Regional President Speech

EMBARGOED UNTIL Thursday, February 6, 2014 at 5:45 PM Eastern Time OR UPON DELIVERY “Underutilization in U.S. Labor Markets” Eric S. Rosengren President & Chief Executive Officer Federal Reserve Bank of Boston New College of Florida “New Topics New College” Speaker Series Sarasota, Florida February 6, 2014 * EMBARGOED UNTIL Thursday, February 6, 2014 at 5:45 PM Eastern Time OR UPON DELIVERY * “Underutilization in U.S. Labor Markets” Eric S. Rosengren President & Chief Executive Officer Federal Reserve Bank of Boston New College of Florida “New Topics New College” Speaker Series Sarasota, Florida February 6, 2014 It is a great pleasure to be at New College of Florida to see where my friend and colleague Bill Dudley, president of the New York Federal Reserve Bank, spent some of his formative years. I would also note that New College is consistently ranked among the top colleges for the combination of academic excellence and affordability – no small achievement in this day and age. Turning to my topic today, I would first like to note – as always – that the views I share are my own, not necessarily those of my colleagues on the Federal Open Market Committee (FOMC) or the Federal Reserve’s Board of Governors. 1 * EMBARGOED UNTIL Thursday, February 6, 2014 at 5:45 PM Eastern Time OR UPON DELIVERY * The New College location in sunny Florida seems appropriate for my message today, given that there has been some sunnier news on the economy recently. Last week, we learned that the preliminary estimate of real GDP growth in the fourth quarter of 2013 was 3.2 percent, and thus for the second half of 2013 real GDP growth averaged 3.7 percent – well above the 2.4 percent average over the recovery period. Some of this growth reflects factors that will likely prove temporary, such as a rapid build-up of inventories and a surge in net exports. But excluding inventory fluctuations, final sales grew three-quarters of a percentage point faster in the second half of 2013 than in the preceding 16 quarters. Not only has there been good news on economic growth, but the unemployment rate has also continued to decline – and I hope we will see another decline in tomorrow’s employment report. It is certainly good news that the national unemployment rate has fallen from a peak of 10 percent in the post-financial crisis recession to 6.7 percent as of December. Unfortunately, not all reductions in the unemployment rate are alike. For example, some are caused by rapid employment growth, while others are due to people giving up looking for a job. Distinguishing which is occurring is often quite difficult. One warning sign for economists is that while overall this has been a long, slow recovery, the recent decline in the unemployment rate has occurred more quickly than many expected. The fact that other labor market indicators have not been as favorable as the unemployment rate suggests that the rate may be falling for both reasons. Gauged by these broader measures of labor market utilization, we remain far from conditions experienced prior to the so-called “Great Recession.” 2 * EMBARGOED UNTIL Thursday, February 6, 2014 at 5:45 PM Eastern Time OR UPON DELIVERY * Currently, over 20 million Americans are either unemployed, are marginally attached to the labor force (including discouraged workers no longer looking for work because they believe no jobs are available for them), or are working part time when they would prefer full-time work. For these Americans, it is all too painfully apparent that labor markets remain far from their state prior to the recession. The concern that the unemployment rate alone does not fully capture the state of labor market conditions is reflected in the Federal Reserve’s current stance of monetary policy. In the January FOMC statement,1 the Committee reaffirmed that it likely will be appropriate to maintain the current target range for the federal funds rate well past the time that the unemployment rate declines below 6.5 percent (especially if inflation remains so subdued). Given that the unemployment rate is now approaching this 6.5 percent threshold, I want to highlight why, in my view, labor-market conditions remain far from where they would need to be to justify raising short-term rates. Furthermore, I will argue that everyone following the economy should be placing weight on the broader measures of labor market underutilization. A key point that I will expand on in a moment is that these measures remain far above where they were when the Federal Reserve began raising short-term rates in the previous recovery. As I have often noted, a patient approach to removing accommodation is further justified because the inflation rate remains far below the Fed’s 2 percent target.2 Very subdued inflation gives policymakers latitude to accommodate a more rapid improvement in labor markets, by keeping short-term rates lower for longer than is typical for recoveries. 3 * EMBARGOED UNTIL Thursday, February 6, 2014 at 5:45 PM Eastern Time OR UPON DELIVERY * Measures of Underutilization in Labor Markets Since we are gathered today on a college campus, I hope you will indulge me a brief tutorial on labor market statistics. The topic may sound a bit dry, but the distinctions in the measures have major implications for policies that affect millions of Americans. The U.S. Bureau of Labor Statistics (BLS) publishes a broader set of unemployment rates in order to capture underutilization in labor markets3 – indeed, the categories were redefined in 1994, so there is a consistent series going back that far.4 Figure 1 charts these various BLS measures of underutilization in labor markets over time. Let me walk you through the chart, beginning at the bottom.  The dark blue reflects those Americans who are unemployed. This is the measure most frequently cited from the labor report. The BLS calls this its “U-3” definition of unemployment, which includes people who do not have jobs but have actively searched for work over the previous four weeks.5 (In a moment you will be conversant in all the “U” measures of labor market slack, from U-3 to U-6.)  The light blue area reflects so called “discouraged workers” who, together with the unemployed, make up the “U-4” measure of labor market utilization. Discouraged workers are not included in the narrower definition of the unemployed because they have not searched for work over the previous four weeks. They have searched for work over the previous year, but do not think 4 * EMBARGOED UNTIL Thursday, February 6, 2014 at 5:45 PM Eastern Time OR UPON DELIVERY * that jobs fitting their skills or background are currently available. As of December, the BLS measured 917,000 of these discouraged workers in the U.S., compared to only 363,000 at the beginning of the recession in December of 2007.6  The green area reflects other persons whom the BLS classifies as “marginally attached to the labor force,” but who are not classified as discouraged workers. The BLS’s “U-5” measure adds these people to the discouraged and unemployed workers. The other marginally attached workers are available to work and have searched for work in the previous year, but have not searched for work in the past four weeks and are therefore outside of the labor force. They do not state that their reason for being out of the labor force is an absence of available jobs.7 Issues such as family responsibilities or transportation challenges may be playing a role. The BLS estimates that there are currently about 1.5 million Americans in this category, compared to 981,000 in December of 2007.  The dark red area on the chart reflects persons who are working part time for so-called “economic reasons.”8 They are available to work full time, but have had their hours reduced – or could only secure part-time work. The BLS’s “U-6” measure adds these people working part time for economic reasons to the people in the U-5 measure. There are currently 7.8 million workers who 5 * EMBARGOED UNTIL Thursday, February 6, 2014 at 5:45 PM Eastern Time OR UPON DELIVERY * are working only part time for economic reasons, up sharply from the 4.6 million in December of 2007. In sum, Figure 1 highlights why, for many Americans, the labor market remains very difficult – and why using the narrow, widely reported unemployment rate alone to summarize labor market conditions could suggest a misleadingly optimistic state of affairs. While the traditional unemployment rate is an important measure, its decline does not capture the wider difficulties in the labor market that are being felt by so many Americans. My sense is that evaluation of the labor market situation requires much more attention to these broader measures. Using the Broader Measures of Labor Market Utilization Figure 2 provides the traditional unemployment rate along with a vertical line at June 2004 – the first tightening of monetary policy during the previous recovery, on June 30, 2004. The unemployment rate at that time was 5.6 percent, and the PCE inflation rate over the previous year was 2.8 percent. It is revealing to contrast those with current conditions. We have a much higher unemployment rate (6.7 percent as of December) and a much lower inflation rate (1.1 percent as of December). Also on Figure 2, I have a horizontal line showing the average unemployment rate from January 1994 to December 2007. We are currently more than one and a half percentage points or about 2.5 million people above that average of the most basic and narrow measure of unemployment. 6 * EMBARGOED UNTIL Thursday, February 6, 2014 at 5:45 PM Eastern Time OR UPON DELIVERY * Figure 3 provides a similar chart, but uses the next broader definition of labor market underutilization. The December U-4 unemployment rate, which includes discouraged workers, was 7.2 percent. This is almost two percentage points above the average for 1994 to 2007, and well above the peak after the previous recession. Similarly, in Figure 4 the U-5 measure of unemployment (adding in all other persons marginally attached to the labor force) shows an even wider gap, 2.1 percentage points, from the pre-recession average (1994 through 2007). I find Figure 5, using the broadest measure (U-6) that adds in those employed part time for economic reasons, particularly striking. Even at this point in the recovery, the measure is still nearly four and a quarter percentage points or 6.5 million people above the pre-recession average – a gap that is much wider than those for the other measures of unemployment.9 The large number of individuals working part time for economic reasons highlights, in my view, the need for much more improvement in labor markets. Figure 6 compares the current reading of all these measures to their level at the time of the first monetary policy tightening during the previous recovery (in June 2004). It highlights how much higher all the unemployment measures are now than they were in June 2004, when the FOMC first tightened rates during the previous recovery. While all measures are well above the rates in June 2004, the broadest measure of unemployment is particularly far from its level then. Figure 7 plots the measures of labor market underutilization on one chart. Two important features stand out to me. First, the gaps between the traditional, narrow U-3 measure of unemployment and all the broader measures remain wider than they were 7 * EMBARGOED UNTIL Thursday, February 6, 2014 at 5:45 PM Eastern Time OR UPON DELIVERY * prior to the recession. Second, while all the measures of labor market utilization have improved from their peaks, you can see that the broader the measure, the further the current reading is from its pre-recession level. Unfortunately, the BLS data series of broader measures of labor market utilization, as currently defined, go back only to 1994. And despite the clear story I think they tell, there are some important questions that have yet to be answered. For example, will the broader measures of labor utilization return to pre-recession levels, or does this represent a more lasting change in labor force dynamics? If it is the former, it will require much more economic growth to absorb this large pool of underutilized workers. Also, how large an impact do these pools of underutilized workers have on wages and prices? On the one hand, growth in wages and prices has been very low by historical standards; on the other hand, wage and price inflation did not fall as low as many predicted, given the earlier upheaval in labor markets. Finally, inflation of late has remained stubbornly low, even as labor market conditions have improved, to the surprise of many forecasters. So we need to explore the link between these labor market indicators and wage and price inflation, in part to understand the likely evolution of inflation, and in part to understand to what degree we can expect these measures to return to their pre-recession norms. In other words, whether we have seen more permanent “scarring” of the unemployed, which might imply longerlasting effects, broadly construed. I will expand on this theme in a moment. As the economy continues to improve gradually, monetary policymakers need to carefully evaluate labor market conditions. My own view is that for now we will need to 8 * EMBARGOED UNTIL Thursday, February 6, 2014 at 5:45 PM Eastern Time OR UPON DELIVERY * place greater weight on these broader measures of labor market utilization – measures that are not returning to “normal” as quickly as is the more narrow U-3 measure of unemployment. Labor Market Slack and Inflation Most forecasters have expected that as labor market slack diminished, inflation would move back to levels closer to the Fed’s 2 percent target. However, the actual inflation experience has been more subdued than forecast over the past two years. At least one potential reason for this forecasting “miss” is the presence of substantial labor market slack, as shown in the broader unemployment measures. Figure 8 provides the midpoint of the central tendency of the inflation forecasts for 2013, provided by the Federal Reserve governors and presidents in our quarterly Summary of Economic Projections. Back in November of 2011, the midpoint of the central tendency of the forecasts showed that Fed policymakers expected both PCE and core PCE inflation to be just under 1.8 percent. As the forecasts have been updated for subsequent Committee meetings, the forecast for the year-over-year inflation rate for 2013 has continued to decline (shown by the midpoint of the central tendency of the forecasts, plotted on the chart). In fact, the PCE inflation rate of 1.1 percent in December was much lower than was expected, even at the beginning of 2013. A similar pattern emerges with core PCE inflation. Figure 9 shows that private sector forecasts of inflation have been no more accurate. The Survey of Professional Forecasters reported a median expectation that the PCE inflation rate would be 2 percent as of November 2012. Much like the SEP, as 9 * EMBARGOED UNTIL Thursday, February 6, 2014 at 5:45 PM Eastern Time OR UPON DELIVERY * economic data have continued to come in weaker, the Survey’s inflation forecast was reduced further. Even in 2013, the forecast reductions have been sizable. The Blue Chip consensus forecasts show a similar pattern. Some argue that so-called structural unemployment has risen, by which I mean that the recession has caused changes in the labor market that imply a long-term increase in unemployment that will not recede as the economy recovers. Many who believe that much of our unemployment is structural argue that there is little that monetary policy can do if weak labor markets reflect structural rather than cyclical forces. While we cannot categorically rule out the presence of some structural unemployment, I would note that if structural unemployment is now much higher, then it is difficult to explain why growth in prices has continued to be low, both in absolute terms and relative to expectations. In short, one potential explanation for persistently low price (and wage) growth may be that the weakness we have just seen in the broader measures of labor markets better reflects the state of the labor market than the traditional, narrow U-3 measure of unemployment. Concluding Observations In summary and conclusion, I would just reiterate that the broader measures of the labor market show less improvement than the more narrowly constructed U-3 measure of unemployment, which is widely followed and has been used in the Federal Reserve’s forward guidance on policy. The broader measures of unemployment are much higher than they were at their peaks following the previous recession, and remain unusually high relative to the traditional U-3 measure of unemployment. If we take these measures as 10 * EMBARGOED UNTIL Thursday, February 6, 2014 at 5:45 PM Eastern Time OR UPON DELIVERY * providing important complements to the U-3 measure of labor market slack, then we must acknowledge and address an even larger amount of remaining labor market slack than the narrower measure indicates. This is, of course, a discussion of data points and measures, but we cannot lose sight of matters of human toll and potential, as well. The implications for monetary policy seem clear – if one believes some of the signals from the broader measures of labor market slack that I have walked you through today, then we remain very far from historical average levels of labor utilization. As a result, I firmly believe that monetary policymakers should remain quite patient in removing accommodation. This perspective, I would add, is consistent with the recent FOMC statement indicating that the Committee expects short-term interest rates to remain low well past the time that the unemployment rate declines below the 6.5 percent threshold. This policy position is also bolstered by the fact that the inflation rate has been well below what most forecasters expected and well below the Fed’s 2 percent target. In short, in light of too much labor market slack and very low inflation rates, monetary policy should continue to be highly accommodative. Thank you. NOTES: 1 See http://www.federalreserve.gov/newsevents/press/monetary/20140129a.htm 2 For inflation measured by the Personal Consumption Expenditures index or PCE. 11 * EMBARGOED UNTIL Thursday, February 6, 2014 at 5:45 PM Eastern Time OR UPON DELIVERY * 3 See Table A-15 in the BLS’s monthly Employment Situation reports, which shows these broader measures of underutilization. BLS definitions include the following: “Persons marginally attached to the labor force are those who currently are neither working nor looking for work but indicate that they want and are available for a job and have looked for work sometime in the past 12 months. Discouraged workers, a subset of the marginally attached, have given a job-market related reason for not currently looking for work. Persons employed part time for economic reasons are those who want and are available for full-time work but have had to settle for a part-time schedule.” 4 In January 1994, the Current Population Survey was redesigned to obtain more accurate and comprehensive information. Among the biggest changes were changes to the definitions of discouraged and part-time workers. 5 The technical notes in the BLS’s monthly Employment Situation define unemployed as follows: “People are classified as unemployed if they meet all of the following criteria: they had no employment during the reference week; they were available for work at that time; and they made specific efforts to find employment sometime during the 4-week period ending with the reference week.” The reference week is “generally the calendar week that contains the 12th day of the month.” 6 When available, the figures quoted are the seasonally-adjusted figures which appear in the Employment Situation published by the BLS. The BLS does not seasonally adjust the figures for discouraged workers and other marginally attached workers, so these quoted figures are not seasonally adjusted. However, for use in Figure 1, the two series have been seasonally adjusted to appear with the seasonally-adjusted series for unemployed and part-time workers which have strong seasonal components. 7 A BLS “Issues in Labor Statistics” publication from 2009 notes characteristics of the marginally attached: see http://www.bls.gov/opub/ils/pdf/opbils74.pdf, especially Table 2. 8 The BLS definition of Part-Time for Economic Reasons is as follows: “Refers to those who worked 1 to 34 hours during the reference week for an economic reason such as slack work or unfavorable business conditions, inability to find full-time work, or seasonal declines in demand.” 9 The current U-6 rate was 13.1 percent for December, in contrast to 8.8 percent in December of 2007. 12 

Eric Rosengren (2014, February 5). Regional President Speech. Speeches, Federal Reserve. https://whenthefedspeaks.com/doc/regional_speeche_20140206_eric_rosengren

@misc{wtfs_regional_speeche_20140206_eric_rosengren,
  author = {Eric Rosengren},
  title = {Regional President Speech},
  year = {2014},
  month = {Feb},
  howpublished = {Speeches, Federal Reserve},
  url = {https://whenthefedspeaks.com/doc/regional_speeche_20140206_eric_rosengren},
  note = {Retrieved via When the Fed Speaks corpus}
}