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
November 01, 2019
New Evidence Points to Mounting Trade Policy Effects on U.S. Business Activity
Trade worries remain at the forefront of economic news. Average tariffs on Chinese imports now stand at 21 percent, up from 3 percent in March 2018. Earlier this month, President Trump suspended plans for further tariff hikes on Chinese goods. Also this month, the U.S. is rolling out new tariffs on $7.5 billion worth of imports from Europe. On another front, fears are growing that Congress may not approve the U.S.-Mexico-Canada Trade Agreement, the intended successor to the North American Free Trade Agreement. Data from the folks at policyuncertainty.com say that articles about trade policy uncertainty in U.S. newspapers were more than 10 times as numerous in the third quarter of 2019 as the average from 1985 to 2010.
Trade policy worries extend beyond the newswires. We hear concerns about trade policy in reports from Main Street firms in the Sixth District collected through our Regional Economic Information Network and, more broadly, in the Federal Reserve's Beige Book. Amid reports of softening manufacturing conditions in the U.S., slowing growth in payroll employment, and a drop-off in business investment, it's natural to wonder whether trade policy is at least partly to blame. Professional forecasters seem to think so. For instance, the International Monetary Fund (IMF) forecasts that the U.S.-China trade dispute will shave roughly three-fourths of a percent from global output by 2020, which, as the IMF's managing director noted, is "equivalent to the whole economy of Switzerland."
Over the past year and a half, we have been keenly interested in how trade policy worries affect business decision making. In August 2018, we reported that trade concerns prompted about 1 in 5 firms to re-evaluate their capital investment decisions. At the same time, only 6 percent of the firms in our sample had then decided to cut or defer previously planned capital expenditures in response to trade policy developments. Early this year, we noted that the hit to aggregate investment from trade tensions and tariff worries was modest in 2018, but firms believed the impact would increase in 2019.
As U.S.-China trade tensions escalated during the third quarter of this year, we went back into the field, posing another set of trade-related questions to panelists in our Survey of Business Uncertainty (SBU). This time around, we asked both backward- and forward-looking questions about the perceived effects of trade policy, and we expanded the scope of our questions to cover employment and sales in addition to capital expenditures.
Overall, our results say that the negative effects of trade policy developments on U.S. business activity have grown over time, particularly for firms with an international reach. Trade policy effects on the business sector as a whole remain modest but larger than we saw six or 12 months ago.
Twelve percent of surveyed firms reported cutting or postponing capital expenditures in the first six months of 2019 because of trade tensions and tariff worries (see exhibit 1). That's twice the share when we asked the same question a year earlier. Given the capital-intensive nature of manufacturing, it is perhaps more concerning that one in five manufacturing firms now report cutting or postponing capital expenditures because of trade policy tensions.
We also find that tariff hikes and trade policy tensions now exert a larger negative impact on gross U.S. business investment. Exhibit 2 uses SBU data on whether firms changed their capital expenditures due to trade policy tensions and, if so, by how much and in which direction. Column (1) reports the average percentage impact in the sample, where we weight each firm's response by its capital stock value. To estimate the dollar impact of trade policy developments in column (2), we multiply the weighted-average percent change by actual U.S. business investment in the first half of 2019, which yields an estimated effect on U.S. business investment of about minus $40 billion.
This estimated trade policy hit to aggregate investment is modest but roughly double what we previously found for the second half of 2018. Our results say that investment is hardest hit in manufacturing and construction, though perhaps for different reasons. The larger response for manufacturing is likely due to its higher international exposure, both in direct goods trade and across the supply chain. For construction, the impact is likely due to an increased cost of imported materials and equipment.
Exhibit 3 reports the estimated effects of tariff hikes and trade policy tensions on private sector employment and sales in the first half of 2019. According to our results (reached by using the same procedure as in Exhibit 2), these developments subtracted about 40,000 jobs per month from nonfarm payrolls and about $259 billion in sales over the first half of the year. Though this employment impact is sizable, it is not estimated very precisely (one standard error corresponds to about 24,000 jobs per month). The estimates for the impact of tariff hikes and trade policy tensions imply about $110,000 in lost sales per lost job.
Notes on Exhibit 3: In Panel A, column (1) reports the employment-weighted mean response to questions about whether tariff hikes and trade policy tensions caused the firm to alter its employment level in the first half of 2019 and, if so, by what percentage amount. We deleted three questionable responses to the employment question that we could not verify. To obtain the aggregate employment impact in column (2), we multiplied the column (1) value by the average nonfarm private sector payroll employment in the first half of 2019. The "Reweighted" row reflects a re-weighting of the SBU data to match the one-digit industry distribution of private sector payroll employment. In Panel B, column (1) reports the sales-weighted mean response to questions about whether tariff hikes and trade policy tensions affected the firm's sales in the first half of 2019 and, if so, by what percentage amount. To obtain the aggregate sales impact in column (2), we multiplied the column (1) value by Nominal Gross Output: Private Industries. According to the U.S. Bureau of Economic Analysis, gross output is, "principally, a measure of an industry's sales or receipts. These statistics capture an industry's sales to consumers and other final users (found in GDP), as well as sales to other industries (intermediate inputs not counted in GDP). They reflect the full value of the supply chain by including the business-to-business spending necessary to produce goods and services and deliver them to final consumers." The "Reweighted" row reflects a re-weighting of the SBU data to match the one-digit industry distribution of private sector gross output. Standard errors are reported in brackets.
We also asked forward-looking questions to assess whether firms think trade policy worries will continue to dampen their business activities in the second half of 2019. Exhibit 4 summarizes our findings in this regard. SBU respondents anticipate that the impact of trade policy on their second-half sales revenue will be similar to what they reported for the first half of 2019, but they anticipate somewhat larger negative effects on their capital expenditures and employment. Across the private sector as a whole, SBU respondents see their capital expenditures as down by 3.8 percent in the second half of 2019 due to tariff hikes and trade policy tensions.
In sum, as trade policy tensions escalated in the first half of 2019, our results say that businesses took a hit to their sales and backed off on hiring and investment. Moreover, firms anticipate that the negative effects will continue during the second half of 2019. Our estimated impact magnitudes are rising over time but remain modest.
We should also note that our estimates do not capture certain effects. For instance, they don't capture the pass-through of tariff hikes to American consumers in the form of higher prices or to American companies in the form of compressed margins and lower profits. Tariff hikes and trade policy tensions also slow growth in the global economy, with negative effects on the U.S. economy. These blowback effects are also outside the scope of our investigation.
September 26, 2019
Digging into Older Americans’ Flat Participation Rate
The rate of labor force participation (LFP) by people age 55 and over had been rising during the decade leading up to the Great Recession. But more recently, as the following chart shows, the share of older individuals engaged in the labor market has barely budged. (We should note that for all the charts in this post, the data are from the Current Population Survey from the U.S. Bureau of Labor Statistics and the authors' calculations.) What behavioral and demographic factors could be underlying these trends?
We can use the Atlanta Fed's Labor Force Dynamics web page to explore why the 55-and-over population has a relatively flat LFP rate of late. One factor working to depress older Americans' LFP rate is the increase in the share of those 55 and older who are retired: from 46.9 percent in the second quarter of 2014 to 47.8 percent in the second quarter of 2019 (see the chart).
This change in the overall retirement rate of the 55-plus cohort is actually the result of the union of two opposite forces—one demographic and one behavioral. From the perspective of demographics, a greater share of the population is reaching the typical retirement age threshold of 65. For instance, the share of people aged 65 and older has increased from 18 percent in the second quarter of 2014 to 20.3 percent in the second quarter of 2019 (see the chart).
This demographic shift is important because the retirement rate is much higher for those 65 and older compared to those from 60 to 64 years old. For example, the retirement rate in 2019 for those 60–64 is around 25 percent, but it's 56 percent for those aged 65–69, and it's 81 percent for those 70 and older (see the chart).
However, the retirement rate among those 60–64 and those 65–69 has also declined in recent years. This change in behavior within older age groups is partly offsetting the downward pressure on participation coming from having a larger share of population over 65.
A second factor that has worked against the overall retirement effect and helped push up the LFP rate of the 55-plus population has been a decline in the share of older individuals saying they are not participating in the labor force because of disability or illness. This rate has decreased from 8.6 percent in the second quarter of 2014 to 8.1 percent in the second quarter of 2019 (see the chart).
As the chart below shows, this shift is largely a demographic effect. Nonparticipation because of disability/illness drops off significantly as people turn 65, so the fact that a greater share of people 55 and older are now 65 and older decreases the overall share of nonparticipation for disability/illness reasons as well. However, it's important to note that a lower disability/illness rate for those 65 and older doesn't mean that this older population is actually less disabled or sick. It is more that individuals in the data fall into only one category, and those 65 and older are more likely to say they are retired than to say they are disabled or ill.
A third force helping push up the LFP rate is the decline in the share of older individuals on the sidelines of the labor market—those who are not participating but nonetheless want a job—from 1.8 percent in the second quarter of 2014 to 1.6 percent in the second quarter of 2019 (see the chart).
This “shadow labor force” effect on participation is mostly the result of changes in behavior (that is, a reduced propensity to remain on the sidelines within age groups) rather than from demographic changes because the rate has declined within age groups, whereas the levels are roughly similar across age groups (see the chart).
To put these various pieces together, the following chart summarizes the overall contributions of demographic and behavioral forces on the LFP rate among the 55-plus population between the second quarter of 2014 and the second quarter of 2019. The chart shows that the contributions stemming from changes in demographics and behavior have largely offset each other.
As I've already described, the biggest demographic effects come from having more people at an age with two specific characteristics: a relatively high rate of nonparticipation because of retirement and a relatively low rate of nonparticipation because of disability or illness. But as the following chart shows, from the second quarter of 2014 to the second quarter of 2109, the retirement demographic dominates, so the overall demographic LFP effect is a negative one.
Conversely, the largest behavioral shifts are, first, a lower propensity within older age groups to stay out of the labor force because of retirement and, second, a lower share of older people wanting a job but not looking for one. As the following chart shows, from the second quarter of 2014 to the second quarter of 2019, these behavioral changes combine to push up LFP by enough to nearly offset the demographic shifts.
It seems reasonable to presume that the aging population will continue to be an important source of downward pressure on the LFP rate of older Americans over the next few years. What will be telling is whether or not the behavioral shifts we have seen will persist as strongly as they have up to this point and continue to provide a countervailing positive influence on participation.
The tools on the Atlanta Fed's Labor Force Dynamics web page are very useful for understanding what's behind changes in LFP for different demographic groups. In addition to cuts for different age groups, you can look at differences between men and women and among different racial and ethnic groups and levels of educational attainment. You can download the chart data—the charts in this blog are downloaded images from the web page—and even download the underlying Current Population Survey microdata from the Kansas City Fed's CADRE web page if you want to create your own cuts. Check it out!
August 05, 2019
What the Wage Growth of Hourly Workers Is Telling Us
The Atlanta Fed's Wage Growth Tracker has shown an uptick during the past several months. The 12-month average reached 3.7 percent in June, up from 3.2 percent last year. But in 2016, it depicted acceleration that eventually reversed course. So is this recent increase real or illusory?
Although using a 12-month average quiets much of the noise in the monthly data, it is possible that the smoothed series still may exhibit some unwanted variation due to the way the Wage Growth Tracker is constructed. For example, how the monthly Current Population Survey reports individual earnings might be a factor introducing unwanted noise into the Tracker. Specifically, some people directly report their hourly rate of pay, and some report their earnings in terms of an amount per week, per month, or per year.
Relative to those paid an hourly rate, there are at least a couple of reasons why using the earnings of nonhourly workers might introduce additional variability into the Wage Growth Tracker's overall estimate of wage growth. First, reported nonhourly earnings include base pay as well as any overtime pay, tips, and commissions earned, and hence can vary over time even if the base rate of pay didn't change. For a worker paid at an hourly rate, reported earnings exclude overtime pay, tips, and commissions and so are not subject to this source of variation. Second, the method we use to convert nonhourly earnings to an hourly rate is likely subject to some margin of error since it involves using the person's recollection of how many hours they usually work. These two factors suggest that the earnings of workers paid at an hourly rate might be a somewhat cleaner measure of hourly earnings.
To investigate whether this distinction actually matters in practice, we created the following chart comparing the 12-month average Wage Growth Tracker since 2015 (depicted in the green line) with a version that uses only the earnings of those paid at an hourly rate (blue line).
As the chart shows, the 12-month average of median wage growth for hourly workers generally tracks the overall series—both series are about a percentage point higher than at the beginning of 2015. However, the hourly series is a bit less variable, making the recent uptick in wage growth more noticeable in the hourly series than in the overall series. This observation suggests that as we monitor shifts in wage pressure, the hourly series could complement the overall series nicely. Versions of the Wage Growth Tracker series for both hourly and nonhourly workers are now available on the Wage Growth Tracker page of the Atlanta Fed's website.
If you would like to use the Wage Growth Tracker's underlying microdata to create your own versions (or to conduct other analysis), follow this link to explore the data on the Atlanta Fed's website. See this macroblog post, "Making Analysis of the Current Population Survey Easier," from my colleague Ellyn Terry to learn more about using this dataset.
July 15, 2019
Making Analysis of the Current Population Survey Easier
Speaking from experience, research projects often require many grueling hours of deciphering obtuse data dictionaries, recoding variable definitions to be consistent, and checking for data errors. Inevitably, you miss something, and you can only hope that it does not change your results when it's time to publish the results. It would be far less difficult if data sets came prebuilt with time-consistent variable definitions and a guidebook that makes the data relatively easy to use. Not only would research projects be more efficient, but also the research would be easier to replicate and extend.
To this end, we have worked closely with our friends at the Kansas City Fed's Center for the Advancement of Data and Research in Economics (CADRE) to produce what we call a harmonized variable and longitudinally matched (HVLM) data set. This particular data set uses the basic monthly Current Population Survey (CPS) data published by the U.S. Census Bureau and the Bureau of Labor Statistics. The HVLM data set underlies products such as the Atlanta Fed's Wage Growth Tracker and the various tools on the Atlanta Fed's Labor Force Participation Dynamics web page.
You may be wondering how this data set is different from the basic monthly CPS data available at IPUMS. Like the IPUMS-CPS data, the HVLM-CPS data set uses consistent variable names and includes identifiers for longitudinally linking individuals and households over time. Unlike the IPUMS-CPS data, the HVLM-CPS also has time-consistent variable definitions. For example, the top-coded values for the age variable in the IPUMS-CPS is not the same in all years, whereas the HVLM-CPS age variable is consistently coded by using the most restrictive age top-code. As another example, the number of race categories is not the same in every year in the IPUMS-CPS (having increased from 3 to 26), while the race variable in the HLVM-CPS data set is consistently coded by using the original three race categories. Applying these types of restrictions means that the resulting data set can be more readily used to make comparisons over time.
The screenshot below shows how accessible the HVLM-CPS data are. For a visual of each variable over time, click on Charts at the top to see a PDF file of time-series charts. Code Book is an Excel file containing the details of how each variable has been coded. You can see in the screenshot how each variable ends with two numbers. These two numbers correspond to the first year that variable is available. For example, mlr76 is coded with consistent values (1 = employed, 2 = unemployed and 3 = not in labor force) from 1976 until today. The Data File is a Stata (.dta) format file with variable labels already attached. For users wishing to use the panel structure of the CPS survey, lags of many variables are provided on the data set already—for example, mlr76_tm12 is an individual's labor force status from 12 months ago).
Clicking on the c icon under Code Book opens a screen with the values of the corresponding variable. The screenshot shows lfdetail94 and nlfdetail94 as examples. The first variable, lfdetail94, contains a large amount of detail on those engaged in the labor market, while nlfdetail94 contains detailed categories for those not engaged in the labor market.
The HVLM-CPS data set is freely available to download and is updated within hours of when the CPS microdata are published, thanks to sophistical coding techniques and the fast processors at the Kansas City Fed. To access the data, go to the CADRE page (using Chrome or Firefox). At the top right, select Sign in, then Google Login. Then, under schema, select Harmonized Variable and Longitudinally Matched [Atlanta Federal Reserve] (1976–Present).
- New Evidence Points to Mounting Trade Policy Effects on U.S. Business Activity
- Digging into Older Americans’ Flat Participation Rate
- What the Wage Growth of Hourly Workers Is Telling Us
- Making Analysis of the Current Population Survey Easier
- Mapping the Financial Frontier at the Financial Markets Conference
- The Tax Cut and Jobs Act, SALT, and the Blue State Blues: It's All Relative
- Improving Labor Force Participation
- Young Hispanic Women Investing More in Education: Good News for Labor Force Participation
- A Different Type of Tax Reform
- X Factor: Hispanic Women Drive the Labor-Force Comeback
- November 2019
- September 2019
- August 2019
- July 2019
- June 2019
- May 2019
- March 2019
- February 2019
- January 2019
- December 2018
- 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