View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

For release on delivery
10:00 a.m. EDT (8:00 a.m. MDT)
August 22, 2014

Labor Market Dynamics and Monetary Policy

Remarks by
Janet L. Yellen
Chair
Board of Governors of the Federal Reserve System
at the
Federal Reserve Bank of Kansas City Economic Symposium
Jackson Hole, Wyoming

August 22, 2014

In the five years since the end of the Great Recession, the economy has made
considerable progress in recovering from the largest and most sustained loss of
employment in the United States since the Great Depression.1 More jobs have now been
created in the recovery than were lost in the downturn, with payroll employment in May
of this year finally exceeding the previous peak in January 2008. Job gains in 2014 have
averaged 230,000 a month, up from the 190,000 a month pace during the preceding two
years. The unemployment rate, at 6.2 percent in July, has declined nearly 4 percentage
points from its late 2009 peak. Over the past year, the unemployment rate has fallen
considerably, and at a surprisingly rapid pace. These developments are encouraging, but
it speaks to the depth of the damage that, five years after the end of the recession, the
labor market has yet to fully recover.
The Federal Reserve’s monetary policy objective is to foster maximum
employment and price stability. In this regard, a key challenge is to assess just how far
the economy now stands from the attainment of its maximum employment goal.
Judgments concerning the size of that gap are complicated by ongoing shifts in the
structure of the labor market and the possibility that the severe recession caused
persistent changes in the labor market’s functioning.
These and other questions about the labor market are central to the conduct of
monetary policy, so I am pleased that the organizers of this year’s symposium chose
labor market dynamics as its theme. My colleagues on the Federal Open Market
Committee (FOMC) and I look to the presentations and discussions over the next two
days for insights into possible changes that are affecting the labor market. I expect,

1

Nonfarm employment contracted by 6.3 percent from its peak in 2008 to its trough in early 2010,
compared with a 5.2 percent loss in the 1948-49 recession, previously the largest since the 1930s.

-2however, that our understanding of labor market developments and their potential
implications for inflation will remain far from perfect. As a consequence, monetary
policy ultimately must be conducted in a pragmatic manner that relies not on any
particular indicator or model, but instead reflects an ongoing assessment of a wide range
of information in the context of our ever-evolving understanding of the economy.
The Labor Market Recovery and Monetary Policy
In my remarks this morning, I will review a number of developments related to
the functioning of the labor market that have made it more difficult to judge the
remaining degree of slack. Differing interpretations of these developments affect
judgments concerning the appropriate path of monetary policy. Before turning to the
specifics, however, I would like to provide some context concerning the role of the labor
market in shaping monetary policy over the past several years. During that time, the
FOMC has maintained a highly accommodative monetary policy in pursuit of its
congressionally mandated goals of maximum employment and stable prices. The
Committee judged such a stance appropriate because inflation has fallen short of our
2 percent objective while the labor market, until recently, operated very far from any
reasonable definition of maximum employment.
The FOMC’s current program of asset purchases began when the unemployment
rate stood at 8.1 percent and progress in lowering it was expected to be much slower than
desired. The Committee’s objective was to achieve a substantial improvement in the
outlook for the labor market, and as progress toward this goal has materialized, we have
reduced our pace of asset purchases and expect to complete this program in October. In
addition, in December 2012, the Committee modified its forward guidance for the federal

-3funds rate, stating that “as long as the unemployment rate remains above 6-1/2 percent,
inflation between one and two years ahead is projected to be no more than a half
percentage point above the Committee’s 2 percent longer-run goal, and longer-term
inflation expectations continue to be well anchored,” the Committee would not even
consider raising the federal funds rate above the 0 to 1/4 percent range.2 This “threshold
based” forward guidance was deemed appropriate under conditions in which inflation
was subdued and the economy remained unambiguously far from maximum employment.
Earlier this year, however, with the unemployment rate declining faster than had
been anticipated and nearing the 6-1/2 percent threshold, the FOMC recast its forward
guidance, stating that “in determining how long to maintain the current 0 to 1/4 percent
target range for the federal funds rate, the Committee would assess progress--both
realized and expected--toward its objectives of maximum employment and 2 percent
inflation.”3 As the recovery progresses, assessments of the degree of remaining slack in
the labor market need to become more nuanced because of considerable uncertainty about
the level of employment consistent with the Federal Reserve’s dual mandate. Indeed, in
its 2012 statement on longer-run goals and monetary policy strategy, the FOMC
explicitly recognized that factors determining maximum employment “may change over
time and may not be directly measurable,” and that assessments of the level of maximum
employment “are necessarily uncertain and subject to revision.”4 Accordingly, the

See paragraph 5 in Board of Governors of the Federal Reserve System (2012), “Federal Reserve Issues
FOMC Statement,” press release, December 12,
www.federalreserve.gov/newsevents/press/monetary/20121212a.htm.
3
See paragraph 5 in Board of Governors of the Federal Reserve System (2014), “Federal Reserve Issues
FOMC Statement,” press release, March 19,
www.federalreserve.gov/newsevents/press/monetary/20140319a.htm.
4
See paragraph 5 in Board of Governors of the Federal Reserve System (2012), “Federal Reserve Issues
FOMC Statement of Longer-Run Goals and Policy Strategy,” press release, January 25,
www.federalreserve.gov/newsevents/press/monetary/20120125c.htm.
2

-4reformulated forward guidance reaffirms the FOMC’s view that policy decisions will not
be based on any single indicator, but will instead take into account a wide range of
information on the labor market, as well as inflation and financial developments.5
Interpreting Labor Market Surprises: Past and Future
The assessment of labor market slack is rarely simple and has been especially
challenging recently. Estimates of slack necessitate difficult judgments about the
magnitudes of the cyclical and structural influences affecting labor market variables,
including labor force participation, the extent of part-time employment for economic
reasons, and labor market flows, such as the pace of hires and quits. A considerable body
of research suggests that the behavior of these and other labor market variables has
changed since the Great Recession.6 Along with cyclical influences, significant structural
factors have affected the labor market, including the aging of the workforce and other
demographic trends, possible changes in the underlying degree of dynamism in the labor
market, and the phenomenon of “polarization”--that is, the reduction in the relative
number of middle-skill jobs.7

5

The central role of labor market conditions in monetary policy deliberations has also been apparent
abroad. Last year the Bank of England announced its intention not to raise its policy rate at least until the
unemployment rate reached 7 percent, subject to conditions on inflation and financial stability. Since that
time, the unemployment rate in the United Kingdom has dropped unexpectedly rapidly, prompting
policymakers to consider data beyond this single indicator when assessing the extent of spare capacity in
the U.K. economy. As in the United States, an unexpectedly swift decline in unemployment has raised
questions about the structural and cyclical effects of a severe recession.
6
For a discussion of important differences in the evolution of labor market conditions during the Great
Recession relative to typical postwar patterns, see Henry S. Farber (2011), “Job Loss in the Great
Recession: Historical Perspective from the Displaced Workers Survey, 1984-2010,” NBER Working Paper
Series 17040 (Cambridge, Mass.: National Bureau of Economic Research, May).
7
For convenience, the analysis here is presented as if cyclical factors and structural factors can be neatly
delineated. In reality, the line between the two may be indistinct. Moreover, what begins as cyclical
weakness may evolve into structural damage. For a discussion of the strategic issues that arise when
policymakers believe such evolution from cyclical to structural to be an important feature of the economy,
see Dave Reifschneider, William Wascher, and David Wilcox (2013), “Aggregate Supply in the United
States: Recent Developments and Implications for the Conduct of Monetary Policy,” Finance and
Economics Discussion Series 2013-77 (Washington: Board of Governors of the Federal Reserve System,
November), www.federalreserve.gov/pubs/feds/2013/201377/201377abs.html.

-5Consider first the behavior of the labor force participation rate, which has
declined substantially since the end of the recession even as the unemployment rate has
fallen. As a consequence, the employment-to-population ratio has increased far less over
the past several years than the unemployment rate alone would indicate, based on past
experience. For policymakers, the key question is: What portion of the decline in labor
force participation reflects structural shifts and what portion reflects cyclical weakness in
the labor market? If the cyclical component is abnormally large, relative to the
unemployment rate, then it might be seen as an additional contributor to labor market
slack.
Labor force participation peaked in early 2000, so its decline began well before
the Great Recession. A portion of that decline clearly relates to the aging of the baby
boom generation. But the pace of decline accelerated with the recession. As an
accounting matter, the drop in the participation rate since 2008 can be attributed to
increases in four factors: retirement, disability, school enrollment, and other reasons,
including worker discouragement.8 Of these, greater worker discouragement is most
directly the result of a weak labor market, so we could reasonably expect further
increases in labor demand to pull a sizable share of discouraged workers back into the
workforce. Indeed, the flattening out of the labor force participation rate since late last
year could partly reflect discouraged workers rejoining the labor force in response to the
significant improvements that we have seen in labor market conditions. If so, the cyclical
shortfall in labor force participation may have diminished.

See Shigeru Fujita (2014), “On the Causes of Declines in the Labor Force Participation Rate,” Federal
Reserve Bank of Philadelphia, Research Rap, special report, February 6,
http://philadelphiafed.org/research-and-data/publications/research-rap/2013/on-the-causes-of-declines-inthe-labor-force-participation-rate.pdf.
8

-6What is more difficult to determine is whether some portion of the increase in
disability rates, retirements, and school enrollments since the Great Recession reflects
cyclical forces. While structural factors have clearly and importantly affected each of
these three trends, some portion of the decline in labor force participation resulting from
these trends could be related to the recession and slow recovery and therefore might
reverse in a stronger labor market.9 Disability applications and educational enrollments
typically are affected by cyclical factors, and existing evidence suggests that the elevated
levels of both may partly reflect perceptions of poor job prospects.10 Moreover, the rapid
pace of retirements over the past few years might reflect some degree of pull-forward of
future retirements in the face of a weak labor market. If so, retirements might contribute
On disability, see Mark Duggan and Scott A. Imberman (2009), “Why Are the Disability Rolls
Skyrocketing? The Contribution of Population Characteristics, Economic Conditions, and Program
Generosity,” in David M. Cutler and David A. Wise, eds., Health at Older Ages: The Causes and
Consequences of Declining Disability among the Elderly (Chicago: University of Chicago Press),
pp. 337-79; and David H. Autor (2011), “The Unsustainable Rise of the Disability Rolls in the United
States: Causes, Consequences, and Policy Options,” NBER Working Paper Series 17697 (Cambridge,
Mass.: National Bureau of Economic Research, December). For a focus on developments within the Great
Recession, see David M. Cutler, Ellen Meara, and Seth Richards-Shubik (2012), “Unemployment and
Disability: Evidence from the Great Recession,” NBER Retirement Research Center Paper Series
NB 12-12 (Cambridge, Mass.: National Bureau of Economic Research, September). On school
enrollment, see Bridget Terry Long (2013), “The Financial Crisis and College Enrollment: How Have
Students and Their Families Responded?” working paper (Cambridge, Mass.: Harvard University, July),
http://isites.harvard.edu/fs/docs/icb.topic1232989.files/BLong%20%20The%20Financial%20Crisis%20and%20College%20Enrollment%20-%20July%202013.pdf.
10
For surveys of students who report job prospects as an important factor for attending or prolonging
school, see John H. Pryor, Kevin Eagan, Laura Palucki Blake, Sylvia Hurtado, Jennifer Berdan, and
Matthew H. Case (2012), The American Freshman: National Norms Fall 2012 (Los Angeles: Higher
Education Research Institute at the University of California, Los Angeles),
http://heri.ucla.edu/monographs/TheAmericanFreshman2012.pdf. On the cyclicality of college enrollment,
see Andrew Barr and Sarah Turner (2013), “Down and Enrolled: An Examination of the Enrollment
Response to Cyclical Trends and Job Loss,” paper presented at the PERC Applied Microeconomics
workshop, held at Texas A&M University, College Station, Texas, March 20,
https://econweb.tamu.edu/common/files/workshops/PERC%20Applied%20Microeconomics/2013_3_20_S
arah_Turner.pdf. For research showing that the high numbers of workers seeking disability status is
correlated with sectoral employment declines and demographics and not correlated with the rate of
workplace injuries, see Norma B. Coe and Matthew S. Rutledge (2013), “Why Did Disability Allowance
Rates Rise in the Great Recession?” Center for Retirement Research paper 13-11 (Chestnut Hill, Mass.:
Center for Retirement Research at Boston College, August), http://crr.bc.edu/wpcontent/uploads/2013/08/IB_13-11-508.pdf; and John Merline (2012), “The Sharp Rise in Disability
Claims,” Federal Reserve Bank of Richmond, Region Focus (Second/Third Quarter), pp. 24-26,
www.richmondfed.org/publications/research/region_focus/2012/q2-3/pdf/feature3.pdf.
9

-7less to declining participation in the period ahead than would otherwise be expected
based on the aging workforce.11
A second factor bearing on estimates of labor market slack is the elevated number
of workers who are employed part time but desire full-time work (those classified as
“part time for economic reasons”). At nearly 5 percent of the labor force, the number of
such workers is notably larger, relative to the unemployment rate, than has been typical
historically, providing another reason why the current level of the unemployment rate
may understate the amount of remaining slack in the labor market. Again, however,
some portion of the rise in involuntary part-time work may reflect structural rather than
cyclical factors. For example, the ongoing shift in employment away from goods
production and toward services, a sector which historically has used a greater portion of
part-time workers, may be boosting the share of part-time jobs. Likewise, the continuing
decline of middle-skill jobs, some of which could be replaced by part-time jobs, may
raise the share of part-time jobs in overall employment.12 Despite these challenges in
assessing where the share of those employed part time for economic reasons may settle in

11

The effects of the Great Recession on retirements are difficult to identify. During the recession and
immediately after, the losses in wealth may have put upward pressure on labor force participation; the
persistently weak labor market may have subsequently contributed to more retirements and thus put
downward pressure on participation. Perhaps as a result of these confounding forces, early research on the
effects of the Great Recession on retirement finds unclear results. For example, see Alan L. Gustman,
Thomas L. Steinmeier, and Nahid Tabatabai (2011), “How Did the Recession of 2007-2009 Affect the
Wealth and Retirement of the Near Retirement Age Population in the Health and Retirement Study?”
NBER Working Paper Series 17547 (Cambridge, Mass.: National Bureau of Economic Research,
October). For a discussion of these developments, see Richard W. Johnson (2012), “Older Workers,
Retirement, and the Great Recession” (Stanford, Calif.: Russell Sage Foundation and the Stanford Center
on Poverty and Inequality, October), http://web.stanford.edu/group/recessiontrends/cgibin/web/sites/all/themes/barron/pdf/Retirement_fact_sheet.pdf.
12
See Tomaz Cajner, Dennis Mawhirter, Christopher Nekarda, and David Ratner (2014), “Why Is
Involuntary Part-Time Work Elevated?” FEDS Notes (Washington: Board of Governors of the Federal
Reserve System, April 14), www.federalreserve.gov/econresdata/notes/feds-notes/2014/why-isinvoluntary-part-time-work-elevated-20140414.html; and Murat Tasci and Jessica Ice (2014), “Job
Polarization and the Great Recession,” Federal Reserve Bank of Cleveland, Economic Trends (May 28),
www.clevelandfed.org/research/trends/2014/0614/01labmar.cfm?WT.oss=murat tasci&WT.oss_r=380.

-8the long run, the sharp run-up in involuntary part-time employment during the recession
and its slow decline thereafter suggest that cyclical factors are significant.
Private sector labor market flows provide additional indications of the strength of
the labor market. For example, the quits rate has tended to be pro-cyclical, since more
workers voluntarily quit their jobs when they are more confident about their ability to
find new ones and when firms are competing more actively for new hires. Indeed, the
quits rate has picked up with improvements in the labor market over the past year, but it
still remains somewhat depressed relative to its level before the recession. A significant
increase in job openings over the past year suggests notable improvement in labor market
conditions, but the hiring rate has only partially recovered from its decline during the
recession. Given the rise in job vacancies, hiring may be poised to pick up, but the
failure of hiring to rise with vacancies could also indicate that firms perceive the
prospects for economic growth as still insufficient to justify adding to payrolls.
Alternatively, subdued hiring could indicate that firms are encountering difficulties in
finding qualified job applicants. As is true of the other indicators I have discussed, labor
market flows tend to reflect not only cyclical but also structural changes in the economy.
Indeed, these flows may provide evidence of reduced labor market dynamism, which
could prove quite persistent.13 That said, the balance of evidence leads me to conclude

13

For an analysis documenting declines in the rates of hiring, layoffs, and quits, along with lower job
creation and destruction, see Steven J. Davis, R. Jason Faberman, and John Haltiwanger (2012), “Labor
Market Flows in the Cross Section and over Time,” Journal of Monetary Economics, vol. 59 (January),
pp. 1-18. For a review of a range of evidence and possible explanations, see Henry R. Hyatt and James R.
Spletzer (2013), “The Recent Decline in Employment Dynamics,” IZA Discussion Paper Series 7231
(Bonn, Germany: Institute for the Study of Labor (IZA), February), http://ftp.iza.org/dp7231.pdf. These
authors suggest that much additional work is needed to understand the role of different factors in changes in
labor market dynamism. For an analysis that raises the possibility that some of these shifts reflect better
job matches, see Raven Molloy, Christopher L. Smith, and Abigail K. Wozniak (2014), “Declining
Migration within the U.S.: The Role of the Labor Market,” NBER Working Paper Series 20065
(Cambridge, Mass.: National Bureau of Economic Research, April).

-9that weak aggregate demand has contributed significantly to the depressed levels of quits
and hires during the recession and in the recovery.
One convenient way to summarize the information contained in a large number of
indicators is through the use of so-called factor models. Following this methodology,
Federal Reserve Board staff developed a labor market conditions index from 19 labor
market indicators, including four I just discussed.14 This broadly based metric supports
the conclusion that the labor market has improved significantly over the past year, but it
also suggests that the decline in the unemployment rate over this period somewhat
overstates the improvement in overall labor market conditions.
Finally, changes in labor compensation may also help shed light on the degree of
labor market slack, although here, too, there are significant challenges in distinguishing
between cyclical and structural influences. Over the past several years, wage inflation, as
measured by several different indexes, has averaged about 2 percent, and there has been
little evidence of any broad-based acceleration in either wages or compensation. Indeed,
in real terms, wages have been about flat, growing less than labor productivity. This
pattern of subdued real wage gains suggests that nominal compensation could rise more
quickly without exerting any meaningful upward pressure on inflation. And, since wage
movements have historically been sensitive to tightness in the labor market, the recent

Among the indicators in the “labor market conditions index” are the labor force participation rate,
workers classified as part time for economic reasons, hires, and quits. The index does not include the
JOLTS job openings series but instead uses the Board staff’s composite help-wanted index, which has a
longer history; the two measures generally track each other closely. See Hess Chung, Bruce Fallick,
Christopher Nekarda, and David Ratner (2014), “Assessing the Change in Labor Market Conditions,”
FEDS Notes (Washington: Board of Governors of the Federal Reserve System, May 22),
www.federalreserve.gov/econresdata/notes/feds-notes/2014/assessing-the-change-in-labor-marketconditions-20140522.html. For a closely related index of labor market conditions, see Craig S. Hakkio and
Jonathan L. Willis (2013), “Assessing Labor Market Conditions: The Level of Activity and the Speed of
Improvement,” Federal Reserve Bank of Kansas City, Macro Bulletin, July 18,
www.frbkc.org/publicat/research/macrobulletins/mb13Hakkio-Willis0718.pdf.
14

- 10 behavior of both nominal and real wages point to weaker labor market conditions than
would be indicated by the current unemployment rate.
There are three reasons, however, why we should be cautious in drawing such a
conclusion. First, the sluggish pace of nominal and real wage growth in recent years may
reflect the phenomenon of “pent-up wage deflation.”15 The evidence suggests that many
firms faced significant constraints in lowering compensation during the recession and the
earlier part of the recovery because of “downward nominal wage rigidity”--namely, an
inability or unwillingness on the part of firms to cut nominal wages. To the extent that
firms faced limits in reducing real and nominal wages when the labor market was
exceptionally weak, they may find that now they do not need to raise wages to attract
qualified workers. As a result, wages might rise relatively slowly as the labor market
strengthens. If pent-up wage deflation is holding down wage growth, the current very
moderate wage growth could be a misleading signal of the degree of remaining slack.
Further, wages could begin to rise at a noticeably more rapid pace once pent-up wage
deflation has been absorbed.
Second, wage developments reflect not only cyclical but also secular trends that
have likely affected the evolution of labor’s share of income in recent years. As I noted,
real wages have been rising less rapidly than productivity, implying that real unit labor
costs have been declining, a pattern suggesting that there is scope for nominal wages to
accelerate from their recent pace without creating meaningful inflationary pressure.
However, research suggests that the decline in real unit labor costs may partly reflect

See Mary Daly and Bart Hobijn (2014), “Downward Nominal Wage Rigidities Bend the Phillips Curve,”
Working Paper Series 2013-08 (San Francisco: Federal Reserve Bank of San Francisco, January),
www.frbsf.org/publications/economics/papers/2013/wp2013-08.pdf.
15

- 11 secular factors that predate the recession, including changing patterns of production and
international trade, as well as measurement issues.16 If so, productivity growth could
continue to outpace real wage gains even when the economy is again operating at its
potential.
A third issue that complicates the interpretation of wage trends is the possibility
that, because of the dislocations of the Great Recession, transitory wage and price
pressures could emerge well before maximum sustainable employment has been reached,
although they would abate over time as the economy moves back toward maximum
employment.17 The argument is that workers who have suffered long-term
unemployment--along with, perhaps, those who have dropped out of the labor force but
would return to work in a stronger economy--face significant impediments to
reemployment. In this case, further improvement in the labor market could entail
stronger wage pressures for a time before maximum employment has been attained.18

For a recent study of the decline in labor’s share, see Michael W.L. Elsby, Bart Hobijn, and Aysegul
Sahin (2013), “The Decline of the U.S. Labor Share,” Working Paper Series 2013-27 (San Francisco:
Federal Reserve Bank of San Francisco, September), www.frbsf.org/economic-research/files/wp201327.pdf. The notion that the labor share of income is a good measure of slack was prominent in the
empirical literature on the New-Keynesian Phillips Curve (for example, see Jordi Galí and Mark Gertler
(1999), “Inflation Dynamics: A Structural Econometric Analysis,” Journal of Monetary Economics, vol.
44 (October), pp. 195-222), and the connections (or lack thereof) between the labor share and traditional
measures of slack (in the statistical sense) were highlighted in, among others, Michael T. Kiley (2007), “A
Quantitative Comparison of Sticky-Price and Sticky-Information Models of Price Setting,” Journal of
Money, Credit and Banking, vol. 39 (February), pp. 101-25; and Michael T. Kiley (2013), “Output Gaps,”
Journal of Macroeconomics, vol. 37 (September), pp. 1-18. Moreover, recent research has highlighted the
challenges that swings in the labor share have presented for the interpretation of inflation developments (for
example, Marco Del Negro, Marc P. Giannoni, and Frank Schorfheide (2014), “Inflation in the Great
Recession and New Keynesian Models,” NBER Working Paper Series 20055 (Cambridge, Mass.: National
Bureau of Economic Research, April)).
17
See Glenn D. Rudebusch and John C. Williams (2014), “A Wedge in the Dual Mandate: Monetary
Policy and Long-Term Unemployment,” Working Paper Series 2014-14 (San Francisco: Federal Reserve
Bank of San Francisco, May), www.frbsf.org/economic-research/publications/working-papers/wp201414.pdf.
18
For example, see Alan B. Krueger, Judd Cramer, and David Cho (2014), “Are the Long-Term
Unemployed on the Margins of the Labor Market?” paper presented at the Brookings Panel on Economic
Activity, held at the Brookings Institution, Washington, March 20-21,
www.brookings.edu/~/media/Projects/BPEA/Spring%202014/2014a_Krueger.pdf; and Robert J. Gordon
16

- 12 Implications of Labor Market Developments for Monetary Policy
The focus of my remarks to this point has been on the functioning of the labor
market and how cyclical and structural influences have complicated the task of
determining the state of the economy relative to the FOMC’s objective of maximum
employment. In my remaining time, I will turn to the special challenges that these
difficulties in assessing the labor market pose for evaluating the appropriate stance of
monetary policy.
Any discussion of appropriate monetary policy must be framed by the Federal
Reserve’s dual mandate to promote maximum employment and price stability. For much
of the past five years, the FOMC has been confronted with an obvious and substantial
degree of slack in the labor market and significant risks of slipping into persistent belowtarget inflation. In such circumstances, the need for extraordinary accommodation is
unambiguous, in my view.
However, with the economy getting closer to our objectives, the FOMC’s
emphasis is naturally shifting to questions about the degree of remaining slack, how

(2013), “The Phillips Curve Is Alive and Well: Inflation and the NAIRU during the Slow Recovery,”
NBER Working Paper Series 19390 (Cambridge, Mass.: National Bureau of Economic Research, August).
For research highlighting potential alternative interpretations, see Michael T. Kiley (2014), “An Evaluation
of the Inflationary Pressure Associated with Short- and Long-Term Unemployment,” Finance and
Economics Discussion Series 2014-28 (Washington: Board of Governors of the Federal Reserve System,
March), www.federalreserve.gov/pubs/feds/2014/201428/201428pap.pdf; and Christopher Smith (2014),
“The Effect of Labor Slack on Wages: Evidence from State-Level Relationships,” FEDS Notes
(Washington: Board of Governors of the Federal Reserve System, June 2),
www.federalreserve.gov/econresdata/notes/feds-notes/2014/effect-of-labor-slack-on-wages-evidence-fromstate-level-relationships-20140602.html.
The interaction of labor force participation and inflationary pressures has been understudied, in part
because the strong trends in participation due to demographic factors have implied that it is difficult to
identify the cyclical component. For an important example of a study demonstrating, within the core
macroeconomic framework widely used in research, that movements in participation should be considered
in models of wage and price determination, see Christopher J. Erceg and Andrew T. Levin (2013), “Labor
Force Participation and Monetary Policy in the Wake of the Great Recession,” IMF Working Paper
WP/13/245 (Washington: International Monetary Fund, July),
www.imf.org/external/pubs/ft/wp/2013/wp13245.pdf.

- 13 quickly that slack is likely to be taken up, and thereby to the question of under what
conditions we should begin dialing back our extraordinary accommodation. As should be
evident from my remarks so far, I believe that our assessments of the degree of slack
must be based on a wide range of variables and will require difficult judgments about the
cyclical and structural influences in the labor market. While these assessments have
always been imprecise and subject to revision, the task has become especially
challenging in the aftermath of the Great Recession, which brought nearly unprecedented
cyclical dislocations and may have been associated with similarly unprecedented
structural changes in the labor market--changes that have yet to be fully understood.
So, what is a monetary policymaker to do? Some have argued that, in light of the
uncertainties associated with estimating labor market slack, policymakers should focus
mainly on inflation developments in determining appropriate policy. To take an extreme
case, if labor market slack was the dominant and predictable driver of inflation, we could
largely ignore labor market indicators and look instead at the behavior of inflation to
determine the extent of slack in the labor market. In present circumstances, with inflation
still running below the FOMC’s 2 percent objective, such an approach would suggest that
we could maintain policy accommodation until inflation is clearly moving back toward 2
percent, at which point we could also be confident that slack had diminished.
Of course, our task is not nearly so straightforward. Historically, slack has
accounted for only a small portion of the fluctuations in inflation. Indeed, unusual
aspects of the current recovery may have shifted the lead-lag relationship between a
tightening labor market and rising inflation pressures in either direction. For example, as
I discussed earlier, if downward nominal wage rigidities created a stock of pent-up wage

- 14 deflation during the economic downturn, observed wage and price pressures associated
with a given amount of slack or pace of reduction in slack might be unusually low for a
time. If so, the first clear signs of inflation pressure could come later than usual in the
progression toward maximum employment. As a result, maintaining a high degree of
monetary policy accommodation until inflation pressures emerge could, in this case,
unduly delay the removal of accommodation, necessitating an abrupt and potentially
disruptive tightening of policy later on.
Conversely, profound dislocations in the labor market in recent years--such as
depressed participation associated with worker discouragement and a still-substantial
level of long-term unemployment--may cause inflation pressures to arise earlier than
usual as the degree of slack in the labor market declines. However, some of the resulting
wage and price pressures could subsequently ease as higher real wages draw workers
back into the labor force and lower long-term unemployment.19 As a consequence,
tightening monetary policy as soon as inflation moves back toward 2 percent might, in
this case, prevent labor markets from recovering fully and so would not be consistent
with the dual mandate.
Inferring the degree of resource utilization from real-time readings on inflation is
further complicated by the familiar challenge of distinguishing transitory price changes
from persistent price pressures. Indeed, the recent firming of inflation toward our
2 percent goal appears to reflect a combination of both factors.
These complexities in evaluating the relationship between slack and inflation
pressures in the current recovery are illustrative of a host of issues that the FOMC will be

19

See Rudebusch and Williams, “A Wedge in the Dual Mandate,” in note 17.

- 15 grappling with as the recovery continues. There is no simple recipe for appropriate
policy in this context, and the FOMC is particularly attentive to the need to clearly
describe the policy framework we are using to meet these challenges. As the FOMC has
noted in its recent policy statements, the stance of policy will be guided by our
assessments of how far we are from our objectives of maximum employment and
2 percent inflation as well as our assessment of the likely pace of progress toward those
objectives.
At the FOMC’s most recent meeting, the Committee judged, based on a range of
labor market indicators, that “labor market conditions improved.”20 Indeed, as I noted
earlier, they have improved more rapidly than the Committee had anticipated.
Nevertheless, the Committee judged that underutilization of labor resources still remains
significant. Given this assessment and the Committee’s expectation that inflation will
gradually move up toward its longer-run objective, the Committee reaffirmed its view
“that it likely will be appropriate to maintain the current target range for the federal funds
rate for a considerable time after our current asset purchase program ends, especially if
projected inflation continues to run below the Committee’s 2 percent longer-run goal, and
provided that longer-term inflation expectations remain well anchored.”21 But if progress
in the labor market continues to be more rapid than anticipated by the Committee or if
inflation moves up more rapidly than anticipated, resulting in faster convergence toward
our dual objectives, then increases in the federal funds rate target could come sooner than
the Committee currently expects and could be more rapid thereafter. Of course, if

See paragraph 1 of Board of Governors of the Federal Reserve System (2014), “Federal Reserve Issues
FOMC Statement,” press release, July 30,
www.federalreserve.gov/newsevents/press/monetary/20140730a.htm.
21
See paragraph 5 in Board of Governors, “FOMC Statement” (July 2014), in note 20.
20

- 16 economic performance turns out to be disappointing and progress toward our goals
proceeds more slowly than we expect, then the future path of interest rates likely would
be more accommodative than we currently anticipate. As I have noted many times,
monetary policy is not on a preset path. The Committee will be closely monitoring
incoming information on the labor market and inflation in determining the appropriate
stance of monetary policy.
Overall, I suspect that many of the labor market issues you will be discussing at
this conference will be at the center of FOMC discussions for some time to come. I thank
you in advance for the insights you will offer and encourage you to continue the
important research that advances our understanding of cyclical and structural labor
market issues.