Pathways for Transformative Economics and Politics

EDI Primary logo color.

A Comment on Dani Rodrik’s “New Paradigm for Economic Policy”

by James K. Galbraith

profile image
This article forms part of Rubric 7
Aerial of a river

Galbraith argues that the neoclassical doctrine has taken over mainstream economics to the point where the two cannot be separated.  With this in mind, he challenges Rodrik's argument that his "productivism" can be "mainstream," except to the extent that the ideas are in fact neoclassical, which in key respects they are.  He further argues that Rodrik's goals -- good jobs at fair wages -- can be pursued effectively only outside the mainstream/neoclassical paradigm, with or without its productivist variation.

The central methodological claim in Professor Rodrik’s essay is that mainstream economics is “paradigm-free.” By this, Rodrik means that the policy implications of the mainstream worldview are “indeterminate.” This is a most interesting claim. Yet, in his contribution, Rodrik does not clearly define “mainstream economics,” so let me begin by venturing such a definition. We can then ask whether it is possible to separate “mainstream” from “neoclassical” – and then to follow Professor Rodrik to his offering of a “productivist” version of mainstream economics.

The metaphor of “mainstream” is hydrological. The word distinguishes the primary channel – let’s say of a river – from tributaries and diversions. It is by definition where most of the water will be found. Let’s therefore define “mainstream economics” as theories shared by most academic economists at what they claim to be “leading” universities, published in what they describe as the “top” journals, as ranked by… well… by mainstream economists. This is a fairly well-defined community. It would be illuminating to discover that its members have no theoretically-determined policy views. Is it true?

To be sure, “mainstream” in economics does not necessarily imply “neoclassical.” To take a canonical case, the Harvard economics department has always been mainstream. It has not always been neoclassical. Until perhaps 50 years ago, Harvard economics was eclectic. Among the traditions that were then well-established, New Dealers believed that the mixed economy had supplanted classical capitalism. Keynes had established an indispensable managerial role for the state. The most prominent Harvard economist was the former director of war-time price control and an apostle of big organizations and countervailing power. Socialism and planning were legitimate alternatives to capitalism; the Soviet Union was studied as an alternative system. Marxism was represented in the senior and junior ranks. From the standpoint of method an earlier figure, Joseph Schumpeter, though vehemently pro-capitalist, had been closer to Marx than to Rodrik. Apart from the young radicals who were purged in 1971, practically all of the Harvard economics faculty at that time would have been considered mainstream.

Today, apart from a rare survivor of that era – Stephen Marglin comes to mind – none of these tendencies can be found at Harvard. Nor, in the “top” departments, anywhere else. To put this in concrete terms, could any modern mainstream economist advocate the “Chinese model” – except implausibly by claiming that China has become a capitalist state? Could any advocate the Meidner-Rehn model of trade-union centralization and egalitarian wages practiced with great success for decades in Sweden? Not to mention the harsh-but effective industrialization of the Soviet Union under Stalin. They could not, and there are no such examples in any “leading” economics department.

Today, the mainstream in economics is indisputably and uniformly neoclassical. It is not without some differences, among neoclassicals, over the policy implications of that stance. An elementary textbook that highlights “rival perspectives” in economics distinguishes “liberal” and “conservative” variants of “conventional” or mainstream economics. Both are built from the same neoclassical basics: households, firms, rationality, profit-seeking, supply, demand, price adjustment, equilibrium. They differ mainly over flaws, frictions, externalities and other caveats and qualifications. As a core matter of policy, “both believe that capitalism is the greatest economic system.”1 This minimalist definition may be taken as uncontroversial.

In practice, neoclassical economics specifies far more than this in most cases, including a set of supposedly mainstream modeling techniques (such as CGE and DSGE) and econometric methods. To say that today’s mainstream economics is paradigm-free is therefore absurd. The entire project of mainstream economics for decades has been to narrow intellectual horizons – to specify a paradigm. Within the paradigm, there is an allowed range for disputes, but it is narrow.2 Though some departures are permitted, the bias is always toward markets, competition, flexibility, openness. No mainstream economist privileges planning, monopoly, controls or autarky.

Rodrik would like to separate himself from neoclassical dogma, but his commitment to the present-day neoclassical mainstream can be adduced in several ways. Most important, his decision-maker is always the “firm.” Firms make “employment, investment, and innovation decisions, in the language of economists.” Firms make “decisions on how many workers to employ, how much to pay, what kind of technology to deploy, and how to organize work [to] effect not just the bottom line, but the life chances of prospective employees and their communities.”

In Rodrik’s paradigm, government exists to provide what “firms need: a reliable and skilled workforce, good infrastructure, an ecosystem of suppliers and collaborators, easy access to technology, and a ‘sound’ regime of contracts and property rights.” Conspicuously missing from this list: democratic control, public purpose, social insurance, human and civil rights, wage and labor standards. These are not in the paradigm. Nor does his model mention unions, collective bargaining, or the non-profit sector. (One might ask, within his paradigm, could Harvard University exist?) The model here is not of real-existing capitalism. It is of a neoclassical vision with an unrealistically limited definition of government and a set of institutions purged of many that necessarily exist in the real world.

Now we come to Rodrik’s proposed innovation, “productivism.” Though the word and its cognates occur repeatedly in his essay, it is not entirely clear to me what they mean. As a matter of logic, if some economic activities are productive, others must be unproductive. This distinction is readily found in classical political economy – it is the subject of an entire chapter in Adam Smith’s Wealth of Nations. But Rodrik does not appear to be specifying a distinction between tangible and intangible outputs, as Smith did. For him, it is clear, services are not less productive than the creation of material objects.

Instead, it would appear that Rodrik’s concept of what is productive is related not to outputs but to the quality of the jobs that are provided. Thus, a “productive” economy is one that generates “good jobs” that “pay sufficiently well to allow for a reasonable living standard.” This is the crux of Rodrik’s policy problem. If, as Rodrik specifies, “firms” generate jobs, under what conditions could they be induced to ensure that jobs are plentiful, and that they meet Rodrik’s criteria for “good jobs”?

Here, let me suggest, Professor Rodrik has to struggle to align his thinking with the logic of mainstream economics, or even with the common meaning of words in economics generally. “Productivity” is normally defined as a ratio of output to labor (Y/L), where Y is measured in physical units and L in labor-time. Productivity goes up when the labor component is driven down – that is, when jobs are eliminated. If the firm is the decision-maker, it is in the logic of the profit-seeking firm to replace costly labor with machines, computers, and artificial intelligence.

If the definition of productivity is given in money terms, then productivity is essentially profit, which goes up when wages are cut relative to prices. Either way, productivity is not intrinsically connected to the provision of good jobs, so long as firms get to decide on technologies and pay. This is why, in real-existing capitalism, governments and unions make many decisions about pay. It is also why, in real-existing capitalism, governments and the non-profit sector are on hand to provide jobs, to the extent that private firms are unwilling or unable to do so.3 Proposals like the job guarantee are thus not radical departures; they are rather improvements on current and past policy practice.

From this I would argue that to reach the objective that Professor Rodrik advocates – let’s summarize it as good jobs at decent wages – there is no alternative to laws, institutions, unions, wage standards and countervailing power. The problem is not to align the firm to this objective through market incentives; it is to regulate and overrule the market in the interest of a stable and prosperous society. To advocate this position coherently, therefore, there is no alternative, except to repudiate today’s mainstream.

To come back to the top of Rodrik’s essay, he states that his new paradigm is “suspicious of large corporations,” exhibits “skepticism toward technocrats” and is “less instinctively hostile to populism.” On one reading, these features may seem to evoke the neoclassical textbook model of perfect competition. But are they mainstream? Does Rodrik really propose to produce automobiles in small shops or steel in backyard furnaces? Does he propose to produce aircraft or electricity without engineers? Does he propose to approve new drugs without scientific research? This is all quite curious. I suspect few other “leading” economists would regard such positions as mainstream. If they would not, then Professor Rodrik’s effort to link these rather special features of his proposed paradigm to the mainstream – neoclassical or otherwise – would appear to be unsupported.

But then, one might ask, is there a heterodox tradition that might take him in? I do not see an obvious home for such ideas in the Institutionalist, Keynesian, Post Keynesian or ecological or environmental schools. They are unconnected to issues of debt, credit and money, and so not relevant to Modern Monetary Theory. Though my father had roots in the avowedly populist United Farmers of Ontario – back in the 1920s – these aspects of Rodrik’s ideas are, of course, highly opposed to those of The Affluent Society and The New Industrial State.

There is a lingering tradition of neo-populist antitrusters. They are very active, these days, in the political space, but they are not – so far as I’m aware – strongly represented within any present-day academic school of economic thought. And, possibly, for good reason.



1Voices on the Economy, Amy S. Cramer and Laura Markowitz, Second Edition, 2023, Vol. 1, 124.

2In 1987 Paul Krugman wrote, “If there were an Economist’s Creed, it would surely contain the affirmations ‘I understand the Principle of Comparative Advantage’ and ‘I advocate Free Trade'” — before going on to cast doubt on both propositions – and yet winding up as a free trader anyway. This is a nice example of policy discipline in mainstream/neoclassical economics.

3In services the definitions of “productive” and “productivity” are even trickier. Services output in the national accounts is measured by the revenue generated. (A highly-paid professor, or concierge, by definition, makes a larger contribution to GDP than a poorly-paid one.) Since services use little capital, most services revenue goes to labor – either as wages or salaries. Productivity is therefore either (a) money revenue divided by factor payments (equal to one by definition and unable to be improved) or (b) money revenue divided by labor hours – in which case “productivity” rises when wages or salaries and therefore prices rise. Within the firm, revenue can be distributed to wages and salaries unevenly, and usually is. If the owner makes the decisions – as Rodrik assumes to be the normal (indeed, universal, paradigmatic) case, then there is a flat conflict of interest between the “firm” and the provision of “good jobs.” The owner pays himself well and everyone else as poorly as possible, unless – once again – wage standards and collective bargaining intervene. But again, these are not in the paradigm of mainstream economics as neoclassical economics.


  • Cramer, Amy S and Laura Markowitz. 2023. Voices on the Economy: How Open-Minded Exploration of Rival Perspectives Can Spark Solutions to Our Urgent Economic Problems, Second Edition, Volume 1. Tucson: Voices on the Economy.
  • Galbraith, John Kenneth. 1958. The Affluent Society. Cambridge: Houghton Mifflin
  • Galbraith, John Kenneth. 1967. The New Industrial State. Cambridge, Houghton Mifflin.
  • Krugman, Paul. 1987 Is Free Trade Passé? The Journal of Economic Perspectives, Vol. 1, No. 2 (Autumn), 131-144.
  • Rodrik, Dani. November 2023. A New Paradigm for Economic Policy“.
  • Smith, Adam. An Inquiry into the Nature and Causes of the Wealth of Nations. Many editions.

Related material

James K. Galbraith teaches at The University of Texas at Austin. He holds a PhD in Economics from Yale.

Back to top