CARTS: Chicago Fed Advance Retail Trade Summary Background

The Chicago Fed Advance Retail Trade Summary (CARTS) tracks the U.S. Census Bureau's Monthly Retail Trade Survey (MRTS) using a mixed-frequency dynamic factor model to combine high-frequency time series from five private companies (Consumer Edge, Facteus, Morning Consult, Safegraph, and Womply) as well as two federal agencies (the U.S. Energy Information Administration and the U.S. Census Bureau). By using multiple sources of information, we predict the MRTS in real time more accurately than consensus forecasts, baseline autoregressive models, and other high-frequency data. For details about how we construct the index and test its accuracy, see “Tracking U.S. Consumers in Real Time with a New Weekly Index of Retail Trade.”

Contact Us

Ezra Karger

E-Mail

In the top left panel below, we compare CARTS with the MRTS, showing that our weekly index closely tracks the monthly level data from MRTS. In the other five panels we show how retail spending (both the CARTS and MRTS) responded to the waves of Covid-19 cases in 2020 and 2021. This figure highlights the value of having a weekly retail spending index. Monthly indexes mask substantial week-to-week variation in retail spending. For example, in March 2020 our weekly index identifies a stockpiling effect, which is not noticeable in the averaged monthly spending data from the MRTS.

Figure 1 features six panels, each being a line chart, showing the paths of the CARTS and MRTS from January 2018 through the latest month for which we have MRTS data. The first panel (top left) overlays the CARTS and MRTS, showing that they closely match each other over time. Both the CARTS and MRTS rise gradually from January 2018 through February 2021 before declining sharply in March 2021 and then quickly recovering. The remaining five panels plot the CARTS and MRTS, along with gray shading indicating daily Covid-19 cases, over different time periods—specifically, for February 2020 through May 2020 (top right), June 2020 through September 2020 (middle left), October 2020 through January 2021 (middle right), February 2021 through May 2021 (bottom left), and June 2021 through the latest month for which we have MRTS data (bottom right).

Note: The left-hand scale in all six panels is index value in millions of U.S. dollars; the right-hand scale in the final five panels is the number of new daily Covid-19 cases.
Sources: Chicago Fed staff calculations based on data from the U.S. Census Bureau and U.S. Energy Information Administration, Consumer Edge, Womply, Facteus, Morning Consult, and SafeGraph; and University of Oxford, Our World in Data, from Haver Analytics.

We can also compare our index to other measures of retail spending from the U.S. Bureau of Economic Analysis (BEA) and Opportunity Insights (see “Tracking U.S. Consumers in Real Time with a New Weekly Index of Retail Trade”). Our index tracks retail spending with less volatility than the BEA measure and provides a less biased measure of retail spending than other measures constructed using retail spending data from a single source.

Figure 2 features two panels, each being a line chart. The top panel shows the CARTS and the BEA’s index of retail spending from January 2020 through the latest month of data. The CARTS and BEA index show no clear trend in retail spending until March 2020, when retail spending in both indexes drop by around 20 percent before recovering to typical levels by July 2020. Both measures of retail spending then remain at typical levels until January 2021, when retail spending increases to between 5 and 10 percent above trend before gradually declining back to typical levels in March 2021. The bottom panel shows the same patterns over the same time period, but compares CARTS with two measures of retail spending from the Opportunity Insights research group—one index includes grocery spending and the other excludes grocery spending. The Opportunity Insights index that includes grocery spending shows a shallower 10 percent decline in retail spending in March 2020 before spending recovers to 10 percent above typical levels by July 2020. The Opportunity Insights index that excludes grocery spending shows a 20 percent decline in retail spending in March 2020 and then recovers to 10 percent above typical levels by July 2020. As in the top panel, CARTS falls by 20 percent in March 2020 before recovering to typical levels by July 2020.

Note: The U.S. Bureau of Economic Analysis uses card data from Fiserv First Data, and Opportunity Insights uses card data from Affinity Solutions.
Sources: Chicago Fed staff calculations; and U.S. Bureau of Economic Analysis (BEA) and Opportunity Insights data from Haver Analytics.

Lastly, we can compare the recent (2020–21) values of the CARTS against historical trends, highlighting the sharp decline in retail spending during the Covid-19 pandemic and the subsequent rebound, concentrated around the three rounds of stimulus checks in 2020 and 2021.

Figure 3 is a line chart showing week-to-week percent changes in CARTS in January 2020 through the most recent month of data relative to historical averages. Historically, CARTS does not change from week to week. But in March 2020, CARTS fell about 10 percent week-over-week in two consecutive weeks before rebounding in mid-April, showing week-to-week growth through July 2020. CARTS was then largely unchanged until January 2021, when the index jumped about 5 percent week-over-week before returning to normal levels. This spike in retail spending in January corresponds to the timing of the second round of Covid-19 stimulus checks.

Note: The dashed vertical lines represent initial disbursals of Economic Impact Payments.
Sources: Chicago Fed staff calculations based on data from the U.S. Census Bureau, U.S. Energy Information Administration, U.S. Department of the Treasury, Consumer Edge, Womply, Facteus, Morning Consult, and SafeGraph.