The history of income and wealth is always deeply political, chaotic, and unpredictable. How this history plays out depends on how societies view inequalities and what kinds of policies and institutions they adopt to measure and transform them. No one can foresee how these things will change in the decades to come. (Piketty, Capital 35)
In fact, neither economic liberalization that began around 1980 nor the state interventionism that began in 1945 deserves such praise or blame. France, Germany, and Japan would very likely have caught up with Britain and the United States following their collapse of 1914-45 regardless of what policies they had adopted (I say this with only slight exaggeration.) The most one can say is that state intervention did no harm. Similarly, once these countries had attained the global technological frontier, it is hardly surprising that they ceased to grow more rapidly than Britain and the United States or that growth rates in all of these wealthy countries more or less equalized....Broadly speaking, the US and British policies of economic liberalization appear to have had little effect on this reality, since they neither increased growth nor decreased it. (Piketty 99)
The first quoted passage urges us to take political control ("transform") over our economic destiny (see also p. 234 where Piketty endorses the claim that "political forces were central.") In Piketty's book 'chaotic' tends to be used to describe unstable, ad hoc political decision-making in context of great disruptive, social upheaval (war, inflation, "shocks," etc.). (His 'chaotic' is unconnected to the stochastic view of randomness beloved by mathematical economists.) It's precisely because politics has an influence on the economy that the world becomes unpredictable. The italicized passage is the central dogma of the political historian's craft.*
Yet, in the second quoted passage, Piketty embraces a form of technological determinism on growth in which much macro-policy has no discernible influence at all on growth. I am not claiming that the two passages contradict each other. For, Piketty's position seems to be that (a) macro-economic policy is largely impotent on growth; (b) governments can make macro-economic mistakes (embrace too much autarky; embrace too much capital liberalization (71);+ (c) governments can impact the distribution of wealth; (d) ensuring the diffusion of technology and education is key to long-term growth (71; 572; see also the emphasis on productivity at 234). So, while governments can invest in education (307), the impotence of traditional macro-economic policy on growth is also emphasized in the conclusion: "countries at the world technological frontier--and thus ultimately for the planet as a whole--there is ample reason to believe that the growth rate will not exceed 1-1.5 percent in the long run, no matter what economic policies are adopted." (572)
Clearly, the crucial concept in Piketty's analysis here is the 'technological frontier.' Amazingly enough I have just quoted two out of four mentions of the term in the whole (long) book, and these are the most detailed descriptions of it. Curiously, the term is introduced on p. 97 without a footnote or explanation. It is one of the few places in his book where Piketty assumes a background in economics in his reader. But its meaning is clear from context: (i) a country's (world's) technological frontier clearly limits the possibilities for growth. (As we philosophers like to say, it's a modal concept.) And (ii) growth can be faster toward the frontier than being on the frontier, where a "relatively slow pace" is "characteristic" (97).
Now, to see the significance of this, I quote Pikkety on his “fundamental inequality" which is "r > g (where r stands for the average annual rate of return on capital, including profits, dividends, interest, rents, and other income from capital, expressed as a percentage of its total value, and g stands for the rate of growth of the economy, that is the annual increase in income or output)" it "will play a crucial role in this book. In a sense, it sums up the overall logic of my conclusions." (25) If g is doomed to a relatively slow pace at the technological frontier, Piketty's crucial arguments go through rather easily.
Now, that Piketty introduces the concept of a technological frontier goes against the official spirit of the book. For that concept is a sub-set of the so-called production function. I mention this because Piketty has a trenchant methodological criticisms of the so-called Cobb-Douglas production function (217-9) and the so-called "Cambridge Capital debate" in which Solow and Samuelson "defended the production function with substitutable factors," 231). His criticisms can be summed up as follows: [i] these functions can be easily used for political manipulation/ideology (the treatment of Samuelson's embrace of "a fairly peaceful and harmonious view of the social order" (218) is very important)** and [ii] that absent a great deal of empirical data these disarmingly "simple" (218) functions only "confuse" the debate (232). There is a good reason to be suspicious of these functions--they abstract too much away from history and institutions (recall).
Oddly, two pages before he first introduces the very idea of a 'technological frontier,' Piketty writes, "it is as difficult to predict the pace of future innovations as to predict future fertility," but he immediately adds "the history of the past two centuries make it highly unlikely that per capita output in the advanced countries will grow at a rate above 1.5 percent per year." (95) So, ultimately, this is straight historical induction (from the period 1820-2012). But note that even if we grant that the induction is an induction about the technological frontier (and this is by no means obvious), the data being used is not and cannot be informative about what determines the particular modal features of the frontier.++ These macro-economic data are uninformative about the possible (institutional) mechanisms (or laws) that determine the technological frontier.*** As Piketty himself notes in strikingly modal terms, "the inequality r>g should be analyzed as a historical reality dependent on a variety of mechanisms and not as an absolute logical necessity." (361)
So, the idea of a technological frontier is a kind of window dressing to give a historical induction a theoretical veneer. It is especially striking that Piketty would resort to it because he is extremely sensitive to the fact that economists will use precisely such constructs in the service of political ideology, not science; note not just the comment on Samuelson above, but recall Piketty's comments on Kuznets in my earlier post on his book.**
Now, Piketty has other arguments for his pessimism about r>g; I don't take my post as a refutation of his general framework. Rather, all I claim is that (a) Piketty's technological determinism on growth is assumed, and (b) that claims about the possibilities for future growth are as strong as any social scientific historical induction I have seen (that cuts two ways). More importantly, (c) Pikett's book does not provide us with a theory to develop research on the mechanisms of growth or the modal properties of the technological frontier. So, to connect this to the theme of my first post, there is a sense in which Piketty's framework returns us to a way of thinking about historical political economy that I associate with Hume and Adam Smith (as practiced by Kuznets/Friedman--again no small achievement), but he does not offer hope yet on an understanding of the nature and causes of economic growth that goes beyond those available in the eighteenth century.
*For all I know this dogma may be a universal empirical truth.
+I return to Piketty's analysis of the political consequences of international capital flows on poor countries before long.
++ Modal economics is no laughing matter, after all.
**In effect, Piketty claims that in Kuznets it's fairly easy to distinguish the ideological from the scientific, but that this much less so in Samuelson. I may be projecting my own prejudices here, but I agree with Piketty's claim. (Some other time I will explore this issue more fully in the context of Samuelson's work.) I postpone to another post, Piketty's engagement with Solow.
***The economist, Steve Durlauf, was kind enough to share slides of his commentary on Piketty with me; Durlauf also points to absence of mechanisms in some of Piketty's other crucial claims.
I am not familiar with Gary Becker's actual work but from what I remember of Parson's claims about Marhsall's claim of imperialism seemed to resonate around the claim that the rationalist methodological homogeneity of objects of observation is not just a methodological decision but descriptive of actuality. Is Becker's imperialism the explicit ontological continuation of economics into other fields with the normative focus , the implicit ontological continuation of economics into other fields or something different?
Posted by: Aaron Alvarez | 05/10/2014 at 09:18 AM
Interesting that your last post on Piketty seemed so sympathetic to the search for long-run economic 'constants', while this one seems hostile. I am more in the spirit of this post.
Posted by: Marcus Stanley | 05/11/2014 at 12:08 AM
Marcus, I am, in fact, sympathetic to the search long-run economic 'constants,' as a means to the development of better theory and/or understanding the mechanisms/laws of capitalist economies/structural change (etc.). In this post, however, I note that Piketty also helps himself to concepts that he knows to be flawed on some level (and that prevent such a search). And in the post I understate it a bit, because Piketty is (correctly) quite critical of the Samuelson textbook on Russia, which was flawed precisely because it relied on production frontier.
Posted by: eric Schliesser | 05/11/2014 at 08:21 AM
Aaron, I suspect you intended this response to the post on Becker/Parsons. Parsons has a lot of criticisms of Marshall. (Marshall plays a very big role in his Structure.) You can read about Becker's economic imperialism in some of the links that I provide, but the short answer is 'yes.'
Posted by: eric Schliesser | 05/11/2014 at 08:25 AM
But its meaning is clear from context: (i) a country's (world's) technological frontier clearly limits the possibilities for growth. (As we philosophers like to say, it's a modal concept.) And (ii) growth can be faster toward the frontier than being on the frontier, where a "relatively slow pace" is "characteristic",.
Eric, I still think you are mistaken in assimilating the concept of the technological frontier to the concept of technological determinism. And your use here of the phrase "a country's technological frontier", it seems to me, is connected with the problem. I believe that for Piketty and others, the concept of a technological frontier is an inherently global notion. It is used to explicate the phenomenon of "catch-up growth" and the tendency for rates of catch-up growth in rapidly developing countries to decline eventually.
Let's be optimistic and say that some new combination of entrepreneurial innovation and macroeconomic policy jolts the entire world economy into a sustained 5% global growth rate. There would still be a technological frontier in that world, in the sense that countries at a lower level of economic development would be able to achieve growth levels significantly higher than 5% by engaging in catch-up growth, but when they had achieved an equivalent level of economic development, they would not be likely to grow faster than other developed and technologically advanced countries.
Now maybe this understanding of catchup growth is right and maybe it is wrong, but it is a quite different concept than the concept of technological determinism. Even if there is no roughly "determined" pace of global technological development, and human beings have the ability to achieve significantly higher or lower levels of global growth and development based on their aggregate policy choices, there is still going to be a technological frontier given any pace of global development defined by where the "leading edge" is.
Posted by: Dan Kervick | 05/19/2014 at 12:28 AM
Dan, first you are not engaging with a crucial feature of my argument (which was, in part, about Piketty's claim about the impotence of macro-economic policy), and you ascribe a misunderstanding where there is none. I am not at all denying the part you are stressing about catch up growth. (It's in the original post -- recall (ii) about growth being faster toward the frontier [that's your catch up] than on the frontier.) Nor am I denying that it is global--recall that in the post I also allowed that it can a global/world frontier (not just a country one). Why do you treat what I say as somehow denying that? You are merely, partially repeating Piketty's position that I have described above. And you, too, treat the 'technological frontier' as a modal concept that limits possibilities; on your view *if you are on that 'frontier' you somehow have a certain potential for growth*--it's this thought (there is a frontier, at the frontier one can grow a certain limited amount, etc.) that I associate with technological determinism. Even if you (and Piketty) think this is not technological determinism as such (maybe you associate the phrase 'technological determinism' with something else?), then what is it? (The frontier determines maximum possible growth rates if you are on the frontier.) Moreover, in doing so, you just recycle a version of neo-classical economics and its thinking about the production possibility frontier. This despite the fact that in your blogging you often claim you reject neo-classical economics. (In the post I explain why Piketty offers the right sort of criticism(s) of production possibility frontier.)
Now, this is not to deny that Piketty may well reject technological determinism in my sense or some other sense. It's certainly not -- as he wrote me in an email that I will discuss before long -- the point of his book to advocate technological determinism; I fully grant that. But if you deploy this 'frontier' concept in the way you do, then you do instantiate a version of technological determinism. Now, I may be wrong about that, but, perhaps, you could offer an argument rather than just a restatement of the view?
By the way, what is your evidence that there is a technological frontier defined by the leading edge, etc.? What supplies your confidence on this matter? Samuelson's text-book?
Posted by: Eric Schliesser | 05/19/2014 at 12:54 AM
Sorry, Eric, I feel like you are throwing a lot at me at once that I can't possibly respond to. I really don't know what "neoclassical" means in this context, nor do I care. Nor can I recall ever having formulated a previous opinion in my blogging on technology and growth, so that seems irrelevant. Can't we just stick to this one issue?
First I agree with you in wanting to push on Piketty's apparently pessimistic attitudes about 21st century growth. I just don't think the concept of the technological frontier, as he uses it in the few places he does in the book, is germane to that issue.
Let's put it this way: Consider countries X, Y and Z. X and Y are among the most technologically advanced countries in the world. Z is not nearly as advanced economically but has abundant people and resources. Would you say X, Y and Z all have the same growth potential?
Also, why is the fact that "potential" is a modal concept significant? There are modal concepts permeating almost every field of thought, and they always play some kind of role where practical reasoning and policy are concerned. It would be shocking if economics and economic policy could get by without them.
When Piketty says, "Similarly, once these countries had attained the global technological frontier, it is hardly surprising that they ceased to grow more rapidly than Britain and the United States or that growth rates in all of these wealthy countries more or less equalized", do you think that is a reasonable statement or not? Do you think it is surprising that growth in France, Germany and Japan slowed down after they had caught up with the US and Britain? If not, how would you describe what happened?
Posted by: Dan Kervick | 05/19/2014 at 02:31 AM
It's great that economics has modal concepts. The issue is what are the mechanisms/laws (etc.) that underwrite these.
We can't answer the questions you ask without understanding the mechanisms/law that drive growth or 'technological advance'. One can't claim something is or isn't surprising then.
What is the evidence that there has been 'catch up'? What you and Piketty do, is notice a relative pattern of growth and then describe that outcome pattern in terms of catch up. But without measures or proxies that track this so-called frontier, you are just describing a pattern that seems intuitive to you (and may rely on tacit assumptions about diminishing rates of return).
Finally, the technological frontier is a technical term in economics; Piketty uses it in that sense. His data, however, have no measures to track it.
Posted by: Eric Schliesser | 05/19/2014 at 09:36 AM