« Economic base case: spreading slowdown & wage squeeze on profits | Main | Monetary / leading indicator signals still negative »

An example of the misuse of economic statistics

Posted on Wednesday, September 5, 2018 at 09:59AM by Registered CommenterSimon Ward | CommentsPost a Comment

A colleague was sent the first chart below, which shows that US after-tax corporate profits peaked six quarters before the onset of the 2008-09 recession and 15 quarters before the 2000-01 recession. With profits reaching a new record in the second quarter of 2018, according to data released last week, the suggestion is that any recession is unlikely before late 2019 at the earliest.

Why does this chart represent a misuse of statistics? An immediate question is why it covers only two recessions. The second chart extends the sample period back to 1985 to encompass the 1990-91 recession. Profits peaked in the quarter immediately preceding the start of that recession. Allowing for a one-quarter reporting lag, there would have been no recession warning from a fall in profits.

The chart, moreover, uses the current vintage of profits data. Profits numbers are frequently revised heavily, making it important to check that real-time data showed a similar profile. The third chart adds real-time profits* as reported in the month immediately preceding the start of the 2008-09 and 2000-01 recessions (i.e. December 2007 and March 2001).

Unlike the current vintage data, real-time profits had not fallen when the 2008-09 recession began and had declined for only one quarter before the 2000-01 recession.

Contrary to the suggestion of the creator of the chart, the new high in profits reported for the second quarter offers no reassurance that a recession is distant.

*Source: St Louis Fed ALFRED database.

PrintView Printer Friendly Version

EmailEmail Article to Friend

Reader Comments

There are no comments for this journal entry. To create a new comment, use the form below.

PostPost a New Comment

Enter your information below to add a new comment.
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>