession, it has remained over 9.5 percent longer than at any time since the Second World War. And the consensus outlook for a relatively slow recovery in economic output — growth in the 2.5- to 3-percent range in the next year — suggests that progress toward more desirable rates of unemployment may continue to be slow.

"Maximum employment" is one of the Fed's long-term objectives and part of its so-called "dual mandate" from Congress. The other is price stability. Inflation recently has been quite low. Overall inflation, as measured by the year-over-year change in the price index for personal consumption expenditures, has been around 1.5 percent since the middle of this year. It has been as high as 4.5 percent when energy prices spiked in mid-2008, and as low as minus 0.5 percent in mid-2009, as energy prices bottomed out. Core inflation, which strips out the volatile food and energy components, has been less volatile, with year-over-year rates over the last two years between 1.8 and the most recent reading of 1.2, down from around 2.5 percent as of mid-2008.

Most monetary policymakers talk about price stability goals in terms of the low rate of inflation that they would like to see achieved on average over the longer term. If monetary policymakers do their jobs well, inflation will fluctuate around that rate, but deviations will be temporary and not too large. I've stated my preference for a target rate of 1.5 percent. But many of my FOMC colleagues have stated preferences for inflation a bit higher — between 1-¾ percent and 2 percent. This range of views can be seen in the economic projections released by the FOMC four times a year. In addition to their outlook for the inflation, growth and unemployment over the next three years, FOMC participants submit their longer term projections, meaning "the rate to which each variable would be expected to converge over time under appropriate monetary policy and in the absence of further shocks." In the case of inflation, these long-term projections are most naturally interpreted as members' views on the average rate of inflation that is most consistent with the Fed's statutory goals.

The long-term projections for unemployment are harder to interpret. While some might think of them in terms of a "natural rate" of unemployment that doesn't change very much over time, others would emphasize that the attainable level of unemployment is itself a changing number and that, especially over the longer run, unemployment is ultimately determined by factors beyond the central bank's control. It's interesting to note that in the Committee's projections released after its June meeting — the most recent projections available — the range of members' opinions about long run unemployment is quite wide — wider than the range of their near term projections. By contrast, for inflation, the range of variation is notably smaller for the long term projections.

Given current economic conditions — with inflation running below levels viewed by most policymakers as mandate consistent, and with unemployment stubbornly high as the economic recovery proceeds slowly — the FOMC voted November 3 to further expand its balance sheet through the purchase of long-term U.S. Treasury securities. In its statement, the Committee noted that progress toward lower employment has been "disappointingly slow." That observation makes the important distinction that it is not the high level of unemployment alone that motivated the action, but rather the slow pace of improvement and the belief that further monetary stimulus could help.

The minutes of the November meeting — which also will include a new round of participants' economic projections — will provide a fuller account of the analysis that went into the decision, as well as elements of the debate within the Committee². Rather than foreshadow the minutes, which will be released November 23, I want to use our time together to look back half a century to another time when the use of monetary — and fiscal — policy to fight unemployment was hotly debated.

Fifty years ago, inflation was around 1.5 percent, right where it is today. In fact, inflation averaged just under 1.5 percent over the six years from January 1959 through December 1965³. During that period, inflation never strayed above 1.9 percent and never fell below 0.5 percent. Unemployment also was high 50 years ago, although not quite as high as it is now. At the end of the recession of 1960-61, the unemployment rate topped 7 percent. At the time, many economists were coming to view 4 percent unemployment as the benchmark for "full employment" that macroeconomic policy should strive to achieve. This thinking was heavily influenced by the work of Paul Samuelson and Robert Solow on the "Phillips curve" trade off in the United States⁴. They interpreted the empirical correlation between price inflation and unemployment — in which a lower unemployment rate is associated with higher inflation — as something of a menu for policymakers. In the absence of so-called "cost-push" inflation shocks, such as increases in oil prices or business and union pricing power, lower unemployment could be achieved by tolerating somewhat higher inflation. The large social costs associated with unemployment were viewed as justifying making maximum employment the primary goal of macroeconomic policy.

The work of Samuelson and Solow was very influential in the Kennedy administration's Council of Economic Advisors, led by Walter Heller. The 1962 Economic Report of the President embraced the new view, and stated that "four percent is a reasonable and prudent full employment target for stabilization policy."⁵ The need for economic stimulus to reduce unemployment motivated the tax cut proposed by President John Kennedy in the spring of 1963. After Kennedy's assassination in November 1963, President Lyndon Johnson pushed for early passage and signed it into law in February 1964. The Heller-led Council argued that it was safe to provide macroeconomic stimulus as long as unemployment exceeded the full employment mark, and generally opposed Fed tightening. Congress also pressured the Fed, arguing that rate increases would "vitiate" the stimulative effects of the tax cut⁶. The Federal Reserve, led by Chairman William McChesney Martin, felt interest rates had to rise when growth picked up in order to head off inflationary pressures before they emerged — a now-widely-accepted strategy known as "pre-emption."

The year 1965 would prove pivotal for U.S. macroeconomic history. Inflation registered 1.4 percent as the year began, and the unemployment rate stood at 5 percent⁷. Unemployment fell to 4 percent by the end of the year, while inflation crept up and rose above 2 percent at the beginning of 1966. From there it drifted up over the remainder of the 1960s, and was not to fall below 2 percent again until early in the 1990s.

Over the course of 1965, there was growing recognition within the FOMC of the need to tighten policy. The administration began planning for a significant expansion of U.S. armed forces in Viet Nam, and had no intention of cutting back on Johnson's Great Society programs. The resulting growth in the federal deficit would be inflationary without offsetting restraint from monetary or fiscal policy. Johnson delayed introducing a tax increase proposal because of opposition in Congress, where some preferred to cut spending on social programs. This left it up to the Fed to check inflation by raising interest rates. This was met with strong opposition from Johnson, who pressured Chairman Martin to delay any discount rate hike. In November of 1965, Martin felt he could hold off no longer, and the Board of Governors voted to raise the discount rate. This resulted in Martin's famous visit to Johnson's Texas ranch, where the President expressed his anger in characteristically strong terms⁸.

Inflation steadily deteriorated thereafter, ultimately breaching 5 percent in early 1970. As that happened, the inflation rate that forecasters and market participants expected to prevail gradually rose as well. Unfortunately, the Fed did not act forcefully during the late 1960s to bring inflation back down. Johnson's tax increase was not passed until June 1968, but was expected to have significant contractionary effects, which the Fed was pressured to offset with easier monetary policy. The rise in inflation expectations made restoring price stability much more difficult. After Martin retired as Fed Chairman in early 1970, near the end of the 1969-70 recession, his successor, Arthur Burns, concluded that maintaining an unemployment rate sufficiently high to bring down inflation would be politically intolerable, and advocated direct wage and price controls instead. A control program was adopted in late 1971, but proved ineffective, and when it was dismantled the inflationary spiral resumed. It was not until the strenuous efforts of the Volcker FOMC after 1979 that inflation was brought back under control.

The history of the pursuit of full employment in the 1960s and 70s provides several important lessons for us now. The first is the risk of presuming that we know more than we really do about what the unemployment rate can or should be at any moment. An economy in recession is responding to shocks that have disrupted the normal process of economic growth. The ability of the economy to quickly reemploy its workforce may depend on the nature of the shocks and the nature of the adjustments businesses and households must make in order to redeploy labor and capital between sectors. A permanent increase in energy prices, for instance, will shift demand away from energy-intensive goods and services. The required resource reallocations will be quite different following a collapse in residential construction resulting from the buildup of a substantial oversupply of homes, which is arguably the situation we are in right now. Historical data can be useful in understanding how different parts of the economy have moved together in response to various shocks, but are imprecise guides for normative judgments about whether unemployment is too high or too low given the most recent shocks.

A second lesson is the danger of overemphasizing the pursuit of "maximum employment." Numerous accounts from participants in the policy deliberations of the 1960s demonstrate that reducing unemployment was viewed as the primary objective of macroeconomic policy, and containing inflation was a secondary objective. Moreover, some academic economists advocated a policy framework that implied that any arbitrary unemployment rate could be sustained if society was only willing to tolerate a somewhat elevated inflation rate. This is now widely recognized as a fallacy, as was pointed out in 1968 by Milton Friedman, among others⁹. Monetary policy can alter unemployment only temporarily. Trying to keep unemployment permanently lower than it otherwise would be, as was the objective in the second half of the 1960s, is a recipe for continually accelerating inflation.

A third, and related, lesson is that it can be very costly to bring inflation down once it has become elevated. As the inflation rate creeps up, consumers and businesses can start to believe that monetary policy will continue to generate elevated inflation. They then build into their decision making the expectation that inflation will continue. The process of restoring price stability and re-establishing some semblance of monetary policy credibility following the inflationary spiral of the 1970s was a painful and costly experience. In hindsight, it would have been far better to have prevented the initial upward creep in inflation in the first place.

A fourth, and final, lesson is to avoid entanglements with fiscal policy. Attempting to fine-tune monetary policy to offset shifts in the stance of fiscal stimulus risks subordinating monetary policy to short-term political considerations, to the detriment of independence, credibility and the stability of inflation expectations.

We may not have learned all that we can from that period in our economic history, and future research will likely continue to yield new insights. But there is broad consensus that the experience of those turbulent times has improved our understanding of both the limits on the ability of monetary policy to achieve sustained reductions in unemployment and the paramount importance of preventing an erosion in inflation and inflation expectations.

These lessons are fully understood by the FOMC, I believe. In his speech at Jackson Hole in August, Chairman Bernanke clearly rejected the idea that the Fed should raise its inflation objective, even temporarily, in the pursuit of improved employment outcomes. And the Committee's statement November 3 confirms this commitment by emphasizing its intent to ensure that inflation remains consistent, over time, with its price stability mandate.

So I am confident that we can and will avoid the inflation outcomes that resulted from the flawed pursuit of full employment a half century ago. But risks remain, especially those associated with inadvertently creating false expectations that the Fed is preoccupied with achieving a specific level of the unemployment rate. Our ability to manage those risks will depend on when and how we choose to tighten policy, as eventually we must. To wait until unemployment reaches some predetermined level, as the Martin FOMC did in the 1960s, is likely to mean waiting too long. That strategy proved bitterly disappointing for Martin and his colleagues, and I expect it would prove disappointing for us as well. At some point in the not-too-distant future, we are likely to face an economy growing in a self-sustaining way while the unemployment rate is still relatively high by historical standards. The decisions we make at that time will be the true test of whether we've learned our lessons.