The pandemic brought about serious upheaval in the health care sector. Services directly connected to the COVID-19 virus were overwhelmed at times, while non-essential services came close to a standstill. Our question for today is, in sum, how has the health care sector fared?

The FRED graph above shows personal consumption expenditures on health care. This measure probably isn’t a good indicator to answer our question, as it doesn’t include any of the expenses paid by other entities, such as businesses and various levels of government. So we need a measure that encompasses all of the health care sector.

The second FRED graph gets closer to what we want, although with data that are not as current as the data in the first graph.

We see that total revenue of health care establishments (in blue) dipped severely at the start of the pandemic, but then got back on track with its previous trend. This pattern is actually not that much different from the rest of the economy. Health care employment (in red) shows a different picture: Employment is still far below the peak before the pandemic and, hence, even further below the long-term trend.

But one has to be careful here: The revenue numbers aren’t adjusted for inflation. So, one more graph…

Our last FRED graph (below) shows that this storyline still holds: The health care sector has had to deliver more than what had been expected before, but with far fewer employees.

How these graphs were created: First graph: Search for and select “health care expenditures.” Second graph: Search for and select “health care revenue.” From the “Edit Graph” panel, use the “Add Line” tab to search for and select “health care employment.” Use the “Format” tab to put the y-axis on right for the second series. Reduce the sample period to focus on the last years. Third graph: Starting with the second graph, use the “Edit Graph” panel to add a series to the first line by searching for “GDP deflator” and applying formula a/b*100.

Suggested by Christian Zimmermann.

The FRED Blog has discussed the seasonality of food prices, labor markets, interest rates, and e-commerce. So, perhaps it’s no surprise to learn that there are also seasonal ups and downs for some import flows.

The FRED graph above shows two data series produced collaboratively by the Census Bureau and the Bureau of Economic Analysis: the monthly U.S. dollar value of goods imported from China (in green) and from Canada (in red). This price does not include import duties, freight, insurance, and other charges related to bringing the merchandise into the U.S.

The repeating up-and-down monthly pattern of U.S. imports from China contrasts with the comparatively steadier pattern of U.S. imports from Canada: Many Chinese goods arrive at U.S. ports and shipping centers in October, while far fewer do in February and March; Canadian goods, in comparison, arrive in fairly similar numbers throughout the year.

The data don’t offer a breakdown by type of imported good, so we can’t say if the Chinese goods are available for purchase only in the latter part of each year. But the timing of the imports is just ahead of the busiest retail season of the year, the Christmas holiday; that suggests these are consumer goods likely to be purchased as gifts. Perhaps Santa Claus prefers to shop in China rather than Canada. But we’ll need to do more research on that.

How this graph was created: Search for and select “U.S. Imports of Goods by Customs Basis from China.” From the “Edit Graph” panel, use the “Add Line” tab to search for and select “U.S. Imports of Goods by Customs Basis from Canada.” To change the graph type and style of the series use the “Format” panel.

Suggested by Diego Mendez-Carbajo.

The FRED Blog recently featured an ALFRED graph to discuss a revision to the methodology used by Realtor.com to report housing inventory data. Today, we return once more to ALFRED, the archive of historical versions (or vintages) of FRED data, to discuss revisions to employment data.

The ALFRED graph above shows 12 different vintages of total nonfarm employment data from the U.S. Bureau of Labor Statistics. The graph doesn’t display the legend, to leave more room for the data (graph with legends). The solid black line shows, as of January 7, 2022, the monthly changes in total nonfarm employment for each month in 2021. The dotted red lines and circles show the initially announced and subsequently revised monthly changes in total nonfarm employment.

Consider January 2021: While employment growth was initially estimated at 49,000 persons, during the next two months that figure was revised first to 166,000 and then to 233,000 persons. Revisions to employment data are a matter of common practice. As the BLS states: “Monthly revisions result from additional reports received from businesses and government agencies since the last published estimates and from the recalculation of seasonal factors.” In fact, during 2021, the BLS revised up eight monthly employment figures and it revised both up and down two monthly figures. (Note: At the time of this writing, data revisions for November and December had not yet been completed.)

The second ALFRED graph shows employment data revisions during 2020. In the year of the COVID-19-induced recession, the BLS revised up eight monthly employment figures, it revised down one monthly figure, and revised both up and down three others. Those revisions are harder to see in a graph because the reduction in total employment recorded during April 2020 was staggeringly large.

However, the takeaway from this post is that employment data are always revised. Economic policymakers are aware of economic data revisions and can see through “the fog of numbers” described by Jordà, Kouchekinia, Merrill, and Sekhposyan.

How this graph was created: Search ALFRED for “All Employees: Total Nonfarm Payrolls.” By default, ALFRED shows a graph with two sets of bars: the most recent vintage and the prior vintage. To change the data units, in the “EDIT BAR 1” tab, select “Units: Change.” To customize the graph type and style of the series use the “Format” panel. Add additional vintages by using the “Add Line” tab and select the date of the desired vintage from the “or select a vintage” dropdown menu.

Suggested by Diego Mendez-Carbajo.