Pleasantly Contentious, Perhaps Even Usefully Dangerous

by Henry on May 16, 2007

This piece by Mark Levinson in Dissent on Paul Krugman and John Kenneth Galbraith touches on something I blogged about last year. When I read my way through Krugman’s early 1990s book, Peddling Prosperity, Galbraith came in for a surprising amount of flak. This gave me the impression that Krugman was being a little defensive, the sotto voce measage being that yes, perhaps Krugman too was an economist who could write wittily and well for a popular audience, but unlike Galbraith, he was a real economist, who had imbibed the lessons of Samuelson et al. and did equations and stuff. Peddling Prosperity is as much as anything an effort to re-create the boundary between real economists and those whom Krugman perceived as populist hacks; Galbraith is awkward to fit into that classification, as he wasn’t a mathematically rigorous economist, but was a past-president of the American Economics Association.

The interesting bit of Levinson’s piece is his discussion of how Krugman has morphed over time into the kind of economist that JKG wanted to see.

Galbraith insisted that power—which he defined as “the ability of persons or institutions to bend others to their purposes”—is decisive in understanding what happens in the world. He went on: “If we accept the reality of power . . . we have years of useful professional work ahead of us. And since we will be in touch with real issues, and since issues that are real inspire passion, our life will again be pleasantly contentious, perhaps even usefully dangerous.” … It’s hard to think of a better description of Krugman. His discovery of the abuse of power now seems to influence not only his op-ed pieces for the Times but also his more serious economic writing. … In the late 1990s and early 2000s, when Krugman spoke about causes [of inequality], he usually said something like this (from an interview in 1999): “Looking at the numbers makes it clear that this [inequality] is . . . [caused by] some combination of technological change and more complicated factors.” Now his explanation incorporates power and politics: “The government can tilt the balance of power between workers and bosses in many ways—and at every juncture this government has favored the bosses.”



tm 05.16.07 at 3:22 pm

I would say PK has not yet morphed into the type of economist JKG wanted to see, as PK still keeps his “rigorous” economics and his op-ed work compartmentalized, whereas JKG considered his books, such as The Affluent Society and its successors, to be his professional economics.


engels 05.16.07 at 3:33 pm

In the late 1990s and early 2000s, when Krugman spoke about causes [of inequality], he usually said something like this (from an interview in 1999): “Looking at the numbers makes it clear that this [inequality] is . . . [caused by] some combination of technological change and more complicated factors.” Now his explanation incorporates power and politics

I dunno. This was written in 1996, I believe: “In brief, much of the polarization of American society can be explained in terms of power and politics.”


stostosto 05.16.07 at 4:36 pm

I am not sure Galbraith is someone to hail. Definitely not as an economist.


Luis Alegria 05.16.07 at 4:50 pm

Mr. Stostosto,

I agreen completely. Back in the day I devoured Galbraiths books, because he was in fact a great writer. His record in terms of the outcomes of policy prescriptions, or of foretelling the future (what else do we read such people for ?) was not very good at all.

As for the issue of “power” in economics – it seems like:

a.) An unecessary and unscientific digression into the roots of policy, for the kind of economics that studies the effects of policy. What do the reasons for x vs y income tax rates have to do with understanding the results of x vs y tax rates ?

b.) A rather narrow, political and most likely unenlightening wandering into the paths of “culture”.


harry b 05.16.07 at 4:55 pm

Luis — economists are completely barred from studying the real world, then? I agree much economics is entirely theoretical, and don’t take any issue with the importance or legitimacy of that kind of work. But people who study actual economies can be economists, right? And they are expected to ignore power? (of which Galbraith gives an unsubtle, but adequate-for-the-moment, working definition).


dsquared 05.16.07 at 5:32 pm

His record in terms of the outcomes of policy prescriptions, or of foretelling the future (what else do we read such people for ?) was not very good at all

It was certainly no worse than any other economists’. He established that strategic bombing in WW2 had not worked, was one of the leaders of the Keynesian revolution in the USA, called the dot com bubble, spotted that the US involvement in Vietnam would be an economic, social and humanitarian disaster and predicted that the majority of foreign aid distributed in the 1960s and 1970s would not work. Just naming the ones off the top of my head.


notsneaky 05.16.07 at 7:56 pm

strategic bombing in WW2 had not worked – not economics

one of the leaders of the Keynesian revolution in the USA – as proponent and policy advisor, not really as a contributor

called the dot com bubble – feh

spotted that the US involvement in Vietnam would be an economic, social and humanitarian disaster – not economics

predicted that the majority of foreign aid distributed in the 1960s and 1970s would not work – ok maybe you got something here.

When you think of Krugman you think of actual contributions to economics. Trade under increasing returns, trade with market power, trade among similar countries, trade and growth and transport costs, theory of free trade areas, why and how financial/BOP crisis happen, trade and geography, geography and growth and many other things. Just off the top of my head you know.

When you think of Galbraith’s contribution to economics you think of… a misguided advocacy for wage and price controls, the essentially empty assertion that “power matters”, and the shocking idea that firms actually try to get consumers to buy their products. Which isn’t to say that he didn’t excel in other roles.


John Emerson 05.16.07 at 8:49 pm

God, I hate economists. They’re so arrogant and so self-pitying. They want to be loved, even though they continue to explain that everyone else is an idiot. (“Learn to think like an economist!”)

Give it up, guys. You won, we lost, and we don’t like you or trust you. I know that you believe that you’re the good guys, but not everyone agrees. Get used to being lords of the universe.

And I know that it hurts your feelings when we take your shittiest representatives to be typical, but they leave a pretty big footprint.

It’s a bittersweet pleasure to see the Bush administration fuck things up so bad that even economists don’t like it, but I also remember economists’ extraordinarily enthusiastic participation in Clinton’s triangulation against labor. This has been explained as a wonderful two-part deal of which we only got one part, ha ha, but the second part is an imaginary pony. We haven’t seen it and we probably never will.


John Emerson 05.16.07 at 9:44 pm

Despite my longtime disagreement with the Crooked Timber clique on many important questions, I strongly and unconditionally condemn whoever it was that blew up their palace.


Colin Danby 05.16.07 at 11:11 pm

Good grief, Luis, culture has such terrible cooties you have to put it in quotes?

Power is part of the social world, and material processes operate in that social world. This is not to say that we cannot occasionally entertain ourselves by thinking in models that lack power, but as Harry notes it’s weird to say that if power is there in the world an economist can have nothing to say about it.

The notion that we read social scientists in order to foretell the future is even more bizarre. A little more of this and John will sound completely reasonable.


John Emerson 05.16.07 at 11:18 pm

But Colin, I am completely reasonable.


Colin Danby 05.16.07 at 11:20 pm

That’s what worries me.


Luis Alegria 05.17.07 at 12:07 am

Mr. Harry B,

Of course economists have to study the real world – but “power” as defined by people like Galbraith seems a bit off the subject – wouldn’t that be Political Science or some other less useful field ?

Mr. Notsneaky,

Just so. You can add that he was very unhelpful as a source of economic advice for the third world (unlucky India). And then there was the failed “triumvurate” prediction. And then there was the silly “convergence” thing between capitalist and socialist economies, the economic success of the Soviet Union, etc.

Mr. Emerson,

Er, who is “we” ? And why is everything so bad ? Things seem to be going very well all in all, no small thanks to the current economic consensus. BTW, I have a fine argument on the success of the 2003 tax cuts ! Come to and have it out.

Mr. Danby,

Well, culture does seem to have cooties, in spite of the probability that somewhere in that cootie-filled box are the most important factors in predicting economic performance. It needs to be taken seriously and analyzed via metrics, not through assertions of power relations or subordinate status or whatever. This sort of thing is no use to anyone.

Of course you use social scientists to predict the future. The reason we have these things at all is to inform public policy, and predict success or failure. If it didn’t have this value then the social scientists might as well move to the philosophers ghetto and help them waste their time.


notsneaky 05.17.07 at 12:15 am

This paper (via Dani Rodrik) shows you how to deal with things like power and institutions in a proper way:

JKG’s statements though were just vague assertions. Or to put it in another way, you can explain almost anything with an appeal to “power” or “culture” or “change in tastes”. Which doesn’t mean it’s meaningless or wrong, just that you got to come up with some data and numbers to back up your claims.


John Emerson 05.17.07 at 12:23 am

So forty or fifty years later Galbraith’s work is superceded, so it means that he never did anything worthwhile?


Robert 05.17.07 at 12:55 am

I believe there have already been arguments at this blog about the worth of Galbraith’s work. I’ve had something to say myself.

As to Colin Danby’s point – I think it should be worrisome that one can point to mathematical mistakes demonstrated decades ago in most mainstream textbooks.


Colin Danby 05.17.07 at 1:00 am

You’ll have to do better than that, notsneaky. The two JKG books on my shelf, _The Great Crash_ and _ The New Industrial State_, don’t fit your caricature (what are you quoting?) and contain plenty of data. He’s open to criticism, but the institutionalist tradition he sprang from was if anything too fixated on data. JKG is among the early popularizers of national income accounting. Take some actual text and make your criticism properly before you lecture others on proper ways to do things.


notsneaky 05.17.07 at 1:46 am

Yeah it was a bit of a caricature but still – even you yourself point to the fact that he was a “popularizer” which I think was his main strangth relative to being a “contributor” like Krugman.


notsneaky 05.17.07 at 1:53 am

…and my spelling really sucks.


Colin Danby 05.17.07 at 2:42 am

Especially if you choose to manifest yourself as a childish pseudonym, notsneaky, I’m less interested in what you think than in what you can show. JKG’s work is not exactly unavailable to anyone who wants to make a real critique. Is this an admission that you were bullshitting in #s 7 and 13?

I hate to be swatting flies here, but there’s good reason for John’s outrage — people turn up claiming to speak for economics who are arrogantly uninformed. I’ve run into this “when you think of” trope more than once, in which people take their own ignorance of a literature as evidence about that literature. It’s used with regularity against heterodox economics.

For anyone interested in informed discussion, here’s a quick description which situates JKG in American Institutionalism:


notsneaky 05.17.07 at 3:25 am

No, colin, no admission of bs-ing. I will admit however that it’s been quite awhile since I’ve last read JKG (8 or 9 years) and I’m relying on how I remember my impressions back then and before and my subsequent thinking about him. I remember being impressed by the quality of the prose and style but also somewhat disapointed by… let’s say the moderate depth of his ideas. Also I’m not exactly sure why is it up to me to show anything here. You’re trying to frame the debate in a way that’s advantegous to yourself. Why shouldn’t you be trying to convince me that the old Krugman was wrong?

Here’s an example – the focus of JKG on advertising and how it affects people’s preferances and market structure. What exactly came out of that? What literature did it inspire? What insights followed, aside from the trivial observation that firms try to actively sell their product and that this has consequence for competition and prices? How much consequence? Is it a completely negative phenomenon or is there some positive aspects to it?… All these questions actually have been addressed by a fairly substential economics literature which took off in the early 70’s (though the antecedents date back to Hotelling and the 20’s). But I’m pretty sure that most of the economists who have been working on these questions would not cite JKG as their inspiration or as a precedent or anything of the sort. Perhaps this is a condemnation of the economics profession but then we’re getting circular here (Economic profession sucks because they don’t pay attention to a great man like JKG. The fact that it doesn’t pay attention proves he was a great man. QED)

I think in the end you gotta face the fact that JKG was a (very good) political polemicist first, a popularizer of some economic ideas second, and maybe a contributor to economics a distant third. This is the Veblen Effect (the other Veblen effect) (I might as well get all anti-Institutionalist here) – the high quality prose makes it seem like there’s more to something then there really is.

So lay off my childish nickname.


notsneaky 05.17.07 at 3:40 am

And here’s a Krugman aside from the Slate article that is linked to somewhere in here, which I think addresses the point that Galbraith’s books containt lots of data:

And by the way, throwing around lots of statistics is not the point: It’s a question of thinking hard about how the statistics fit together.

And no, I’m not confusing James and John, just found it relevant.


Robert 05.17.07 at 9:13 am

“notsneaky” is the commentator formerly known as “Radek”.

I haven’t looked at Krugman’s professional economics work myself. As I understand it, he, like other leading economists, pretends whole literatures do not exist. Barkley Rosser, Jr., says something like this in his 1996 review of Krugman’s Development, Geography, and Economic Theory.


engels 05.17.07 at 11:35 am

I hate to be a bore by interrupting the Galbraith flame war this thread is fast becoming but does anyone have an opinion on the accuracy of the linked essay? It seems to me, based partly on the article I linked to in #2, that contrary what Levinson appears to be saying Krugman has always emphasised “power and politics” as drivers of inequality, alongside generally accepted factors like technological change, and so it is wrong to portray this, as Levinson’s note does, as a late ‘Galbraithian’ turn of Krugman’s.


John Emerson 05.17.07 at 1:49 pm

Maybe the decision of economists to narrow the definition of economics in such a way as to exclude Galbraith was not a good decision. This was a pervasive movement in American academia during the 50s, 60s and 70s, in many departments besides economics.


notsneaky 05.17.07 at 5:44 pm

Engels, I think what’s happened is that Krugman decided that you have to know how to pick your battles. I seriously doubt that his opinion of JKG has changed since the 90’s, rather he just seems to think that these kinds of arguments are not important in this day and age, given the current political climate. Why waste time arguing over who is and who isn’t a real economist, when Bush is busy screwing everything up?
But yes, I think he always talked about “power”, though perhaps he didn’t emphasize it as much as he does now.

DRR makes a similar point over at Mark Thoma.


Colin Danby 05.17.07 at 9:04 pm

notsneaky is quite right on this point, Engels. If John Edwards became President I think Krugman’s apparent positioning would change.

As a matter of background, neoclassical economics, as it came into existence as a self-conscious and rapidly expanding school mid-century, to a great extent defined itself in opposition to institutionalism, which had been dominant early in the century.

By 1959 you can find Paul Samuelson boasting of a “new uniformity” in economic analysis and singling out the crushing of the institutionalists as evidence of progress toward becoming a proper science. John Kenneth Galbraith was for several decades the last really prominent institutionalist left in the United States, and slagging him off was a ritual act for neoclassicals.

Samuelson, Paul. 1959 Social Science: Whence and Whither? in Daniel Lerner, ed., The Human Meaning of the Social Sciences. New York, Meridian Books.


John Emerson 05.17.07 at 11:52 pm

Back in 1967 or 1968 I met an econ PhD candidate through a friend. I mentioned that I’d just been reading Polanyi and was quite impressed. He did one of those little double takes academics do when they realize they’re talking to an idiot, recovered, and said something lik “Ah…. the direction of the fields is toward mathematical economics…..political economy isn’t really studied any more.”

From the way he talked I knew that he would never think about any topic discussed by Polanyi unless it was also discussed by the new orthodoxy. His not-thinking would be active; he know vaguely what Polanyi talked about (though he’d probably never read Polanyi), and he had made up his mind never to think about those things.

Some people assure me either that econ has improved, thoug others say that Polanyi is crap. What’s particularly infuriating to me is that it wasn’t factual or theoretical errors of Polanyi’s that were being rejected; it was certain kinds of topics, and whole areas of study.


engels 05.18.07 at 5:20 am

Umm, I can’t quite tell if you guys think you are arguing with me but that’s basically what I was saying.


dsquared 05.18.07 at 7:47 am

[You can add that he was very unhelpful as a source of economic advice for the third world (unlucky India). ]

really? I have a strong intuition that you don’t know what his advice to the developing world was, Luis. (The same intuition is telling me that notsneaky doesn’t know very much about Galbraith’s contribution to the strategic bombing survey if he refers to it as “not economics”).


Luis Alegria 05.18.07 at 3:58 pm

Mr. Dsquared,

Yes I do actually, with respect to India back in the 1950’s-early 1960’s. Basically it was to ignore Milton Friedman with respect to nationalizations, economic planning and open markets. He may even have specifically discouraged the Indians from obtaining Friedmans advisory services, I am not sure of that bit.

India became even more statist later into the 60’s -early 1970’s.

Note the acceleration of Indian growth in recent years as they have adopted more liberal policies.


notsneaky 05.18.07 at 4:09 pm

In so far as it looked at the impact of the bombing on the German war economy I guess you could say it was economics. Or you could say it was applied statistics. I don’t really know which parts of the report (available here: are JKG and which are not JKG and to the extent that JKG toned down his personal views, but overall the report is actually fairly supportive of the strategic bombing and argued that it was decisive. Most of JKG’s criticisms of strategic bombing actually came later.

Anyway, to be meaningful, your #6 should be expressed in percentage terms. For a man that did as much opinin’ as JKG did, one would hope that he’d happen to be right on some things. And of course he was wrong about a lot.

Engels, yeah I’m agreeing with you.


spencer 05.18.07 at 4:13 pm

Interesting. Here we have a discussion about the influence of books published 50 years ago and claiming they are of no influence. If that is true why are we still having this discussion?


notsneaky 05.18.07 at 4:23 pm

Spencer, the proper question is “Influence on whom and what?”.


Colin Danby 05.18.07 at 6:31 pm

You have some references on that, Luis? Post-1947 Indian economic planning and policy is usually traced through people like Mahalanobis and Nehru; none of the accounts I’ve read even mention American institutionalists. I’m sure JKG took an interest in Indian economy while he was ambassador and his views were received politely, but that doesn’t establish influence.

There *was* an interesting institutionalist/historicist strain (e.g. Radhakamal Mukherjee) in Indian economic thought before independence, which influenced what came to be called “Gandhian economics,” and those folks were certainly aware of U.S. institutionalism. But they and their policy vision seem to have been ignored post-independence by Nehru’s technocrats.


Luis Alegria 05.19.07 at 3:08 am

Mr. Danby,

I have no idea to what degree if any Galbraith influenced Indian economic policy, as you say it was born in a statist mold to begin with.

I recall this question about liberalization in reference to Galbraith from a Asian Development Bank seminar perhaps 25 years ago. This was at a time when established closed-economy orthodoxies were being challenged, including ours in the Philippines. I was working for a professional firm assisting foreign investors obtain investment permissions against the grain of Philippine government regulations. The ADB was on our side in these efforts.


Colin Danby 05.19.07 at 5:03 am

So Luis you withdraw the Galbraith-India comments in 13 and 31? I’m just keeping track of how many charges against JKG so far have been sustained. It would be most interesting though to know if anyone has written up his influence in the Philippines.

I’m just noticing the rest of your assertions in 13. Quickly (a) there are plenty of rigorous ways to examine power — are you really denying that power exists in the social world? Sociology, anthropology, psychology offer an abundance of approaches. (b) Social science is about understanding society, period. If you want to apply it to policy, fine, but even that does not logically entail making predictions. If you believe you can predict social phenomena, that’s nice.


Luis Alegria 05.19.07 at 3:35 pm

Mr. Danby,

I don’t see why I should withraw anything. I have read much more than enough of Galbraith to understand what his approach to public policy was.

There definitely aren’t rigorous ways to examine “power”, or much else in that sphere. There are lots of approaches, but is there anything that can be used to design, as in engineering, policies that will produce predictable results ? A structural engineer can depend on having solid data and reliable formulas to be able to specify that a given structure and materials will be able to take a given load. A social engineer (should we ever develop such a creature) cannot reliably predict that a population will improve its educational performance by a given amount given a certain educational strategy, and this is in the most metricised sub-specialty of social science.

Why should we care ? Simply because these things are important only to the extent that they are useful. Structural engineers are useful, they can create designs and (almost) guarantee their future performance. The entire technologial world we have is based on being able to make these predictions.

What good does it do to “understand society” if it doesn’t ultimately help us make informed decisions ? To be useful it must make accurate predictions, just like those structural engineers. If it never gets there then it is just an eccentric hobby. In the meantime we have trillions upon trillions of dollars at stake on decisions that could be guided by a better knowledge in these fields. An ability to make accurate predictions and engineer policy on the basis of reliable formulas is a vital necessity.


John Emerson 05.19.07 at 5:47 pm

Simply because these things are important only to the extent that they are useful.

I thought I was a fanatical pragmatist. If it ain’t engineering it ain’t shit, I hear Luis saying.


Colin Danby 05.19.07 at 7:20 pm

The weird part, John, is when people who claim to be anti-statist conservatives embrace social engineering. If only Milton Friedman were alive.


Luis Alegria 05.19.07 at 9:56 pm

Mr. Emerson,

If it ain’t engineering (or somewhere on the road to becoming engineering) then its just entertainment, thats right.

Mr. Danby,

There are conservatives and there are conservatives and there are conservatives. Who of us is pure anything ? Certainly not conservatives. There is no more varied bunch, ideologically speaking. In practice most of us pull a little from here and a little from there. Complaining about a conservative being unorthodox is missing the entire point about conservatism.

The way I see it is we will have public policy anyway, so it might as well be good public policy.


Richard 05.19.07 at 9:58 pm

I thought I wouldn’t have anything to say in this thread… but then I saw that I should repeat what I always say:

What good does it do to “understand society” if it doesn’t ultimately help us make informed decisions ? … If it never gets there then it is just an eccentric hobby. In the meantime we have trillions upon trillions of dollars at stake

But structural engineering, dollars and culture are just eccentric hobbies – for all that lives depend on them. They’re all motivated by value systems just like those that support the power of some persons over others. Economics is not acultural, and to the extent that it relies on or describes cultures, societies and interested actors it will find itself tangled up somehow with those timewasting philosophers.


Luis Alegria 05.19.07 at 10:20 pm

Mr. Richard,

I appreciate what you are saying, but it just doesn’t make sense to me. A valid social science should be able to predict under any cultural situation, just as structural engineering or thermodynamics work across cultures. Inca civil engineering can be evaluated by modern Western standards – a bridge is safe or not, road drainage is effective or it isn’t. If a social science cannot be understood and be made to work (make good predictions) for a given culture by a person from a different culture then there is something very wrong with it.

If those timewasting philosophers have something to add to improve the formulas then they aren’t timewasting at all. But unfortunately they haven’t done anything so useful so far.


Richard 05.20.07 at 7:41 am

Surprising. I think I agree with almost all of this ststement – I’m just left wondering: how would such a useful science develop? I suspect there’s something in the old cliche that social sciences are now about where the alchemists were in the development of chemistry: we can see a few things working, but our explanations are based on hope + metaphors drawn from other areas of thought. Under this model, having a rich pool of ideas would be one part of the recipe for eventually developing such a theory (and hypothesis-production and logical argument are part of philosophical work). The other vital element: reproducible circumstances from which to derive observations, has been a bit backward in coming forward.

BTW: in light of recent events, the bridge safety example would seem to demonstrate rather how these things are never known except in retrospect.


Luis Alegria 05.20.07 at 3:30 pm

Mr. Richard,

Part of the problem I think is the constraint of thinking of the social sciences only as science, rather than as technology. Physical science developed through the interaction of empirically derived technologies (i.e., engineering) mainly through the accumulation of “best practices” by generations of craftsmen, which exploited phenomena nd experience without requiring hypotheses or deep understanding, and the intellectual work of the scientific method. Both tracks contributed to each other, advancing both. It wasn’t scientists who created the tools and technologies for the precision instruments that drove the scientific revolution.

In the social sciences it seems to me there is much too little of the technology approach, there is too little consistent R&D and development of empirically-based “best practices” even in the fields where this is possible – and there are many cases where we have reproducible circumstances and plenty of scope for an R&D process, education for one.

I think the good old Marks Mechanical Engineers Handbook is a fine example of these best practices, some derived or informed by physical science, but most of which are based on pure experience. I recommend a look at one of these, and in fact Marks Manual is still in print –

There will be no real social science until there are social science equivalents to the Marks Handbook, unscientific as it is.

Comments on this entry are closed.