A good rule of thumb when engaging in media criticism is to remember that the journalist usually didn’t write the tweet. So I can’t in good conscience blame Noah Smith for this tweet of his recent Bloomberg Opinion column:

The problem with this graph is that it is, in all likelihood, wrong, because the underlying data are, in all likelihood, wrong. Smith chose the wrong metric, and the tweet implies a level of certitude that is not justified by the evidence (the article itself is less credulous). The graph above gets its data from the private Payscale Index, which shows a nearly 2 percent collapse in real wages during the second quarter of 2018. We don’t yet have official numbers from the Bureau of Labor Statistics on second-quarter wage growth, but the BLS does provide monthly earnings measures and an employment-cost index that serves as a decent proxy for wage data. And nothing in either of these metrics suggests the Payscale data is anywhere near correct:

This is not to say that we should be satisfied with the current rate of wage growth. Real wages barely rose in June per BLS data, and the Tax Cuts and Jobs Act did not provide as much in the way of working-class tax relief as it could have. Many at NRO, including yours truly, have argued that the Republican party should be more concerned about the sluggish wage gains that have characterized the 2010s economic expansion than it is.

But the idea that wages are plummeting simply is not supported by credible evidence, as several economists have noted online. And if you needed a reminder that the “This one chart proves a sweeping conclusion” model of journalism has pratfalls, this should be it.