Anti-empiricism is not humility

"Faith is the substance of things hoped for, the evidence of things not seen."
- Hebrews 11:1

Empirical economics is taking over the profession. It's very hard to make it in the field these days without doing a hefty amount of empirical work. Lots of job market papers are still theory papers (cough! signaling! cough!), but the number of economists who can make it as pure theorists is shrinking to a rarefied, brilliant sliver.

I see that as a very good thing. That's what natural science looks like - a small number of theory papers, supported by a very large base of applied theory and empirical work. It's the sign of a mature field. I also think it's going to be very important in helping the economics profession recapture some of the public respect that it's lost over the last decade. When people start to see economists as fact-driven scientists grounded in observable reality, rather than mathematical philosophers dispensing Olympian received wisdom, the profession will lose much of its accumulated stigma. And almost every young economist I talk to thinks the same - everyone's excited about new data sources. The kids these days seem to want to know facts about the world, instead of just "organizing their thinking" with models. The future looks bright.

But not everyone is on board. A few older folks, who grew up during econ's Age of Theory, are not so happy about the change. One of these is Russ Roberts, host of the excellent podcast EconTalk. In a recent blog post, Russ explains at length why he thinks the new empirical economics is overrated:

A lot of professional economists...will tell you how many jobs will be lost because of an increase in the minimum wage or that an increase in the minimum wage will create jobs. They will tell you how many jobs have been lost because of increased trade with China and the amount that wages fell for workers with a particular level of education because of that trade... 

[T]here is no simple way to resolve differences in analysis done by professional economists...[T]there is no way of knowing reliably if the consensus reflects the truth...Most economics claims are really not verifiable or replicable... 

I am arguing that the math and science of economic predictions and assessments are nothing like the math and science of space travel. Economics provides the illusion of science, the veneer of mathematical certainty...

He even goes further, and says that empirical economics isn't even really economics at all:

[M]ost of the people I am talking about are not economists. They are really applied statisticians. Economics is primarily a way of organizing one’s thinking in considering incentives and costs and the interactions between individuals that we call a market but is really emergent behavior with feedback loops.

Adam Ozimek has a patient and reasonable response to Russ, noting that even when empirical economics doesn't settle questions definitively or provide reliable point estimates, it narrows the scope of debate and rules out obvious wrong answers. That's certainly true. But I want to go further than Adam. The alternative to empiricism in economics is not agnostic humility, but intuitionism - the idea that we can know about the world by thinking about how it works, and that exposure to evidence will only pollute the truths that we divine from our own minds. And that's something I think economists need to avoid.

Consider the minimum wage issue. Suppose that a city like Seattle is considering hiking the minimum wage. How can we - economists, policymakers, and the general public - predict what the effect of the hike will be?

One approach would be to use theory. Basic Econ 101 labor supply-and-demand theory tells us that the effect will depend on the elasticities of labor supply and demand, which have to be estimated empirically. An economic geography theory might predict that the effect will be overcome by the strength of agglomeration effects, and therefore small. A search theory might predict that search frictions will preclude any sort of large short-term effect in labor markets.

How about stylized facts? Russ says that stylized facts are the only things that economists can really "know":

It is useful to know that 40% of the American work force was in agriculture in 1900 and now the number is 2%. It is useful to understand that that transition (which was most faster in the first half of the 20th century than the last half) did not lead to mass unemployment and starvation. There are indeed roughly 5 million fewer manufacturing jobs today than in 2000.

OK. So what do the stylized facts tell us about the minimum wage? Well, they tell us that places that raise the minimum wage don't tend to lose jobs. Look throughout American history. You won't find any cases where there was a big minimum wage hike and the unemployment rate soared. If we rely on stylized facts rather than careful controls and natural experiments, we'd conclude, as minimum wage proponents do, that the minimum wage isn't dangerous.

A third option is to rely on the kind of empirical studies Russ pooh-poohs. Most empirical studies say the short-term impact of the minimum wage on employment is small. 

A fourth option is to rely on casual intuition - not really theory, but a sort of general gestalt idea about how the world works. If we're of a free-market sort of persuasion, our casual intuition would tell us that minimum wage is government interference in the economy, and that this is bound to turn out badly. Russ seems to be advocating for this when he writes that "economics is primarily a way of organizing one’s thinking in considering incentives and costs." "Organized thinking" seems like just another term for intuition. 

As I see it, the fourth option is by far the worst of the bunch. Theories can be wrong, stylized facts can be illusions, and empirical studies can lack external validity. But where does casual intuition even come from? It comes from a mix of half-remembered theory, half-remembered stylized facts, received wisdom, personal anecdotal experience, and political ideology. In other words, it's a combination of A) low-quality, adulterated versions of the other approaches, and B) motivated reasoning. 

If we care about accurate predictions, motivated reasoning is our enemy. And why use low-quality, adulterated versions of theory and empirics when you can use the real things?

As I see it, a rational predictor should use a combination of theory and empirics. But theory should also be informed by data - there are lots of theories, and in general they can't all apply to the same situation, so you need evidence to tell you which one(s) to use. So a rational predictor's predictions should always be tied as closely as possible to empirical evidence. Discounting empirical evidence, as Russ does, seems inevitably to lead to the use of casual intuition (or to even worse things, like pure ideology).

Anyway, just in case you were curious, Seattle went ahead and hiked the minimum wage, and whether you measure by stylized facts or carefully controlled empirical studies, any negative effect on employment was small or zero. Of course, if you want, you can say that the empirical studies weren't controlled well enough, and the stylized facts are illusions, and the minimum wage hike must have hurt employment because government intervention always hurts employment la la la I can't hear you, but if you say that, who's going to respect you intellectually?

Now I want to turn to a second claim: the idea that discounting evidence represents "humility". Russ writes:

We economists should be more humble and honest about the reliability and precision of statistical analysis.

John Cochrane, in a blog post praising Russ' post as an exercise in "economic humility", writes:

[L]et's call [Russ' attitude] Hayekian humility. This is the hardest one for so many economists to admit, as we all like to play central planner.

This seems to be a bit of a change from when John wrote that "the stars in their 30s are scraping data off the internet." Or when he himself got famous and respected partly for doing careful empirical studies of asset prices. But anyway.

We'd all like economists to be more humble, right? Sure, count me in. But discounting empirical evidence in favor of "organized thinking" is probably not what most people have in mind when they call for economists to be more humble. 

Which is more humble: To try as hard as you can to assess the facts? Or to throw up your hands and say we'll never know the facts for sure, so we should rely on our own intuition about what people's incentives are? 

Russ writes:

[A]n economist when considering a policy of banning autonomous vehicles...would think about...how such a ban will effect the incentives to discover future innovation that might also people out of work. We would think about how putting more power in Washington would encourage lobbying for protection...These ideas are not rocket science. But they come easily to economists and not so easily to non-economists. Thinking like an economist is very useful.

Does that sound humble to you? To me it sounds like the exact opposite of humility. To say that an economist has special insight into simple ideas sounds to me like the opposite of humility. To say that an economist's intuition can yield an understanding of the incentives governing innovation, or the effect of lobbying, and that checking this intuition against the facts would only pollute the truth it yields, sounds to me like the opposite of humility.

Anyway, one final point. Russ cites an empirical disagreement between David Autor and Jonathan Rothwell over the impact of trade on jobs. He writes:

Is Rothwell correct? I have no idea. Here is what I do know. There is likely to no way of knowing which view is correct with anything close to reliability or certainty.

No! No, Russ, you do not know that there is no way of knowing who's right. How could you possibly know that it's impossible to know something?? You can't prove a negative! This is the argument-from-ignorance fallacy. Just because a matter isn't settled doesn't mean it can't be settled.

But even worse than argument-from-ignorance would be an argument-from-personal-ignorance. It doesn't sound to me like Russ has tried very hard to determine the particulars of the Autor-Rothwell dispute. It doesn't sound like he has read the papers closely, studied and understood the statistical methodology, or done anything other than observing that the two researchers disagree. I don't want to put words in Russ' mouth here, but "two people disagree, so there must be no way to tell who's right" is pretty anti-rational.

Imagine if two researchers did experiments to determine the mass of the electron. The first researcher says the mass is 9.1e-31 kg, and the second says it's 4.6e-31 kg. After hearing these two conflicting results, do you say "Here is what I do know. There is likely no way of knowing the mass of the electron with anything close to reliability or certainty."???

No. That is not what you say. Not if you're rational, at any rate. If you're rational, you might say "Let me take a look at these two experiments and see if one of them got something wrong." Or you might say "I'm going to wait until scientists figure out which one of these two experimenters got something wrong, and defer judgment until then." Or you might even say "I trust one of these labs, since they have a great track record, so I'll tentatively favor their result until more evidence comes out." But what you would not say is "Huh, it must be impossible for physicists to determine the mass of the electron."

So I believe economists can do a lot better than Russ seems to think. They can do better than relying on intuition and throwing up their hands at any empirical disagreement. And by and large, they are doing better. Let's hope that trend continues, and doesn't regress.