April 05, 2013
Labor Market Update: Muddy Waters Continue to Run Deep
Earlier this week, Atlanta Fed President Dennis Lockhart gave a speech in Birmingham, Alabama, focused on labor markets, risks to the outlook, and current monetary policy. One of the things President Lockhart noted was that the picture for the labor market remained muddy. Specifically:
"The fact is that conditions in the broad labor market are quite mixed. While some indicators of labor market health have improved a lot since the recession, others have not improved much at all or have even worsened. As I said, net job creation is picking up. Initial claims for unemployment insurance have fallen. But the official rate of unemployment remains high, many discouraged workers have left the labor force, and there are many people working part-time jobs who want to work full time."
Today's labor report did little to clarify improvement in labor market conditions, with March payrolls estimated to have grown by a much less than expected 88,000 workers and the jobless rate falling one-tenth of a percent, to 7.6 percent, on the back of a decline of 496,000 in the size in the labor force. Updated with today's data, below is the spider chart we have previously offered as one way to simultaneously track and visualize "conditions in the broad labor market."
As a reminder, we've taken the approach of dividing a set of 13 indicators of labor market conditions into four segments:
- Employer Behavior includes indicators related to the hiring activities of employers.
- Confidence includes indicators of employer and worker confidence in the labor market.
- Utilization includes measures related to available labor resources.
- Leading Indicators shows data that typically provide insight into the future direction of overall labor market activity.
The circle at the perimeter of this chart represents labor market conditions that existed just before the recession. We have dated this as late 2007. The inner circle represents the state of affairs when payroll employment reached its trough in late 2009. The oddly shaped red figure inside the perimeter depicts where each of the indicators was in March 2011 relative to the benchmarks. The purple figure depicts the state of the labor market in March 2012. Finally, the blue figure shows where the indicators were as of March 2013. All of the indicators are scaled so that outward movement represents improvement. The progression of these point-in-time snapshots provides us with a picture of how labor market conditions have evolved over the past four years.
As you can see, substantial improvement has arguably been achieved in the leading indicator series. As a group, these data points are approaching their prerecession levels. Employer hiring behavior and confidence are slowly moving outward but remain quite weak relative to their prerecession benchmarks. Finally, the labor utilization measures are very weak and, notably, have hardly improved at all over the past two years.
In the macroblog post that introduced the spider chart, we noted that there are a couple of immediate issues that arise in using this graphic to interpret market improvement, substantial or otherwise:
First, a variable such as the level of payroll employment will eventually exceed its pre-recession level, and grow consistently over time as the population grows. A variable like "hiring plans"—which is the net percentage of firms in the National Federation of Independent Business survey expecting to hire employees in the next three months—cannot grow without bound....
Second, it is not obvious that 2007:IVQ levels are necessarily the best benchmarks for all (or even any) of the variables we are monitoring [in the spider chart].
Of these two issues, the second one is potentially the more problematic. The spider chart invites you to think of the inner circle as the starting point and the outer circle as the "goal," or a representation of "normal" labor market conditions. In assessing improvement in variables such as payroll employment, using the prerecession level as a reference point makes some sense as a minimal standard. But for some, "minimal" may be the operative word. If we are interested in questions of how employment is doing relative to some concept of "full employment," it might be appropriate to assess the data relative to some measure of trend. For example, payroll employment is still about 3 million below the prerecession peak, but the labor force, despite the drop in March, is by about 1 million higher than before the recession. When measured relative to the size of the labor market, progress on employment is less impressive than it would appear by just looking at growth in employment itself.
One way to address both of the caveats noted above is to scale variables that involve numbers of jobs or people—such as payroll employment or the number of unemployed—by the size of the population or the labor force. Doing so ensures that variables measured in numbers of jobs or people do not grow without bound. It also helps in assessing progress in these variables relative to a (back-of-the-envelope) measure of the trend in labor resources.
We take this approach in the following chart, which reproduces the spider chart but divides the variables that are counts of people by the size of the labor force. (The labor force has grown more slowly than the population since the end of the recession, but a generally similar picture emerges if the variables are instead deflated by the population.)
Not surprisingly, for most of the indicators, labor market progress is a bit more subdued relative to postrecession growth in the labor force than growth in the indicators alone would suggest. These adjustments definitely would not alter our view that the labor market picture is "quite mixed." President Lockhart's recent comments on CNBC—in which he said he would like to see more positive data before declaring that sustained improvement has taken hold—seem especially prescient in light of today's job numbers:
By John Robertson, vice president and senior economist in the Atlanta Fed's research department, and
Dave Altig, executive vice president and research director of the Atlanta Fed
TrackBack URL for this entry:
Listed below are links to blogs that reference Labor Market Update: Muddy Waters Continue to Run Deep:
March 28, 2013
The Same-Old, Same-Old Labor Market
In his March 24 Wall Street Journal piece on declining government payrolls, Sudeep Reddy offers up a key observation:
The cuts in the public-sector workforce—at the federal, state and local levels—marked the deepest retrenchment in government employment of civilians since just after World War II... down by about 740,000 jobs since the recession ended in June 2009. At the same time, the private sector has added more than 5.2 million jobs over the course of the recovery.
As the Journal article notes, the story of shrinking government employment combining with private-sector payroll expansion has been remarkably consistent for much of the recovery.
About a year ago, we provided a graphical illustration of postrecession employment patterns using payroll-employment "bubble charts." These charts measure postrecession average monthly employment changes by sector relative to the changes in the prerecession period from December 2001 through October 2007. Not a lot in that chart has changed over the intervening year (just as not a lot had changed in 2012 compared with 2011):
The stability in the employment picture across private industries, both relative to one another and relative to the precrisis pace of job gains, is just as notable as the changing fortunes of private versus public employment. In fact, the charts offer some pretty clear impressions:
- Virtually all private-sector industries have moved into positive employment-growth territory. That movement includes the construction and financial activities sectors, which have generally lagged the improvement in the rest of the economy. The only broad category still shedding jobs in the private sector has been the information industry—which includes publishing, motion picture production, telecommunications, data processing, and the like—an industry that was also shrinking in terms of employment over the decade leading up to the financial crisis.
- All of the private-sector bubbles in the charts are now close to, on, or above the 45-degree line, meaning that the average pace of monthly job creation in each sector is near, equal to, or greater than what prevailed during the last recovery.
- As the Reddy piece emphasizes, government employment has been in decline since early 2010, though the government sector as a whole retains its status as the sector with the largest employment share. (The size of the bubbles in the charts above represents the share of employment in each sector at the end of the period for which the graph is drawn.) But the chart also illustrates another key point of Reddy's article: To date, the decline in government employment has been concentrated in state and, especially, local government jobs. Until recently, job creation by the federal government, which is relatively small in the bigger scheme of things, has not deviated much from its prerecession pattern.
The last point brings us to this observation, from the WSJ article:
How the rest of the private sector responds to a shrinking of the federal government could play a bigger role in determining how the budget fight hits the workforce.
"The private sector in the U.S. is growing so much stronger than anyone had expected," said Bernard Baumohl of the Economic Outlook Group. "This organic growth is going to significantly offset the effect of the sequester in terms of economic output and employment."
It is worth pointing out that the monthly average of 17,000 state, local, and federal government jobs lost since March 2010 has been nearly matched by average monthly increases of better than 14,000 jobs in manufacturing, a sector that persistently shed jobs in the previous recovery. The replacement of private- for public-sector employment has generated 175,000 to 185,000 net jobs per month in both 2011 and 2012. To put one perspective on that figure, at current labor force participation rates (along with some other assumptions and caveats), that pace would be sufficient to reach the Federal Open Market Committee's 6.5 percent unemployment threshold by sometime in spring 2015 (as you can verify yourself with the Atlanta Fed's Jobs Calculator). That calculation raises the stake somewhat on the matter of how "the rest of the private sector responds."
By Dave Altig, executive vice president and research director of the Atlanta Fed
TrackBack URL for this entry:
Listed below are links to blogs that reference The Same-Old, Same-Old Labor Market:
March 19, 2013
Being Ahead of the Curve: Not Always a Good Thing
Our friends at the New York Fed have a nifty interactive graphic that compares the unemployment rate, labor force participation rate, and employment-to-population ratio over the last five business cycles. You can even break these indicators down by gender, by age, or by a particular business cycle. (For a deeper dive, check out this post at Liberty Street Economics by Jonathan McCarthy and Simon Potter.) And though it’s not exactly late-breaking news, no matter which of the three indicators you look at, you can’t help but conclude that the most recent recession is an outlier.
The Beveridge curve is a fourth and particularly useful graphical representation of a steady-state economy showing how, in theory, one might expect the vacancy rate to change, given an unemployment rate. It depicts the relationship between job openings and the unemployment rate. (The Atlanta Fed’s magazine, EconSouth, discussed the Beveridge curve.) It, too, has been standing out over the course of the most recent recovery, so much so that we think it warrants at least a second glance. There are a number of ways to estimate a Beveridge curve (see, for example, methods described by Gadi Barlevy of the Chicago Fed here and by the Richmond Fed’s Thomas Lubik here).
We use the method described by Barnichon et al. (2012) to estimate the solid curve used in the first chart below. The square plots represent actual vacancy rate (y-axis) and unemployment rate (x-axis) combinations by month from December 2000, when the Job Openings and Labor Turnover Statistics (JOLTS) data series from the U.S. Bureau of Labor Statistics (BLS) series begins, to January 2013, the most recent month of data available for both series.
Blue squares represent data from December 2000 to December 2009, when the “errors” between actual plots and the curve estimation were below 2 percentage points, and red squares represent data since January 2010, where data suddenly seem to jump higher than the predicted Beveridge curve to the tune of 2 percentage points or greater (see the chart below).
In June 2012, Regis Barnichon and his coauthors concluded that the unemployment rate’s lackluster performance so far in the recovery was attributable to a shortfall in hires per vacancy. Since then, the vacancy rate has climbed its way back to its June 2008 level of 2.7 percent. However, the unemployment rate has clearly not returned to either its June 2008 level (5.6 percent) or where the Beveridge curve says it should be given this vacancy rate, which one might predict to be 5.5 percent using the methodology of Barnichon et al.
This “ahead of the curve” phenomenon has not gone unnoticed and has prompted some explanations. In a March 6, 2013, article in The New York Times (which also has some cool charts), Catherine Rampell posits that available positions are staying unfilled longer, while interview processes have become lengthier.
The next day, Rampell went into more detail about why we’re going “off the curve” in a New York Times Economix post. She cites skills mismatch and a skills atrophy effect of the long-term unemployed affecting the ability of employers to fill positions (which she explains aren’t full explanations, yet we would expect to see wages for highly coveted positions rise significantly).
Rampell goes back to the explanation many of us continue to hear from business contacts: employers are unwilling to fill vacant positions because of economic and fiscal policy uncertainty. She quotes Stephen Davis of Chicago’s Booth School: “They’re taking longer to fill vacancies because they just feel less need to fill jobs now,” Davis said. “They recognize that in a slack labor market, there is an abundance of viable candidates. If something happens, and if they need to hire quickly, they know they can do that. That’s harder in a tight labor market.”
So maybe as labor markets “tighten up,” or perhaps if the speed by which they tighten up quickens, we’ll get back on the Beveridge curve. Only time, and several BLS releases, will tell.
By Patrick Higgins, an economist at the Atlanta Fed, and
Mark Carter, a senior economic analyst at the Atlanta Fed
TrackBack URL for this entry:
Listed below are links to blogs that reference Being Ahead of the Curve: Not Always a Good Thing:
February 15, 2013
Promoting Job Creation: Don't Forget the Old Guys
In a provocative article posted this week, the American Enterprise Institute's James Pethokoukis concludes that the state of entrepreneurship in the United States is, disturbingly, weaker than ever. In particular, Pethokoukis documents a decline in jobs created by establishments less than one year old, a trend that began before the 2001 recession and has continued more or less unabated since. He specifically cites the following symptoms of trouble:
- Had small business come out of the recession maintaining just the rate of start-ups generated in 2007, according to McKinsey, the U.S. economy would today have almost 2.5 million more jobs than it does.
- There were fewer new firms formed in 2010 and 2011 than during the Great Recession.
- The rate of start-up jobs during 2010 and 2011—years that were technically in full recovery—were the lowest on record, according to economist Tim Kane of the Hudson Institute.
That last point appears to be all the more ominous given this observation from Tim Kane:
"...that startups create essentially all net new jobs. Existing employers, it turns out, tend to be net job losers, averaging net losses of 1 million workers per year."
Pethokoukis makes his case with political commentary that we don't endorse and don't find particularly helpful. But we won't argue with his conclusion that more entrepreneurial start-up activity would be a good thing. Nonetheless, we get a little concerned when the conversation jumps from data on net job creation and the role of start-ups and early life-cycle firms, and moves on to policy conclusions that seem to disproportionately focus on that class of businesses specifically.
Here is the source of our concern: Though it is also tempting to lump all "existing employers” into the basket of net job destroyers, there are existing firms that create jobs, and a few are doing so on a very large scale.
Take 2006, for instance. Based on data from the Commerce Department called Business Dynamics Statistics (BDS), new firms (businesses with a payroll that existed in March 2006 but not in March 2005) had about 3.5 million employees. This is the large net job creation by new firms reported by Kane. However, over the same year, expanding firms more than 10 years old added a whopping 11 million jobs—about three times as many jobs as created by new firms. Of course, some older firms were downsizing or closing—contracting mature firms destroyed an estimated 10 million jobs. So the net number of jobs created by older established firms looks somewhat less impressive than the record of those young start-ups. But in the overall picture, were the 11 million jobs created by the expanding older businesses really less important than 3.5 million created by the newbies?
It turns out that older firms also account for a large fraction of the job creation occurring in fast-growing firms, arguably a better characterization of entrepreneurism than newness. We found some compelling evidence reported in recent research by Akbar Sadeghi, James Spletzer, and David Talan. Using data from the U.S. Bureau of Labor Statistics' Business Employment Dynamics (BED), Sadeghi, Spletzer, and Talen find that older firms (those at least 10 years old) accounted for more than 40 percent of the employment created by high-growth firms (those with at least 20 percent annual employment gains between 2008 and 2011). A similar conclusion about the role of older, fast-growing firms is found in this earlier Kauffman Foundation report based on BDS data looking at the 1 percent of fastest-growing firms in the United States.
The point is not that start-up entrepreneurial activity is unimportant. It is vitally important. But in larger terms, we should recognize that all entrepreneurial activity is important, no matter what the age of the firm in which it occurs. Atlanta Fed President Dennis Lockhart highlighted this point in a speech delivered earlier this week at Instituto de Empresas in Madrid, Spain:
My bank's experience in trying to understand the role of small businesses, small-growth businesses, young businesses, and mature-growth businesses in job creation illustrates a key point, I think. In the pursuit of economic growth and increased employment, there is no silver bullet. Rather, the policy community should be pursuing an effective mix of policy elements (with focus in areas such as new business formation, labor rules, and regulatory efficiency, to name a few) that together catalyze a virtuous circle of innovation, growth, and employment.
Certainly, entrepreneurial risk-taking, whether by large, mature businesses or start-ups aimed at becoming growth companies, is part of the solution.
When it comes to promoting job creation, forgetting to throw mature businesses into the mix with start-ups is surely not the path to finding the best policy solutions.
TrackBack URL for this entry:
Listed below are links to blogs that reference Promoting Job Creation: Don't Forget the Old Guys:
February 05, 2013
2013 Business Hiring Plans: Employment, Effort, Hours, and Fiscal Uncertainty
How much is fiscal uncertainty holding back hiring? The answer seems to depend on whom you ask. Early in January, the Atlanta Fed spoke to 670 businesses in the Southeast about employment. Conditional on the respondents’ 2013 hiring plans (expand, hold steady, or contract), the following set of charts summarizes the results for how the businesses viewed activity relative to their own interpretation of “normal” along three dimensions: their current employment level, the amount of effort required from their staff per hour, and the average hours worked per employee. These questions were modeled on questions asked in the Atlanta Fed’s December 2012 Business Inflation Expectations Survey. In the following three charts, the green bars represent firms that said they planned to expand employment in 2013. The grey bars represent firms that said they did not plan to change their employment level in 2013, and the red bars represent firms that planned to reduce employment in 2013.
The first chart shows the results for current employment. Regardless of hiring plans over the next 12 months, most firms said they were currently at or below normal employment levels. Those planning on increasing employment over the next 12 months were a bit more likely to say they have already surpassed normal levels of employment than other firms, while those looking to shed employees were very likely to say their employment level is below normal employment levels.
Chart 2 shows that businesses are generally pushing hard along the effort dimension. Firms were quite likely to say that their staff’s effort per hour worked was currently at or above normal, whether or not they were planning to change employment in 2013.
Chart 3 shows that firms planning to expand were very likely to say that average hours worked were at or above normal (28 percent said hours were above normal, 60 percent about normal), whereas firms planning to contract were more likely to say that hours were at or below normal (48 percent about normal, 39 percent below normal).
Taken together, these results suggest that some firms are approaching the limit of how far they can go along the intensive margins of effort and hours before they have to hire more workers. With effort elevated, as more firms increase average hours worked to above-normal levels, one might expect more hiring to follow.
Each business was also asked how uncertainty about future fiscal policy was affecting its hiring plans. Firms planning to reduce employment tended to cite fiscal uncertainty as having a negative impact on their hiring plans. However, for those firms, hours also tended to be well below normal, so it is unlikely that removing fiscal uncertainty would move many of those firms into expansion mode (although it may help stabilize their outlook).
In contrast, fiscal uncertainty was generally viewed as having less impact by those planning to expand employment and those planning to hold employment levels steady. Presumably, reducing fiscal uncertainty would move some of the firms planning to hold steady into expansion mode, and those planning to expand would do so a bit more. To get some idea of this potential, Chart 4 shows the responses by firms who reported above-normal effort per hour and above-normal average hours worked. About 40 percent of those businesses said that fiscal uncertainty had caused them to scale back their hiring plans.
It is unclear whether eliminating fiscal uncertainty would have a big impact on the hiring plans of these firms. But these results suggest that it sure couldn’t hurt.
By John Robertson, vice president and senior economist, and
Ellyn Terry, a senior economic analyst in the Atlanta Fed's research department
TrackBack URL for this entry:
Listed below are links to blogs that reference 2013 Business Hiring Plans: Employment, Effort, Hours, and Fiscal Uncertainty:
November 05, 2012
Reading Labor Markets
When the September employment report was released on October 5, the top-line payroll employment gain for the month, as reported in the U.S. Bureau of Labor Statistics' (BLS) establishment survey, logged in at 114,000. Under standard assumptions, a number of this magnitude would be barely enough to absorb the growth of the labor force and keep the unemployment rate constant. In contrast, in that same October 5 report we learned from the BLS household survey that the measured unemployment rate fell from 8.1 percent in August to 7.8 percent in September.
According to Friday's BLS report on the employment situation for October, the top-line payroll employment gain for the month from the establishment survey was 171,000. At that pace—which is also the current average gain for the past three months—the Atlanta Fed jobs calculator suggests the unemployment rate should fall another one-half of a percentage point over the next year. At the same time, according to the BLS household survey, the unemployment rate rose from 7.8 percent in September to 7.9 percent in October.
This is as good an illustration as any to explain why, on November 1, Atlanta Fed President Dennis Lockhart said the following in a speech to the Chattanooga Tennessee Downtown Rotary Club:
In its post-meeting statement on September 13, the FOMC said, "If the outlook for the labor market does not improve substantially, the Committee will continue its purchases of agency mortgage-backed securities, undertake additional asset purchases, and employ its other policy tools as appropriate until such improvement is achieved in a context of price stability."...
For policy purposes, I think it's appropriate to be cautious about relying on a single indicator of labor market trends—for example, the unemployment rate—to determine whether the condition of "substantial improvement" has been met.
As the FOMC went into its September meeting, the official BLS statistics indicated that net U.S. job creation in August was a mere 92,000. That number is below the “all else equal” threshold of about 100,000 jobs required to keep the unemployment rate from rising, and that information is what Fed policymakers had in hand when they met on September 12–13 and decided on the policy action described by President Lockhart.
On Friday, after two revisions, the BLS told us jobs expanded by 192,000 in August, well above the average for the jobs recovery that started in early 2010, 100,000 jobs (more than double) above the initial estimate.
Looking through month-to-month variations is not a lot of help in real-time tea-leaf reading. Here is the 12-month moving average of employment gains, the blue line indicating the way things looked in September, the red line showing the way they look today:
Over time, it remains the case that monthly employment gains are pretty consistently coming in at 150,000 to 160,000 jobs created per month, and that rate has been enough to generate relatively steady declines in the unemployment rate:
That could change, of course, and the last four months of data have generally shown an acceleration in the job-growth trend. But the data definitely were not indicating that trend as it was happening, an unfortunate reality that isn't likely to change. One way to soften the blow of that problem, emphasized in the Lockhart speech, is to keep an eye on as a broad a set of signals as possible:
... let me share a qualitative framework for defining "substantial improvement."
The starting point certainly should be the headline unemployment rate and the payroll jobs number. The interpretation of movements in these two statistics would be enriched and reinforced by a review of additional data elements.
I added the emphasis there, as I think the point bears highlighting. President Lockhart goes on to give examples of what he would look for in determining whether the substantial improvement threshold has been met. Things like reductions in the numbers of marginally attached and discouraged workers, growing labor force participation rates, declining numbers of people who want full-time work but have to settle for part-time, and positive forward indicators like falling initial claims for unemployment insurance.
The Calculated Risk blog continues to be a one-stop shop for a lot of this information—here and here, for example—and overall nothing much overturns the picture of steady, but slow, progress. That would suggest the acceleration of the past several months is probably not a new trend, but a continuation of the same-old same-old. But then again, the track record painfully demonstrates how hard that is to know in real time.
By Dave Altig, executive vice president and research director at the Atlanta Fed
TrackBack URL for this entry:
Listed below are links to blogs that reference Reading Labor Markets:
October 10, 2012
Divergent Jobs Reports: Will the Real State of the Labor Market Please Stand Up?
The September employment report from the U.S. Bureau of Labor Statistics (BLS) was predestined to create a significant amount of buzz. But the confluence of headline jobs growth at a modest clip of 114,000 and a surprisingly large 0.3 percentage point reduction in the unemployment rate has made the report more buzz-worthy than we (here at macroblog) expected. Although some of the commentary has been more heat than light, there have been some particularly good reminders of the difference between the establishment survey data, from which the headline jobs figure is derived, and the household survey data, from which come the unemployment statistics. The discussions on Greg Mankiw's blog and by Catherine Rampell (at The New York Times's Economix blog) are especially useful. Or, perhaps even better, you can go to the source at the BLS.
It's important to remember that both surveys are subject to error and, because of its much smaller sample size, the household survey can be subject to particularly sizeable swings. Specifically, the standard error of the household survey's monthly change in employment is 436,000(!). Based on the most extreme assumptions about flows in and out of unemployment and in and out of the labor force, understating or overstating actual employment by 436,000 would imply a measured unemployment rate ranging from 7.5 percent to 8.1 percent. (The BLS estimate of the standard error for unemployment puts a range on September's number of 7.6 percent to 8 percent.)
In his post, Greg Mankiw makes reference to a Brookings Institution paper by George Perry from a few years back that offers what is probably good advice: since both the payroll and household surveys are subject to error, and since the errors in each are likely unrelated to one another, the clearest picture about what is happening to employment in real time can be gleaned by combining information from both.
In fact, in a directional sense, both the household and payroll surveys are giving the same signals. In the table below, we compare the recent trends in monthly job gains measured in both surveys. The coverage in the two reports is slightly different. Unlike the payroll count, the household survey includes the self-employed and counts multiple jobs held by a single person as a single instance of employment. Because of this, the BLS also reports an adjusted version of the household survey, called the payroll concept adjusted employment measure. This payroll-consistent measure is designed to control for definitional differences across the household and establishment reports and also makes statistical adjustments for changes to the population controls in various years. So we include the data from this measure in the last column of the table.
Overall, all three measures suggest a weaker trend over the last six months than over the last nine months. All three measures also indicate that things were somewhat stronger on average in the last three months than in the prior three months. The bottom line in our view is that, though the employment levels can be quite different across the three measures, all suggest that the jobs picture has improved somewhat in the past three months.
The suggestion in George Perry's Brookings paper—combining the household and establishment data—can be implemented by constructing a weighted average of the two surveys. In our variation we put weights in proportion to the inverse of the sampling variability of the payroll and household surveys, which would roughly imply an 80 percent weight on the establishment measure and 20 percent on the payroll-consistent household measure. The estimates using these weights are reported in the last column of the table above. Because the component employment measures display directionally similar trends in recent months, the weighted average does as well.
In a speech given a few weeks ago, Atlanta Fed President Dennis Lockhart, our boss here, offered the opinion that
Taking a two-year view, the trend rate of gains in employment has been roughly 150,000 per month. This pace would be sufficient, at current levels of participation in the workforce, to sustain a steady, gradual reduction of the unemployment rate.
As the September employment reports show, predicting the unemployment rate month to month can be tricky business, and there may be better ways than just extrapolating from the jobs data. But thus far we are inclined to think that slow but steady progress on the jobs front is still the best story.
By Dave Altig, executive vice president and research director at the Atlanta Fed
TrackBack URL for this entry:
Listed below are links to blogs that reference Divergent Jobs Reports: Will the Real State of the Labor Market Please Stand Up?:
September 20, 2012
Examining the Recession’s Effects on Labor Markets
Four years after the onset of the Great Recession, labor market outcomes in the U.S. remain depressed. The fraction of 16- to 64-year-old individuals who are employed fell from above 72 percent in 2007 to less than 67 percent in 2009 and remains stuck there. The unemployment rate rose from 4.5 percent to 10 percent and still hovers above 8 percent. And the fraction of unemployed workers who have been looking for a job for more than six months has increased to a share not seen in the United States in at least 60 years. The Atlanta Fed's Center for Human Capital Studies hosted a conference last weekend, organized by Richard Rogerson (Princeton University), Robert Shimer (University of Chicago) and the Atlanta Fed's Melinda Pitts that explored why the employment losses were so large and why the labor market recovery has been so weak. Examining these questions is important because different hypotheses about the nature of the recession suggest that different policy interventions may help to accelerate the recovery.
The paper "On the Importance of the Participation Margin for Labor Market Fluctuations" by Michael Elsby, Bart Hobijn, and Ayşegül Şahin offered some suggestions on how to think about the disparate behavior of the unemployment rate and labor force participation rate during the last couple of years. While the unemployment rate has steadily fallen back towards its historic levels, labor force participation has fallen, keeping the employment-population ratio constant. At some level, this movement suggests that the decline in labor force participation has acted as a relief valve for the unemployment rate. Using evidence on the gross flows of workers between employment, unemployment, and out-of-the-labor-force, Elsby and his coauthors question that interpretation. Instead, relatively few unemployed workers have dropped out of the labor force during the recovery, reflecting the high desire to work among the current stock of unemployed individuals.
A number of papers offered specific hypotheses about the reason for the large and persistent deterioration in labor market outcomes and tested those hypotheses using a variety of methodologies and datasets. For example, the paper "What Explains High Unemployment? The Aggregate Demand Channel" by Atif Mian and Amir Sufi explored the implications of the negative shock to household balance sheets that followed the collapse in house prices. They document that employment in the nonconstruction, nontraded sector declined most in U.S. counties that experienced the largest adverse shock to house prices, while the decline in the traded goods sector occurred equally nationwide. If wages and prices were flexible, we would expect the balance sheet shock to reduce the demand for nontraded goods and raise the supply of labor and hence employment in the traded good sector. The fact that this did not happen is evidence that wages and prices have not adjusted. They infer that roughly two-thirds of the total employment losses can be attributed to the balance sheet shock, in combination with wage and price rigidities.
A second hypothesis is that the recovery has been so weak because of underlying adverse trends in the U.S. labor market. "Manufacturing Busts, Housing Booms, and Declining Employment: A Structural Explanation" by Erik Hurst, Matt Notowidigdo, and Kerwin Charles shows how the ongoing decline in the demand for less educated men in manufacturing has generated a negative trend in labor market outcomes for these workers for three decades. This trend continued unabated during the years after the 2001 recession but was masked by the housing boom, which lifted employment for less-skilled workers for another five years. This observation is relevant for how one interprets the time series changes in labor market outcomes. If we view the housing boom as an aberration that is unlikely to resume, it is inappropriate to compare current labor market outcomes with those just preceding the onset of the Great Recession.
The paper "The Trend is the Cycle: Job Polarization and Jobless Recoveries" by Nir Jaimovich and Henry Siu focuses on a related but distinct long-term phenomenon in the U.S. labor market: job polarization. This refers to the fact that the U.S. labor market increasingly consists of low- and high-paying jobs with relatively few middle-income jobs. While this ongoing change has been noted by other researchers, Jaimovich and Siu show that this long-term evolution has not been occurring at a slow and steady rate but rather has been concentrated during aggregate downturns. They argue that the recent phenomenon of jobless recoveries is simply a reflection of the fact that these are the periods in which middle income jobs are disappearing, never to be brought back.
On the other hand, "The Labor Market Four Years Into the Crisis: Assessing Structural Explanations" by Jesse Rothstein explores and finds little direct evidence for a number of specific structural channels that might explain the weak recovery. For example, there are no identifiable sectors of the U.S. economy with strong wage growth, which suggests that the shortage of suitable workers is probably not a large constraint on employment growth.
A third hypothesis is that the weak recovery reflects an increase in economic uncertainty, which induces firms to wait rather than hire and invest. "Measuring Economic Policy Uncertainty" by Scott Baker, Nicholas Bloom, and Steve Davis proposes a novel methodology for quantifying the overall level of economic uncertainty and the portion of uncertainty that is induced by economic policy. They show that both measures of uncertainty have been elevated since the onset of the Great Recession and have scarcely recovered during recent years. "Uncertainty, Productivity and Unemployment in the Great Recession" by Edouard Schaal examines how an increase in uncertainty affects labor market outcomes in the context of a job search model. He focuses on one measure of uncertainty, the cross-sectional variability of sales growth rates across business establishments, which increased sharply in 2008 but has since subsided. Because of this finding, Schaal finds that the model can account for a large deterioration in labor market outcomes at the time of the shock but that it cannot explain why the deterioration has been so persistent.
A final hypothesis is that the weak recovery reflects disincentive effects of new tax and transfer programs that have been introduced since the onset of the recession. One aspect of this that has attracted particular attention is the extension of unemployment benefits. "The Effect of Unemployment Insurance Extensions on Reemployment Wages" by Johannes Schmieder, Till von Wachter, and Stefan Bender uses evidence from Germany to explore this hypothesis. They show that extending unemployment benefits by six months causes approximately a one-month increase in the amount of time it takes an individual to return to work. This extension has two effects on the wage of workers when they return to work. On the one hand, the additional time to look for a job allows workers to find better jobs. On the other hand, workers' skills tend to decline during an unemployment spell. On net, these effects roughly cancel so extended benefit programs do not have a large impact on average wages.
The framework that most economists use to study the behavior of unemployed workers is search theory. Robert Hall's paper "Viewing the Observed Acceptance Decisions of Job-Seekers through the Lens of Search Theory" analyzes detailed data on the job finding process for a sample of unemployed workers in New Jersey from 2009 in the context of this theory to assess how well the theory can provide a consistent explanation for observed behavior. Previous work had suggested that this framework has problems in accounting for observed job acceptance decisions, but Hall shows that with a few simple modifications, the framework offers a consistent explanation of how workers behave given labor market conditions.
The discussions at the conference questioned the usefulness of labels like deficient demand, structural unemployment, and cyclical unemployment. These terms mean different things in different contexts and do not clarify the key causal factors. Explanations such as "employment is slow because uncertainty is high" could easily fit under any of these banners. Instead, isolating the key changes that have taken place in the U.S. economy, and then scrutinizing the factors that have influenced how those changes have affected the labor market, would be more conducive to arriving at answers.
By Richard Rogerson of Princeton University and Robert Shimer of the University of Chicago, both advisers to the Atlanta Fed’s Center for Human Capital Studies, and Melinda Pitts, a research economist and associate policy adviser in the Atlanta Fed's research department
TrackBack URL for this entry:
Listed below are links to blogs that reference Examining the Recession’s Effects on Labor Markets:
June 07, 2012
The skills gap: Still trying to separate myth from fact
Peter Capelli has looked at the skills gap explanation for labor market weakness and sees more myth than fact:
"Indeed, some of the most puzzling stories to come out of the Great Recession are the many claims by employers that they cannot find qualified applicants to fill their jobs, despite the millions of unemployed who are seeking work. Beyond the anecdotes themselves is survey evidence, most recently from Manpower, which finds roughly half of employers reporting trouble filling their vacancies.
"The first thing that makes me wonder about the supposed 'skill gap' is that, when pressed for more evidence, roughly 10% of employers admit that the problem is really that the candidates they want won't accept the positions at the wage level being offered. That's not a skill shortage, it's simply being unwilling to pay the going price."
To some extent, the issue is semantic:
"But the heart of the real story about employer difficulties in hiring can be seen in the Manpower data showing that only 15% of employers who say they see a skill shortage say that the issue is a lack of candidate knowledge, which is what we'd normally think of as skill. Instead, by far the most important shortfall they see in candidates is a lack of experience doing similar jobs. Employers are not looking to hire entry-level applicants right out of school. They want experienced candidates who can contribute immediately with no training or start-up time..."
In the language of economists, Capelli is defining skill as the possession of generalized human capital, while businesses are defining skill as the possession of firm- or job-specific human capital. In more familiar language, Capelli appears to be focused on innate skill levels and education, while businesses are looking for the types of skills that would be attained through past on-the-job training. In even more colloquial language, Capelli wants businesses to appreciate book-learning, and businesses prefer those who have already survived the school of hard knocks.
We have recently completed our own version of the Manpower survey Capelli references. Our results are based on the responses of about 100 businesses in the Sixth Federal Reserve District represented by the Atlanta Fed, and we do not claim that they are conclusive. But we do think they are instructive.
Of those firms that said they experienced an increase in hiring difficulty over the last year, our poll respondents confirm the notion that businesses are looking for candidates with specific skills:
The lack of technical skills is the only factor that really jumps out as an issue that businesses have with the pool of job applicants. We often hear anecdotal complaints about job seekers' lack of "soft skills," or the difficulty in finding applicants who can pass required background checks. But only 14 percent of all selections indicated too few applicants with required interpersonal skills, and only 7 percent indicated a problem with applicants passing screening requirements like drug-use or credit checks.
On the other hand, our poll found scant support for Capelli's claim that businesses are "unwilling to pay the going price." Only 9 percent of respondents reported that too few applicants would accept the offered compensation package.
Despite the fact that we see some evidence consistent with skill mismatch, it is far from clear that this issue is the smoking gun that explains the current anemic state of job growth. When asked if a dearth of skilled applicants is a persistent problem, our survey respondents overwhelmingly answer "yes." But when asked if they have had more difficulty hiring over the past 12 months, the overwhelming majority answered "no":
Even among the minority of businesses that report recent hiring difficulties, only half indicate that this difficulty is restraining growth:
We infer a couple of lessons from all of this information. First, it does appear that there is a long-term skill level problem in the U.S. economy. Adopting Capelli's definition of skill does not mean the existence of skill mismatch is a myth.
But turning to the short run, we've been pretty sympathetic to structural explanations for the slow pace of the recovery. Nonetheless, we have yet to find much evidence that problems with skill-mismatch are more important postrecession than they were prerecession. We'll keep looking, but—as our colleagues at the Chicago Fed conclude in their most recent Chicago Fed Letter—so far the facts just don't support skill gaps as the major source of our current labor market woes.
By Dave Altig, executive vice president and research director at the Atlanta Fed, and
John Robertson, vice president and senior economist in the Atlanta Fed's research department
TrackBack URL for this entry:
Listed below are links to blogs that reference The skills gap: Still trying to separate myth from fact:
June 01, 2012
Will labor force participation continue to rise?
The labor force participation rate ticked up in May, as did the rate of unemployment. As we have noted in the past, the near-term trajectory of the unemployment rate depends critically on what happens to the participation rate. So the question is, can we expect further upward changes in the participation rate? The answer depends a lot on the labor market attachment of those that are currently out of the labor force.
A few weeks ago, my frequent coauthor, Julie Hotchkiss, wrote about what we can gain from detailed labor market data about the activities of people who have exited the labor force. In her posting, she discussed the overall increase in exits from the labor force, with a focus on 25–54 year olds. Her work concluded that while people identified "Household Care" as the dominant activity for those not in the labor force, there has been a significant upward shift since the recession in those indicating "School" or "Other" as their primary reason for not being in the labor force. A supposition is that at least those that indicated they were in school would reenter the labor force at some point, doing so with a higher level of skills or, at least, with skills that are better aligned with labor demand. However, because we know little about those in the Other category, the future labor market attachment for them is less clear.
This post explores data on transitions into the labor force, primarily for those in the Other category. As in the earlier blog, the focus is on individuals aged 25–54, as retirement dominates the activity of older individuals not in the labor force and schooling dominates the activity of younger individuals not in the labor force.
One indicator of whether those in the Other group are planning to reenter the labor force is whether the individuals in this group are classified as marginally attached to the labor force. A nonparticipant who is marginally attached indicates they want employment or are available for employment. Also, they indicate having looked for a job in the previous year but not actively looking for a job at present. Using monthly data from the Current Population Survey (CPS) that are matched year over year, we see that the marginally attached workers do transition back into the labor force at twice the rate of all individuals who are not in the labor force, as chart 1 illustrates. These rates are relatively stable over time.
As chart 2 shows, a much higher proportion of individuals in the Other category are marginally attached to the labor force, compared to other types of nonparticipants. Moreover, the percentage of these marginally attached nonparticipants has increased from around 20 percent to 30 percent over the last three years.
This higher probability of marginally attached workers returning to the labor force combined with the significantly increased share of marginally attached workers in the Other category suggests that we should expect to find a higher share of those in the Other group returning to the labor force than we've seen in the past. But it turns out that this expected development is not what has happened. The Other group also includes individuals who are not marginally attached to the labor market, and their transition rates into the labor market have declined. On net, while the transition rate to employment is highest for the Other category (reflecting the large of share of marginally attached), the transition rate into the labor force does not fully reflect the increased level of marginal attachment to the labor force.
The group with the next highest transition rate to employment is in the School category, which reflects the inherent transitory nature of that activity. However, it is noteworthy that the school transition rate is lower than it was before the recession. This development reflects an increase in the share of individuals continuing to indicate that school is their primary reason for not participating in the labor force from one year to the next. And it suggests that the lower opportunity cost of attending school is influencing the decision to remain in school longer.
While these trends suggest that we could expect to see higher rates of return to the labor force going forward, this potential development will likely require a much better showing of jobs numbers than were seen today before kicking in.
By Melinda Pitts, research economist and associate policy adviser