Jim Manzi and a Conversation With My Doctor
“Thanks for seeing me, doctor. I’ve been having this persistent pain in my left hip, sometimes extending down the leg to the ankle. Could you help me with that?”
“How long have you had the problem?”
“It’s been several months, and it’s been getting worse over time.”
“Did you have any trauma to your leg that you recall?”
“No. I had knee surgery a couple of years ago, but that was the other leg.”
“Has there been any change to your regular activities that you think could have affected your left leg?”
“Not that I can think of.”
“Does any activity you engage in exacerbate or alleviate the pain?”
“Driving makes it worse, definitely, and sitting often does in general. Walking or flexing my gluteal muscles sometimes helps, but usually only temporarily.”
“How are you treating the pain when it occurs? Does the treatment help?”
“Usual with ibuprofen, and yeah, it often helps, but not always, and the pain keeps recurring.”
“Well, let’s take a look.”
[Various manipulations of leg, knee, hip, interspersed with “does this hurt” and “how about now.”]
“My best guess is it’s an inflammation of the bursa. I’m going to prescribe a cortisone injection. It’ll probably take a few days to know if it’s been effective, so why don’t you call the office a week from today and let me know if the pain has recurred. It would be particularly helpful for you to drive the day before you call to see if the pain recurs under conditions that have tended to exacerbate it in the past.”
“Thanks, doc.”
Jim Manzi is fond of making unfavorable comparisons between economics and physics, and those sorts of comparisons are pretty good for making economics look bad. But I think a much better comparison is of economics to medicine. How does economics stack up in that comparison?
Well, is medicine a science? Doctors certainly have a lot of scientific knowledge. But there’s also a great deal they don’t know. And much of their actual practice involves operating in the area where knowledge is limited.
Let’s take my example above. Turns out the cortisone injection didn’t work. Does that mean I don’t have bursitis. It might mean that. But it might not: the injections don’t always work. What else could I have? Could be sciatica. Could be . . . some other pain of unknown etiology. Is there any way to determine conclusively? Not really. Are there other treatments that could be tried? Sure: physical therapy, different pain medications, acupuncture. Wouldn’t we need a definitive diagnosis to prescribe any of these therapies? Not really – none of them are dangerous, so we can just try them and see what works. But how can we know if they did work? Well, we’ll know if the pain goes away, but we won’t know definitely if the therapy did the trick, or how it could have done so – in some cases (e.g., acupuncture) we have only very limited understanding of how the therapy works at all, but doctors still prescribe it because in their extensive experience it “works sometimes for some people.”
So if the doctor doesn’t know for sure what I have or how to treat it, why should I even consult him as an expert? Presumably because I believe that his extensive training in his field, his training in medical science, that is, qualifies him especially well to make good judgments even in the numerous areas where there is no scientifically rigorous answer to a given question, only better and worse guesses as to the probable answer.
The technical expertise a doctor has is not the same as the technical expertise a physicist has – and, more importantly, we rely on the doctor developing a general capacity for good judgment in his field in a way that we do not rely on a physicist to develop in hers. But the doctor is also different from a historian. I would be very surprised if anybody actually relies on historians to help them make specific decisions. Rather, we rely on historians as repositories of a certain kind of knowledge that, we believe, is important for a society to retain and develop: for the sake our own social cohesion, to develop an appreciation of the texture of other societies, and because it is useful for developing practical wisdom in general (and I’m sure there are other good reasons – I’m not trying to be exhaustive here).
Economics is frequently taught and described as if it is physics, which would make policymakers more like engineers applying well-understood and tested principles in specific situations. But I would argue that it’s more like medicine – another field where what we know with the kind of scientific rigor that would pass muster for a physicist is dwarfed by what we know only imperfectly if at all, but where decisions have to be made constantly, and where we have strong reasons to believe that the knowledge and experience of the experts is very useful in helping us make those decisions.
Of course, medicine is very different today than it was a hundred years ago, to say nothing of two hundred or more years ago. That’s not to say that medicine had no meaning before the advent of modern antibiotics – surgery, for example, had some of the character of a scientific discipline even in antiquity. But plenty of discarded practices – bleeding, cupping, etc. – were once medically mainstream. I would interpret Manzi’s skepticism about, for example, the effectiveness of the stimulus to be akin to someone’s skepticism about whether some modern aspect of medicine – psychopharmacology, let’s say – isn’t, in fact, utterly without foundation (and there are certainly such skeptics). There’s a place for that kind of skepticism, for tough questions about whether this particular practice is actually founded on any kind of knowledge. But that, it seems to me, is the right way of formulating the question. Not, “can you prove this is true to the standards a physicist would accept” but “if I look at the kinds of evidence for those aspects of the field that are overwhelmingly agreed upon and relied upon regularly and routinely in all kinds of decision making, how far are we from those standards when we talk about contested propositions like this one about the stimulus?” Asking that question requires a considerable understanding of what is generally accepted as knowledge within the field – it’s not really a generally skeptical question at all.
And I don’t see a lot of point to a general skepticism about economics as a practical science, any more than I see the point of general skepticism about medicine. If you have a pain in your leg, you have to do something about it – even doing nothing is doing something. In general, I’d argue that the right thing to do if the pain is persistent and serious is to go to the doctor. If you are the head of the Federal Reserve, you have to have a monetary policy; if you are the Secretary of the Treasury, you have to have a fiscal policy. You can’t not have one. In formulating those policies, I should think the most important recourse you have is to economists (one of whom you are likely yourself to be).
With respect to the stimulus, the first question, it seems to me, is not “do we know that this bill will improve conditions with a sufficient degree of certainty” because we do not know with sufficient certainty whether not passing the bill will be superior to passing it – setting a barrier whereby “no action will be taken unless certainty is above a given level” is just status quo bias with some fancy windowdressing. This, I think, is what is really annoying to some of Manzi’s interlocutors: the stimulus bill wasn’t a peer-reviewed paper, but a practical decision, and it’s not at all clear that a strong status-quo bias has a theoretical justification in the face of a massive change in objective conditions. Moreover, I strongly suspect Manzi doesn’t believe the status quo is optimal. (Quick: which constitutes the status quo, a world in which the Bush-era tax cuts are extended, or a world in which they are allowed to expire? Answer without reference to theories about economic expectations that cannot be tested with the rigor of a physicist.)
Rather, the starting question should be “what is our best guess of the optimal fiscal policy given general economic conditions”? If we are in a recession (quick: are we? answer without reference to any data, such as the level of unemployment, or the velocity of money, or the change in the price level, that is insufficiently well-defined for a physicist) then should we run a deficit or a surplus and, if either, how big should it be? The answer to that question is very unlikely to be “our policy should be whatever it was last year – don’t change a thing.” At least, there is no particular reason to believe that last year’s policy is economically optimal this year. And answering that question without reference to economic theories strikes me as completely impossible – all a general skepticism about economics gets you is an inability to do the job. (Quick: we’ve got a treasury auction coming up. What term bonds should we issue? Answer without any reference to current economic conditions or theories about how those conditions may change in the future or how government finance decisions may affect said future conditions.)
Once we have our best judgment about optimal fiscal policy, we can move to a discussion of practical politics: what are the levers available to move toward the optimal policy, and what will be the likely collateral economic costs of said moves, and on that basis make (again) a judgment call about whether a particular piece of legislation is “going in the right direction” and, if so, whether it is “worth it” given the ancillary costs.
Would you want to consult a variety of economists on any point where practitioners generally differ? Of course, particularly if the course of action being contemplated has significant risks. Wouldn’t you want an opinion from a second physician in similar circumstances? But, again, that’s a very different conclusion from saying, “these doctors don’t know anything” and throwing up your hands.
A couple of final points.
First, Jim Manzi himself makes decisions with reference to general principles that he cannot prove, and for which there are strong counter-arguments. For example, he has a fondness for the principle of subsidiarity. There are a bunch of different ideas packed into that preference. Smaller government entities will be closer to the governed and hence in a better position to make informed decisions. The scope for error will be less in a smaller entities (and the law of large numbers will start to work in your favor with respect to larger entities encompassing a diverse array of smaller entities). And small entities will compete with each other, and competition will force them to converge on a closer approximation of optimal policy. But there are strong arguments on the other side of the ledger. Smaller entities are more prone to capture by local interests. (This was Madison’s argument in favor of large republics as more stable and effective than small ones because large republics could balance factions that could take over smaller republics.) Smaller entities are also more prone to sorting effects that both reinforce inequality and reduce the incentive for competition that drives progress. (See, for example, the American education system.) And, of course, smaller entities must duplicate a variety of functions that might be accomplished more effectively in larger entities because of economies of scale. The point is: these kinds of debates in political science are even further abstracted from any kind of testable scientific proposition than are the kinds of debates that economists might have about the stimulus. (Quick: is there such a thing as economy of scale? Can you prove it?)
Second, I don’t believe that anyone short of a Socrates can actually maintain radical skepticism about as pervasive an intellectual edifice as economics – or medicine. Rather (and this generalization is based on my anecdotal observation of people I know personally who claim to be such skeptics; it’s not something I can prove), what professed skepticism frequently amounts to is a vulnerability to crank theories. The people I know who tend “not to trust” doctors would, of course, go to a doctor if they broke their legs. But they believe that cancer is best treated with herbal remedies – or that vaccinations cause autism – or that mental disorders like depression have no chemical basis – and so forth. That is to say: where scientific knowledge is limited, they prefer the advice of those opposed to the establishment. That’s their heuristic. And I see the same sort of thing with respect to economics. None of the people I know who profess to believe that “fiat money” is a fraud and that therefore we’re all doomed – and I know, many, many such people, including professional investors – actually behave in their private or professional life in the way that they would if they really believed such a thing. But professing such a believe does make them vulnerable to specific truth claims from specific fellow-adherents in areas where scientific knowledge is limited (such as debates about what the optimal monetary policy might be). Again: the heuristic is: when we are in the gray areas, I listen to the cranks. In neither case does this strike me as a particularly defensible heuristic when exposed to the light of day.
Radical critiques can often be extremely useful, exposing hidden assumptions behind the existing intellectual orthodoxy, posing fundamental questions that, it turns out, that order cannot adequately answer, thereby forcing that order to change or to shore up those shaky foundations. Radical skepticism, though, I have less truck with.
Noah,
Knowing what you know now, would you have advised someone in the 19th century to visit a doctor to treat what you understood to be a non-life threatening illness?
I think your analogy is apt. But I also think I would answer no to my question. I’m not sure Macroeconomics as a “therapeutic” discipline is more like medicine as practice in the 21st Century than like medicine as practiced in the 19th. At minimum, I am hesitant for the following reasons:
(1) there is still significant disagreement over both diagnosis and treatment among the best practitioners in the field (we could limit ourselves to Nobel Prize winners and this would still be the case)
(2) there is an insufficient track record of successful treatment by macroeconomists as compared with unsuccessful or harmful treatment attempts during the short time period that macroeconomic theory has been in a position to make a difference; and
(3) medicine, while an art, is a practice of applying increasingly confident theories about reality to conditions about which doctors have imperfect knowledge, and such theories — anatomical, biological, chemical, physiological — are quite scientific. Macroeconomics suffers in comparison both with respect to our knowledge of the condition as well as our knowledge of the underlying economic processes and realities.
— Jay Daniel · Dec 6, 07:29 PM · #
This is why my dad (a doctor) advises that one should, as much as humanly possible, stay the hell away from doctors!
— Tony Comstock · Dec 6, 07:29 PM · #
I don’t believe that anyone short of a Socrates can actually maintain radical skepticism about as pervasive an intellectual edifice as economics
JUST TRY ME, MILLMAN!
I would point out that, in my experience, doctors are quite good at taking the specific, personal evidence from a given case (as in your anecdote) and applying accumulated medical knowledge to solve or ameliorate the problem. But they aren’t, I often find, good at evaluating different kinds of medical knowledge and medical data that hasn’t yet been absorbed into the larger context of practical medical information. This is part of the reason why the “vaccines cause autism” myth has gotten such purchase; people bring it up to their pediatricians and GPs, and often find a sympathetic audience. (This is part of the service-y element of contemporary medicine, which is a whole other issue.) They trust their doctors because they’ve applied medical knowledge effectively in the past. But that’s a very different task than reading and evaluating the latest medical science scholarship on a contentious issue.
— Freddie · Dec 6, 07:46 PM · #
Jay Daniel – But even if everything you say is true, Noah still is correct to point out that economic decisions nevertheless must be made, that policies must be written. And even if economics is imperfect, it still makes sense to turn to economists for guidance. Is there an alternative?
— E. D. Kain · Dec 6, 08:08 PM · #
Jay:
Good question! I’d say: it depends on the malady. Again, in the 19th century surgery already had hundreds of years of history, and there was real practical and theoretical knowledge brought to bear. On the other hand, antibiotics hadn’t yet been discovered. By the 1950s, anti-biotics had long-since revolutionized medicine. But there was still essentially no understanding of or proper treatment for mental illness.
But here another point. A doctor in 1900 might well know more anatomy and more biology than he could practically use. The same is true of economists today. There may be more consensus around statements that are the equivalent of “if we don’t bring that fever down, she’ll die” than there is consensus around how to cure the patient, but that doesn’t mean there’s no knowledge.
But the larger point, once again, is: we don’t have the option of not going to the doctor. We can’t not have a fiscal or monetary policy. It’s not actually possible.
It might be possible to design such policies entirely around trying to avoid human error or political interference, making them as automatic as possible. But that’s a very different thing from not having a policy at all – you’d still need to come to some kind of decision as to what policies are optimal. I don’t know what recourse you have in doing so other than to economists.
— Noah Millman · Dec 6, 08:16 PM · #
Tony: my father and wife are also both doctors. And yes: they tend to avoid going to doctors. But that’s because (a) they think they can diagnose themselves (and prescribe as needed), and (b) they have a healthy appreciation for the limits of what doctors actually know. When they suspect that the doctor is just going to be guessing with her answer, they’ll tend not to bother to go. But when they know there’s something wrong that needs to be diagnosed, they go. They don’t actually want to go around with a walking pneumonia or an abdomenal hernia or what-have-you without getting it looked at.
— Noah Millman · Dec 6, 08:21 PM · #
Freddie: I don’t know – the people I know who believe in the whole vaccines-cause-autism thing came to it very much on their own and are well aware that they are going against the medical mainstream. And all the doctors I know think it’s terrible anyone would fail to vaccinate their child because of such fears. So I’m skeptical of your “they believe it because their pediatricians are sympathetic” thesis.
In my experience with doctors, the patients want more certainty than doctors are generally willing to give – and often hear more certainty than the doctor said. That is probably true of policymakers and economists as well.
— Noah Millman · Dec 6, 08:24 PM · #
My dad’s view of doctors was more of the “They don’t know shit and if you get near them they’ll probably try to get you to do something that will either be ineffective or make things worse” variety.
Don’t forget the Boogie Woogie Flu. That one’s nasty!
— Tony Comstock · Dec 6, 08:27 PM · #
IMHO, you don’t want to slide too far in either direction. Radical skepticism would be to say that doctors are no more likely to be correct about areas within their field of expertise than lay people.
Radical “opposite of skepticism” would be to say that a doctor is as likely to be correct in correcting human ailments as a structural engineer would be in designing a bridge.
I think Manzi is right to point out that we often make the mistake of assuming that since a macroeconomist is more likely to be correct on an issue of macroeconomics than a bricklayer, that means we should just act as if the macroeconomist is certain to be correct. IMHO, that insight has two important implications:
1) As I said in the Manzi thread, given the importance of the subject matter, we should be committing more resources to developing ways to test important but hard to verify sciences like macroeconomics and climate science.
2) You also need to have a handle on the degree of uncertainty. If your doctor proposes something radical — aggressive chemotherapy, say, or a heart transplant — you look a lot more closely at the costs and benefits of each course of action, as well as the level of certainty that you have that the doctor is right. You’re correct, of course, that it wouldn’t make sense to approach the doctor with radical skepticism, but the higher the stakes, the more you want an accurate idea of just how sure the doctor is, and why.
— J Mann · Dec 6, 08:27 PM · #
Someone smart should rank the professions according to how reliable their advice is. Why stick to just economists, doctors and physicists? What about lawyers? Architects? Engineers? Accountants? Graphic designers?
— Kevin Lawrence · Dec 6, 08:47 PM · #
Noah and E.D. —
You are both obviously correct that we are at a point in our federal government’s evolution where we have people who have been given the power to manipulate the macroeconomy, and we therefore need to give them policy goals and a rubric for doing so (whether or not their power is a good thing is, I think, debatable).
I also agree that these policy goals will have to be rooted in some sort of macroeconomic theory. The 19th Century doctor may have been terrible at treating an infection or mental illness, but as you point out, he was pretty decent at surgery, and much better at it than a carpenter or a blacksmith.
However, I disagree that fed members should attempt to implement a more fulsome macroeconomic theory in their roles, or that they be given the authority to do so. That’s where I think the analogy to a 19th Century doctor attempting to treat an infection or brain disease becomes much more apt.
Prior to the federal reserve era, the U.S. economy suffered various often deep recessions/depressions when it was essentially taking an unguided trip on the business cycle rollercoaster. Since we have had the federal reserve system, we have had several recessions (and a few depressions) that rivaled those of the 19th Century despite significant advances in both economics and the markets themselves. Monetary policy is harder than it looks. I think Coase would have something to say about this.
What I would propose instead is a minimal and exclusive set of ex ante goals which generally promote macroeconomic economic stability — like maintaining inflation between 1-3% — and the federal reserve should use the minimum monetary policy levers to achieve those goals. There is broad agreement on the basic conditions of economic stability. Even when those goals come into conflict with the macroeconomic theory du jour, the fed should should stay the course. This would provide an macroeconomic safety valve and predictability for the market.
Finally, the federal stimulus (and Krugman’s bemoaning) illustrates the folly of the legislature getting into the macroeconomic game, at least directly. There, we had two levels of folly. The first was obviously among economists, who did not agree as to the correct approach. But the second was among the quite diverse and mostly ignorant legislators, who would almost certainly manage to bastardize any macroeconomic intervention, even if it had been rooted in a well-grounded macroeconomic theory.
— Jay Daniel · Dec 6, 09:38 PM · #
Jim Manzi’s skepticism of public policy that isn’t first tested by experiment is clearly linked to his experience as a founder of a test marketing company, Applied Predictive Technologies, that runs the kind of controlled experiments for business clients that he advocates government should do.
I’m not saying Jim’s views are financially biased, but that his professional experience plays an obvious role in his public policy views. Personally, I share many of his views, in part because I share with him similar professional experience. I worked for a marketing research company that had great success in the early 1980s introducing scientific control group real world test marketing of supermarket products. For example, in isolated small towns, we would make up two random panels of 3000 households who had bought identical amounts of Product X in the last 12 months, then show one panel a new commercial for Product X for a year on their own cable TVs in their own homes and show the other random panel the old commercial for Product X for a year in their own homes. We’d determine which commercial sold more of Product X by tracking individual purchases at supermarkets. (We had recruited thousands of shoppers to voluntarily identify themselves at every supermarket checkout in town.) It was a scientifically bulletproof test market service.
On the other hand, the company I worked for then expanded beyond conducting experiments by going into tracking vast amounts of naturally occurring supermarket sales data so that clients could make faster decisions without waiting around for testing, in part because our clients pointed out to us the limitations of the kind of textbook experimental design we were peddling. In particular, marketers didn’t like to wait around a year to find out if their new commercial was a good idea. Conditions changed too fast. Indeed, our test marketing business dropped off after the mid-1980s because of these kind of complaints from our clients.
The analogy to the stimulus debate is clear. On the one hand, the ideal experiment would have been to run the stimulus on one randomly chosen panel for a year while not running the stimulus on another random control group for a year. On the other hand, I don’t know how to set up such an experiment, and even if I did, we’d have to wait around for a year (plus set up and analysis time) to find out the results.
Not surprisingly, policymakers under stress are more likely to shoot from the hip in order to respond quickly.
— Steve Sailer · Dec 6, 10:57 PM · #
No, that’s wrong. It’s exactly because medical and pseudo-medical professionals gave so much leeway to the autism/vaccinations theory— even if only to say “more study is needed” or “it’s a theory”— that it didn’t wither on the vine beyond your average Williamsburg food co-op in the first place.
— Freddie · Dec 6, 11:50 PM · #
You want to know about your spine by using objective medical evidence, i.e., an MRI. If you have a bulging disc, that could be causing sciatica and your recurrent, one-sided pain. Then the options include stretching, building stomach and back muscles (try swimming), acupuncture, anti-inflammatories. Your youth is in favor of recovery. As for economics, it would be helpful if companies, banks, Fannie and Freddie, and investment houses did not commit fraud and, lo and behold, instead of getting criminally prosecuted and sued civilly for fraud, they get supported by governments all over the world because it is better to make destitute the average taxpayer than to let the superrich and the too-big-to-fail banks and investment houses go bust. There is no economic cure for fraud.
— LDM · Dec 7, 03:15 AM · #
I actually think that your comparison of economics to medicine is more apt than Manzi’s to historians. However, he seems to be lamenting what happens in the 5% of hard cases, not the 95% of “day to day” economics.
To me, Manzi’s complaints have more to do with the rhetoric surrounding modern macroeconomics debates in the media and blogosphere. The Krugmans and DeLongs portray their best-guess diagnoses and prescriptions for once-in-a-lifetime, seldom-seen symptoms as absolute, unassailable facts questioned only by ignorant Luddites. It’s the push by high-profile economists to turn these hard cases into easy ones, and the lack of humility on the part of these professionals. (this is done by both Right and Left, it’s just easy to hammer the Left given current policy solutions)
Something along the lines lf “economics as medicine” combined with a Kling-like skepticism of people we exalt as experts is, to me, the right way to treat it. Doctors can fall prey to the same choice problem that every expert faces. Notably, recently, the “we have to do something; x is something; therefore we have to do x” problem.
— Kevin · Dec 11, 11:07 PM · #
These are wonderful! Thank you for sharinghttp://www.caps2011.com
— red bull hats · Dec 13, 10:39 AM · #