Misleading Economic Reporting

In today’s Wall Street Journal I found one of the worst examples of economic reporting I have seen in a long time – bad enough that I just had to get this out on a Saturday night.

The topic was the Labor Department’s announcement for employment in November. The actual one-month change in total employment was about 245,000 workers. One would think that in the middle of one of the worst economies we can remember, that the press would stand up and applaud 245,000 more jobs in one month.

The article wrote about “fitful” efforts to pass another aid passage were now even more necessary. The article correctly pointed out that the 245,000 in November was smaller than half of the 610,000 added in October. The article quoted unnamed pols who cried that the 245,000 number is “further impetus for quick action." Apparently some at the WSJ know how to calculate halvesies.

Unfortunately, those people don’t know a thing about the real world. For example, follow some of the numbers over the last few years.

2018 and 2019 were fairly normal years. In 2018 employment increased by about 2.3 million persons. In 2019 it increased by another 2.1 million workers. Divide those numbers by 12 months and you get a monthly average of around 180,000 more jobs each month. So, in pretty typical years we could think of 180,000 jobs increase as pretty good or at least average. Compared to 180,000 November’s 245,000 doesn’t seem too shabby . It doesn’t seem like that number presages cataclysm.

But wait, the article crowed that 245,000 was a lot less than Octobers’ 610,000 increase. But what they don’t either understand or don’t want to say is that neither the 610 nor the 245 should be compared to previous averages or to each other.

In March and April, US employment fell by a total of 22 million jobs. A decline of 22 million jobs in two months is nothing short of crazy. The Covid-associated shutdown helped to create an impossible number to plot on any graph. Of course – of course my dear WSJ friends – after a loss of 22million jobs in two months, you might expect something equally weird to follow that.

And it did. Between May and December of 2020, employment bounced back by 12.3 million jobs. That did not make up the whole decline and that makes sense because we do not think we are clear of Covid. The 610,000 in November was simply part of an expected partial bounce back. The months directly before November showed ever larger increases. For example, in June of 2019, employment increased in that one month by almost 5 million jobs.

What’s the point? The point is that November’s number of 245,000 increased jobs tells us almost nothing. It is not a bad number as monthly averages go. It is not enough of a number to help us to get back to were we were before Covid. It is also not a terrible number that should make us to want more and more stimulus from Washington. It is not alarming in any sense. What's wrong is that we have Covid plaguing us. November was not anything to cry over. If anything, it was reason for some optimism. 

Why do journalists have to misinterpret reality? Is it that hard to sell advertising dollars that they have to turn respectable vehicles of news into purveyors of economic porn? I can see the New York Times publishing rubbish like this as part of its ideological thrust. But I thought the WSJ was above this kind of tomfoolery.