Brad DeLong's Semi-Daily Journal: A Weblog

I feel trapped in an iron cage. Economics, as it is practiced, has become the strongest, most rigid iron cage around. True enough, many of the "tools" we have developed can create significant insight into many problems. That doesn't change the fact that economics is overwhelmingly dominated by the Ricardian vice and ruthlessly suppresses dissent.

[[How is it that we economists make
the heaviest use of formal theory,
and yet don't--or I at least don't--feel
trapped in an iron cage by our
discipline?]]

Well, you might start by addressing those of us who don't think that there are actually "markets" everyone academic (or non-academic, formal) economists thing that there are markets. Or who think that some of the fundamental axioms on which formal economics is based are not correct (humans do not have transitive preferences, for example).

Cranky

PS The new comment feature is a bit sub-optimal.

This is going to get very quickly into the realm of deep theory of science. And I have no time to defend my claim. But let me throw it out there and see if it makes sense anyway.

I think that economics, despite the adjective "dismal," is much closer to being a real "science" of the type discussed at length by Thomas Kuhn, et. al., than are fields such as history, poli sci, etc.

As such, if Kuhn is correct (which I, for one, think he is at least 90%, but of course I won't try to argue the point here), economics has a real, functioning paradigm. [Let's fight about the accuracy of the word 'paradigm' in another context, ok?] Brad described the economic paradigm well in his post. This paradigm is clearly functional. It clearly delimits the sorts of questions that are to be asked -- and provides clear ways to try to answer them.

History (and the other fields that range from the softer sciences to the humanities) does not have a paradigm. Thus the range of permissible questions and answers is much broader.

At first blush, this seems backward. If history is a broader field, why would people feel MORE constrained in this field than they do in economics?

Part of the answer is that the structure of economics not only enables certain sorts of questions, it also cuts off entirely other ones, pushing them into other fields! Economists are perfectly happy leaving to non-economists many questions that might be called "economic" in some sense! By contrast, people who have a sense that the current modes of historical thought are not asking the right questions can ask new types of questions, in new ways, and still be DOING history.

Economists feel that they can leave certain questions aside, and pursue only the ones they care about, using the methods they prefer, precisely because the science of economics has a paradigm. Like physicists, biochemists, and others, they feel secure in the knowledge that applying THIS type of method to THIS type of problem will yield good, interesting results. Historians (et. al.) lack that confidence. So individual historians regularly get the sense that the whole field is off-track (if it even has a track to be on or off). Other historians may well share this sense! But it is rare that two historians can agree about the WAY IN WHICH the whole field is off track.

Timothy Burke's post looks to me like it he is trying to put forward a really new way of thinking historically. ("it seems to me that Johnson is operating well within the norms that are part of the problem, not the solution....") In other words, Mr. Burke wants to pose new questions in new ways that haven't been thought of before. If he were an economist, he would probably be told that what he's trying to do isn't economics at all. And he would be less likely to say it in the first place, because most economists do not believe that the entire field is off-track!

OK, I think I'm beginning to repeat myself, so I must be done. :-)

As a working historian, I feel a great sense of freedom and lightness when I look at some other disciplines, such as literature. There it seems that every point of view has to be made a theory and every theory has to be tossed out every decade at least. Could be exaggerating! However, in history, there are plenty of old-fashioned fields and perspectives alive and well and still finding publishers, along with the trendy and sometimes exciting stuff.

I think that partly because fields outside economics do not have a paradigm, as Kent just said, there is more arbitrariness in imposing quasi-paradigms on people, or as far as that goes hiring and promoting people for arbitrary reasons ("who you want to have lunch with for the next 20 years" has been named as an important class of reason).

And in history and English especially, there are fewer jobs and fewer alternatives to academic careers (i.e., not-quite-successful economists can get good jobs in finance or public administration, but there's not so much for historians or lit people). So the arbitrariness is given free play.

As I see it, economics has achieved scientific success by choosing to limit themselves to problems which they can deal with successfully (again, as Kent says).

To me problems arise when some real-world question has an economic aspect and a non-economic aspect. Since econ is the more successful science, the economic aspect tends to be overvalued, regardless of its functional importance. My analogy is some stage of medicine when (say) cardiology was well-developed, and endocrinology poorly developed. If the cardiologists get away with pulling rank on the endocrinologists, poor medical care will result. Practitioners of well developed scientists often speak of practitioners of poorly-developed scientists as if they know nothing of any importance at all.

Successful scientists also do not consider that maybe other scientists (or scholars) are less successful simply because the thing they're studying is harder to understand. Ilya Prigogine has said that one of the reasons for the success of early modern science is that they were picking the low-hanging fruit.

Some economists believe that everything is economics in the end, but Becker's attempt to describe the family as a three-cornered contract-and-exchange relationship strikes me as batshit crazy. It's very hard to figure out how raising kids is economically rational in our world, except by inventing fictitious fudge-factor exchanges of imaginary goods.

With all due respect, Brad may not feel constrained by the system because he is one of its masters.

Brad and Martin Feldstein can talk to each other because they are a lot closer methodologically than politically. But not every self-styled economist is in the neoclassical tradition. Is John Commons or Thorstein Veblen an economist? I dunno--but they were smart guys who were considered economists in their time. Their followers (if any exist) aren't getting any jobs in economics departments, as far as I know.

Heck, I know you can make the same argument for chemists and alchemists: one sharpened by the passage of time. Are the disciplines best able to purge their heretics the most scientific? I will agree with Brad only on the converse--the real sciences are good at purging heresy. But I'm not sure economics is a real science, and I know that one cannot infer a proposition from its converse.

Markets fail? I certainly haven't heard any politically conservative economists admit this.

Just to echo Chip Poirot, the smug passage starting "Economists share a language and a set of standard operating procedures" presents a sadly limited vision of what economics is about. The four questions that follow sound like a dissertation formula for an unimaginative grad student in an orthodox program. Cranky's comments about leaving questions aside also apply, as does Lerner's famous remark about solved political problems. (Additionally, as a heterodox economist, I take exception to Brad's blithe appropriation of "economist" for this narrow vision.) Progress in social science may require a transdisciplinary -- even counterdisciplinary -- approach that does not take so many things for granted.

I suppose that there is some requirement of professional courtesy which forces Brad to continue to take Martin Feldstein seriously on the subject of Social Security, but that doesn't mean that the rest of us have to.

Feldstein didn't originally cast his arguments about Social Security in the form of a statement about the equity risk premium. He started off in 1975 by saying that Social Security was responsible for low savings rates and thus needed to be abolished in order to raise savings. When this argument was shown to have a big flaw in it (IIRC, unrealistic assumptions about the effect of unemployment on savings), he moved on to saying that Social Security was bust and needed to be privatised. When it was pointed out that his model was using an assumption for long term GDP growth which was hardly consistent with 8% real returns on equities, he moved on to the current formulation.

The constant in Feldstein's career has been the need to get rid of Social Security; his arguments have simply changed in order to provide a rationalisation for that view that didn't look too transparent at the time. I have not a single idea in my head why he is being used as a paradigm example of an economist reaching a conclusion by arguing from first principles.

"How is it that we economists make the heaviest use of formal theory, and yet don't--or I at least don't--feel trapped in an iron cage by our discipline?"

Stockholm Syndrome?

The problem with Kent's comment (or one of the problems) is his identification of "the economic paradigm" as "economics". It would be more accurate to identify "the" Neo-Classical paradigm as a paradigm within economics. This really gets to the heart of how the bars on the iron cage are constructed and how the gate is locked. At least some economists aren't content to not explore the economic issues that Neo-classical economics "leaves" to other disciplines. But by exploring these issues, one is ruled as outside the "economics" paradigm.

Since we are getting deep into philosophy of science, we would do better to focus on Lakatos' concept of "research program". Kuhn overemphasizes subjectivity, discontinuity and paradigm incommensurability in my opinion. Lakatos provides a way of allowing for overlap between rival research programs, steady accretion of knowledge and objectivity in judging research programs. We should judge whether research programs are "progressive" or "degenerating" and whether there are valid alternatives available. Mark Blaug does a good job of at least laying out how Lakatos can be applied to economics in "Economic Theory in Retrospect" (even if I disagree with some of his conclusions).

The really curious thing is that we find ourselves in a position in economics where Nobel Prize winners like Douglas North openly renounce the rational choice framework, and yet rational choice remains the operative assumption underlying most of economic theory. Stiglitz showed that the welfare conclusions of Walrasian General Equilibrium theory are impossible, and yet the concept of Pareto efficiency remains the organizing principle for most of economics.

And these are only the dissents from within the neo-classical research program.

As I said in my first post-the Neo-Classical research program can indeed address many questions. The problem is definint that one research program as the entire disciplinary breadth of economics. Economics is now the only mono-research program discipline in the social sciences. All other social sciences are pluralist in conception and practice.

Let me take another angle on this. Freeman Dyson makes a useful distinction between what he calls "Athenian" and "Manchesterian" sciences. Athenian sciences are like the kind Kuhn was talking about when he came up with the idea of paradigms. They study relatively invariant states of affairs that are not greatly influenced by reactive interactions. This means that formalized, deductive theories can be used successfully to describe the subjects and the theories can suggest critical experiments that will validate them. It also means that technical means can be much more easily mustered to manipulate the relationships discovered, leading more readily to engineering applications.

Manchesterian sciences, on the other hand, involve reactive, historically conditioned states of affairs. This means that theories are probablistic at best and do not lend themselves to formalization without simplifying assumptions that severly limited the theories descriptive capabilities for all states of affairs concerned. This means, in turn, that the theories will be more inductive, historical, and based on empirical descriptions of process. Theory testing is based on statistical controls and episodic "natural" quasi-experiments. Consequently, the application of technical means to manipulate relationships is much more restricted and the "engineering" applications much less successful.

While no science fully conforms to both models, it should be obvious that the social sciences are much more Manchesterian than Athenian in scope and method. That is why it so difficult for paradigms to become established in them and, I might add, why attempts to do so make people feel that they are in an iron cage. Brad surely knows that this feeling of analytical confinement is widespread in economics and that, while he might feel comfortable with his list of questions, many of the members of his profession, especially the younger ones, don't. The same frisson is present in my discipline, political science.

Personally, I don't feel bad about this at all. Attempts to establish more than middle range paradigms in all the social sciences have generally been empirical failures. If we have a clear idea of what kind of science we are doing, the limitations of our endeavors will be much easier to accept and we can begin to concentrate on analyzing the data correctly. One sound empirical result is worth 100 formalizations.

On another subject: Brad, I like the site still, but I think it would be easier to navigate if you went back to serif fonts (Times?) you used before. The sudden absence of your topic classified archive also caused considerable panic. I know it must be a bitch to generate, but it sure is handy for dredging up your always useful views on things.

I don't know about Mr Burke, but many conservatives who object to the prevailing "liberal" orthodoxy in a given field are simply unwilling or unable to entertain the possibility that the world may not, in fact, conform to their prejudices.

One must remember that prejudice, not reason, is the animating principle of conservative thought. The more honest ones, who are unfortunately pretty scarce in government and academia, are quite upfront about this.

I grasp what Brad is saying about economics. There is one sense in which it seems much looser than the humanities, even history, which is already more open in its feel, as Stevem notes, than literary studies. All the economists I know have a kind of pragmatism in their views on what makes good and bad work that is much more politically disengaged and neutral in many respects. On the other hand, as many commentators note, the constraints on what can be asked and said are sometimes much tighter, and the areas of almost universal exclusion in the research paradigm of the discipline are sometimes striking. In my work for the SSRC, we've occasionally wondered why there are relatively few economists who plan to do sustained observational fieldwork of some kind given its potential benefits. It's not so much a malicious exclusion as it is something that is simply Not Done, except off in the borderlands of experimental economics or behavioral economics. You could argue that this is a question of defintion--that such fieldwork is the province of anthropology, or sociology, or some other discipline than economics, that this isn't what economics does. Which is kind of the point I was making on this subject.

As to my own arguments and their reception, in part, I would say, "Wait for the book". (Both to see the arguments and see whether they fall on deaf ears.) I have found, however, that some preliminary versions of my claims in the current project don't meet with an actively hostile or antagonistic reception as much as they simply get remapped as something much less unsettling or more conventional within the highly particular disciplinary AND interdisciplinary specializations I'm operating within (comparative colonialism, British imperial history, modern African history, black studies).

On the other hand, Brad's mild skepticism may be justified in that I may be assuming an oppositional or closed groupthink when in fact it's simply a lack of clarity in my own arguments, or even a lack of stridency. I have a hard time being too crude in picking a fight precisely because it seems to me that one obligation of scholarship is to privilege complexity and subtlety in our interpretations. Even when I think I'm breaking ranks, perhaps I'm not so much, or I'm being so cautious about saying it that my arguments come off simply as a shading of the orthodox historiography on colonialism in Africa. Which isn't too far from the truth: I'm not intending to tear down the temple or throw out four decades of careful, meticulous work, merely to try and probe what strike me as unasked questions and quietly evaded complexities.

It could also be that what I perceive in the literature is just a cover-your-ass theoretical gloss that many scholars put in as a way to protect themselves within a competitive job market, and that their internal sense of what they're saying and writing is actually very close to mine. I was at a conference on British imperialism a few years back where the historian Bernard Porter presented a paper designed to argue that imperialism was much less a part of Edwardian British life and culture than has been commmonly thought or claimed in a huge range of recent scholarship. I could see he was sort of preparing himself to get blasted but in fact just about everyone agreed with him, very matter of factly. Maybe that's what will happen to me as well. This is the other problem of the overly shrill complaints about academia: they tend to overestimate the systematic beastliness of academics. This is not to say that academia is not capable of extraordinarily vile behavior, both from individuals and collectively, but maybe we remember those incidents too much or infer their systematicity wrongly.

Not to intrude, (since this has been to my pre-graduate eyes, a very illuminating comment section), but I had to take a little umbrage at:

"In my work for the SSRC, we've occasionally wondered why there are relatively few economists who plan to do sustained observational fieldwork of some kind given its potential benefits. It's not so much a malicious exclusion as it is something that is simply Not Done, except off in the borderlands of experimental economics or behavioral economics."

Not to completely disagree, but I know of at least one program that requires field work as part of the Masters' curriculum, the International and Development Economics program at University of San Francisco. Admittedly, the fieldwork portion seems to be only a few weeks (about a month, maybe two) but I have talked with other development economists at Berkeley and they certainly did extensive fieldwork. I think the World Bank requires it when doing a country study. Perhaps to the non-economist, the inter-disciplinary differences are hard to see? Alternatively, perhaps other types of economists don’t use the tool development economists’ use? I’m still getting a hold of the differences within the disciple (and pleased to see some heterodox economists posting), so am not surprised that we all look alike to non-econ eyes.

yours,
devgirl
budding development economist

p.s. the new comment section is amaingly painful to use

I grasp what Brad is saying about economics. There is one sense in which it seems much looser than the humanities, even history, which is already more open in its feel, as Stevem notes, than literary studies. All the economists I know have a kind of pragmatism in their views on what makes good and bad work that is much more politically disengaged and neutral in many respects. On the other hand, as many commentators note, the constraints on what can be asked and said are sometimes much tighter, and the areas of almost universal exclusion in the research paradigm of the discipline are sometimes striking. In my work for the SSRC, we've occasionally wondered why there are relatively few economists who plan to do sustained observational fieldwork of some kind given its potential benefits. It's not so much a malicious exclusion as it is something that is simply Not Done, except off in the borderlands of experimental economics or behavioral economics. You could argue that this is a question of defintion--that such fieldwork is the province of anthropology, or sociology, or some other discipline than economics, that this isn't what economics does. Which is kind of the point I was making on this subject.

As to my own arguments and their reception, in part, I would say, "Wait for the book". (Both to see the arguments and see whether they fall on deaf ears.) I have found, however, that some preliminary versions of my claims in the current project don't meet with an actively hostile or antagonistic reception as much as they simply get remapped as something much less unsettling or more conventional within the highly particular disciplinary AND interdisciplinary specializations I'm operating within (comparative colonialism, British imperial history, modern African history, black studies).

On the other hand, Brad's mild skepticism may be justified in that I may be assuming an oppositional or closed groupthink when in fact it's simply a lack of clarity in my own arguments, or even a lack of stridency. I have a hard time being too crude in picking a fight precisely because it seems to me that one obligation of scholarship is to privilege complexity and subtlety in our interpretations. Even when I think I'm breaking ranks, perhaps I'm not so much, or I'm being so cautious about saying it that my arguments come off simply as a shading of the orthodox historiography on colonialism in Africa. Which isn't too far from the truth: I'm not intending to tear down the temple or throw out four decades of careful, meticulous work, merely to try and probe what strike me as unasked questions and quietly evaded complexities.

It could also be that what I perceive in the literature is just a cover-your-ass theoretical gloss that many scholars put in as a way to protect themselves within a competitive job market, and that their internal sense of what they're saying and writing is actually very close to mine. I was at a conference on British imperialism a few years back where the historian Bernard Porter presented a paper designed to argue that imperialism was much less a part of Edwardian British life and culture than has been commmonly thought or claimed in a huge range of recent scholarship. I could see he was sort of preparing himself to get blasted but in fact just about everyone agreed with him, very matter of factly. Maybe that's what will happen to me as well. This is the other problem of the overly shrill complaints about academia: they tend to overestimate the systematic beastliness of academics. This is not to say that academia is not capable of extraordinarily vile behavior, both from individuals and collectively, but maybe we remember those incidents too much or infer their systematicity wrongly.

Development economics is an odd field. It's almost interdisciplinary. It's a good angle to approach economics from because it asks questions about how markets come to be, how they are historicaly situated, what the relation between markets and societal structures are, what role technology plays and so forth.

Unfortunately development economics has mostly given very very bad advice to developing nations. The sad truth is that with the exception of city states and Britain, no country has been succesful in developing by using the orthodox advice of western development experts.

Development economics is the edge of the discipline and shows that the discipline's weaknesses in starkest form. Use that view from the horizon well so that when you're 60 you can honestly say "I did more good than harm to those in my charge." Most development economists of older generations can only say, "my intenions were good, but it did not work out as I had hoped."