The Post Hoc Fallacy

According to How to Lie with Statistics, written by Darrell Huff, The Post Hoc Fallacy goes like this:

If B follows A, then A caused B.

Interesting and possible, for sure, but not certain. An example he shares is that of college graduates and their higher annual income, on average. He points out that it is quite possible that it is quite possible that these graduates may have earned more even if they did not attend college.

What is not thought of and considered is often as important as what is.

Now what about unemployment and inflation. It often follows that when unemployment is low by historical measures, inflation tends to rise soon after. This is caused by employees getting new, better paying jobs due to the fact that there are less unemployed people around. Low supply of quality workers = an increase in wages.

What are we not considering, though? What we are surely missing is the fact that many of the job gains in recent years have been of the low quality variety.  As the graphic below shows, wages are, in fact, starting to pick up.


The bottom panel goes on to display the U6 (underemployment+unemployment) and U3 (unemployment). What we are seeing is a draw-down from the underemployed. This is nearing its mean. When steadily below the average for some time and rising, this shows a possible indicator for looming economic disappointment. As of now, and according to this alone, it looks like the skies are clear.

To be clear, despite “fantastic” and “terrific” headline economic numbers, it pays to look even just below the surface to gain insight into considerations we may lack. While the popular U3 unemployment rate may seem to lead to inflation, it is perhaps the fact that future engineers and teachers are waiting tables on the U6 report, waiting for their chance to create wage growth instead of their elder counterparts who have not adequately saved for their golden years.



Sources: WSJ Daily Shot