Federal Reserve Economic Data

When unionized workers go on strike, the total amount of hours worked in the economy decreases; but the level of employment reported each month in the Current Employment Statistics survey may not show a similar change. The impact of labor stoppages on the number of people on payrolls depends on the timing and duration of those collective bargaining actions.

The FRED graph above shows the employment level in the motor vehicles and parts industry reported by the Bureau of Labor Statistics between January 1990, the earliest available data, and the time of this writing. Two major labor stoppages stand out: the lockdown caused by the COVID-19 pandemic in April 2020 and the 54-day strike by the United Auto Workers (UAW) union against General Motors in July 1998.

The Bureau of Labor Statistics publishes a monthly strike report of labor stoppages of 1,000 workers or more who were idle during a complete pay period while the Current Employment Statistics survey was conducted. Depending on the industry, the pay period may be weekly, biweekly, monthly, or semimonthly. The reference date for conducting the CES survey is the 12th day of the month.

To learn more about this topic, read “Understanding strikes in CES estimates” by John P. Mullins in the Monthly Labor Review. You can check the industry data referenced in Figure 1 of that publication in this FRED graph. For more information on current UAW contract negotiations, look to the Chicago Fed.

Other large-scale labor actions, such as the August 1983 telecommunication workers strike against AT&T, are also visible in FRED. However, the ongoing Screen Actors Guild-American Federation of Television and Radio Artists (SAG-AFTRA) strike against the Alliance of Motion Picture and Television Producers is not immediately noticeable in the recent overall employment figures for that industry. That may well be a case where the full story is truly behind the numbers and a data graph is unable to tell it.

How this graph was created: From FRED’s main page, search for and select “All Employees, Motor Vehicles and Parts.”

Suggested by Diego Mendez-Carbajo.

The FRED graph above shows the exchange rate of two currencies with the US dollar: the Swiss franc and the Colombian peso. We chose these two for their obvious contrasting history.

• The Swiss franc is considered to be among the strongest currencies—meaning that it tends to appreciate with respect to many other currencies.
• The Colombian peso is the opposite, with continuing depreciation with respect to strong currencies. (One peculiar benefit of using this currency is that it was never rebased—i.e., never had a few zeroes removed from its high face value. This numeric consistency avoids potential issues with displaying the peso’s exchange rate across various definitions of the currency.)

The graph shows that, over the longer run, the Swiss franc has become stronger than the dollar while the Colombian peso has gotten significantly weaker than the dollar. There are considerable variations at shorter horizons, which can be driven by many factors related to the expectations about the currencies’ respective economies. (This recent FRED Blog post covers this topic.) But back to the long-run changes…

The second graph takes the same exchange rates and adjusts them by the inflation rates in the US, Switzerland, and Colombia. These are so-called real exchange rates. Note how the lines are much flatter, especially as you compare the scale of the vertical axes in both graphs.

A large part of these long-run exchange rate movements can indeed be explained by inflation differentials: Inflation is typically low in Switzerland, while it is typically high in Colombia. The lines are not completely flat, though. First, the consumer price index or the GDP deflator may not be the appropriate price index to use here, as other factors such as taxes, tariffs, and other trade impediments may matter. Finally, most currency exchange is not performed to buy foreign goods, but rather for purely financial transactions. Thus, a currency can be more or less attractive depending on economic or political developments.

How these graphs were created: Search FRED for “Colombia exchange rate” and take the option with the longest time range. Click on “Edit Graph,” open the “Add Line” tab, and search similarly for “Switzerland exchange rate.” Open the “Format” tab and put the legend for the second line on the right. You have the first graph. For the second, take the first graph, click on “Edit Graph,” add series by searching for “US CPI,” then again for “Colombia CPI,” in both cases making sure the series is as long as possible and in levels, not growth rates. Apply formula a*b/c. Repeat for the second line and “Switzerland deflator” (the CPI series is too short).

Suggested by Christian Zimmermann.

Summer is ending. As the new school year gears up, some areas of economic activity will get seasonal boosts—such as increases in retail sales of office supplies as incoming students and their families buy what they need for the classroom. Female employment also picks up at this time of year. Recent research on labor markets finds that the childcare services provided by formal schooling drive this increase in employment.

The FRED graph above replicates Figure 1 in a related piece of research: the NBER Working Paper, “The Summer Drop in Female Employment,” by Brendan Price and Melanie Wasserman. The graph shows the non-seasonally-adjusted labor force participation rates among males (orange line with triangles) and females (blue line with dots) between 25 and 54 years of age. (The values are normalized to zero in December 2019.*) A close examination of this graph shows that, every summer, women’s labor force participation drops sharply, whereas men’s participation remains comparatively stable.

Why? During the summer, women reduce the amount of time they work outside the home; they are more likely than men to step in and provide some of the childcare services required while school is out for the summer. Vacations, summer school, and camps—supplemented by informal childcare by relatives, for example—do not add up to the six hours per weekday that children spend in school most of the year.

The research by Price and Wasserman helps answer the seasonal puzzle that the FRED Blog described last year, which helps tell the bigger story behind the numbers.

* The data in the NBER paper are two series from the Bureau of Labor Statistics available in FRED through the Organization for Economic Co-operation and Development’s Main Economic Indicators Release. Borrowing from Geoffrey Chaucer: All roads lead the data user to FRED.

How this graph was created: In FRED, search for “Activity Rate: Aged 25-54: Males for the United States.” From the “Edit Graph” panel, use the “Add Line” tab to search for and select “Activity Rate: Aged 25-54: Females for the United States.” Use the “Edit Line 1” tab to customize the units by selecting “Index (Scale value to 100 for chosen date)” and enter “2009-12-01” in the date box. Click on “Copy to all” to apply the unit transformation to both series.

Suggested by Diego Mendez-Carbajo and Mary Clare Peate.