Dude, stay away from those essences


People in and around markets love to mistake abstractions for things that actually exist and then argue endlessly about the “true” measure of that presumed thing.  For lack of a better term, and the deeper knowledge I wish I had, I am going to call this pervasive error essentialism.

I apologize to people with knowledge of philosophy for probably misusing that expression. I touched on it with equal amateurism in my Distorted post.  It is huge in macro and in politics.  Don’t let the fact that I am an amateur in it fool you.

My favorite example of this is the tedious, usually motivated, debate about what the real inflation rate is. Inflation is meant to measure the trend change of the general price level. The general price level is the ratio of money expenditure to the aggregate utility purchased by it. Utility is not comparable across people or therefore able to be aggregated, so there is no such thing as THE general price level or THE inflation rate.  Drop the mic.

What we call the general price level reflects a particular way of pretending to aggregate utilities, which is laden with value judgments, reasonable ones chosen in part to allow tractability, but value judgments just the same. When people say the measured inflation rate is wrong, they are often just saying they don’t accept those value judgments.  Ok. I am not going to argue about value judgments. There are also the crazies who follow Shadow Stats, from whom you should preserve your sanity by just ignoring. Let’s not go there.

This is not to say that we should ignore attempts at measuring inflation. The Consumer Price Index is used to determine social security checks and some wage rates in the unionized sector. It is also the means of indexing TIPS and inflation swaps. Meanwhile, the Fed claims to try to deliver an average rate of inflation in the PCE deflator of 2% over time.

Knowing how the the CPI and PCE deflator are tracking, then, is quite important. But once the decision to index on CPI or target PCE is made, the veracity of these measures is actually not that relevant. The veracity matters only to the extent it is perceived and influences the decisions to index with and target respectively.

Most recently, the most popular abstraction to argue about is the unemployment rate.  Within economic theory, unemployment is an aspect of the labor market that is meant (variously) to correlate with the state of the business cycle, influence people’s willingness to quit or fire, affect the relative bargaining power of employers and employees, and either influence or reflect (the surprise component of) what we call inflation.  It probably does other things too that I have forgotten to list or don’t know about.

We might be willing to believe that there is such a thing as higher or lower unemployment.  Some schools of economics would actually deny this, which means that the very idea of unemployment is model-specific.  I choose the models that say unemployment is a thing – or, more honestly, I skip the middle step and go straight to acceptance. But the idea that there is a “true” measure is essentialism, i.e., wrong or unhelpful.

Recently, a largely fake and pointless debate about what the “true” measure of unemployment is has heated up.  The causes of this are not hard to identify.  First, and practically, the Fed is threatening to do a tightening program partly on the grounds that it is near hitting the employment side of its dual mandate.

Second, it is a presidential election year and the BLS’s highlighted measure of unemployment, U3, is exactly equal to the CBO’s estimate of the “natural” rate of unemployment.  So some people are saying, yay, full employment.  I did that myself, although with implications that were meant to be a negative for the stock market, taken in isolation. Others are doing it for their own political reasons.  I would if I were them too, but need not get into that there.

The leading contender for the GOP nomination to president, Donald Trump, wants none of that and has claimed that the real unemployment rate is 42%.  The number is not all that wrong.  If you take the ratio of all adults not working to the adult population, you get near 42%, although about a year ago.

Hard-line, liberal super-doves who claim the Fed cannot raise interest rates because the employment/population ratio is too low are actually basing their argument on The Donald’s measure. So, (fellow) liberals, please spare us the claim that his (and your) number is crazy.


Trump’s choice of number is obviously politically motivated. An honest Trump – imagine a parallel universe – would put that number in the context of its own history and get into a discussion of why inexorable demographics have depressed it. But the number is not all that wrong. What is wrong is the idea that there even exists a correct number. More egregiously, Trump is confusing apples and oranges, using a measure that obviously does not correlate with what people think of when they hear the expression “unemployment.” Shame on him.

Paul Singer, who is Israel’s chief ambassador to America’s political donor class and the Master of Marco Rubio, claimed last summer that the true unemployment rate was above 8% and therefore that the Fed could not and would not raise interest rates. Oops.

The Fed did raise interest rates. And while I am with Ambassador Singer on the idea the rate rise need not go far, this theme has been extremely important in macro for the past several months.  That the Fed had reason to go has created financial market reactions that have fed back to the case for the Fed not going too quickly.  Bad thing to have missed that hugely important and reflexive dynamic.

This guy puts politics ahead of macro analysis, which is fine so long as you know, and he makes his money outside macro, and impressively so.  I find what he does with his money somewhat irritating and am hoping others will soon too. But the guy makes tons of money, just very obviously not by understanding how the economy works.

Update from later in the day: Phew! Good thing I conceded the guy knows how to make money.

Anyhow, when he mentioned 8%, he was seemingly referring to something somewhere between the U5 and U6 measures of unemployment.  I apologize for not knowing which and admit this is my ignorance, not Singer’s.  For the sake of argument, I will use U5 just because it generates a number closest to what Singer mentioned when he made the reference in my presence.  (Full disclosure: this choice strengthens my point, although only slightly.)  U5 is the augmented unemployment rate, which includes discouraged and other marginally-attached workers in both the numerator and denominator, unlike U3.

A really good estimate of U5 is U3 + 1 percentage point, as the chart below shows.  Note that the two scales are offset by a percentage points and that the lines sit atop each other.  It is actually kind of laughable.

If your interest is how changes of the labor market might influences changes of business cycle conditions that might interest you (like monetary policy, or profits, or inflation, or the stock market), there is virtually no value in the distinction between U3 and  U5.  People who reference U5 – or to a slightly lesser extent U6 – who do not tell you this are either ignorant or trying to fool you.

Ben Casselman covers economics for Nate Silver’s Five Thirty Eight blog, which is supposed to apply Silver’s rigorous empirical approach to politics and sports to public affairs more generally.  Unsurprisingly, Casselman does not make claims as egregiously misleading as those of Trump or Singer, but he really should know better than to ask in his slug “What is the Real Unemployment Rate?” or to answer that “my own dive into the numbers finds that the true unemployment rate has fallen significantly over the past two years and is now under 6 percent.”

Casselman’s piece seems to me to do a good and honest job of going over the technical issues influencing the headline unemployment rate U3 as well as the legitimate controversy around how much of the recent decline of the labor force participation rate is cyclical and likely to be reversed (allowing for quicker than otherwise employment growth) as the labor market continues to tighten.

He spots the benefit of the doubt there to the doves who claim that there is some cyclical compression of participation rate.  That is not unreasonable, particularly given that Casselman is clear that all agree that the majority of the decline of the participation rate is structural.  He is head and shoulders above the other two mentioned above.

On the other hand, he should be, because Five Thirty Eight claims that being numerate is their thing and that they are better than the hopelessly innumerate mainstream media. Holding Casselman to this standard, I would criticize him on two grounds.  First, he should not so emphasize the result of his own “dive” into this issue.  Rather, he should emphasize that others with much greater expertise have dove much deeper and much longer and surfaced with quite different answers.  They all agree that the labor market has tightened, and I would emphasize that practically speaking that is the main thing, but plenty of better informed people would disagree with Casselman.

Second, and more to my point, his use of the terms real and true shows that he is falling for essentialism.

The point raised in this post is admittedly “philosophical,” perish the thought. And a lot of people will flatter themselves to be focused on much more practical issues. Fine, more for me, assuming I eventually get back in the game, if you think obsessing over the correct measure of something that does not even exist is a good use of your time.

It seems there is a quote from Keynes for most things, so let me conclude with this insight from the master:

Practical men who believe themselves to be quite exempt from any intellectual influence, are usually the slaves of some defunct economist. Madmen in authority, who hear voices in the air, are distilling their frenzy from some academic scribbler of a few years back.

I kid you not when I assert that people are harkening back to Plato, who still does major mischief to how people talk about public policy issues. I just wish I were smarter and better educated so I could nail the point. But this ain’t about me. Essentialism is a thing, man. Stay away from it.