The FRED® Blog

Tracking inflation with sugar and sweets?

Inflation reflects how prices of goods and services in the economy are changing. One measure of inflation is the consumer price index (CPI), which is the common headline number reported in the media.

The numbers for the CPI are released monthly, so it can be hard to tell what the CPI is doing on a daily or weekly basis. But what if there were prices you could observe directly that would serve as a closely related proxy for the CPI?

The FRED graph above shows that the CPI for all items and the CPI for sugar and sweets move similarly. Another way to say this is that these two price indices are highly correlated. Since 1947 the correlation is 0.99, and since the middle of 2020 the correlation is 0.97. Having a positive correlation that’s this close to 1 means that, as the CPI for sugar and sweets increases, so does the CPI for all items.

So, are price changes for sugary treats an indication of price changes overall? Well, this correlation between the two indices might be spurious. As an earlier FRED Blog post described it, “because time series can exhibit a common trend, it becomes difficult to interpret whether there is a relationship between them beyond that common trend.”

One way to investigate potential spurious correlation is to see if there is also a correlation in the growth rates of the two variables. We do this in our second graph by plotting the growth rate from the month prior for both price indices.

The correlation between the growth rates is 0.97 since 2020, which suggests this correlation is sound and not spurious. So, you could track the prices of candy or cookies on a daily or weekly basis at the grocery store to gauge what’s going on with the CPI for all items.

How these graphs were created: First graph: Search FRED for and select the “CPI for All items” series. In the “Edit Graph” panel, use the “Add Line” tab to search for and select “CPI Sugar and Sweets.” To normalize the data, selecte the last option in the the “Units” dropdown menu: “Index (Scale value to 100 for chosen date)”; choose 1990-01-01 and click “Copy to All.” Display the graph since 1947-01-01, which is the common start of the two series. Second graph: From the first graph, choose Units “Percent Change,” click “Copy to All,” and display the graph since 2000-01-01.

Suggested by B. Ravikumar and Amy Smaldone.