Monetary Policy, Economics and the Recovery
I will focus my remarks this morning on three areas: First, I will take a brief look at the extensive impact of the recession on the workforce and capital investment. Second, I will discuss my view on the cyclical and structural aspects of the recovery. I will then conclude with a discussion on the Federal Reserve's monetary policy responses.
The Recession: Key Impacts
Let me start by highlighting a couple of key indicators to illustrate just how extraordinary the impact of the recession has been on the American economy and the workforce. We are all painfully aware of the sharp pullback that occurred in the fall of 2008 and its significant impact on the labor market. In 2007, over 63 percent of the American population was working. During the recession, the percentage of the working population dropped by more than 5 percentage points. In terms of people, getting back to 63 percent of the population working would translate into about 12 million more people working today. I'm sure we would agree that's an extraordinary number of people. Despite some employment growth, the percentage of our population that is working has hardly budged during the recovery.
Let's look at a second major economic indicator - capital spending. Capital spending is also a critical contributor to national output, yet it makes the headlines far less often than unemployment statistics. Capital spending is the purchase of equipment, software, buildings, and other items needed to produce final goods and services. Like the size of the labor force, it is a key determinant of our economic potential, which is the maximum amount of goods and services our economy can sustainably produce.
Over the course of the recession, capital spending shrank by almost 25 percent, or $400 billion. No other post-World War II recession has seen nearly as large of a pullback. The decline in investment and the closure of facilities actually reduced the total U.S. capital stock for the first time since the Great Depression. And, although firms have been increasing investment activity for more than two years, real spending on business fixed investment has not yet surpassed its 2007 level. Data show that U.S. investment spending is 7 percent below its pre-recession pace. Anecdotal reports and my regular conversations with company leaders and bankers continue to indicate that there are a lot of financial resources sitting on the sidelines, waiting for opportunities to invest in new productive capital.
The significant declines in labor and capital spending during the recession naturally led to a decline in American output, which at its worst point fell about 5 percent from its peak in 2007. Again, this decline was the worst since the Great Depression.
While the National Bureau of Economic Research tells us that the recession officially ended in June of 2009, the nation is producing just a little more total output today than we were at the end of 2007. However, the cutbacks in both labor and capital spending have only partially recovered.
This very slow path to recovery, particularly in terms of labor and investment, is an important element to consider when setting monetary policy. Would more monetary policy accommodation help to speed up the economic recovery? Or should monetary policymakers be more concerned about inflation in today's environment? Among the many things we need to understand to answer these questions is whether the forces keeping this recovery relatively modest are cyclical or structural.
Economic Impacts of the Recession: Structural or Cyclical
Many of you are trained economists and are very familiar with the concept of dividing economic activity into structural and cyclical components. Broadly defined, cyclical impacts of a recession will be recovered as the economy works through the business cycle and as economic conditions improve. Such impacts rise and fall, but they do not persist. In contrast, impacts that persist even after the economy has been expanding for years are considered structural.
This distinction is often applied to unemployment rates, but it can also be applied to spending on plants and equipment. The underlying question in either case is, How much of the reduced demand for labor and capital will persist even when the U.S. economy has fully recovered from the recession?
Let's look first at the labor market. Structural unemployment happens when workers are either unwilling or unable to make the changes needed to meet the changing requirements of employers. For example, structural unemployment could result from workers' inability to relocate. Jobs might be increasing in North Dakota, the state with the lowest unemployment rate, but unemployed workers from Nevada, the state with the highest unemployment rate, are unwilling or unable to relocate. The inability or unwillingness of people to move from Nevada to North Dakota for jobs would boost structural unemployment.
Structural unemployment can also result from changing demands for workers' expertise. Nursing and healthcare skills might be in high demand, but relatively unavailable in some parts of the country. On the other hand, construction workers and skills might be abundant across all regions of the country, but not in demand.
Let's examine a couple of sectors of the economy where some analysts have contended that our unemployment problem is primarily structural.
The U.S. construction sector was at the bull's-eye of the recession. The impact on construction workers was immense and the thinking was that, even after the economy recovered, we would need far fewer construction workers than before the steep downturn. Unemployed construction workers are often cited as examples of structural unemployment because of a belief that many of the pre-recession construction jobs were never coming back. Given that construction workers typically have relatively low levels of formal education and relatively high levels of specialized skills that they acquire on the job, some believed their prospects for employment in other sectors were limited.
The statistics tell a different story. The recession increased the number of unemployed workers in the construction sector by 1.1 million, raising the unemployment rate of construction workers to a peak of roughly 20 percent. Nonetheless, recent employment statistics indicate that the number of unemployed construction workers is now just 300,000 higher than it was in 2007, even as the construction sector has continued to shrink since the end of the recession. Indeed, the number of unemployed workers from construction has declined more rapidly than the total number of unemployed workers in the United States.
Another sector of the economy often cited for evidence of structural unemployment is manufacturing. Today, manufacturing requires more specialized skills than ever, and some manufacturing activities have moved to other parts of the world. The recession increased the numbers of unemployed workers in the manufacturing sector by 1.2 million people. Here again, unemployment figures in manufacturing have declined more rapidly than economy-wide numbers. The number of unemployed manufacturing workers is now just 300,000 higher than it was in 2007. The employment data for these two large sectors of the economy in part reflects the flexibility of U.S. workers to move out of one sector into another and demonstrates that there has been less structural unemployment than many had thought.
A major part of the ongoing unemployment problem today is the elevated numbers of young unemployed people. The number of unemployed people in the early-career ages of 20 to 34 rose by 2.6 million in the recession. The recovery has been less effective at shrinking this pool of unemployed workers, with 1.8 million young and recently trained workers still looking for work. While it is hard to categorically rule out structural unemployment, people in this age group are typically very flexible about which sector or region they work. In addition, this generation is more educated than prior generations, and that education should be more relevant to current employment opportunities. The employment weakness in this age group points more to a broad-based lack of demand for labor - that is, a cyclical shortfall in employment.
There has been progress in the number of job openings over the past two years, yet according to the Bureau of Labor Statistics, all major industry categories, from manufacturing to government, are still reporting fewer openings than they were in 2007. That is not to say that there are no openings that go unfilled today. Many do get filled, but it always takes time for employers to find the best candidate for an open position.
What is extraordinary about today's economy is the number of unemployed people compared to the number of job openings. Before the recession, there was a little over one job seeker for every open position. Today, despite progress, there are about three people looking for work for every job opening. The only long-run solution to the unemployment problem is to increase the number of job openings through more growth in overall economic activity.
Together, these facts make a lot of today's unemployment look more cyclical than structural to me, although persistent cyclical unemployment runs the risk of translating into structural unemployment through the loss of skills. I encourage you to read our latest Annual Report, just published, for a thorough discussion of the employment side of the Federal Reserve’s dual mandate.
I've talked about labor, so let me now shift to capital as another area to consider for the question of structural versus cyclical impacts. One key indicator from the capital side of production is capacity utilization -- that is, how much of a manufacturer's capacity to make a product is actually being utilized. In the depths of the recession, manufacturing capacity utilization figures fell from 79 percent to below 64 percent - a record low. Since then, the utilization rate has climbed almost all the way back to the pre-recession level. This performance is due not only to increased production, but also to an outright reduction in the amount of manufacturing capacity. In other words, capital, like labor, was taken out of production and reduced the potential for economic growth in our economy.
The deep recession led to plant closings. Current Federal Reserve estimates indicate that our nation's manufacturing capacity is about 6 percent lower than 2007 levels. According to the Bureau of Labor Statistics (Quarterly Census of Employment and Wages), about 15 percent of the more than 100,000 larger U.S. manufacturing plants that were operating in 2007 were closed over the past five years. These plant closings could point to a loss of potential output.
As economic conditions improve, however, there is room to rebuild capacity through investments in new and existing plants. Business fixed investment has grown over the recovery, but has remained weaker than typical for a recovery period. Investment often increases as capacity utilization tightens, but many companies have only recently reached the point where those decisions are more immediately relevant. Unfortunately, I continue to hear anecdotal reports of a reluctance to invest. My business contacts cite a variety of reasons for this reluctance, including the economic situation in Europe, uncertainty about U.S. tax policies and regulations, and just general caution. That said, business leaders in selective industries are seeing some opportunities to invest in domestic production.
How business investment activity develops in the coming months bears our close attention. If the uncertainties restraining investment diminish with a stronger expansion, then we may see investment activity pick up, which is the typical cyclical response. In previous expansions, our economy has generated double-digit growth rates in investment spending, which has driven an expansion in industrial capacity. If, on the other hand, the primary restraints on capacity expansion are more persistent, then this reduced-investment environment may be more of a structural limit. I find it harder to assess the balance between cyclical and structural factors affecting manufacturing capacity than I do for labor markets.
In addition to the direct examination of changes in labor markets and investment patterns, pricing data can also be used to assess structural versus cyclical impacts. If today's high unemployment were largely due to structural factors, then the labor market would be tighter than it appears to be, and we should see some wage pressures. This has not been the case to date, as growth in the Employment Cost Index has remained quite subdued. Similarly, if plant utilization levels have been pushed too high, we should see businesses more willing to raise prices beyond their input costs. However, producer price data have shown little evidence of broad pricing pressures connected to capacity constraints.
To wrap up this discussion of structural versus cyclical elements, while there may be some structural changes in the use of both labor and capital, there appears to be plenty of opportunity to add labor and capital as the economy expands.
U.S. Monetary Policy: Responsive and Measured
This brings me to the issue of how monetary policy should respond when faced with both an uncertain outlook for the economy and uncertainty about structural changes. Monetary policymakers are being challenged in some quarters for not doing enough to support a speedy recovery to absorb the vast numbers of cyclically unemployed workers. Other critics, who believe that the economy's problems are primarily structural, worry that the FOMC's efforts to bolster growth pose an unacceptably high inflation risk. My monetary policy decisions are based on my outlook, the risks to the outlook, and an assessment of the costs and benefits of our policy options.
Currently, I am projecting the economy to grow at a moderate rate, slightly above 2-1/2 percent this year and around 3 percent in 2013 and 2014. At this pace of growth, I expect that it could take as long as four to five years for the unemployment rate to fall to the 6 percent rate I judge to be consistent with maximum employment. I also project inflation to run very close to the Federal Reserve's longer-run objective of 2 percent as measured by the PCE price index through 2014. My inflation outlook, although it is close to the 2 percent objective, is based on an economy that is working through a significant amount of cyclical weakness over the projection horizon. My outlook for both economic activity and inflation relies on monetary policy remaining accommodative. Therefore, I have voted in favor of the FOMC's policy statements and actions, including the statement that economic conditions "are likely to warrant exceptionally low levels for the federal funds rate at least through late 2014." This date is not a commitment; rather, it conveys the FOMC's collective judgment of when economic conditions would warrant an increase in the federal funds rate. If there is a substantial change in the economic outlook, or risks to the outlook, then the guidance would change appropriately.
It is important to remember that monetary policy is not something that the FOMC should just "set and forget" - it must evolve with the FOMC's outlook, including its estimates for structural and cyclical impacts of the recession. Above all, in these uncertain times, I think it is important to keep an open mind and take a balanced approach to meeting our dual mandate. In summary, the recession had a huge impact on our economy. The prolonged nature of the recovery has thrown into question what our productive capacity will be when the recovery is complete. Today I have offered some examples to illustrate why it is challenging to assess how much of the weakness in labor markets and tightness in capacity utilization is structural and, therefore, not reversible with broad-based economic growth. Nonetheless, my current assessment is that the real economy continues to show considerable cyclical weakness. This assessment, along with my outlook for moderate growth and subdued inflation, calls for today's highly accommodative monetary policy.