A Perspective on the Future of U.S. Monetary Policy
Thank you for that kind introduction. It’s a great pleasure to be here. Over the past three years, we have been dealing with a crisis that has reverberated around the world, inflicting severe damage to financial markets and economies. The response of policymakers in the U.S. and elsewhere was swift and unprecedented in its scope and magnitude. It was also well coordinated across the community of central banks. We set up swap lines to ease liquidity strains throughout the world and lowered target rates jointly at the height of the crisis. Such coordinated action was very helpful in stabilizing the global financial system.
Today, as we progress with economic recovery, policymakers in all countries are faced with a new set of challenges and issues. Recognizing that they face a distinct set of challenges, I look forward to hearing my co-panelists discuss the issues facing European policymakers as they deliberate the future of monetary policy. For my part, I would like to give you my perspective on some of the issues we face in the U.S. Before I proceed further, let me stress that I will be sharing my personal views with you, and not necessarily offering the views of my colleagues on the FOMC, or the Federal Reserve System.
The U.S. was the epicenter of the financial crisis and the deep recession that followed. The decline in our economic activity and the damage to financial markets in the past three years are a reflection of the magnitude of the shock. Over the course of the recession, U.S. real GDP declined by more than 4 percent, over 8 million jobs were lost, the unemployment rate doubled within two years and the household sector – which accounts for the lion’s share of the U.S. economy – lost more than $13 trillion in wealth.
Over the past year, however, we’ve seen significant improvements. Financial markets have stabilized and have progressed toward full repair. Aided by accommodative monetary policy, fiscal stimulus and improvements in financial markets, real GDP has recorded four quarters of positive growth and employment has begun to rise. However, given the depth of the recession, we still have a long road ahead before we catch up to the level of activity we would have achieved in the absence of the crisis, or any other shock. Moreover, we appear to have lost some of our forward momentum in recent months. Real GDP growth slowed markedly from an annual rate of 3.7 percent in the first quarter to 1.6 percent in the second quarter. More recent data suggest that growth in the second half of the year will be similarly weak – and softer than what I had expected just a few months ago. At the same time, measures of underlying inflation have fallen to very low levels -- levels that are below the 2 percent level that I, and most Fed policymakers, consider to be consistent with effective price stability.
Looking ahead, I expect output growth will strengthen in 2011 and 2012. Nonetheless, the pace of growth I currently anticipate is quite moderate given the severity of the recession we experienced and the potential growth rate of output. As a result, I expect the unemployment rate to remain well above its pre-crisis level over the next two to three years. Given the current and the anticipated future level of resource slack and subdued long-term inflation expectations, I also expect inflation to remain below desirable levels over any reasonable forecast horizon.
With such an outlook for the economy, as a macroeconomist and policymaker, I think we face two key issues in the near and medium term. First, to what extent do structural factors explain the very high unemployment rate we currently have? Some have suggested that the financial crisis and the accompanying recession precipitated a seismic shift in the structure of labor markets, raising the natural rate of unemployment significantly above its pre-crisis level. If, as they suggest, labor market frictions rose dramatically over the past two years, then monetary policy is not the appropriate tool to address the ramifications of such a change. If, on the other hand, structural factors can only explain a modest part of the rapid rise in the unemployment rate, and aggregate demand deficiencies account for remainder, then monetary policy may be able to play a constructive role.
This brings me to the second key issue facing policymakers. If further monetary policy accommodation is desirable, what is the appropriate policy action when short-term interest rates are already at zero? Has extreme risk aversion by businesses and consumers put us in what can be described as a liquidity trap? And if so, what can we do about it?
Let me first elaborate on the unsatisfactory progress in employment gains, and what it implies for monetary policy. There are several reasons to think that the natural rate of unemployment has risen over the last couple of years. It is possible that the extension of unemployment insurance benefits during the recession, while cushioning unemployed workers from the adverse effects of lost income, might have reduced some workers’ job search intensity, or kept others from leaving the labor force. To the extent that such incentives are present, the natural rate of unemployment would increase while the extended benefits are in effect. However, reasonable estimates of the effect of the extension of unemployment benefits range one-half to one percentage points – far from sufficient to explain the nearly 5 percentage point increase over the past two years. Moreover, given the current schedule for the expiration of these benefits, the resulting increase in the natural unemployment rate will fade away over the next two years – leaving us still with an unsatisfactorily high rate of unemployment at the end of my forecast horizon.
It is also possible that the shocks that reverberated through the economy created a mismatch between the skills sought by employers and the skills the unemployed workers have. For instance, it is conceivable that the recession affected the different regions and sectors of the economy unevenly. Moreover, the recession may have severed an unusually large number of long-term employment relationships, making for an especially difficult transition for affected workers. The unusually long spells of unemployment experienced during the recent recession and potential erosion of skills during that time are additional factors that might have magnified labor market frictions. The sharp decline in home values and tight credit conditions might have reduced the ability of unemployed workers to sell their homes and move to regions where jobs are available. Taken together, these developments might have eroded the efficiency of matching between workers and jobs, and raised the natural rate of unemployment.1
The key question is: Are these possible structural changes in the labor market sufficient to explain the current unemployment rate? The Beveridge curve that describes the relationship between the unemployment rate and the job vacancy rate is a useful tool for addressing this question. The unemployment-vacancy relationship through the end of 2009 is captured very well by a simple, stable Beveridge curve and a constant-returns Cobb-Douglas matching function. So, the relationship between job openings and unemployment through the end of 2009 has been relatively stable, and does not suggest a dramatic increase in labor market frictions. It is only recently that we have seen an improvement in job openings that was not matched by a correspondingly large reduction in unemployment. Based on these data, some have suggested that most of the increase in the unemployment rate over the past two years is due to skills mismatch.
However, it seems likely that much of the apparent conflict between unemployment and vacancy data may be purely an issue of timing. At this stage of an economic recovery, it is not unusual for the vacancy rate to increase ahead of reductions in the unemployment rate – we have often seen such loops in the Beveridge curve at the end of past recessions.
But even if we take the job openings data at face value, it doesn’t suggest an increase in the natural rate to anything like the current rate of unemployment, which stands at 9.6 percent. Making some plausible assumptions, my staff estimates that the typical level of unemployment associated with a stable Beveridge curve passing through the recent data is likely to be about 7 percent. This includes the effects of increased unemployment insurance benefits that I already discussed.2
I am not suggesting that 7 percent is a good estimate of the current natural rate. As I said, there are reasons to discount some of the recent improvement in the vacancy data. Rather, I want to point to out that, even if one takes the vacancy data at face value and accepts that the natural rate has risen to 7 percent, we are still left with a very large amount of slack relative to the current rate of unemployment and the rate most analysts expect to see at the end of 2012.3
The size of the unemployment gap, combined with the fact that inflation has been running below the level I consider consistent with long-term price stability, suggests that it would be desirable to increase monetary policy accommodation to boost aggregate demand and achieve our dual mandate.
Should the FOMC judge that further monetary policy accommodation is appropriate based on our economic outlook, what is the optimal level of additional accommodation and what policy tools should be employed to deliver the additional stimulus?
During a typical period of policy accommodation, the answers to these questions would be straightforward. The FOMC would lower the target federal funds rate based on our economic outlook and the historical relationship between policy actions and their impact on the economy. For instance, were the current fed funds rate at 3 percent, my forecast would call for a substantial decline in the target rate. Such a reduction in the nominal fed funds rate would be consistent with several versions of the Taylor rule, which would call for negative interest rates.4 However, at roughly zero, the fed funds rate is as low as it can go.
As a result, the current economic environment poses unusual challenges to policymakers. In assessing the current state of the economy and considering the optimal policy response, a key issue that concerns me is the possibility that we might be in a liquidity trap.
As I assess the incoming data and talk to my business contacts, I see that executives are very cautious in their outlook and spending plans. They appear to be content to post strong profits generated by unprecedented cost-cutting, rather than growing their top-line revenues by expanding capital investments and hiring. Very conservative attitudes reign and cash is still king – even after the improvements in financial markets and strong bond issuance by businesses. Firms are sitting on the cash generated by profits and funds raised in capital markets. Very few are planning to grow their workforce. Although some contacts point to uncertainties raised by regulatory actions and government policies to explain their reluctance to invest, most admit that they would increase spending if stronger demand conditions prevailed.
Households are similarly cautious and gun-shy in their spending. Given the millions of jobs lost during the recession, the job insecurity faced by those employed, trillions of dollars in lost wealth and the balance-sheet repair that households have undertaken, consumers are displaying significant risk aversion. They have raised their savings rate, even though those savings earn very little interest income.
These are the classic symptoms associated with a liquidity trap: the supply of savings that outstrip the demand for investment even when short-term nominal rates are at zero.
The modern economic theory of liquidity traps indicates that the optimal policy response at zero-bound is to lower the real interest rate, almost surely by employing unconventional policy tools. Theory also indicates that, in the absence of such policy stimulus, the factors that generate high risk aversion could very well stifle a meaningful recovery, keep unemployment high and reinforce disinflationary pressures – clearly an undesirable equilibrium.
So, in the coming weeks and months, as I assess the incoming data, update my forecast and deliberate on the best monetary policy approach, I will be pondering two key issues: How much more should monetary policy do to reduce the shortfalls in meeting our dual mandate responsibilities for employment and price stability; and what tools should we use? Thank you.
1 See Petrongolo and Pissarides (2001) for a review of aggregate matching functions in labor markets and the related literature.
2 As I noted before, unless unemployment insurance benefits and the Emergency Unemployment Compensation program are extended further, they are scheduled to expire in 2011, at which point any impact they had on the natural rate of unemployment will begin to fade away.
3 There are other challenges for the view that the increase in the unemployment rate over the past two years reflects primarily structural factors: the dearth of sectors reporting strong demand for hard-to-find skilled workers; the lack of inflation pressures that we would expect to observe if the natural rate were higher and resource slack were smaller; and the unusually low quit rates and job-to-job transitions during the recession that are indicative of weak labor markets (De Wolfe and Klemmer, 2010; unpublished updates of the employment-to-employment transitions in Fallick and Fleischman, 2004).
4 For instance, the guideline suggested by Taylor (1999) would call for the federal funds rate to be negative 3.5 percent given the current CBO output gap of negative 6 percent and core CPI inflation rate of 1 percent, assuming that the FOMC considers 2 percent inflation to be consistent with price stability. The guideline outlined in Taylor (1993), which puts a smaller weight on the current output gap, would suggest that the federal funds rate target should be negative 0.5 percent.
ReferencesDe Wolf, M., and K. Klemmer, 2010, “Job openings, Hires, and Separations Fall during the Recession,” Monthly Labor Review, Washington, DC: Bureau of Labor Statistics, Vol. 133, No. 5, May; available online.
Fallick, B., and C. A. Fleischman, 2004, “Employer-to-Employer Flows in the U.S. Labor Market: The Complete Picture of Gross Worker Flows.” Finance and Economics Discussion Series, Washington: Board of Governors of the Federal Reserve System, No. 2004-34; available online.
Petrongolo, B., and C. Pissarides, 2001, “Looking into the Black Box: A Survey of the Matching Function,” Journal of Economic Perspectives, Vol. 39, June, pp. 390-431.
Taylor, J. B., 1993. “Discretion versus Policy Rules in Practice,” Carnegie-Rochester Conference Series on Public Policy, Vol. 39, June, pp. 195-214.
Taylor, J. B., 1999, “A Historical Analysis of Monetary Policy Rules,” in Monetary Policy Rules, John B. Taylor (ed.), Chicago: University of Chicago Press, pp.319-341.