The Atlanta Fed's macroblog provides commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues.
- BLS Handbook of Methods
- Bureau of Economic Analysis
- Bureau of Labor Statistics
- Congressional Budget Office
- Economic Data - FRED® II, St. Louis Fed
- Office of Management and Budget
- Statistics: Releases and Historical Data, Board of Governors
- U.S. Census Bureau Economic Programs
- White House Economic Statistics Briefing Room
September 27, 2016
Back to the '80s, Courtesy of the Wage Growth Tracker
Things have been a wee bit quiet in macroblog land the last few weeks, chiefly because our time has been devoted to two exciting new projects. The first is a refresh of our labor force dynamics website, which will feature a nifty tool for looking at the main reasons behind changes in labor force participation for different age groups. More on that later.
The other project has been adding more history to our Wage Growth Tracker. The tracker's current time series starts in 1997. The chart below shows an extended version of the tracker that starts in 1983.
Recall that the Wage Growth Tracker depicts the median of the distribution of 12-month changes of matched nominal hourly earnings. In the extended time series, you'll notice two gaps, which resulted from the U.S. Census Bureau scrambling the identifiers in its Current Population Survey. For those two periods, you'll have to use your imagination and make some inferences.
As we have emphasized previously, the Wage Growth Tracker is not a direct measure of the typical change in overall wage costs because it only looks at (more or less) continuously employed workers. But it should reflect the amount of excess slack in the labor market. This point is illustrated in the following chart, which compares the Wage Growth Tracker with the unemployment gap computed from the Congressional Budget Office's (CBO) estimate of the long-run natural rate of unemployment.
As the chart shows, our measure of nominal wage growth has historically tracked the cyclical movement in the unemployment rate gap estimate fairly well, at least since the mid-1980s. We think this feature is potentially important, because the true unemployment rate gap is very hard to know in real time and hence is subject to potentially large revision. For example, in real time, the unemployment rate was estimated to have fallen below the natural rate in the fourth quarter of 1994, but it is now thought to have not breached the natural rate until the first quarter of 1997—more than two years later. The Wage Growth Tracker is not subject to revision (although it is subject to a small amount of sampling uncertainty) and hence could be useful in evaluating the reliability of the unemployment rate gap estimate in real time.
This also is important from a monetary policy perspective if we are worried about the risk of the economy overheating. For example, President Rosengren of the Boston Fed described why he dissented at the most recent Federal Open Market Committee meeting in favor of a quarter-point increase in the target range for the federal funds rate. His dissent, he said, arose partly from his concern that the economy may overheat and drive unemployment below a level he believes is sustainable.
Currently, the CBO estimate of the unemployment rate gap looks like it is plateauing at close to zero. The fact that the Wage Growth Tracker for the third quarter slowed a bit is consistent with that. But it's only one quarter of data, and so we'll closely monitor the Wage Growth Tracker in the coming months to see what it suggests about the actual unemployment rate gap. We'll discuss what observations we make here.
August 15, 2016
Payroll Employment Growth: Strong Enough?
The U.S. Bureau of Labor Statistics' estimate of nonfarm payroll employment is the most closely watched indicator of overall employment growth in the U.S. economy. By this measure, employment increased by 255,000 in July, well above the three-month average of 190,000. Yet despite this outsized gain, the unemployment rate barely budged. What gives?
Well, for a start, there is no formal connection between the payroll employment data and the unemployment rate data. The employment data used to construct the unemployment rate come from the Current Population Survey (CPS) and the payroll employment data come from a different survey. However, it is possible to relate changes in the unemployment rate to the gap between the CPS and payroll measures of employment, as well as changes in the labor force participation (LFP) rate, and the growth of payroll employment relative to the population.
The following chart shows the contribution of each of these three factors to the monthly change in the unemployment rate during the last year.
A note about the chart: The CPS employment and population measures have been smoothed to account for annual population control adjustments. The smoothed employment data are available here. The method used to compute the contributions is available here.
The black line is the monthly change in the unemployment rate (unrounded). Each green segment of a bar is the change in the unemployment rate coming from the gap between population growth and payroll employment growth. Because payroll employment has generally been growing faster than the population, it has helped make the unemployment rate lower than it otherwise would have been.
But as the chart makes clear, the other two factors can also exert a significant influence on the direction of the unemployment rate. The labor force participation rate contribution (the red segments of the bars) and the contribution from the gap between the CPS and payroll employment measures (blue segments) can vary a lot from month to month, and these factors can swamp the payroll employment growth contribution.
So any assumption that strong payroll employment gains in any particular month will automatically lead to a decline in the unemployment rate could, in fact, be wrong. But over longer periods, the mapping is a bit clearer because it is effectively smoothing the month-to-month variation in the three factors. For example, the following chart shows the contribution of the three factors to 12-month changes in the unemployment rate from July 2012 to July 2013, from July 2013 to July 2014, and so on.
Gains in payroll employment relative to the population have helped pull the unemployment rate lower. Moreover, prior to the most recent 12 months, declines in the LFP rate put further downward pressure on the unemployment rate. Offsetting this pressure to varying degrees has been the fact that the CPS measure of employment has tended to increase more slowly than the payroll measure, making the decline in the unemployment rate smaller than it would have been otherwise. During the last 12 months, the LFP rate turned positive on balance, meaning that the magnitude of the unemployment rate decline has been considerably less than implied by the relative strength of payroll employment growth.
Going forward, another strong payroll employment reading for August is certainly no guarantee of a corresponding decline in the unemployment rate. But as shown by my colleagues David Altig and Patrick Higgins in an earlier macroblog post, under a reasonable range of assumptions for the trend path of population growth, the LFP rate, and the gap between the CPS and payroll survey measures of employment, payroll growth averaging above 150,000 a month should be enough to cause the unemployment rate to continue declining.
July 29, 2016
Men at Work: Are We Seeing a Turnaround in Male Labor Force Participation?
A lot has been written about the long-run decline in the labor force participation (LFP) rate among prime-age men (usually defined as men between 25 and 54 years of age). For example, see here, here, here, and here for some perspectives.
On a not seasonally adjusted basis, the Bureau of Labor Statistics estimates that the LFP rate among prime-age males is down from 90.9 percent in the second quarter of 2007 to 88.6 percent in the second quarter of 2016—a decline of 2.3 percentage points, or around 1.4 million potential workers.
Many explanations reflecting preexisting structural trends have been posited for this decline. But how much of the decline also reflects cyclical effects and, in particular, cyclical effects that take a while to play out? We don't really know for sure. But one potentially useful approach is to look at the Census Bureau's Current Population Survey and the reasons people give for not wanting a job. These reasons include enrollment in an educational program (especially prevalent among young individuals), family or household responsibilities (especially among prime-age women), retirement (especially among older individuals), and poor health or disability (widespread). In addition, there are people of all ages who say they want a job but are not counted as unemployed. For example, they aren't currently available to work or haven't looked for work recently because they are discouraged about their job prospects.
To get some idea of the relative importance of these factors, the following chart shows how much each nonparticipation reason accounted for the total change in the LFP rate among prime-age males between 2012 and 2014 and between 2014 and 2016. The black bars show each period's total change in the LFP rate. The green bars are changes that helped push participation higher than it otherwise would have been, and the orange bars are changes that helped hold participation lower than it otherwise would have been.
A note on the chart: To construct the contributions derived from changes in nonparticipation rates, I held constant the age-specific population shares in the base period (2012 and 2014, respectively) in order to separate the effect of changes in nonparticipation from shifts in the age distribution.
Notice that the decline in the prime-age male LFP rate between 2012 and 2014 has essentially fully reversed itself over the last two years (from a decline of 0.53 percentage points to an increase of 0.55 percentage points, respectively). The positive "want a job" contribution in both periods clearly reflects a cyclical recovery in labor market conditions. But the most striking change between 2012–14 and 2014–16 is the complete reversal of the large drag attributable to poor health and disability. Other things equal, if nonparticipation resulting from poor health and disability had stayed at its 2012 level, prime-age male participation in 2014 would have only declined 0.10 percentage points. If nonparticipation due to poor health and disability had stayed at its 2014 level, prime-age male participation in 2016 would have increased only 0.14 percentage points.
The incidence of self-reported nonparticipation among prime-age men because of poor health or disability has been declining recently. According to the Current Population Survey data, this reason represented 5.4 percent of the prime-age male population in the second quarter of 2016. Although this is still 0.7 percentage points higher than in 2007, it is 0.3 percentage points lower than in 2014. Some of this turnaround could be the result of changes in the composition of the prime-age population. But not much. Around 90 percent of the LFP rate change because of poor health and disability is due to age-specific nonparticipation rather than shifts in the age distribution, suggesting that some of the turnaround in the incidence of people saying they are "too sick" to work is a cyclical response to strengthening labor market conditions. We've yet to see how much longer this turnaround could continue, but it's an encouraging development.
For those interested in exploring the contributions to the changes in the LFP rate by gender and age over different time periods, we're currently developing an interactive tool for the Atlanta Fed's website—stay tuned!
July 07, 2016
Is the Labor Market Tossing a Fair Coin?
How important is tomorrow's June employment report? In isolation, the answer would surely be not much. The month-to-month swings in job gains can be quite large, and one month does not a trend make.
And yet, there seemed to be a pretty significant reaction to the May employment number, a reaction that did not escape the attention of MarketWatch's Caroline Baum:
So yes, the Fed does seem to be altering its macro view on potential growth (slower) and the neutral funds rate (close to zero) as a hangover from the Great Recession becomes an increasingly inadequate explanation for persistent 2% growth.
What comes across to the observer is a bad case of one-number-itis. The monthly jobs report does contain a lot of important information, including hiring, wages and a proxy for output (aggregate hours index). But the Fed talks out of both sides of its mouth, cautioning against putting too much weight on a single economic report, and then doing just that.
I get it. I don't speak for the Fed, of course—above my rank—but I am in fact one of those who regularly cautions against putting excessive weight on one number. And I am also one of those taken aback by the May employment report, so much so that my view of the economy changed materially as a result of that report.
Let me check that. My view of the risks to the economy, or more specifically the risks to my assessment of the strength of the economy, changed materially.
Here's an analogy that I find useful. Flip what you assume to be a fair coin. The probability of getting a heads, as we all know, is 50 percent. And if you weren't too traumatized by the statistics courses in your past, you will recall that the probability of two heads in a row is 25 percent, dropping to just about 13 percent of the coin coming up heads three times in a row.
Now, 13 percent is not zero, but it may be getting low enough for you to begin to wonder about your assumption that the coin is actually fair. If you have some stake in whether it is or isn't, you might want to take one more toss to get a little more evidence (since the odds of getting four heads in a row is, while not impossible, pretty improbable).
The point is that it wasn't just the May statistic that was striking in last month's report, but also the fact that the March and April numbers were revised downward to the tune of nearly 60,000 jobs. And if you step back a bit, you will see that the rolling three-month average of monthly job gains has been declining through the first half of the year (as the chart shows), even adjusting the May number for the Verizon strike:
Strike-adjusted, the May job gains were the lowest since December 2013. The three-month average (again strike-adjusted) was the lowest since the middle of 2012. In other words, although the year-over-year pace of jobs gains has been holding up, momentum in the labor market is decidedly softer—at least when measured by payroll employment gains.
I have been assuming that the U.S. economy will, for a while yet, continue to create jobs at a pace greater than necessary to maintain the unemployment rate at a more or less constant level. That pace is generally believed to be about 80,000 to 140,000 jobs per month, depending on your assumptions about the labor force participation rate. Another jobs report (including revisions to past months) that counters that assumption would, I think, cause a reasonable person to reassess his or her position.
Based on today's ADP report, the odds look good for some decent news tomorrow. On the other hand, if the June employment number does tick up, some observers will no doubt note that it is a pre-Brexit statistic. It may take a few more flips of the coin to determine if that consideration matters.
July 06, 2016
When It Rains, It Pours
Seasonally adjusted nonfarm payroll employment increased by only 38,000 jobs in May, according to the initial reading by the U.S. Bureau of Labor Statistics (BLS), and the total increase for the prior two months was revised down by a cumulative 59,000. Although the May increase was depressed by 35,100 striking workers at Verizon Communications, observers widely anticipated this distortion (the strike started April 13). Nonetheless, the median forecast of the May payroll gain from a Bloomberg survey of economists was 160,000, still well above the official estimate. The disappointing employment gain in May, I believe, is statistically related to the downward revisions to the seasonally adjusted gains made over the prior two months.
In contrast to the revision to the seasonally adjusted data, the nonseasonally adjusted level of payroll employment in April was only revised down by 3,000 in the May report. So most of the downward revision to the seasonally adjusted March and April employment gains was the result of revised seasonal factors (the difference between 59,000 and 3,000). In the chart below, the green diamond (toward the left) is the downward revision of 56,000 that resulted from the revised seasonal factors plotted against the Bloomberg survey forecast error for May (the difference between the actual estimate of 38,000 and the forecast of 160,000). The other diamonds represent corresponding points for reports from January 2006 through April 2016. The data points indicate a clear positive relationship and—based on the May Bloomberg forecast error—a simple linear regression would have almost exactly predicted the total downward revision to the March and April employment gains coming from revised seasonal factors.
To gain some insight into the positive relationship in the above chart, I used a model to seasonally adjust the last 10 years of nonfarm payroll employment data (excluding decennial census workers). Note that although I followed the BLS's procedure of accounting for whether there are four or five weeks between consecutive payroll surveys, I did not seasonally adjust the detailed industry employment data and sum them up, as the BLS does.
According to my seasonal adjustment model, the seasonally adjusted April employment level using data from the May employment report is 60,000 below the seasonally adjusted April employment level estimated with data from the April report. My seasonal adjustment model only using data through April from the May report predicts a nonseasonally adjusted increase of 789,000 jobs in May instead of the BLS's estimated increase of 651,000 jobs. The difference between these two estimates is similar to the Bloomberg survey forecast error noted above.
Further, when I replace the BLS's nonseasonally adjusted estimate for May with the model's forecast, the estimate of seasonally adjusted April employment is only 2,000 less than the model estimated with data from the April employment report. Hence, almost all of the model's downward revision to seasonally adjusted April employment appears to be the result of adding fewer jobs in May than the model expected.
The above analysis illustrates that, when it comes to looking at seasonally adjusted employment data, the number of jobs next month will affect the estimate of the number of jobs this month. This is not a very appealing notion, but when using seasonally adjusted data, it comes with the territory. Fortunately, analyzing the nonseasonally adjusted data allows us to gauge the impact of a surprise in the current estimate of seasonally adjusted employment growth on revisions to the prior two months. So when the June report is released on Friday, we will be paying close attention to both the seasonally adjusted headline numbers as well as the revisions to the nonadjusted data.
June 22, 2016
Was May's Drop in Labor Force Participation All Bad News?
The unemployment rate declined 0.3 percentage points from April to May, and this was accompanied by a similar drop in the labor force participation rate. It is tempting to interpret this as a “bad” outcome reflecting a weakening labor market. In particular, discouraged about their job-finding prospects, more unemployed workers left the labor force. However, a closer look at the ins and outs of the labor force suggests a possibly less troubling interpretation of the outflow from unemployment.
To get a handle on what is going on, it is useful to look at the number of people that transition among employment, unemployment, and out of the labor force. It is not that unusual for an individual to search for a job in one month and then enroll in school or assume family responsibilities the next. In fact, each month millions of individuals go from searching for work to landing a job or leaving the labor force, and vice versa.
The U.S. Bureau of Labor Statistics (BLS) publishes estimates of these gross flows. Analyzing these data shows that there was indeed an unusually large number of unemployed persons leaving the labor force in May. Curiously, the outflow was concentrated among people who had only been unemployed only a few weeks. It wasn't among the long-term unemployed. Therefore, it seems unlikely that discouragement over job-finding prospects was the main factor. Although it is plausible that people who say they are now doing something else outside the labor market feel disheartened, the number of unemployed who said they gave up looking because they were discouraged was largely unchanged in May.
So why was there an increase in the number of short-term unemployed who left the labor force in May? One clue is provided by the fact that the short-term unemployed tend to be relatively younger than other unemployed. Moreover, the single most common reason that unemployed young people leave the labor force is to go to school. Hence, there is a very distinct seasonal pattern in the outflow. It tends to be relatively low around May when school is ending and high around August when school is starting. Seasonal adjustment techniques correct for these patterns by lowering the unadjusted data in the fall and raising it in late spring.
The following chart shows the seasonally adjusted and unadjusted flow from unemployment to departure from the labor force. Although the trend has been declining during the last few years, a relatively large increase in the seasonally adjusted outflow took place in May of this year.
When I looked at the unadjusted microdata from the Current Population Survey (CPS), I found that the number of people who were unemployed in April 2016 but in May said that they were not in the labor force because they were in school did not exhibit the usual large seasonal decline. Therefore, when the seasonal adjustment is applied, the result is an increase in the estimated flow from unemployment to out of the labor force.
Taking the seasonally adjusted data at face value, it's not obvious that this is bad news. We know that people who leave unemployment to undertake further education tend to rejoin the labor force later. Moreover, they tend to rejoin with better job-finding prospects than when they left. Alternatively, it could be just a statistical quirk of the May survey. After all, the CPS has a relatively small sample, so the estimated flows have a large amount of sampling error. Either way, I don't think it is wise to conclude that the decline in the labor force participation in May reflected a marked deterioration in job-finding prospects. In fact, the job-finding rate among unemployed workers improved in May from 22 to 24 percent, contributing to the decline in the unemployment rate.
June 21, 2016
Wage Growth for Job Stayers and Switchers Added to the Atlanta Fed's Wage Growth Tracker
The Atlanta Fed's Wage Growth Tracker (WGT) moved higher again in May—the third increase in a row and consistent with a labor market that is continuing to tighten. At 3.5 percent, the WGT is at a level last seen in early 2009.
As was noted in an early macroblog post, when the labor market is tightening, people changing jobs experience higher median wage growth than those who remain in the same job. Median wage growth for job switchers has significantly outpaced that of job stayers in recent months. For job stayers, the May WGT was 3.0 percent, the same as in April, whereas for people switching jobs the median WGT increased from 4.1 percent to 4.3 percent in May (the highest reading since December 2007; see the chart).
Because these patterns over time can help shed light on the relative strength of the labor market, we have added downloadable job stayer and job switcher WGT series to the Atlanta Fed's Wage Growth Tracker web page.
I should note that it is not possible to completely identify people who are in the same job as a year ago according to data from the Current Population Survey. Instead, we define a "job stayer" as someone whom we observe in the same occupation and industry as a year earlier, and with the same employer in each of the last three months. A "job switcher" includes everyone else (a different occupation or industry or employer). We'll be monitoring these data in coming months to see if discernable trends begin to emerge, and we'll discuss any findings here.
June 16, 2016
Experts Debate Policy Options for China's Transition
After nearly three decades of rapid economic growth, China today faces the challenge of economic rebalancing against the backdrop of slow and uncertain global growth. Although investment and exports have been a motor for growth, China is increasingly experiencing structural issues: widening inequality, overcapacity as a consequence of policy distortions, unsustainable environmental costs, volatile financial markets, and rising systemic risk.
On April 28–29, I attended the First Research Workshop on China's Economy, organized jointly by the International Monetary Fund (IMF) and the Atlanta Fed. The workshop, held at the IMF's headquarters in Washington DC, explored a series of questions that have emerged as China shifts toward a new growth model. Is this the end of the growth miracle? Will the Chinese renminbi one day be as important as the U.S. dollar? Should the rapidly increasing shadow banking activity in China be a source of concern? How worrisome is the rapid rise in China's housing prices?
Panelists shared their views on these and other issues facing the world's second-largest economy (or largest, if measured on a purchasing-power-parity basis). Plans are under way for a second workshop to be held in 2017.
The following is a nice summary of the research discussed at the workshop. It was originally published in the IMF Survey Magazine, and was written by Hui He, IMF Institute for Capacity Development, and Nan Li, IMF Research Department. Thanks to the IMF for allowing me to repost it here.
Is China's economic growth sustainable?
Understanding the source of China's tremendous growth was a recurring theme at the workshop. "China's economy combines enormous dynamism with huge distortions," observed Loren Brandt (University of Toronto). Brandt described his research based on China's firm-level data and emphasized that firm dynamics (entry and exit), especially firm entry, have been the main source of the productivity growth in the manufacturing sector.
Echoing Brandt's message, Kjetil Storesletten (University of Oslo) discussed regional growth disparities and showed that barriers preventing firms from entering an industry account for most of the disparities. Such barriers are more severe for privately owned firms in regions in which state-owned enterprises (SOE) dominate, he said.
In his keynote speech, Nicholas Lardy (Peterson Institute for International Economics) offered an upbeat view on China's transition to a new growth model, one in which the service sector plays a larger role than manufacturing. The bright side of the service sector, he noted, is its continued strong productivity growth. The development of financial deepening and the stronger social safety net are contributing to increased consumption, which helps to rebalance the economy.
However, he emphasized, SOE reforms remain critical as the service sector cannot provide a silver bullet for a successful transition.
Central bank's policy decisions
Several participants tried to discern how the People's Bank of China (PBC) conducts monetary policy. Tao Zha (of the Atlanta Fed's Center for Quantitative Economic Research and Emory University) found that the PBC reacts sharply when the gross domestic product's growth rate falls below its target, increasing the money supply by 11.5 percentage points for every 1 percentage point shortfall.
Mark Spiegel (Center for Pacific Basin Studies) discussed the trade-offs involved in Chinese monetary policy—for example, controlling the exchange rate versus maintaining inflation stability. He also argued that the heavy use of reserve requirements on banks as a monetary policy tool might have an unintentional consequence to reallocate capital from SOEs to more efficient privately owned firms and could therefore offset the resource misallocation caused by the easy credit to SOEs that banks granted in the high growth years.
Renminbi versus the dollar
Eswar Prasad (Cornell University and Brookings Institution) argued that China's capital account will become more open and the renminbi will be used more widely to denominate and settle cross-border transactions. But he also noted that legal and institutional constraints in China were likely to prevent the renminbi from serving as a safe-haven currency as the U.S. dollar does today.
Moreover, he said, the current sequencing of liberalization initiatives—that is, removal of capital account restrictions before appropriate financial market supervision and regulation and exchange rate reform—poses financial stability risks.
Shadow banking and the housing market
Recently, volatile Chinese financial markets and continued housing price appreciation have raised serious financial stability concerns.
Michael Song (Chinese University of Hong Kong) argued that rapidly rising shadow banking activity is an unintended consequence of financial regulation. Restrictions on deposit rates and loan-to-deposit ratios have led to the issuance by banks of "wealth management products" to attract savers with higher returns. Because these restrictions had a greater impact on small banks, the big state banks had more room to undercut the smaller banks by offering wealth management products with higher returns and then restricting liquidity to them in interbank markets, ultimately making the banking system more prone to liquidity distress and runs.
Hanming Fang (University of Pennsylvania) found that, except in big cities such as Beijing and Shanghai, housing prices in China's urban areas between 2003 and 2013 more or less tracked rising household incomes. In his view, the Chinese housing boom is thus unlikely to trigger an imminent financial crisis. He warned, however, that housing prices may fall rapidly if economic growth slows dramatically, and that such a development could, in turn, amplify the economic downturn.
Rising wage inequality
China's rapid growth over the past two decades has been accompanied by rising wage inequality, an issue highlighted by two conference participants. Dennis Yang (University of Virginia) explored the distributional effects of trade openness in China and found a significant impact on wage inequality of China's accession to the World Trade Organization in 2001.
Chong-En Bai (Tsinghua University) argued that the decline after 2008 of the skill premium—that is, the ratio of the skilled labor wage to the unskilled labor wage—can be explained by the Chinese government's targeted credit extension to unskilled labor-intensive infrastructure sector (as part of the fiscal stimulus following the global financial crisis). Such distortionary policies might have short-run growth benefits but could lead to long-run welfare losses, he said, especially when rural-to-urban migration has run its course.
June 02, 2016
Moving On Up
People who move from one job to another tend to experience greater proportionate wage gains than those who stay in their job, except when the labor market is weak and there are relatively few employment options. This point was illustrated using the Atlanta Fed's Wage Growth Tracker in this macroblog post from last year.Given that the Wage Growth Tracker ticked higher in April, it is interesting to see how much of that increase can be attributed to job switching. Here's what I found:
A note about the chart: In the chart, a "job stayer" is defined as someone who is in the same occupation and industry as he or she was 12 months ago and has been with the same employer for at least the last three months. A "job switcher" is everyone else.
The overall Wage Growth Tracker for April was 3.4 percent (up from 3.2 percent in March). For job stayers, the Tracker was 3.0 percent (up from 2.9 percent), and for job switchers it was 3.9 percent (up from 3.7 percent). So the wage gains of job switchers do appear to have helped pull up our overall wage growth measure.
Moreover, unlike the wage growth of job stayers, job switchers are now tending to see wage growth of a similar magnitude to that experienced before the recession. This observation is broadly consistent with the improvement seen during the last year in the quits rate (the number of workers who quit their jobs as a percent of total employment) from the Job Openings and Labor Turnover Survey.
I think it will be interesting to continue to monitor the influence of job switching on wage growth as a further indicator of improving labor market dynamism. An update that includes the May data should be available in a few weeks.
June 01, 2016
Putting the Wage Growth Tracker to Work
The April pop in the Atlanta Fed's Wage Growth Tracker has attracted some attention in recent weeks, resulting in some interesting analysis. What is the tracker telling us about the tightness of the labor market and the risks to the inflation outlook?
We had earlier noted the strong correlation between the Wage Growth Tracker and the unemployment rate. Tim Duy took the correlation a step further and estimated a wage Phillips curve. Here's what he found:
The chart shows that lower unemployment generally coincides with higher wage growth (as measured by the Wage Growth Tracker), but wage growth varies a lot by unemployment rate. In the past, an unemployment rate around 5 percent has often been associated with higher wage growth than we currently have.
If the Wage Growth Tracker increased further, would that necessarily lead to an increase in inflation? Jared Bernstein suggests that there isn't much of an inflation signal coming from the Wage Growth Tracker. His primary evidence is the insignificant response of core personal consumption expenditure (PCE) inflation to an increase in the Wage Growth Tracker in a model that relates inflation to lags of inflation, wage growth, and the exchange rate.
However, I don't think the absence of a wage-push inflation connection using the Wage Growth Tracker is really that surprising. The Wage Growth Tracker better captures the wage dynamics associated with improving labor market conditions than rising labor cost pressures per se. For example, if firms are replacing departing workers with relatively low-wage hires, then the wages of incumbent workers could rise faster than do total wage costs (as this analysis by our colleagues at the San Francisco Fed shows). That said, as Bernstein also pointed out in the Washington Post, it's also pretty hard to find evidence of wage pass-through pushing up inflation in his model using more direct measures of labor costs.
I look forward to seeing more commentary about Atlanta Fed tools like the Wage Growth Tracker and how they can be part of the broader discussion of economic policy.
- Unemployment versus Underemployment: Assessing Labor Market Slack
- Does a High-Pressure Labor Market Bring Long-Term Benefits?
- Net Exports Continue to Bedevil GDPNow
- Examining Changes in Labor Force Participation
- Wage Growth Tracker: Every Which Way (and Up)
- Following the Overseas Money
- The Impact of Extraordinary Policy on Interest and Foreign Exchange Rates
- Using Judgment in Forecasting: Does It Matter?
- Does Lower Pay Mean Smaller Raises?
- Outside Looking In: Why Has Labor Force Participation Increased?
- February 2017
- January 2017
- December 2016
- November 2016
- October 2016
- September 2016
- August 2016
- July 2016
- June 2016
- May 2016
- Business Cycles
- Business Inflation Expectations
- Capital and Investment
- Capital Markets
- Data Releases
- Economic conditions
- Economic Growth and Development
- Exchange Rates and the Dollar
- Fed Funds Futures
- Federal Debt and Deficits
- Federal Reserve and Monetary Policy
- Financial System
- Fiscal Policy
- Health Care
- Inflation Expectations
- Interest Rates
- Labor Markets
- Latin America/South America
- Monetary Policy
- Money Markets
- Real Estate
- Saving, Capital, and Investment
- Small Business
- Social Security
- This, That, and the Other
- Trade Deficit
- Wage Growth