Dismal, yes, but is it science?

As I was driving home from work today, I was listening to Marketplace on public radio. In the middle of a story, reported by Nancy Marshall Genzer, about opponents of health care reform, there was an interesting comment that bears on the nature of economics as a scientific discipline. From the transcript of the story:

The Chamber of Commerce is taking a bulldozer to the [health care reform] bill. Yesterday, the Washington Post reported the Chamber is hiring an economist to study the legislation. The goal: more ammunition to sink the bill.

Ewe Reinhardt teaches economics at Princeton. He says, if the Chamber does its study, it will probably get the result it wants.

EWE REINHARDT: You can always get an economist with a PhD from a reputable university to give a scientific report that makes your case. So, yes, there will be such a study.

It's that comment from Dr. Reinhardt that got my attention. What precisely is he claiming about economics?

Is it that well-trained economists cannot be expected to be able to come to agreement about the economic facts about any given policy? (Why wouldn't we expect that well-trained economists would be able to ascertain a relatively stable set of facts?)

Or that well-trained economists may be able to come to agreement about the economic facts relevant to a given policy, but that they cannot be expected to come to any kind of agreement about what outcomes these facts should lead them to predict should the policy be implemented? (What makes economic predictions so wildly difficult? More than that, what would make different economists predict such wildly divergent things from an agreed upon set of economic facts?)

Or is it that just because someone has a PhD in economics from a reputable university, there's no reason to think he or she is going to generate a scientific report that other economists will find credible?

Dr. Reinhardt describes what an economist might generate in this situation as a scientific report; I'm trying to connect the dots here to figure out just what kind of science economics is presumed to be.

More like this

Uwe Reinhardt is a great guy, superb health economist, and excellent instructor. I accidentally ran into him at Duke Hospital (he's on one of their boards for the med school, univ, or both) and he was cool enough to spend 20 min with me over coffee explaining the concept of quality-adjusted life years (QALYs) as a measure of how some countries (like the UK) have decided to limit public expenditure on health care.

I am NOT an economist but what I understand from him is that economists employ very diverse models toward the "systems biology" of the economy that are of complexity that each can produce highly divergent outcomes. The methods are imperfect. However, you seem to be asking a more political question - my guess is that an economist will be brought onboard whose model system gives the answer that is desired.

btw, a prof of your stature might be able to get a direct response from Uwe - he also writes every week or two at the New York Times Economix blog. Perhaps I'll head over and see if he speaks toward this very issue:


P.S. I also just realized that his March 20 post on QALYs has the figure he drew for me on a napkin in the Duke Hospital cafeteria over two years ago:


Well physics is a hard science, and sociology a soft science - so given the descriptors relate to the physical properties of materials I guess that economics might be best described as a malleable science

As an economist, I would make the following comments:

1. Although economists do know some things and generally agree on some things, some key behavioral parameters and causal links are less firmly established than is true in the natural sciences. This is because it is generally more difficult to do random assignment experimentation in economics, and some of the outcomes are harder to measure. Therefore, economists, compared to natural scientists, have to rely to a greater extent on so-called "natural experiments", in which other factors than what is at issue can cause the observed correlations.

2. Many public policies related to economics depend to a greater degree than in the natural sciences on values and ethical propositions. Two economists may agree on what will be the effects of a public policy, yet totally disagree on whether it is desirable. For example, a policy may be economically "inefficient", in the technical meaning that term has in economics, yet be a good policy because its effects on the distribution of well-being across individuals are thought to be worth the inefficiency, based on a certain set of values.

3. Therefore, as some of the facts are more in dispute, and as so many of the recommendations depend to some degree on ethical values, some economists are sometimes tempted to let their political beliefs shape which factual beliefs they find most convincing. I think most economists try to avoid this, although it is difficult to do so.

Keynes had an interesting quote about economists compared to physicists which seems relevant here:

"Professor [Max] Planck, of Berlin, the famous originator of the Quantum Theory, once remarked to me that in early life he had thought of studying economics, but had found it too difficult! Professor Planck could easily master the whole corpus of mathematical economics in a few days. He did not mean that! But the amalgam of logic and intuition and the wide knowledge of facts, most of which are not precise, which is required for economic interpretation in its highest form is, quite truly, overwhelmingly difficult for those whose gift mainly consists in the power to imagine and pursue to their furthest points the implications and prior conditions of comparatively simple facts which are known with a high degree of precision. "

Keynes then goes on to say the following:

"The study of economics does not seem to require any specialised gifts of an unusually high order. Is it not, intellectually regarded, a very easy subject compared with the higher branches of philosophy and pure science? Yet good, or even competent, economists are the rarest of birds. An easy subject, at which very few excel! The paradox finds its explanation, perhaps, in that the master-economist must possess a rare combination of gifts. He must reach a high standard in several different directions and must combine talents not often found together. He must be mathematician, historian, statesman, philosopherâin some degree. He must understand symbols and speak in words. He must contemplate the particular in terms of the general, and touch abstract and concrete in the same flight of thought. He must study the present in the light of the past for the purposes of the future. No part of man's nature or his institutions must lie entirely outside his regard. He must be purposeful and disinterested in a simultaneous mood; as aloof and incorruptible as an artist, yet sometimes as near the earth as a politician. "

It is not surprising that some economists fall short of having all of these skills in the right balance, and end up using the field to justify their own political biases.

By Tim Bartik (not verified) on 17 Nov 2009 #permalink

As a physical scientist who goes to the occasional economics seminar at a prominent academic institution, I would like to chime in by noting that academic economics seems to be primarily modelers attempting to do the impossible and a minority expressing opinions. Ironically, those expressing opinion based on a deep-seated knowledge of the past appear to have the best shot at predicting the future. In this world, the Austrians appear to be gaining the high ground after years of obscurity.

I expressed concerns about what I was seeing inside the Ivy-covered walls to an acquaintance in the more pragmatic world--a prominent economist at Morgan Stanley--I was told that "academic economists are a woeful bunch." This is a little too sweeping, but the idea comes through.

Well, this is something that happens with almost every application of ANY science, when it does not refer to just laboratory data or/and it has political implications. It's not a problem of economics, but of 'science in the wild'.

Tim Bartik gave an an excellent and detailed explanation. Since I am not an economist, I won't even try to add to that.

Some great reading on the topic is by Nassim Nicholas Taleb - The Black Swan, which anticipated the recent banking problems, and Fooled By Randomness. Both explain many of the problems with economic science and the hubris of attempting to control something with so many variables (many of which are not even agreed upon).

I don't think it says much of anything about economics. For decades, cigarette companies were able to come up with medical researchers who parroted the company line on smoking and health. This didn't show that medicine isn't science. It showed that people can be bought.

Arthur Petersen, a Dutch philosopher of science, has done some very nice work looking at the uncertainties and need to deploy significant assumptions in modeling complex systems, and the way this contributes to the politicization of science. Basically, in a system with lots of different processes that all influence each other in various complicated ways, you have to black-box or guess at what's happening in some of those processes in order to build your model. If you're lucky, you can make highly qualified predictions based on ranges: if variable X is between a and b, then variable Y will most likely be between c and d. If you're unlucky, you'll get models that point in completely different directions, and no easy way to figure out which is right until it's too late to use the model to make a policy decision. And that, in turn, means that people with an interest in one set of policies will be able to point at some models, people with an interest in a completely different set of policies will be able to point at other models, and science doesn't give us a way to settling the matter in the amount of time we have before we have to make a decision.

Okay, that was all very abstract, but it's basically what's being said about economics in this thread: building economic models requires making significant assumptions and simplistic representations of complicated processes, and as a result libertarians can point to various models that say (for example) fiscal stimulus is at best wasteful, while Keynesians can point to various models that say (for example) fiscal stimulus is the only hope for warding off long-term economic stagnation. In other words, I pretty much agree with what Tim Bartik said about economics.

Except that Petersen doesn't work on economics. He has a Ph.D. in climate science, was a member of the IPCC, and is interested in how all the stuff above plays out in the science and policymaking surrounding climate change. Hence, as a result of the sheer complexity of atmospheric chemistry and physics, Bjorn Lomberg (and the libertarians at Reason and the Superfreakonomics authors and so on) can point to various models that say cap-and-trade is at best wasteful, while environmentalists can point to various models that say that cap-and-trade is the best hope of warding off long-term significant warming effects.

It is not surprising that some economists fall short of having all of these skills in the right balance, and end up using the field to justify their own political biases.

More importantly, given the intersection of economics and public policy it is not surprising that it would be politically convenient for certain factions (the Communists are the most famous historical example, but this also applies to the political right) if certain economic propositions (the exact ones depend on faction) were true. As a result, there are so-called think tanks (the Heritage Institute and the American Enterprise Institute are two of the more notorious ones) which will hire economists who share their political disposition to produce studies supporting those propositions. So Dr. Reinhardt's statement is absolutely correct.

@Thomas: I am also a physical scientist, and I don't find it ironic that the empirical economists are more successful at prediction than the modelers. A model cannot be any better than the assumptions that go into it, and as Tim implied, many economists question the reasonableness of many of those assumptions.

By Eric Lund (not verified) on 18 Nov 2009 #permalink

If you don't (or can't) put error bars or confidence limits on your projections, then it isn't science.

The tobacco industry didn't put confidence limits on their assertions that smoking didn't cause cancer. The global warming deniers don't put confidence limits, the Creationists don't use confidence limits, CAM quacks don't use confidence limits.

The most insightful observation I ever heard about economics: (unattributed)

"If you placed every economist in the world end-to-end, they'd never reach a conclusion."


By Roadtripper (not verified) on 18 Nov 2009 #permalink

The report is "scientific" in that it will appear to the outside observer to conform in some sense to the usual anthropological practices known as "science".

Whether the contents ought be considered "scientific" in a philosophical sense? A little more doubtful.

Whether in whatever sense the "science" is of a caliber for doing reliable Engineering with? Very, very, very doubtful.

I think that the comment has more to do with a cynicism about the value of any research which is funded by any entity perceived as having a biased interest in a particular outcome. The way I heard the same comment was a disparagement about the ethics of both the Chamber of Commerce and any economist who might take their study.

By Kristine Carlson (not verified) on 18 Nov 2009 #permalink

Economics is a science if we redefine science as "pick a number and work backwards"

Economics has more to do with theology than science.

By antipodean (not verified) on 18 Nov 2009 #permalink

Physicists can no more agree upon the fate of a crash test dummy in an automobile accident than economists can agree upon the outcome of a particular health care reform. Physicists and engineers smash automobiles because the results of even classical physics theory are endlessly debateable while an experiment is definative.

With enough time, money, mathematics, and assumptions; the "exact science" of physics will not eliminate the need for crash test dummies anytime soon. Likewise healthcare reform is going to have to do the social experiment. And then the physics and economic experimental results will be endlessly interpreted. Did the crash test dummy die because of too weak a bumper or too strong an engine mount?

I don't know about CAM quacks, but quacks in general do use confidence limits: when warrants are issued for their arrest, then the confidence limit for being unable to distinguish their products from placebo has been passed. It's hard to quantify beforehand, though.

Oh. That's not what you meant? Oops.