Federal Reserve Economic Data

FRED recently added 20 data series about work-from-home practices in the United States reported by Jose Maria Barrero, Nicholas Bloom, and Steven J. Davis. The data are organized by industry and worker characteristics and offer a broad range of insights.

The FRED graph above shows the frequency of three types of work arrangements:

• working fully on site (blue line)
• working partly on site and partly remotely (orange line)
• working fully remotely (green line)

Between November 2021 and August 2025, when these data are available, working fully on site was approximately twice as frequent as a hybrid working arrangement and five times more frequent than working fully remote.

Here are some other takeaways from the recently added data:

• The proportion of fully paid workdays worked from home by men and women is slightly different.
• At the time of this writing, the industries with remote work rates between 50% and 30% are (in descending order) finance & insurance; information; professional & business services; arts & entertainment; utilities; real estate; and wholesale trade.
• The industries with remote work rates between 25% and 10% are (also in descending order) health care & social assistance; government; education; manufacturing; retail trade; transportation & warehousing; and hospitality & food services.

For more insights into this topic, check out the work of St. Louis Fed economist Alex Bick and coauthors. They research the trends in work arrangements and provide a technical analysis of the differences in work-from-home rates by sex, education, state of residence, industry, and firm size.

How this graph was created: Search FRED for and select “All Full-Time Wage and Salary Workers: Working Fully on Site.” Click on the “Edit Graph” button and select the “Add Line” tab to search for “All Full-Time Wage and Salary Workers: Working Fully Remote.” Don’t forget to click on “Add data series.” Repeat the last two steps to search for and add “All Full-Time Wage and Salary Workers: Working in Hybrid (Some Days Working from Home, Some Days at Employer or Client Site).”

Suggested by Diego Mendez-Carbajo.

The FRED Blog has discussed how economists use the consumer price index (CPI) to adjust the dollar value of retail sales, credit card debt, and state tax revenue to account for rising prices over time.

Today, we discuss how the Social Security Administration uses a version of the CPI to calculate cost of living adjustments, which affect the benefits recipients receive.

Another way to put it: how SSA uses CPI-W to calculate COLAs.

The FRED graph above shows the annual percent growth in both the CPI and the related CPI-W (from the US Bureau of Labor Statistics) from 2014 through 2024:

• Blue bars represent the all-items price inflation for all households living in cities and urban settings. This CPI inflation metric covers over 90% of the total population, regardless of employment status.
• Green bars represent the all-items price inflation for all urban wage earners and clerical workers,* CPI-W. Those households amount to approximately 30% of the total US population.

As we might expect, the CPI and CPI-W annual inflation rates are very similar. But it’s important to note that neither is consistently higher or lower than the other. Also, in times of higher inflation, the differences are more marked than in times of lower inflation. Different spending patterns among the populations covered by each price index can help explain that.

Since 1975, the dollar value of the Social Security’s general benefits has increased by the average July-to-September annual growth rate of the CPI-W. The amount of adjustment is announced in October. Last year, it was 2.5%. More details are available here.

*NOTE: Urban wage earners and clerical workers are defined as “households in which at least one of the members has been employed for 37 weeks or more during the previous 12 months in an eligible occupation and for which 50 percent or more of the household income must come from wage earnings associated with an eligible occupation.”

How this graph was created: Search FRED for and select “Consumer Price Index for All Urban Consumers: All Items in U.S. City Average.” From the “Edit Graph” panel, use the “Add Line” tab to search for and select “Consumer Price Index for All Urban Wage Earners and Clerical Workers: All Items in U.S. City Average.” Be sure to click “Add data series.” Next, use the “Edit Lines” tab to select each of the two graph lines from the dropdown menu and change the units to “Percent Change from Year Ago” and modify the frequency to “Annual.” Lastly, use the “Format” tab to change the graph type to “Bar.”

Suggested by Nylah Martinez and Diego Mendez-Carbajo.

Life insurance provides a benefit, typically a one-time payout, at the time of death of the insured person. As such, it could be called death assurance. This FRED Blog post seeks to determine the scope of such insurance in the US.

One thing we could look at is the total amount of promised benefits. But imagine if all the promised benefits of all the life insurance policies had to be paid out immediately. How much life insurance companies could pay would be limited by the assets they hold. So that is what our FRED graph above shows. It’s a big number that’s been growing over time, but it’s not a complete picture of what we’re trying to find out.

In our second FRED graph, above, we represent those same life insurance company assets, but this time as a percentage of total yearly US personal income.

We can see that life insurance assets hit a low point in the early 1980s, at 19% of total yearly US personal income; a high point right after the pandemic, at 48%; and it seems to have now settled around 41%. This means that the life insurance industry would now be able to provide the equivalent of a little less than five months of income if everyone had an insurance policy.

Given about 51% of Americans have life insurance, that would amount to an average of about ten months of income as benefits to the insured.

How these graphs were created: Search FRED for “life insurance assets.” The first choice should be our first graph. For the second graph, click on “Edit Graph”; search for “personal income”; take the nominal series, not the real one; apply the formula a/b/10 to get percentages, given the units. Note that this ratio is not perfect, as we divide a series that is not seasonally adjusted by one that is. For looking at longer trends, this is sufficient.

Addendum: Today, our editor George Fortier is celebrating 25 years at the Federal Reserve Bank of St. Louis. He is the one who is giving this blog its special je ne sais quoi while getting credit for very few blog posts. A big thank you for your dedication, and we are looking forward to many more posts.

Suggested by Christian Zimmermann.