What Is the Left-on-Left Debate Among Economists Really About?
-- Posted by Neil H. Buchanan
In a series of recent posts (here, here, here, and here), I have explored a debate that emerged recently between "orthodox left" economists and "heterodox left" economists. The former group's most recognizable practitioner is Paul Krugman, and it includes pretty much every economist who might have had any connection to a Democratic politician in the U.S. in the post-WWII period -- Larry Summers, Bob Solow, Janet Yellen, and so on. (Prior to roughly 2000, even some Republican economic advisors could reasonably have been called "orthodox left" in the sense that I am using that term here. Summers once served in the Reagan Administration, for example.) The heterodox left has been almost completely marginalized, not just by people on the right, but as I discussed yesterday, very much by the orthodox left. The most recognizable name among the heterodox is probably James K. Galbraith, who is an excellent economist, but whose broader name recognition almost surely derives from his late father, the great John Kenneth Galbraith.
Despite having written several thousand words about the recent orthodox/heterodox tussle, I have thus far been somewhat circumspect about describing the actual differences between the two groups. I have mentioned and minimally described "production functions," without really explaining why they matter. And I have also noted that, on the key policy questions of the day, the two groups seem to agree on nearly everything. This last point, in turn, seems to be what is motivating Krugman's unpleasantness (for example, titling a recent blog post "Hangups of the Heterodox"), because he just does not understand why he is being criticized for (as he appears to see it) getting the right answers via a different route than the heterodox economists would prefer. One can practically hear him muttering: "Why don't you jerks shut up?! We agree on what ultimately matters!"
In that blog post, Krugman off-handededly dismisses a recent article in Dissent by Galbraith, calling it an "oddly off-center attack on Thomas Piketty." At best, this is snark. At worst, it is malpractice. Whatever one might think of Galbraith's argument, waiving it away as if it is some kind of unmoored personal attack is simply out of bounds. One might ultimately conclude that Piketty's argument is important, even after accepting Galbraith's critique of Piketty's book. That is my conclusion, and in fact, it seems to be Galbraith's conclusion as well. All one needs to do is read Galbraith's entire article. But dismissing the article as if it is meaningless is simply dishonest.
But the bulk of Galbraith's piece is undeniably a full-on attack on much of Piketty's book. In particular, he faults Piketty for mischaracterizing a famous debate called "the Cambridge Controversies" (aka "the Capital Controversies," "the Two Cambridges," "The Cambridge-Cambridge Debate," and similar variations). Galbraith is right that Piketty is wrong. Here, I will briefly explain that debate, what Piketty got wrong, and how the misrepresentation by Piketty reflects badly on him (and Krugman, and others in the orthodox camp). In a post next week, I will bring this home by describing why it matters, in terms of policy implications.
The word "capital" can refer to financial assets, as in "having lots of capital to invest." For macroeconomic purposes, however, and especially for the purposes of the Cambridge Controversies, capital refers to non-financial assets that are used to produce economic output. Generally, this means the physical assets that capitalists own, such as factories, transportation networks, machinery, and so on.
Economics is only useful if we make simplifying assumptions. Otherwise, as Joan Robinson once put it: "A model which took account of all the variegation of reality would be of no more use than a map at the scale of one to one." Capital is highly variegated, and it is tempting to think that there must be some way to assume away much of the detail, in order to measure it.
Suppose that one lived in an economy with different amounts of drill presses, railroad cars, office chairs, computer printers, and so on. We live in such an economy. Now, what if the amounts of some or all of those things changed? How would we know if we then had "more capital" or "less capital" than we do now? The Cambridge Controversies were, at their core, a disagreement between the American Keynesians (mostly at MIT, whom I am now referring to as the orthodox left) and the British Keynesians (mostly at Cambridge, the heirs of whom constitute the heterodox left) over the question of whether there is a coherent way to measure capital, such that it can be reduced to a single number.
Why is that such a difficult question? After all, we think we can measure labor by "hours worked," even though we know that there are different qualities and types of labor effort. It turns out, however, that even measuring labor in that way is problematic, and the additional complications in measuring capital make matters worse. One cannot measure all capital goods by their original cost, their replacement cost, by their inflation-adjusted cost, or any other dollar-denominated cost, because all of those measures are affected by changes in the amounts of capital in the economy. So the American side of the debate tried to build a measuring rod that would be independent of what it purports to measure.
They failed. Not only did they fail, but their leader admitted it openly and unreservedly. In a famous 1966 article that essentially concluded the debate, Paul Samuelson wrote that his side had lost. Their attempt to claim that capital could consistently be measured had failed, and thus "production functions" -- which presume the ability to measure capital and labor coherently -- make no sense as a matter of economic theory.
Part of what angered Galbraith was Piketty's casual mischaracterization of the Cambridge Controversies, including the claim that the American side's use of production functions "definitively carried the day" in the 1970's. This is true only in the sense that the American side (Samuelson himself honorably excluded) simply chose to act as if they had won the capital debate. They "carried the day" by sheer force, deciding that they would act as if the winning side had made merely an academic point that has no significance.
This is why it was especially galling to read Krugman's blog post yesterday, in which he was attacking right-wing economists for failing a "moral" test, by refusing to admit that they have been proved wrong. Pointing to the clear evidence that the right-wingers have been proven completely wrong, Krugman wrote: "Tests in economics don’t get more decisive; this is where you’re supposed to say, 'OK, I was wrong, and here’s why'." Actually, there is an equally decisive test, which is when a leader says, "OK, we were wrong, and here's why," yet his followers then spend a half-century saying, "Yeah, whatever."
Or, as Krugman put it in his "hang-ups" post: "And what’s going on here, I think, is a fairly desperate attempt to claim that the Great Recession and its aftermath somehow prove that Joan Robinson and Nicholas Kaldor were right in the Cambridge controversies of the 1960s. It’s a huge non sequitur, even if you think they were indeed right (which you shouldn’t.)" There is no need to use the Great Recession to prove that Robinson and Kaldor were right, because Samuelson already admitted that they were right.
In the end, the only way to justify Piketty's and Krugman's (and Solow's, as I will discuss next week) attempt to claim victory in the face of their own side's surrender is by claiming that none of it matters. In other words, Krugman might be saying that one should not think that the British side was "right" in the sense that they believed that winning the Cambridge Controversies otherwise undermined American-style Keynesianism. But they are only wrong if one believes either or both of the following: (1) Might makes right, so that the Brits lost the debate after winning it, when they could not get enough people to follow their path, or (2) The path to the right policy answer does not matter. It is that second issue to which I will return next week.
In a series of recent posts (here, here, here, and here), I have explored a debate that emerged recently between "orthodox left" economists and "heterodox left" economists. The former group's most recognizable practitioner is Paul Krugman, and it includes pretty much every economist who might have had any connection to a Democratic politician in the U.S. in the post-WWII period -- Larry Summers, Bob Solow, Janet Yellen, and so on. (Prior to roughly 2000, even some Republican economic advisors could reasonably have been called "orthodox left" in the sense that I am using that term here. Summers once served in the Reagan Administration, for example.) The heterodox left has been almost completely marginalized, not just by people on the right, but as I discussed yesterday, very much by the orthodox left. The most recognizable name among the heterodox is probably James K. Galbraith, who is an excellent economist, but whose broader name recognition almost surely derives from his late father, the great John Kenneth Galbraith.
Despite having written several thousand words about the recent orthodox/heterodox tussle, I have thus far been somewhat circumspect about describing the actual differences between the two groups. I have mentioned and minimally described "production functions," without really explaining why they matter. And I have also noted that, on the key policy questions of the day, the two groups seem to agree on nearly everything. This last point, in turn, seems to be what is motivating Krugman's unpleasantness (for example, titling a recent blog post "Hangups of the Heterodox"), because he just does not understand why he is being criticized for (as he appears to see it) getting the right answers via a different route than the heterodox economists would prefer. One can practically hear him muttering: "Why don't you jerks shut up?! We agree on what ultimately matters!"
In that blog post, Krugman off-handededly dismisses a recent article in Dissent by Galbraith, calling it an "oddly off-center attack on Thomas Piketty." At best, this is snark. At worst, it is malpractice. Whatever one might think of Galbraith's argument, waiving it away as if it is some kind of unmoored personal attack is simply out of bounds. One might ultimately conclude that Piketty's argument is important, even after accepting Galbraith's critique of Piketty's book. That is my conclusion, and in fact, it seems to be Galbraith's conclusion as well. All one needs to do is read Galbraith's entire article. But dismissing the article as if it is meaningless is simply dishonest.
But the bulk of Galbraith's piece is undeniably a full-on attack on much of Piketty's book. In particular, he faults Piketty for mischaracterizing a famous debate called "the Cambridge Controversies" (aka "the Capital Controversies," "the Two Cambridges," "The Cambridge-Cambridge Debate," and similar variations). Galbraith is right that Piketty is wrong. Here, I will briefly explain that debate, what Piketty got wrong, and how the misrepresentation by Piketty reflects badly on him (and Krugman, and others in the orthodox camp). In a post next week, I will bring this home by describing why it matters, in terms of policy implications.
The word "capital" can refer to financial assets, as in "having lots of capital to invest." For macroeconomic purposes, however, and especially for the purposes of the Cambridge Controversies, capital refers to non-financial assets that are used to produce economic output. Generally, this means the physical assets that capitalists own, such as factories, transportation networks, machinery, and so on.
Economics is only useful if we make simplifying assumptions. Otherwise, as Joan Robinson once put it: "A model which took account of all the variegation of reality would be of no more use than a map at the scale of one to one." Capital is highly variegated, and it is tempting to think that there must be some way to assume away much of the detail, in order to measure it.
Suppose that one lived in an economy with different amounts of drill presses, railroad cars, office chairs, computer printers, and so on. We live in such an economy. Now, what if the amounts of some or all of those things changed? How would we know if we then had "more capital" or "less capital" than we do now? The Cambridge Controversies were, at their core, a disagreement between the American Keynesians (mostly at MIT, whom I am now referring to as the orthodox left) and the British Keynesians (mostly at Cambridge, the heirs of whom constitute the heterodox left) over the question of whether there is a coherent way to measure capital, such that it can be reduced to a single number.
Why is that such a difficult question? After all, we think we can measure labor by "hours worked," even though we know that there are different qualities and types of labor effort. It turns out, however, that even measuring labor in that way is problematic, and the additional complications in measuring capital make matters worse. One cannot measure all capital goods by their original cost, their replacement cost, by their inflation-adjusted cost, or any other dollar-denominated cost, because all of those measures are affected by changes in the amounts of capital in the economy. So the American side of the debate tried to build a measuring rod that would be independent of what it purports to measure.
They failed. Not only did they fail, but their leader admitted it openly and unreservedly. In a famous 1966 article that essentially concluded the debate, Paul Samuelson wrote that his side had lost. Their attempt to claim that capital could consistently be measured had failed, and thus "production functions" -- which presume the ability to measure capital and labor coherently -- make no sense as a matter of economic theory.
Part of what angered Galbraith was Piketty's casual mischaracterization of the Cambridge Controversies, including the claim that the American side's use of production functions "definitively carried the day" in the 1970's. This is true only in the sense that the American side (Samuelson himself honorably excluded) simply chose to act as if they had won the capital debate. They "carried the day" by sheer force, deciding that they would act as if the winning side had made merely an academic point that has no significance.
This is why it was especially galling to read Krugman's blog post yesterday, in which he was attacking right-wing economists for failing a "moral" test, by refusing to admit that they have been proved wrong. Pointing to the clear evidence that the right-wingers have been proven completely wrong, Krugman wrote: "Tests in economics don’t get more decisive; this is where you’re supposed to say, 'OK, I was wrong, and here’s why'." Actually, there is an equally decisive test, which is when a leader says, "OK, we were wrong, and here's why," yet his followers then spend a half-century saying, "Yeah, whatever."
Or, as Krugman put it in his "hang-ups" post: "And what’s going on here, I think, is a fairly desperate attempt to claim that the Great Recession and its aftermath somehow prove that Joan Robinson and Nicholas Kaldor were right in the Cambridge controversies of the 1960s. It’s a huge non sequitur, even if you think they were indeed right (which you shouldn’t.)" There is no need to use the Great Recession to prove that Robinson and Kaldor were right, because Samuelson already admitted that they were right.
In the end, the only way to justify Piketty's and Krugman's (and Solow's, as I will discuss next week) attempt to claim victory in the face of their own side's surrender is by claiming that none of it matters. In other words, Krugman might be saying that one should not think that the British side was "right" in the sense that they believed that winning the Cambridge Controversies otherwise undermined American-style Keynesianism. But they are only wrong if one believes either or both of the following: (1) Might makes right, so that the Brits lost the debate after winning it, when they could not get enough people to follow their path, or (2) The path to the right policy answer does not matter. It is that second issue to which I will return next week.