Federal Reserve Economic Data

The FRED® Blog

Spending on tobacco products and smoking supplies over time and across groups

Last week, we used data from the Consumer Expenditure Surveys to discuss changes in what people spend on reading materials. Here, we use the same data release to look at what people spend on tobacco products and smoking supplies.

Our first FRED graph shows that, between 1986 and 2019, overall spending on tobacco/smoking decreased for the sum total of all surveyed households. We’ve adjusted the annual dollar figures by the consumer price index for these products to account for their changing price over time. (FYI: The nominal figure for these expenditures in 2019 is $320 per household.)

Last week, we sorted the survey data by age group; here, we sort them by educational attainment, shown in our second, colorful FRED graph.

The share of these expenditures decreases as the level of educational attainment increases. In 2012, households with at most a high school degree made up 47.3% of the overall spending, and households with a bachelor’s or postgraduate degree made up 15.4%. Given the consistent sizes of each colored area in the graph, we can see that the relative shares have been fairly constant, at least since data were first collected in 1996.

Be aware that these patterns across groups might cloud some of the socio-demographic factors at play here. Savvy economic research often considers those factors. And speaking of savvy, read the work of Michael Darden, Julie Hotchkiss, and Melinda Pitts for more insight into the connection between smoking, educational attainment, and wages.

How these graphs were created: For the first graph, search for and select “Expenditures: Tobacco Products and Smoking Supplies: All Consumer Units.” From the “Edit Graph” panel, use the “Edit Line 1” tab to customize the data by searching for and selecting “Consumer Price Index for All Urban Consumers: Tobacco and Smoking Products in U.S. City Average.” Next, create a custom formula to combine the series by typing in “a/b*100” and clicking “Apply.”

For the second graph, search for and select “Expenditures: Tobacco Products and Smoking Supplies by Education: Master’s, Professional, Doctorate.” From the “Edit Graph” menu, use the “Add Line” tab to search for “Expenditures: Tobacco Products and Smoking Supplies by Education: Bachelor’s Degree.” Repeat the last step to add the same series for the other education levels (associate degree, high school with some college, high school, and less than high school). Next, use the “Format” tab to change graph type to “Area” and stacked “Percent.”

Suggested by Diego Mendez-Carbajo.

View on FRED, series used in this post: CUSR0000SEGA, CXUTOBACCOLB0101M, CXUTOBACCOLB1303M, CXUTOBACCOLB1304M, CXUTOBACCOLB1305M, CXUTOBACCOLB1306M, CXUTOBACCOLB1308M, CXUTOBACCOLB1309M

What the young and old spent on reading materials: 2019 vs. 1984

Naturally, reading the free FRED Blog doesn't count

FRED now has data from the Bureau of Labor Statistics’ Consumer Expenditure Surveys, which are used to keep the consumer price index (CPI) up to date with the current basket of goods and services bought by households. Here, we use that survey data to analyze spending on reading materials.*

The FRED graph above shows spending on reading for seven age groups in 2019. Hover over the pie chart slices to see how much each age group spent relative to the total for all age groups. In 2019, persons between 25 and 34 spent the least and persons over 75 spent the most.

The second FRED graph shows the same age groups but from 1984. Back then, persons under 25 spent the least, but persons 35 to 44 and 45 to 54 spent the most. Because the dollar figures for these expenditures in each pie chart aren’t adjusted for changes in the cost of living, we can’t compare the dollar amounts in 2019 with those in 1984. To do that, we need to use the formula described at the bottom of this post.

Nonetheless, we can compare the different expenditure shares across age groups. While in 1984 overall spending on reading was more or less evenly divided across persons from age 26 to 74, by 2019 almost a quarter of it was done by persons older than 75. The growth of digital media and its uneven adoption across age groups could help explain the changing spending on reading.

*Reading includes subscriptions for newspapers and magazines; books through book clubs; e-books and digital reading material; and the purchase of single-copy newspapers, magazines, newsletters, books, and encyclopedias and other reference books, per https://www.bls.gov/cex/csxgloss.htm.

How these graphs were created: Search for and select “Expenditures: Reading by Age: Under Age 25.” From the “Edit Graph” menu, use the “Add Line” tab to search for “Expenditures: Reading by Age: from Age 25 to 34.” Repeat the last step to add “Expenditures: Reading by Age: from Age 35 to 44,” “Expenditures: Reading by Age: from Age 45 to 54,” “Expenditures: Reading by Age: from Age 55 to 64,” “Expenditures: Reading by Age: from Age 65 to 74,” and “Expenditures: Reading by Age: Age 75 or Over.” Next, use the “Format” tab to change graph type to “Pie.” To show data from different years, edit the date above the graph canvas.

Suggested by Diego Mendez-Carbajo and Maria Arias.

View on FRED, series used in this post: CXUREADINGLB0402M, CXUREADINGLB0403M, CXUREADINGLB0404M, CXUREADINGLB0405M, CXUREADINGLB0406M, CXUREADINGLB0408M, CXUREADINGLB0409M

Gauging the recovery in retail sales at bars and restaurants

The FRED Blog has discussed how, during the onset of the COVID-19 pandemic, households cut back on eating out and increased food purchases to prepare meals at home. With data from the U.S. Census Bureau, we created a FRED graph with two lines to compare sales at restaurants and bars with sales at grocery stores.* Today, we use the same Census dataset to re-examine the topic from a different perspective.

The red line in our FRED graph today shows the value of retail sales at restaurants and bars as a fraction of retail sales at grocery stores. We added the black dashed line, with a constant value of 1, to make it easy to see when the two categories of retail sales are equal. And they were essentially equal from March 2019 through February 2020.

In March and April 2020, though, consumers sheltered in place to help reduce the spread of the COVID-19 virus and a wide gap opened. Between then and October, retail sales at restaurants and bars slowly climbed back. But in November and December 2020, those sales decreased again. A second wave of virus infections and renewed social distancing, as described by Guillaume Vandenbroucke, could help explain this change in trajectory.

To learn more about the effects of 2020’s second wave of virus infections on overall economic growth, inflation, and unemployment, read the work of William Chen, Marco Del Negro, Shlok Goyal, and Alissa Johnson.

*Technically, we compared inflation-adjusted advanced retail sales from food services and drinking places with inflation-adjusted advanced retail sales from food and beverage stores.

How this graph was created: Search for and select “Advance Retail Sales: Food Services and Drinking Places.” From the “Edit Graph” panel, use the “Edit Line 1” tab to customize the data by searching for and selecting “Advance Retail Sales: Food and Beverage Stores.” Next, create a custom formula to combine the series by typing in “a/b” and clicking “Apply.” Next, use the “Add Line” tab to create a user-defined line. Create a line with start and end values of 1. To change the line colors, use the choices in the “Format” tab.

Suggested by Diego Mendez-Carbajo.

View on FRED, series used in this post: RSDBS, RSFSDP


Back to Top