Why This Counts: Measuring Industry Productivity

At BLS, productivity is the economic statistic that describes the efficiency of production. The productivity statistics you hear about most often in the news are for the entire U.S. economy. But there’s more to the productivity story than just the overall numbers. The economy is made up of hundreds of industries, and each one works in a different way. Productivity data for each industry help us understand how specific types of production have changed over time. Let’s look at a few specific industries to see how labor productivity data can enhance our understanding of their unique production systems.

General Freight Trucking: Technological Innovations

Economic conditions in the general freight trucking industry closely mirror the health of the overall economy. During the 2007–09 recession, both output and hours worked fell dramatically in trucking. Because employment and spending were down nationwide, there was less demand for the transportation of all kinds of goods. After the recession ended, output and hours in trucking picked back up. Output reached prerecession levels by 2014, but in 2018 hours worked were still slightly below their 2007 level.

Dividing output by hours worked yields labor productivity. Because output in trucking has grown faster than hours during the recovery from the recession, labor productivity has increased. This helps us understand the nature of operations in general freight trucking. Innovative technologies such as communications systems, mapping software, and truck-based sensors and monitors known as “telematics” have improved transportation efficiency. These systems allow deliveries to be planned more efficiently with fewer delays, allowing more freight to be delivered without an equivalent increase in worker hours.

General freight trucking, average yearly percent change in output, hours worked, and productivity from 2007 to 2018

Editor’s note: Data for this chart are available in the table below.

Travel Agencies: Digital Transformation

Another industry that has changed the way it operates is travel agencies. Since 2000, output has increased substantially, while hours fell from 2000 to 2010 and have increased only slightly since then. The major transformation for travel agencies has been the Internet. Online tools have allowed clients to make travel reservations with far less help from workers. This increase in efficiency is reflected in the industry’s labor productivity, which has more than tripled from 2000 to 2017.

Travel agencies, average yearly percent change in output, hours worked, and productivity from 2000 to 2017

Editor’s note: Data for this chart are available in the table below.

Supermarkets: Incremental Change

Changes in other industries have been more subtle. Supermarkets are a particularly competitive industry, and firms employ a large number of workers to maintain high levels of customer service. Managing inventories, stocking shelves, checking out merchandise, and staffing specialty stations are all tasks that supermarkets continue to need. But even in supermarkets, productivity has been increasing since 2009, as output has grown faster than worker hours. To continue growing sales with lower costs, many firms in this industry have relied more on labor-saving technology, such as self-checkout machines. This technology increases efficiency by allowing supermarkets to process more transactions with less help from workers.

Supermarkets, average yearly percent change in output, hours worked, and productivity from 2009 to 2018

Editor’s note: Data for this chart are available in the table below.

Cut and Sew Apparel Manufacturing: Establishment Turnover

Productivity declines also can show the changing nature of work. Cut and sew apparel manufacturing has seen much of its production move outside the United States. In 2018, U.S. apparel manufacturers produced less than 15 percent of the output they produced in 1997. Although worker hours also have declined, they have not dropped as much as output, leading to a decline in labor productivity. This indicates a shift over time in the nature of the average apparel manufacturer. While many large establishments moved overseas in search of cheaper labor, the remaining domestic apparel manufacturing establishments are on average smaller and more specialized, requiring more labor-intensive work.

Cut and sew apparel manufacturing, average yearly percent change in output, hours worked, and productivity from 1997 to 2018

Editor’s note: Data for this chart are available in the table below.

To Learn More

BLS industry productivity data help us study the efficiencies of economic activities. Historical trends in productivity provide an important window into each industry’s working conditions, competitiveness, contribution to the economy, and potential for future growth. These data are used by investors, business leaders, jobseekers, researchers, and government decision makers. We have annual labor productivity measures for over 275 detailed industries.

To dive into the data for yourself, check out the BLS webpages on labor productivity. You also can see productivity data in a brand new way using our industry productivity viewer! Even more specialized industry data are on our webpages for hospitals, construction industries, elementary and secondary schools, and urban transit systems. We also have a recent article on productivity in grocery stores.

Average yearly percent change in output, hours worked, and productivity in selected industries
IndustryOutputHours workedProductivity

General freight trucking, 2007 to 2018

1.0%-0.1%1.2%

Travel agencies, 2000 to 2017

4.8-3.08.1

Supermarkets, 2009 to 2018

1.90.71.2

Cut and sew apparel manufacturing, 1997 to 2018

-9.4-7.5-2.1

Improving the Accuracy of the Consumer Price Index

We are in the “hot stove” months of the baseball year, when teams make trades and other decisions to improve their prospects for next season. Even the best teams, like the World Champion Washington Nationals, can’t rest on their laurels. In much the same way the Nationals continue to tinker with a good thing to make it better, we constantly work to improve our gold standard products, including the Consumer Price Index (CPI). There’s a lot going on with the CPI these days, and we’ll use this blog and other publications to share the latest information. You’ll read about how we reflect changes in consumer spending patterns, (including new goods), how we’re using other rich sources of data on prices and spending, how we’re accounting for changes in the quality of goods and services, and much more. So let’s get started.

The CPI is designed to measure the average change over time in the prices paid by urban consumers for a market basket of consumer goods and services. The CPI is used to determine annual cost-of-living allowances for Social Security beneficiaries. The CPI also is used to adjust the federal income tax system for inflation and as the yardstick for U.S. Treasury inflation-indexed bonds. These are just a few of the many uses of the CPI.

The CPI dates back to 1912, when the Washington baseball team was called the Senators and Walter Johnson ruled the mound. Throughout the history of the CPI, there has been debate about the concepts the CPI should measure and whether it might overstate or understate changes in consumer living costs. The CPI has undergone methodological changes both in response to these discussions and to reflect the changing economic environment. If we hadn’t made these changes, transportation, medical care, recreation, and other goods and services would still be combined into one “miscellaneous” category. Taking the long view, we can track major shifts in consumer inflation for more than a century.

Chart showing 12-month percent change in the Consumer Price Index for All Urban Consumers (CPI-U), 1914 to 2019

Editor’s note: Data for this chart are available in our database at data.bls.gov/timeseries/CUUR0000SA0

In the 1960s, a committee commissioned by Congress recommended that BLS move the CPI closer toward a cost-of-living measure. We responded to those recommendations by creating the CPI for all urban consumers (CPI-U). The former index for urban wage earners was relabeled as the CPI-W. Today, the CPI-U represents the spending patterns of about 93 percent of the population, while the CPI-W represents the spending patterns of about 29 percent.

Here are a few more recent milestones in the history of the CPI:

  • In 1988, following direction from Congress, BLS began calculating the CPI for Americans age 62 and older—called the CPI-E—as an experimental index.
  • In the early 1990s, Congress directed another study of the CPI, popularly referred to as the Boskin Commission. This commission estimated the CPI was overstating the rise in the cost of living and recommended changes in the way the CPI is designed and estimated.
  • In response, BLS sponsored a project in 2002 with the National Academy of Sciences, Committee on National Statistics (CNSTAT) to investigate conceptual, measurement, and other statistical issues in the development of cost-of-living indexes. At this point, we have adopted completely, partially, or experimentally almost all of the CNSTAT recommendations. This includes developing and publishing the Chained CPI, which broadly accounts for consumer substitution of goods and services.

But we can’t stop researching and improving. Today, consumers buy goods and services that weren’t even known a decade ago. And we buy things in many different ways, including from the living room sofa. The growth of e-commerce has created enormous opportunities, but also challenges, for measuring inflation. We continue to work on improvements in response to these developments, and we will talk more about them in future blogs and other publications. In addition, we recently sponsored another CNSTAT panel to investigate three key methodological issues for the CPI:

  1. How best to incorporate data on transactions?
  2. How best to integrate other data sources in the indexes for health insurance, owner-occupied housing, and durable goods?
  3. How to lessen certain types of substitution bias, such as when consumers purchase chicken when the price of steak increases? (Our methods already do a good job accounting for shifts between more similar items, such as between steak and ground beef.)

CNSTAT will convene an expert panel and hold a workshop. Both the panel kickoff and the workshop will be open to the public and will be announced in advance on the BLS website. The panel will then spend about a year in internal discussions and preparing a written report for our consideration.

We expect the CNSTAT report in May of 2021—new ideas, to go with the start of a new baseball season. I’ll be back to blog about the results, so be sure to check back here.

Meet Our New Science and Technology Fellow at BLS

Samantha Tyner
Samantha Tyner

Seeing that we are the U.S. Bureau of LABOR Statistics, we go the extra mile to attract the highest quality labor to accomplish our mission. This includes over 2,000 permanent staff scattered around the country. We also partner with state employees on several BLS programs, and we work with contractors and others to get the job done. Further, we look for opportunities to bring in specialized talent to help with some projects, such as the Civic Digital Fellows who joined us this past summer. Today I want to recognize the first-ever Science and Technology Policy Fellow to spend time at BLS — Samantha Tyner.

The Science & Technology Policy Fellowship is a program of the American Association for the Advancement of Science (AAAS). To understand this program in a nutshell, let me quote directly from their website:

“AAAS Science & Technology Policy Fellowships (STPF) provide opportunities to outstanding scientists and engineers to learn first-hand about policymaking and contribute their knowledge and analytical skills in the policy realm. Fellows serve yearlong assignments in the federal government and represent a broad range of backgrounds, disciplines, and career stages. Each year, STPF adds to a growing corps over 3,000 strong of policy-savvy leaders working across academia, government, nonprofits, and industry to serve the nation and citizens around the world.”

This is the first year BLS has worked with AAAS to bring on a Science and Technology Fellow. We are so fortunate that Samantha (Sam) Tyner started in September and will be with us over the next year. Sam, one of about 200 fellows in the current class, earned her Ph.D. in statistics from Iowa State University and was most recently a postdoctoral researcher at the Center for Statistics and Applications in Forensic Evidence. She is working in the BLS Office of Survey Methods Research (OSMR), focusing on interactive data visualization, text mining, and effective communications to wider audiences.

Let’s find out a little bit about Sam and her fellowship. I asked her what drew her to the federal government. She said she knew pretty early on in graduate school that she didn’t want to go the traditional professor route. She also wasn’t particularly interested in working in one of those internet giants, where the statistics are interesting but the focus is on getting people to click more. She wanted to find ways to use her statistical skills to solve real world problems, and government seemed like a good place for that.

Her first impressions of BLS have been positive. “It’s like hanging out with a bunch of professors, but the staff in OSMR is much more laid back.” One of her current projects involves text mining of BLS mentions on Twitter — what are people saying about us. We’ll use this research to learn how we can better serve our customers.

Another project involves BLS data from the Quarterly Census of Employment and Wages. There is so much data each quarter, down to the county level. She is developing an R Shiny app that will graph these data and allow users to do quick searches. I got to see a quick demo — impressive work after only 2 months on the job.

She is an expert in data visualization, so I asked her what she thinks of some of the charts that BLS produces. I think she was a bit reluctant to criticize, but the comment “you do have a lot of bar charts” was very telling. She describes her goal as to “take a sad chart and make it better.” We certainly welcome her guidance and look forward to producing fewer sad charts in the future.

Beyond all the work Sam is doing at BLS, she also provides posts on the AAAS blog, focusing on some practical aspects of her research. A recent blog taps into her expertise on data visualization. She writes about a problem that can sometimes occur when charts provide too much information. We hope we are not making this mistake with BLS charts.

I’m glad that Samantha has gotten a good start to her Fellowship. We are planning to take full advantage of her research and skills to improve BLS products. I asked her what will make this year a success. Her response — a job offer. Maybe at BLS, or at one of many government agencies where she can use her skills. She will be an asset anywhere she goes.

New Data on Balancing Family Needs with Work

Among the many challenges for today’s families is the balance between caregiving and the demands of working outside the home. Some workers are even sandwiched between the need to provide both childcare and eldercare. New information from the Bureau of Labor Statistics shows that about two out of three employees have paid time off available to meet these needs.

Interest among federal, state, and local policymakers in paid time off and other job flexibilities motivated the U.S. Department of Labor’s Women’s Bureau to sponsor an extra set of questions in the American Time Use Survey. The 2017–18 Leave and Job Flexibilities Module gives us data on the characteristics of wage and salary workers who have access to paid and unpaid leave in their jobs. The module also asked questions about workers who work at home and whether they have flexible work schedules. We also know more about workers who do not have access to leave and job flexibilities. Because we collected the data directly from workers, we could ask them about their experiences, such as the reasons they take leave, or don’t take it even when they need to, and why they work at home.

We now know that 66 percent of U.S. wage and salary workers were able to take paid leave from their jobs in 2017–18. Workers were most often able to use paid leave for a vacation and if they were sick or needed medical care. One area of interest is about people who provide unpaid eldercare. The survey showed that 64 percent of eldercare providers who were employed were able to use paid leave to provide elder caregiving. Another 28 percent of these caregivers were not able to take paid leave for this reason, and 8 percent didn’t know if their employer would allow them to use paid leave to provide eldercare.

Percent of workers with access to paid leave who could use it for the following reasons, 2017–18

Editor’s note: Data for this chart are available in the table below.

We also have learned that 36 million workers (25 percent) sometimes worked at home, and they did so for different reasons. Twenty-four percent worked at home because of a personal preference, 23 percent did so to catch up on work, 22 percent worked at home to coordinate their work schedule with personal or family needs, and 16 percent did so because their job required it. Among those who sometimes worked at home, men and women had different reasons for doing so. Women were more likely than men to work at home to finish or catch up on work and to coordinate their work schedule with personal or family needs. Men were more likely than women to work at home because of a personal preference.

Percent of workers who work at home by main reason, 2017–18

Editor’s note: Data for this chart are available in the table below.

We published these results and more in two recent news releases. One news release focused on workers’ access to leave, their use of leave, and an unmet need for leave. The second focused on workers’ job flexibilities and work schedules.

These releases present data on:

  • Access to paid and unpaid time off
  • Use of paid and unpaid time off
  • Needing to take leave from a job but deciding not to take it
  • Flexible work hours
  • Knowing work schedule in advance
  • Working from home

The releases provide information by:

  • Gender
  • Age
  • Race
  • Hispanic or Latino ethnicity
  • Educational attainment
  • Full- or part-time status
  • Earnings

We also have data files that allow researchers to analyze the data and gain even more insights. Following the policies of BLS and the U.S. Census Bureau to protect the privacy of survey respondents, these data files do not have any information that could identify individual participants.

Percent of workers with access to paid leave who could use it for the following reasons, 2017–18
ReasonYesNoDon’t know

Vacation

95%5%0%

Own illness or medical care

9461

Illness or medical care of another family member

78166

Birth or adoption of a child

76159

Errands or personal reasons

70282

Childcare, other than for illness

65314

Eldercare

64288

Note: The estimates for “childcare, other than for illness” are for workers who were parents of household children under age 18. The estimates for “eldercare” are only for workers who were eldercare providers.

Percent of workers who work at home by main reason, 2017–18
ReasonTotalMenWomen

Personal preference

24%27%21%

Finish or catch up on work

232126

Coordinate work schedule with personal or family needs

222025

Job requires working at home

161616

Reduce commuting time or expense

9109

Weather

443

Other

221

New Data on Employment and Wages in U.S. Establishments with Foreign Ownership

Did you know that U.S. establishments at least partially owned by foreign companies employed 5.5 million U.S. workers in 2012? That was 5.0 percent of U.S. private-sector employment. The U.S. Bureau of Labor Statistics recently partnered with the Bureau of Economic Analysis to produce new data on foreign direct investment in the United States. These two agencies created a new, richer dataset on employment, wages, and occupations in U.S. establishments that have at least one foreign owner.

So how do we define foreign direct investment anyway? In the simplest sense, it is when a U.S. establishment has an owner from another country with at least a 10-percent stake. We consider any establishment that does not meet this threshold as domestically owned. The new data are more detailed than any data previously available on foreign direct investment in the United States. This first set of data is for 2012, but the agencies plan to work together to produce more recent data soon.

Nearly two-thirds of jobs in establishments with foreign ownership had European ownership (3.5 million jobs). The United Kingdom accounted for 874,000 of these jobs. Asia accounted for 17 percent (936,000 jobs) of jobs in U.S. establishments with foreign ownership. Canada accounted for 12 percent (671,000 jobs). The remaining world regions together accounted for less than 8 percent.

Now let’s look at how employment in establishments with foreign ownership breaks down within the United States. The map below shows the percent of private employment in establishments with foreign ownership in each state. South Carolina had the largest share of private employment in establishments with foreign ownership, 8.0 percent. Other states with large shares include New Hampshire, Michigan, Connecticut, New Jersey, and Indiana.

Map showing  each state's percent of private employment in establishments with foreign ownership, 2012

Editor’s note: Data for this map are available in the table below.

Each state’s percent of employment in establishments with foreign ownership depends in part on the industry mix in the state. The chart below shows the percent of each industry’s employment in establishments with foreign ownership. In mining, quarrying, and oil and gas extraction, 14.7 percent of employment is in establishments with foreign ownership. A large share of employment in Alaska is in this industry. Alaska’s share of employment in establishments with foreign ownership, 5.7 percent, is above the national average. Alaska’s vast energy resources may play a role in its share of employment in establishments with foreign ownership.

About 13.2 percent of all employees in manufacturing work in establishments with foreign ownership. Michigan has a large share of employment in manufacturing, and also a large share of employment in establishments with foreign ownership.

Chart showing percent of private employment in establishments with foreign ownership, by industry, 2012

Editor’s note: Data for this chart are available in the table below.

Now let’s turn from employment to wages. The map below shows how wages in establishments with foreign ownership compare with wages in domestically owned establishments across the country. We make this comparison by calculating the ratio of what workers make in average wages in establishments with foreign ownership compared to the average wage in domestically owned establishments. Wage ratios greater than one mean the average for establishments with foreign ownership is higher than for domestically owned establishments. The U.S. wage ratio in 2012 was 1.57, and every state had a wage ratio greater than one. The highest wage ratio was in New York, at 1.98. At the other end of the spectrum, Vermont had a wage ratio of 1.05.

Map showing each state's ratio of average wages in establishments with foreign ownership to domestically owned establishments, 2012

Editor’s note: Data for this map are available in the table below.

Does this mean every establishment with foreign ownership pays higher wages than domestically owned establishments? Let’s analyze wage ratios by industry. We see that the health care and social assistance industry had a wage ratio of 0.86 in 2012. All other major industry groups had wage ratios of 1.00 or higher. The finance and insurance industry had a wage ratio of 1.82.

Want to know more about these data? See our Spotlight on Statistics, “A look at employment and wages in U.S. establishments with foreign ownership.”

Chart showing ratio of average wages in establishments with foreign ownership to domestically owned establishments, by industry, 2012

Editor’s note: Data for this chart are available in the table below.

BLS and the Bureau of Economic Analysis hope to continue this interagency collaboration. Our goal is to merge and analyze more recent data from both agencies. When agencies work together to produce new datasets with little increase in cost to the public, all data users benefit. Producing accurate, objective, relevant, timely, and accessible products is the BLS mission. This collaboration to produce new relevant data allows us to improve our service to the American people.

Percent of private employment in establishments with foreign ownership, 2012
StateEmployment share

National

5.0%

Alabama

5.4

Alaska

5.7

Arizona

3.9

Arkansas

4.5

California

4.2

Colorado

4.6

Connecticut

6.5

Delaware

6.0

District of Columbia

3.4

Florida

3.6

Georgia

5.5

Hawaii

6.0

Idaho

2.9

Illinois

5.1

Indiana

6.4

Iowa

4.0

Kansas

5.7

Kentucky

6.2

Louisiana

3.9

Maine

6.1

Maryland

4.7

Massachusetts

6.3

Michigan

6.6

Minnesota

4.0

Mississippi

3.4

Missouri

4.0

Montana

1.8

Nebraska

3.6

Nevada

3.8

New Hampshire

6.9

New Jersey

6.5

New Mexico

3.0

New York

5.8

North Carolina

6.2

North Dakota

3.8

Ohio

5.3

Oklahoma

3.6

Oregon

3.4

Pennsylvania

5.5

Rhode Island

6.1

South Carolina

8.0

South Dakota

2.1

Tennessee

5.5

Texas

5.3

Utah

4.0

Vermont

3.7

Virginia

5.1

Washington

4.0

West Virginia

4.8

Wisconsin

3.5

Wyoming

3.8
Percent of private employment in establishments with foreign ownership, by industry, 2012
IndustryEmployment share

Mining, quarrying, and oil and gas extraction

14.7%

Manufacturing

13.2

Management of companies and enterprises

9.6

Wholesale trade

9.0

Information

7.8

Finance and insurance

7.5

Utilities

7.3

Transportation and warehousing

6.3

Administrative and waste services

6.0

Professional, scientific, and technical services

5.5

Total private

5.0

Retail trade

4.7

Real estate and rental and leasing

2.2

Construction

1.8

Accommodation and food services

1.6

Other services (except public administration)

1.3

Agriculture, forestry, fishing, and hunting

1.0

Health care and social assistance

0.9

Arts, entertainment, and recreation

0.7

Educational services

0.6
Ratio of average wages in establishments with foreign ownership to domestically owned establishments, 2012
StateWage ratio

National

1.57

Alabama

1.44

Alaska

1.63

Arizona

1.28

Arkansas

1.43

California

1.49

Colorado

1.53

Connecticut

1.53

Delaware

1.78

District of Columbia

1.08

Florida

1.52

Georgia

1.36

Hawaii

1.06

Idaho

1.30

Illinois

1.61

Indiana

1.56

Iowa

1.48

Kansas

1.56

Kentucky

1.36

Louisiana

1.67

Maine

1.26

Maryland

1.28

Massachusetts

1.46

Michigan

1.84

Minnesota

1.50

Mississippi

1.63

Missouri

1.55

Montana

1.63

Nebraska

1.35

Nevada

1.47

New Hampshire

1.39

New Jersey

1.64

New Mexico

1.22

New York

1.98

North Carolina

1.47

North Dakota

1.55

Ohio

1.49

Oklahoma

1.40

Oregon

1.41

Pennsylvania

1.43

Rhode Island

1.31

South Carolina

1.43

South Dakota

1.45

Tennessee

1.42

Texas

1.80

Utah

1.45

Vermont

1.05

Virginia

1.23

Washington

1.40

West Virginia

1.33

Wisconsin

1.38

Wyoming

1.72
Ratio of average wages in establishments with foreign ownership to domestically owned establishments, by industry, 2012
IndustryWage ratio

Finance and insurance

1.82

Construction

1.62

Total private

1.57

Accommodation and food services

1.51

Real estate and rental and leasing

1.50

Arts, entertainment, and recreation

1.45

Other services (except public administration)

1.44

Agriculture, forestry, fishing, and hunting

1.40

Wholesale trade

1.39

Professional, scientific, and technical services

1.39

Mining, quarrying, and oil and gas extraction

1.28

Management of companies and enterprises

1.23

Retail trade

1.20

Educational services

1.19

Manufacturing

1.18

Utilities

1.15

Administrative and waste services

1.13

Information

1.05

Transportation and warehousing

1.00

Health care and social assistance

0.86