A Non-Technical Guide to Econometrics

Chris Auld has an excellent piece on his blog regarding interpreting the “competing” Seattle minimum wage studies from the University of Washington and UC Berkeley.  It’s long, but very much worth the read.  In fact, it’s probably the best short introduction to statistics/econometrics I think I’ve read (another great one is Chapter 1 of Robert Abelson’s Statistics as a Principled Argument.  I’m also a big fan of Angrist & Pischke’s Mastering ‘Metrics).

Allow me to highlight two items in particular from this blog:

There is no statistical magic which can fully overcome these fundamental [causal] problems.  We will never be able to “prove” what the effect of the minimum wage was: that’s not the way statistics work in general, and in a case study like `what was the effect of the 2015 increase in minimum wages on employment in Seattle?’ the best we can hope for is to bring some suggestive evidence to the table. [Emphasis added]


In other words, what they Berkeley team means when they report “no effect” on employment is not that there is no effect on employment (yes, that is confusing).  What they mean, again, is that there is no statistically significant effect on employment, whereas the UW team, using different data and somewhat different statistical methods, finds a statistically significant effect.  But the difference between statistically significant and statistically insignificant is often itself not statistically significant.

One team found there were no statistically significant effects on employment, but that result should not be misunderstood as a claim that the study “proves” the effect was actually zero… [original emphasis]

Any additional commentary I add here will only detract.  Read Dr. Auld’s post.  It’s excellent stuff.

H/T: Michael Enz

Minimally Critical

Below is a letter to Ben Zipper and John Schmidt of the Economic Policy Institute:

Dear Sirs,

In your June 26th report on the University of Washington’s minimum wage working paper on Seattle, you claim:

One initial indicator of these problems is that the estimated employment losses in the Seattle study lie far outside even those generally suggested by mainstream critics of the minimum wage (see, for example, Neumark and Wascher [2008])—as the authors themselves acknowledge.

With respect, sirs, this is a rather weak criticism.  In fact, it’s logical that the results of the study are outside the norm given that the Seattle wage hike itself is outside the norm; it’s an unusually strong hike.  Given that the wage hike is higher than previous studies, it’d signal to me methodological or empirical issues if the results weren’t outside the typical range.

In short, your criticism amounts to saying “it cannot be true slamming on the breaks causes an abrupt stop because tapping the brakes causes just a slight deceleration!”


Jon Murphy
George Mason University
Fairfax, VA

Update: an eagle-eyed reader caught a spelling error.  It has been fixed

What-ifs and Whatnot

Two friends are sitting by a pool on a hot day.  One of the friends, Joe, casually says to Smith (the other): “Smith, it sure is a hot day today.  I hope the sun doesn’t dry up all the water!”

“Don’t be stupid!” says Smith.  “The sun doesn’t cause water to evaporate.  It causes the water level to go higher!”

Joe looks at his friend perplexed.  Smith continues:

“It’s real simple.  The sun hits the water, water gets warmer and starts to evaporate, right?  So the pool master comes out and adds more water to the pool.  On net, the water rises!  Ergo, the sun causes a higher water level!”

Joe, still confused, says “No, that’s not true.  The effect of the sun is to evaporate the water.  The pool master coming in is serendipitous; it’s a ‘what-if’.”

Smith laughs.  “Oh Joe.  Don’t be so dogmatic in your thinking!  Always insisting that the sun causes evaporation!  But I have clearly proven that wrong.  These chemists who constantly insist evaporation occurs because of the sun are just ideologues.”

Joe, rolling his eyes, goes back to his book.


And so it is with minimum wage, too.  Minimum wage advocates love to construct all kinds of “what-ifs” to explain why minimum wage has no effect (or even a positive effect) on employment.  But by doing this, they hide the effect of minimum wage behind all sorts of stories and claim, then, they have turned theory on its head.  But constructing what-ifs are easy.  Any storyteller can do it.  But what-ifs and serendipity make poor bases for public policy.