The Atlanta Fed's macroblog provides commentary on economic topics including monetary policy, macroeconomic developments, financial issues and Southeast regional trends.

Authors for macroblog are Dave Altig and other Atlanta Fed economists.

September 01, 2015

Should I Stay or Should I Go Now?

A recent article by Jason Faberman and Alejandro Justiniano at the Chicago Fed shows that there is a strong relationship between quit rates—as a proxy for the pace of job switching—and wage growth. Movements in the quit rate and wage growth are both procyclical. A tighter (weaker) labor market implies workers are more (less) likely to find better employment matches, and employers are more (less) willing to offer higher wages to attract new workers and retain existing workers.

To get some idea of the different wage outcomes of job switching versus job staying, we can use microdata underlying the Atlanta Fed's Wage Growth Tracker from the Current Population Survey. The following chart plots the quarterly private-sector quit rate (orange line) from the Job Openings and Labor Turnover Survey using Davis, Faberman, and Haltiwanger (published in 2012 in the Journal of Monetary Economics) estimates before 2001. Also shown is the median year-over-year wage growth of private-sector wage and salary earners who switched jobs (blue line) or stayed in the same job (green line). Job stayers are approximated by the restriction that they are in the same broad industry and occupation as 12 months earlier and have been with the same employer for each of the last four months. Job switchers do not satisfy these restrictions but were employed in the current month and 12 months earlier.

Private Sector Quit Rate and Wage Growth

The correlation between the quit rate and median wage growth is strongly positive and is slightly higher for job switchers (0.91) than for job stayers (0.88). In most periods, the median wage growth of job switchers is higher than for job stayers. This difference is consistent with the notion that job switching tends to involve moving to a better-paying job. However, during periods when the quit rate is slowing, median wage growth slows for both job stayers and switchers (reflecting the correlation between quits and wages), and the wage-growth premium from job switching tends to vanish.

Since the end of the last recession, the quit rate has been rising and a wage-growth premium for job switching has emerged again. Interestingly, during the last year, the wage growth of job stayers appears to have strengthened as well, consistent with a general tightening of the labor market.

September 1, 2015 in Employment, Labor Markets, Wage Growth | Permalink | Comments (0)

August 21, 2015

No Wage Change?

Even when prevailing market wages are lower, businesses can find it difficult to reduce wages for their current employees. This phenomenon, often referred to as "downward nominal wage rigidity," can result in rising average wages for incumbent workers despite high unemployment levels. Some economic models predict that a period of subdued wage growth can follow, even as the labor market recovers—a kind of delayed wage-adjustment effect.

In her 2014 Jackson Hole speech, Fed Chair Janet Yellen suggested this effect may explain sluggish growth in average wages in recent years, despite significant declines in the rate of unemployment.

This macroblog post looks at evidence of wage rigidity, particularly a spike in the frequency of zero wage changes relative to wage declines. A comparison is made between hourly and weekly wages and between incumbent workers (job stayers) and those who have changed employers (job switchers).

Chart 1 shows the fractions of job stayers reporting the same or a lower hourly or weekly wage than 12 months earlier. These measures are constructed from the Current Population Survey microdata in the Atlanta Fed's Wage Growth Tracker. They include workers who are paid hourly (accounting for about 60 percent of all wage and salary earners). The measures exclude those who usually receive overtime and other supplemental pay and those with imputed or top-coded (redacted) wages. Weekly wage is defined as the hourly wage times the usual number of hours per week worked at that rate. The data are aggregated to an annual frequency (except for 2015, where the first six months of the year are covered).

Job stayers cannot be exactly identified in the data and are approximated by those who are in the same occupation and industry as they were 12 months earlier and the same job as they were in the prior month. Consistent with other studies (see, for example, the work of our colleagues at the San Francisco Fed), we find that the incidence of unchanged hourly wages among job stayers is substantial (although some of this is probably the result of rounding errors in self-reported wages). The measured share of unchanged hourly wages rose disproportionately between 2008 and 2010, and it has remained elevated since. Zero hourly wage changes (the green line in chart 1) have become almost as common as declines in hourly wages (the blue line in chart 1).


Chart 1 also suggests that weekly wages for job stayers show a pattern over time broadly similar to hourly wages. But the fraction of unchanged weekly wages (the purple line in chart 1) is lower. Each year, about 60 percent of those with no change in their hourly wage had no change in their weekly wage (or hours) either. Also, there are relatively more declines in weekly wages (the orange line in chart 1) than in hourly wages—mostly the result of reduced hours worked. On average, a reduction in weekly wages is associated with a four-hour decline in hours worked per week. About 90 percent of those with lower hourly wages also had lower weekly wages, and 20 percent of those with no change in their hourly wage had a lower weekly wage (working fewer hours).

If job stayers show a relatively high incidence of no wage change, we might expect a different story for job switchers, since they are establishing a new wage contract with a new employer. Chart 2 shows the fraction of job switchers reporting the same or a lower hourly or weekly wage than 12 months earlier. Job switchers are approximated by workers who are in a different industry than a year earlier.


Not surprisingly, a smaller share of workers experience no change in their hourly or weekly wage when switching jobs. But the pattern of zero wage change for job switchers over time is generally similar to that of job stayers. It is also true that a decline in hourly and weekly wages is more likely for job switchers than for job stayers, with a significant temporary spike in the relative frequency of wage declines for job switchers during the last recession.

Taken at face value, this analysis suggests the presence of some amount of wage rigidity. Also, rigidity increases during recessions and has remained quite elevated since the end of the last recession—especially for job stayers. The question then becomes whether this phenomenon has important macroeconomic consequences. A prediction of most models in which wage stickiness has allocative effects is that it causes firms to increase layoffs when faced with a decline in aggregate demand. Interestingly, during the last recession—when wage stickiness appears to have increased substantially—the rate of layoffs was not unusually high relative to earlier recessions. What was atypical was the size of the decline in the rate of job creation, and this decline contributed to unusually long unemployment spells. As noted by Elsby, Shin, and Solon (2014), it is not clear that an increase in wage rigidity would constrain the hiring of new workers more than it constrains the retention of existing workers.

On the other hand, persistently high wage rigidity in the wake of the Great Recession is consistent with the relatively sluggish pace of wage increases seen in most measures of aggregate wage growth via the "bending" of the short-run Phillips curve (as described by Daly and Hobijn (2014)). Interestingly, the Atlanta Fed's Wage Growth Tracker is an exception. It has indicated somewhat stronger wage growth during the last year than other measures. It will be interesting to see if that trend continues in coming months.

Photo of John Robertson
By John Robertson, a senior policy adviser in the Atlanta Fed's research department

August 21, 2015 in Employment, Labor Markets, Wage Growth | Permalink | Comments (0)

July 17, 2015

Getting to the Core of Goods and Services Prices

In yesterday's macroblog post, I highlighted an aspect of a recent Wall Street Journal article that concerns how households perceive inflation. Today, I'm going back to the same well to comment on another aspect of that story, which correctly notes that service-sector prices are rising at a faster clip than the price of goods.

Of course, this isn't just a recent event. Core services prices have outpaced core goods prices over the past 50 years, save a few short-lived deviations. What's unusual about the current recovery, as the chart below shows, is how low services inflation has been.

Core Goods and Services Prices

In the nearly six years since the end of the 2007–09 recession, core services prices have risen at an annualized pace of 2.1 percent, a full percentage point below their average during the last expansion. Conversely, the annualized growth rate in core goods prices during the recovery has been 0.5 percent, compared to a decline of 0.6 percent during the last expansion (see the chart below).

Core Goods and Services Prices

To see how broad-based the slowdown across core services prices has been relative to that of core goods prices, let's take a deeper dive into the components. The chart below compares the difference between a particular component's annualized growth rate during the current expansion and its growth rate during the previous expansion. A negative number here means that a component's price is growing more slowly now than it did prior to the recession.

It's evident that the slowdown in core services prices is fairly broad-based (17 of 22 components are exhibiting disinflation relative to their growth rate over the previous expansion). For core goods components, that number is just five of 15 components. So, if we accept the premise of the WSJ article—that trends in services prices more closely reflect "unused domestic capacity"—then it's possible we could be farther away than we think.

July 17, 2015 in Inflation | Permalink | Comments (0)

July 16, 2015

Different Strokes for Different Folks

A recent Wall Street Journal article offered an interesting conjecture. The author stated,"[b]ecause consumers pay service bills more often than they buy most goods other than food and gasoline, perceptions of inflation skew on the high side."

Research supports the idea that inflation perceptions are unusually influenced by particular prices. For example, some authors have noted that inflation expectations appear to be unusually influenced by movements in gasoline prices.

This research by Georganas, Healy, and Li shows that inflation perceptions are affected by how frequently people buy a particular good—so that nondurable goods prices like gasoline affect inflation perceptions more than durable goods.

And recent work by Johannsen at the Federal Reserve Board shows that demographic groups who have a more disperse set of inflation experiences also tend to hold more disperse inflation expectations. One thing I think we can say is that different demographic groups appear to have different inflation experiences, as this research by Hobjin, Mayer, Stennis, and Topa indicates.

For example, let's take a look at the difference between the inflation experiences of two households. The first is a single older female (over 55 years of age) who rents her home and has a relatively low income (less than $30,000 a year). The second is a young couple (younger than 35 years old) who own their home and have a high income (over $70,000 annually). Both households have high school educations. Recently, the difference between the inflation experiences of these two demographic groups has opened up to a sizable 2.0 percentage points (see the chart). Why?

Different Inflation Experiences

Well, the spending habits of these two groups contain a few striking differences. For example, the older female spends a lot more of her household income on food at home, rent, and medical care than the young couple does (see the table). Also, the young couple appears to spend a larger fraction of their income on transportation (a large portion of which is gasoline).

Comparison of myCPI Weights

Average of the previous five years (through December 2014)


A young couple, homeowner, high income, high school education

Older female, renter, low income, high school education

Food at home



Food away from home









Household operations



Household furnishings and equipment









Medical care












Note: "Other" includes personal care, alcohol, tobacco, reading, and miscellaneous goods and services
Source: Author's calculations based on the BLS's Consumer Expenditure Survey

What's the inflation experience for someone in your particular demographic group? Let's find out. We've developed a tool called myCPI. It allows users to track a measure of the cost of living that captures some of the variation that occurs between demographic groups. In less than a minute, you can answer a few questions about your demographic category, and we'll show you the cost-of-living trends for "your" group. And if you want, we'll send you updates of your demographic group's inflation with every consumer price index (CPI) report.

Why not get your myCPI report? And when tomorrow's CPI report is released, we'll send you a note telling your how your group's cost-of-living adjustment compares to the average urban consumer in the headline CPI.

July 16, 2015 in Inflation | Permalink | Comments (0)

July 15, 2015

Have Changing Job and Worker Characteristics Restrained Wage Growth?

In the wake of the Great Recession, nominal wage growth has been subdued. But it is unclear how much of this relatively low wage growth reflects protracted weakness in the labor market versus other factors, such as changes in the composition of the workforce and jobs over time. Wage growth tends to vary across personal and job characteristics, so it stands to reason that changes in the composition of the workforce, alongside demographic and work characteristics, could be an important explanation of overall movements in wage growth.

In this post, we explore the impact of the changing mixture of worker characteristics (by age and education) and types of jobs (by industry and occupation) on the Atlanta's Fed Wage Growth Tracker. We find that composition effects do not account for the low median wage growth experienced in recent years. Holding worker and job characteristics fixed at their 1997 shares raises the median wage growth in 2014 by only about 0.2 percentage point. Our results are consistent with the analysis in a previous macroblog post, which found that changing industry-employment shares could not explain much of the sluggish growth in the average hourly earnings data from the payroll survey.

Median wage growth, composition change by worker characteristics
In terms of demographics, we consider two features: a worker's age and education. As shown in this earlier macroblog post, younger workers tend to experience higher median wage growth than do older workers. Although older workers tend to be paid more based on experience, they are also more likely to be near the top of the wage distribution for their job, so the median older worker experiences less wage growth. The difference is quite large. In 2014, the median wage growth of workers over age 54 was around 1.2 percentage points lower than the overall median.

A person's education can also affect his or her wage growth. Workers with a high school diploma or less tend to have lower median wage growth. In 2014, the median wage growth of less-educated workers was about 0.1 percentage point lower than the overall median, reflecting that these workers are more likely to be earning minimum wage, which does not change very frequently.

In addition, the employment shares by age and education have changed over time. The proportion of workers in the Atlanta Fed's Wage Growth Tracker data who are over age 54 has more than doubled from 12 percent in 1997 to 25 percent in 2014. During the same period, the share of workers without a college degree has declined from 63 percent to 49 percent (see the charts).

Education and Age Distribution Over Time

Wage growth, composition change by job characteristics
In terms of job characteristics, we consider two features: the worker's industry (where they work) and their occupation (what they do). Before 2011, workers in service-producing industries experienced slightly higher (about 0.1 percentage point) median wage growth than all workers. But since then, the trends have flipped. In recent years, median wage growth of individuals working in service-producing industries has been slightly below the median wage growth of all workers.

Nonetheless, workers in professional occupations such as managerial, legal, scientific, and engineering jobs tend to experience relatively higher median wage growth. In 2014, the median wage growth of workers in these professional jobs was 0.2 percentage point higher than the median wage growth for all workers.

The share of workers in service-producing industries and in professional jobs has increased moderately over time. In 1997, 77 percent of workers in the data were employed in service-producing industries. In 2014, the share had increased to 82 percent. During the same period, the share of workers in professional occupations rose from 36 percent to 41 percent.

Composition effects on median wage growth
Individually, an aging workforce is putting downward pressure on wage growth, whereas rising education levels are adding upward pressure. The rising share of workers in professional occupations is also pushing wages up somewhat, although the impact of the rising share of workers in service-producing industries is ambiguous. But how large are these effects when combined?

To get an idea, we conducted two counterfactual experiments. First, we held fixed the age and education distributions at their 1997 levels (the first year in our Wage Growth Tracker data). Second, we held fixed the age, education, industry, and occupation characteristics at their 1997 levels. We used three age groups (16–24, 25–54, and 55-plus years of age), two education groups (college degree and no college degree), two industry groups (service- or goods-producing industries), and two occupation groups (professional and other occupations).

The blue line in the next chart is the median wage growth over time with no adjustments for changes in composition. For example, for 2014, the chart shows median wage growth of workers in the data set with earnings in January 2014 and January 2013, February 2014 and February 2013, etc. This depiction is the Atlanta Fed Wage Growth Tracker, but at an annual frequency. The other two lines show the results of the experiment: demographically adjusted (green) and both demographically and job adjusted (orange).

These experiments suggest that—for our data set, at least—the impact on the median of the wage growth distribution from shifts in the composition of the workforce and jobs over time has increased in recent years, but the impact is not especially large. For example, the unadjusted median wage growth for 2014 is 2.5 percent. Holding fixed all four characteristics at their 1997 levels would have raised this by only 0.2 percentage point. Shifting worker and job characteristics are not a primary explanation of low median wage growth since 2009.


July 15, 2015 in Employment, Labor Markets, Wage Growth | Permalink | Comments (0)

July 01, 2015

Far Away Yet Close to Home: Discussing the Global Economy's Effects

In case you needed any motivation to take interest in the outcome of ongoing negotiations between the Greek government and its international creditors, this excerpt from the Wall Street Journal ought to do it:

Global growth is really important. We are all connected through the financial markets, through foreign-exchange markets," Fed governor Jerome Powell said last week in an interview with The Wall Street Journal. "If global growth weakens, or remains weak, and we get into a trend of that, then yes, that will be a big headwind for the United States economy."

Last week, I participated in the latest edition of our webcast, ECONversations, devoted to the theme "what to make of the first quarter?" (The webcast can be found here). The conversation revolved around the Atlanta Fed staff's view of why 2015 began with such a whimper and ideas on prospects for improvement through the balance of the year.

Not surprisingly, the international context loomed large. Between June 2014 and March 2015, the U.S. dollar appreciated by about 14 percent against a broad basket of currencies, and by about 20 percent against major currencies. The dollar has roughly remained in those neighborhoods since. As to the gross domestic product (GDP) side of the story, arithmetically net exports subtracted almost 2 percentage points off first quarter growth.

A key assumption of our current outlook is that the international environment (including the exchange rate) will stabilize, and smoother sailing without the "big headwind" referenced by Governor Powell is ahead.

That assumption generated some discussion (in the Q&A part of the webcast, and via online questions). With some paraphrasing, here are a few of the comments and questions we received, and my best attempt to respond:

Q: You associate the prior appreciation in the dollar with a several percentage point subtraction from growth in the first quarter. This seems quite large in context of available research on the elasticity of the trade balance to movements in the foreign exchange value of the dollar.

A: In the webcast, I did loosely refer to the trade effect on first quarter GDP as a "dollar effect." But the questioner—Barclay's head of U.S. economics research, Michael Gapen— is completely correct in asserting that standard estimates wouldn't support exchange-rate appreciation as an all-encompassing explanation for the big first quarter trade deficit. Our own estimates imply that four quarters after an exchange rate shock that raises the real broad-dollar index by 10 percentage points, real GDP is about one-half a percentage point lower than it would have been without the shock. This impact is roughly the same as most standard estimates (including Barclay's).

Some analyses might imply a larger GDP impact for the pure dollar effect, but any reasonable estimate would leave a fair amount of the first quarter net export decline unexplained. In any event, exchange-rate movements are both cause and effect, which brings us to:

Q: I have a question regarding the impact of the U.S. dollar (USD) in the economy. We often learn that changes in the real exchange rate affect the economy with a lag. Take Japan, for instance. It had a substantial depreciation in Japanese yen (JPY) real exchange rate but with very minimal impact on Japan's trade performance so far. What makes you so confident that the strong USD has had a strong impact in the U.S. economy in such a short period of time? Wouldn't the negative contribution from net exports more likely be linked to delays in West Coast ports and the sharp slowdown in Asian economies (China, in particular)?

A: Yes, in our analysis (and most we know of), the effects of exchange rates occur with a lag. And, as noted above, only a fraction of the decline in net exports by the end of 2014 and into the beginning of this year can be plausibly attributed to dollar appreciation. But we do think those effects are there, and they are continuing (to a lesser extent) in the current quarter.

Of course, changes in the value of the currency are an effect of other developments as well as a cause of changes in exports, GDP, and the like. All else is not typically equal, which often makes simple correlations (or, in the Japanese case, the lack thereof) difficult to interpret.

One of those "not equal" things could well have been the port delays. We don't have a firm estimate of how the backlogs might have affected the first quarter GDP statistic. If the impact was indeed material, we should see some reversal in the second and third quarters now that things are apparently getting back to normal. We'll count that as an upside risk.

And looking forward?

Q: Shouldn't the economic crisis in Greece dampen the demand for American exports and decrease growth well into the fourth quarter?

A: The good news is that current forecasts suggest 2015 euro-area growth will exceed its 2014 pace (according to the World Bank). In fact, the 2015 forecast strengthened over the course of this year despite the ongoing uncertainty associated with the Greek crisis. By most accounts, Canadian economic activity this year is expected to follow a trajectory similar to the United States (in like a lamb, out like something less lambish).

Mexico, as well, is expected to show more growth this year than last, despite some softening of the outlook since the beginning of the year. Put those three together (expanding the euro area to the entire European Union), and you have the anticipation of some improvement in countries accounting for somewhere in the neighborhood of 55 percent of our export markets.

The bad news is the ongoing uncertainty associated with the Greek crisis. Further, the outlook in emerging economies is growing more downbeat. These realities—a continuing impact of prior dollar appreciation and the fact that better foreign growth still does not equate to great growth—has us reluctant to think that net exports will be a big positive number in this year's GDP calculations. That reluctance notwithstanding, for now we are writing in a smaller trade deficit over the course of the year than what we saw in the first quarter.

If you want to go into the July 4 holiday on a somewhat optimistic note, I'll note that our GDPNow estimates for the second quarter have strengthened substantially with the arrival of more recent data—notably including signals of a much lower trade deficit effect than in the first quarter and today's positive news on manufacturing and nonresidential construction. Those data may not be enough to generate full confidence in our forecast for a much better second half of 2015, but they are moving in the right direction.

July 1, 2015 in Economic Growth and Development, GDP | Permalink | Comments (0)

June 29, 2015

New Atlanta Fed Series Shows Wage Growth Held Steady in May

According to the Atlanta Fed's Wage Growth Tracker, a new series constructed using data from the Current Population Survey, the median increase in wages for individuals working in May 2014 and May 2015 was 3.3 percent (reported as a three-month average).

Wage growth by this measure was essentially unchanged from April and 1 percentage point higher than the year-ago reading. The current pace of nominal hourly wage growth is similar to that seen during the labor market recovery of 2003–04 and about a percentage point below the pace experienced during 2006–07, which was the peak of the last business cycle. You can download the data going back to March 1997 from our website by clicking "export," shown in the upper right of the chart below.

Wage growth differs by job and worker characteristics. For prime-age individuals and full-time employees, for example, the Wage Growth Tracker recorded slightly higher readings than the group overall. The median wage growth of these individuals was 3.5 percent compared to 3.3 percent for all individuals. To see more cuts of the data, check out our website.

June 29, 2015 in Economic conditions, Wage Growth | Permalink | Comments (1)

June 23, 2015

Approaching the Promised Land? Yes and No

Last Friday, we released our June installment of the Business Inflation Expectations (BIE) survey. Among the questions we put to our panel of businesses was a quarterly question on slack, asking firms to consider how their current sales levels compare to what they would consider normal.

The good news is, on average, the gap between firms' current unit sales levels and what they would consider normal sales levels continues to close (see the chart).


By our measure, firm sales, in the aggregate, are 1.9 percentage points below normal, a bit better than when we polled them in March (when they were 2.1 percent below normal) and much improved from this time last year (3.7 percent below normal). For comparison, the Congressional Budget Office's (CBO) estimate of slack on a real gross domestic product (GDP) basis was 2.6 percent in the first quarter (though this estimate will almost certainly be revised to something closer to 2.4 percent when the revised GDP estimates are reported later today). And if GDP growth this quarter comes in around 2.5 percent as economists generally expect, the CBO's GDP-based slack estimate will be 2.2 percent this quarter, just a shade larger than what our June survey data are saying.

Now, as we have emphasized frequently (for example, in macroblog posts in May 2015, February 2015, and June 2013), performance in the aggregate and performance within select firm groups can differ widely. For example, while small firms continue to have greater slack than larger firms, their pace of improvement has been much more rapid (see the table).


Likewise, some industries (such as transportation and finance) see current sales as better than normal. But others, like manufacturers, are currently reporting considerable slack—and findings from this group appear to show a marginal worsening in sales levels over the past 12 months.

Another item that caught our attention this month was the differing pace of narrowing in the sales gap among those firms with significant export exposure (greater than 20 percent of sales) relative to those with no direct export exposure. We connected these dots using responses to this month's special question, in which responding firms specified their share of customers by geographic area: local, regional (the Southeast, in our case), national, and international (see the table).


So things are still getting better for the economy overall, and the small firms in our panel have displayed particularly rapid improvement during the last year. But if you've got exposure to the "soft" export markets, as mentioned in the June 17 FOMC statement, you've likely experienced a slower pace of improvement.

Photo of Mike Bryan
By Mike Bryan, vice president and senior economist,
Photo of Brent Meyer
Brent Meyer, economist, and
Photo of Nicholas Parker
Nicholas Parker, economic policy specialist, all in the Atlanta Fed's research department

June 23, 2015 in Business Cycles, Business Inflation Expectations, Economic conditions | Permalink | Comments (0)

June 19, 2015

Will the Elevated Share of Part-Time Workers Last?

There seems to be mounting evidence that at least part of the elevated share of part-time employment in the economy is here to stay. We have some insights to offer based on a recent survey of our business contacts.

Why are we interested? A higher part-time share of employment isn't necessarily a bad thing, if people are doing so voluntarily. Unfortunately, the elevated share is concentrated among people who would prefer to be working full-time. Using the average rate of decline over the past five years, the part-time for economic reasons (PTER) share of employment is projected to reach its prerecession average in about 10 years.

This is significantly slower than the decline in the unemployment rate, whose trajectory suggests a much sooner arrival—in around a year. The deviation raises an important policy question for measuring the amount of slack there is beyond what the unemployment rate suggests, and ultimately the extent to which policy can effectively reduce it.

What are the drivers? Data versus anecdotes
Researchers (here, here, and here) have pointed to factors such as industry shifts in the economy, changing workforce demographics, rising health care costs, and the Affordable Care Act as potentially important drivers of this shift. But we can glean only so much information from data. When a gap develops, we generally turn to our business contacts who are participating members in our Regional Economic Information Network (REIN) to fill in the missing information.

According to our contacts, the relative cost of full-time employees remains the most important reason for having a higher share of part-time employees than before the recession, which is the same response we received in last summer's survey on the same topic. Lack of strong enough sales growth to justify conversion of part-time to full-time workers came in as a close second.

The importance rating for each of the factors was notably similar to last year's survey, with one exception. Technology was rated as somewhat important, reflecting an uptick from the average response we received last year. We've certainly heard anecdotally that scheduling software has enabled firms to better manage their part-time staff, and it seems that this factor has gained in importance over the past year.

The chart below summarizes the reasons our business contacts gave in the July 2014 and the May 2015 surveys. The question was asked only of those who currently have a higher share of part-time workers than they did before the recession. The chart shows the results for all respondents, whether they responded to one or both surveys. When we limited our analysis to only those who responded to both surveys, the results were the same.

Will the elevated share persist?
The results suggest that a return to prerecession levels is unlikely to occur in the near term.

The chart below shows employers' predictions for part-time employment at their firms, relative to before the recession. About 27 percent of respondents believe that in two years, their firms will be more reliant on part-time work compared to before the recession. About 7 percent do not currently have an elevated share of part-time employees but believe they will in two years. About two-thirds believe their share of part-time will be roughly the same as before, while only 8 percent believe they will have less reliance on part-time workers compared to before the recession.

The majority of our contacts believe their share of part-time employment will normalize over the next two years, but some believe it will stay elevated. Still, 2017 does not mean the shift will be permanent. In fact, firms cited a balance of cyclical and structural factors for the higher reliance on part-time. Low sales growth and an ample supply of workers willing to take part-time jobs could both be viewed as cyclical factors that will dissipate as the economy further improves.

Meanwhile, higher compensation costs of full-time relative to part-time employees and the role of technology that enables companies to more easily manage their workforce can be considered structural factors influencing the behavior of firms. Firms that currently have a higher share of part-time employees gave about equal weight to these forces, suggesting that, as other research has found, both cyclical and structural factors are important explanations for the slow decline in the part-time share of employment.

June 19, 2015 in Business Cycles, Employment, Labor Markets, Unemployment | Permalink | Comments (1)

June 08, 2015

Falling Job Tenure: It's Not Just about Millennials

The image of a worker in the 1950s is one of a man (for the most part) who plans on spending his entire career with one employer. We hear today, however, that "...long gone is the lifelong loyalty to a corporation with steadfast servitude for years on end." One report tells us that "people entering the workforce within the past few years may have more than 10 different jobs before they retire." The reason? "Millennials don't like commitments." Well, the explanation is probably not that simple, but even simply measuring trends in job tenure is also not all that straightforward.

Despite a strong impression that entire careers spent with one employer are a thing of the past, some have declared the image of job-hopping millennials a myth. (You can read some discussions at, CNBC, and Marketwatch, for example.) These reports are all based on a September 2014 news release from the U.S. Bureau of Labor Statistics (BLS) stating that among every employee age group (even the youngest), median job tenure has not declined from when it was reported 10 years earlier. (Median job tenure is basically the "middle" amount of job tenure. If all workers are lined up from lowest tenure to highest tenure, the median tenure would be the amount of time the person in the middle of that line has been with his/her employer.)

Chart 1 illustrates the biennial data on job tenure reported by the BLS and interpreted by the reports mentioned above as indication that job tenure is not falling. Each line represents an age range, from 20- to 30-year-olds at the bottom (the lowest median tenure among all age groups) to 61- to 70-year-olds on the top (the age group with the highest median tenure). It sure doesn't look as though workers at each age group are staying with their jobs for shorter periods.

However, the problem with simply comparing median tenure across time by age group is that different ages at different time periods face different labor market institutions, incentives, and expectations. There are generational, or cohort, differences in what the labor market looks like and has to offer a 25-year-old born in 1923 and a 25-year-old born in 1993. In other words, each generation is represented across the age groups at different points in time.

The different colored points across age groups in chart 1 indicate the range of years the people in that particular year, in that age group, were born (and to what named generation they belong). The labor market facing a 31-to 40-year-old baby boomer in 1996 looks quite different from the labor market facing a 31-to-40-year-old Gen Xer in 2012, and the social, economic, and behavioral differences are even more dramatic the farther apart the generations become.

For example, one of the most dramatic changes facing workers has been the transformation from defined-benefit to defined-contribution retirement plans. The number of years a worker spends with an employer is no longer an investment in the employee's retirement. (William Even and David Macpherson (1996) illustrated the important link between the presence of an employer-sponsored retirement plan and worker tenure in their paper "Employer Size and Labor Turnover: The Role of Pensions.")

Additionally, the share of those 25 and over with a college degree in the United States has increased from 5 percent in 1950 to 32 percent in 2014, according to data from the U.S. Census Bureau. A more educated workforce is one with more general, or transferable, human capital, reducing the need to stay with just one employer to reap a return on one's investment in human capital. The transition of the U.S. economy from a basis in manufacturing to one based in services, supported by technology, also means employers require more general, rather than specific, human capital.

Firms have also changed the way they invest in workers, offering less on-the-job training than they used to, weakening their ties to the worker. And on top of all of this, because of near-instantaneous access to information, movies, and music brought by the digital age, younger cohorts are purported to have shorter attention spans than older cohorts (as reported here). All these factors shape the environment in which workers and employers view the value of longevity in their relationship.

To get a more accurate picture of the lifetime pattern of median job tenure and how it has changed across generations, we use the same BLS data used to produce the chart above to group workers into cohorts, or people who have similar experiences by virtue of when they were born. In other words, we rearrange the data used in chart 1 to line people up by birth year rather than by calendar year in order to illustrate (in chart 2) that median job tenure is indeed declining through the generations.

What we see in this chart—using the 20- to 30-year-olds, for example—is that the median job tenure was four years among those born in 1953 (baby boomers) when they were between 20 and 30 years old. For 20- to 30-year-olds born in 1993 (millennials), however, median job tenure is only one year. Similar—and some even more dramatic—declines occur across cohorts within each age group.

Declining job tenure is not just all about millennials having short attention spans. In fact, there is a greater (five-year) decline in median job tenure between 41- and 50-year-old "Depression babies" (born in 1933) and 41- to 50-year-old Gen Xers (born in 1973). So, just as our colleagues here at the Atlanta Fed discovered with regard to declines in first-time home mortgages, millennials aren't to blame for everything!

So what does declining job tenure mean for the U.S. labor market? From the perspective of the worker, portable retirement savings and, now, portable health insurance mean that workers confront a world of possibilities that our parents and grandparents never dreamt of. Yes, perhaps the days of predictability in one's career is a thing of the past. But so is the "eggs-in-one-basket" loss of retirement savings when your employer goes out of business as well as potentially slower career progression within a single firm.

From the economy's perspective, the flexibility of workers seeking their highest rents and the flexibility of firms to seek better matches for their needed skills mean greater productivity—not to mention growth—all around.

Photo of Julie Hotchkiss
By Julie L. Hotchkiss, research economist and senior policy adviser, and
Photo of Christopher MacPherson
Christopher J. Macpherson, an intern, both in the Atlanta Fed's research department

June 8, 2015 in Employment | Permalink | Comments (3)

Google Search

Recent Posts

October 2015

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31



Powered by TypePad