07 June 2023

Economic disinformation, an example

Justin Wolfers, author of Think Like an Economist, gives a good example of  how politically motivated distortion of economic data often works. This kind of "reporting" (read disinformation) is the source of a lot of predictions of imminent recession, despite the fact that inflation is roughly half what it was a year ago (and at historic norms after a long period of exceptionally low inflation); while unemployment remains low and job creation is at historic highs, when considered over multiple months. Anyway, here's his example. In both Republican and Democratic administrations, going back at least a few decades, monthly job creation numbers are released as early as feasible using statistical extrapolation methods to yield a number, even though some data is not final. This feeds the demand for timely information. And these numbers are always adjusted around a month later when data is more complete. Over longer periods, typically, there are either slight overreads or slight underreads of the initial numbers, depending on what you might call "hidden secular trends." So numbers might be consistently overestimated for, say, six months in a row, and this might reasonably be considered as a sign of what you might call "hidden weakness" in the economy. Generally these numbers really are produced by dispassionate econometric analysts, so, in both Republican and Democratic administrations, these deviations tend to be small and genuinely caused by trends which are inherently difficult to detect. So, if it were the case over, say, the last six months, that the "Biden" jobs numbers were being adjusted downward for each month in succession, one could say, as some of the "Tech Bro" libertarian economists have done, "See, the economy is actually worse than they're reporting. Fair enough, actually. Technically right, in the hypothetical, although the effect is usually very small. But. But. Do these same libertarian axe-to-grind folks say the opposite when, month after month, the jobs numbers are actually adjusted upwards? No, they do not. They say, almost as a chorus, that the numbers are rigged and the administration is corrupt. In fact, this is the situation: jobs numbers, averaged over the last year or so, have pretty consistently been adjusted upwards, and as an indicator honest reportage must conclude that these numbers reflect concealed strength, not weakness, in the economy. (This indicator, obviously, isn't the only one, but inflation numbers are also encouraging since the beginning of the year). 

This is a good example about how having a particular vested interest in a certain view of the economy can and does result in completely misleading information being plastered all over the mainstream press. Unfortunately, economic anxiety, fed by this kind of thing, is somewhat self-fulfilling, and can create negative feedbacks that actually negatively affect economic outlook. Bottom line: stick to facts, and keep a critical view that looks for reliable factual basis in prognostications you see in the media. 

No comments:

Post a Comment

Gyromantic Informicon. Comments are not moderated. If you encounter a problem, please go to home page and follow directions to send me an e-mail.