A detail

My habit is to use the inflation measure FRED calls CPIAUCSL because I memorized the name of it and because it was FRED's most popular inflation measure when I was looking for a name to memorize. (It still is their most popular inflation measure.) But that inflation measure gives the CPI "for all urban consumers". 

I figure most consumers are urban consumers. I used the urban measure for a long time without worrying that it might be different from the national measure. But now I have decided to take a look at it, and that is our adventure today.

On the graph I show the first dataset as a wide, bright blue line and the second one narrow and bright red so I can see how the one line matches the other. But it is easier to see if the image is bigger. To see a bigger version of the graph below, click on the graph image. To access the graph at FRED, click the word "Graph" in the caption.

Graph #1

By eye, red and blue differ a little in 1954, 1962, 1967, and 1970. But the two lines are similar enough that I figure the "urban" measure is a good match to the "national" measure. I should say, though, that I don't really know if the blue is a national measure. It sounds national from the dataset title "Consumer Price Index in the United States" but if it is a national measure, I don't really know.

The values used for the graph are annual.

The red line starts in 1948 and ends with the most recent annual data to date, 2021. The blue starts in 1951 (I don't know why) and ends in 2012; this series was "discontinued". I like it because the red line sticks out at both ends. I think that makes the graph easier to read.

Not that it matters, but FRED's name for the blue dataset is USACPIBLS. I didn't memorize this one, but it is easy enough to parse: For the USA, it is a measure of CPI from the BLS. Sometimes the names are easy like that.

If you remember the 1970s inflation, you know "CPI" is "Consumer Price Index"; it was on the news all the time. With this current inflation I don't know that they emphasize "CPI" so much. 

And BLS is the Bureau of Labor Something.

 

I'm not happy that I don't know if the blue line is supposed to represent "national" data. So I found another dataset that sounds like it might be national: "Inflation, consumer prices for the United States". I don't know if this one represents national data, either, but I figure two weak links are better than one because I'm using them in parallel, not in series: I'm using two chains to carry this analysis, each with one weak link.

FRED's name for this one is FPCPITOTLZGUSA. I didn't even try to parse it.

For this graph I used the one above, at FRED, and just swapped datasets. The colors and line widths worked themselves out automatically.

My second chain:

Graph #2

This time the blue line starts in 1960 and ends with the red one in 2021. And here, blue matches red so well it makes me wonder if they just used the red data and cut off the first few years to make the new dataset look different.

So I still don't know about urban vs national. But I'm going to keep using the "for all urban consumers" measure of inflation, and I'm gonna figure that it works as a measure of inflation nationwide.