Krugman on Friedman: An Austrian Approach

Paul Krugman, in a most recent post, argues “Backward moves the macroeconomic debate” with “the result that our economic discourse is significantly more primitive now that it was 70 years ago.” Per Krugman, this backward movement is apparent in the use by some opponents of active demand management policy, such as Amity Shlaes, and of the “supposed legacy of Milton Friedman.”

He further argues, “The truth, although nobody on the right will ever admit it, is that Friedman was basically a Keynesian — or, if you like, a Hicksian.” While those on the “right” might never admit it, those familiar with the work of Roger W. Garrison would not be surprised that in a key important way, Friedman’s similarity with Keynes, or more correctly Keynesians such as Krugman, are more important than his apparent empirically based disagreements, the level of aggregation chosen for macroeconomic modeling. As early as 1992, Garrison in “Is Milton Friedman a Keynesian?” addresses the question gives and answer to Friedman’s own conclusion (Friedman on J M Keynes , p. 6), “He isn’t, but he is.” Garrison explains in more detail, “This ISLM framework, more broadly called income-expenditure analysis, has in many quarters—but not in Austrian ones—come to be thought of as the analytical apparatus common to all macroeconomic theories. Appropriate assumptions about the stability of investment and money demand, interest elasticities, and price and wage rigidities allow for the derivations of either Keynesian or Monetarist conclusions.”

A major theme of Cochran and Glahe, The Hayek-Keynes Debate: Lessons for Current Business Cycle Research, was that Hayek and the Austrians with their underlying capital theory and market process analysis provided a potentially more useful macroeconomics when compared to the analysis of Friedman, monetarist/classical; old or new, and or Keynesians; old and new. Garrison’s Time and Money, best known for its restatement of the Mises-Hayek Austrian business cycle theory into a capital-structure based macroeconomics which contrasts a saving supported sustainable growth to policy induced unsustainable growth provides strong support for this argument. Garrison is often at his best when he compares and contrasts different approaches to addressing macroeconomic conditions (see chapter 12, “Taxonomy and Perspective”).

More recently Garrison adeptly applies his Friedman/Austrian contrast in the context of the current crisis in a recent Independent Review essay “Alchemy Leveraged: The Federal Reserve and Modern Finance” (see especially pp 442-450). Here he concluded, Time and again Dowd and Hutchinson point to downwardly distorted interest rates and long-term investments as key to our understanding of loose money’s perverse effects. This emphasis, of course, is the Austrian element. Friedman downplays allocation effects of the rate of interest and casts the interest rate as only a minor determinant of the demand for money. Dowd and Hutchinson see the Greenspan Fed’s loose-money policies as an essential element in the story of housing-led boom and subsequent financial crisis. On the occasion of Greenspan’s retirement from the Federal Reserve, Friedman penned a piece for the Wall Street Journal with the title “He Has Set a Standard” (2006). The fact that Greenspan’s reign had seen only mild inflation was evidently enough for Friedman to credit him for doing the right thing, despite the absence of a viable monetary rule.”

Building on Garrison, I recently commented in a similar context (“A Crisis of Authority: Pierre Lemieux’s Somebody in Charge: A Solution to Recessions?” The Independent Review, 596-97),

Without the Fed, the impact of the distortions in the housing market would still have been significant, but they would also have been much more limited. The fact that the “Greenspan Fed adopted a loose monetary stance in the wake of the bust and well into the century’s first decade was a game changer. The accommodation freed the housing sector from having to draw investment funds from other sectors. It fueled an economywide boom—the housing bubble leveraged by practitioners of Modern Finance being the most dramatic aspect of it.” And he concludes, “[T][he fact that the bubble was doubly artificial provided a strong hint about the difficulties inherent in the subsequent recovery” (2012, 449).

Lemieux is correct that the “authorities in charge messed up in more than one way” (p. 131), but he understates the role of a key player. The Fed messed up in a big way, generating back-to-back cycles driven by monetary excesses that “turbo-charged” misdirections of economic activity. The Fed’s actions and expanding power since the onset of the crisis are even more alarming. To criticize the Fed and the Treasury’s response to the financial crisis, John Taylor has coined the term mondustrial policy, which describes “not a monetary framework,” but “an intervention framework financed by money creation” (qtd. in Hilsenrath 2009). Hummel argues that the policy responses to the crisis “resulted in another Fed failure” and have “also resulted in a dramatic transformation of the Fed’s role in the economy. [Ben] Bernanke has so expanded the Fed’s discretionary actions beyond controlling the money stock that it has become a gigantic, financial central planner” (2011, 485–86).

Krugman is partially correct that the debate is more primitive today, but more primitive than 80, the apex of Hayek’s influence as a monetary theorist, not 70 years ago which was the beginning of Keynesian ascendency. The debate is primarily more primitive because while the insights of Hayek and Mises at least appear in some discusions, capital theory is still too much neglected. The debate moved backward, not by the insertion of an unfortunate and perhaps uninformed appeal to authority of Friedman, but to the return of textbook Keynesianism to the policy front.

For some interesting commentary from a non-Austrian persepctive see Uneasy Money, “Was Milton Friedman a Closet Keynesian” at


  1. John

    First, Krugman has a political agenda and Keynesian policy supports that agenda. Everything he says and does is in support of that political agenda. It has absolutely nothing to do with any moral assumption of meritocracy or the common good implied by economics as a tool for assisting in policy decisions. Second, he never uses prewar data or historical examples which would expose his ideas to scrutiny. Third, he argues that the good that comes from Keynesian spending compensates investors and entrepreneurs for the costs. Fourth, he ignores the misallocation of human capital and the long term social consequences of that misallocation – again, because it suits his political agenda.

    Austrians assert that not only are we misallocating capital and human capital, and not only are we creating perverse incentives and moral hazards like confetti at an italian wedding, and not only are we destroying the civic virtues, but that entrepreneurs and investors are not compensated for the impact upon their planning. (Some even make a purely moral argument which I think is specious on all accounts.)

    The problem is, as far as I can tell, we cannot produce a mathematical model for an argument either way. I’m sure that we intuit that we are kicking the can down the road and creating bubbles of every possible kind. But I’m not sure that we can argue (yet) that the use of aggregates and all the implied redistribution that the use of aggregates entails, is either good or bad.

    It’s pretty clear that the conservative (aristocratic classical liberal) social model is being affected. it’s pretty clear that entrepreneurs are being prevented from solving many social problems like education. But these are difficult causal relations to prove. And to many they’re desirable outcomes. Freedom is and always has been the desire of the minority. Everyone else just wants ‘aristotle’s relishes’: to consume without consequence.


  2. I was thinking much of the same thing when I read Krugman’s blog post. I think much of what Leland Yeager discusses in his piece, “The Keynesian Diversion,” is relevant, as well. A lot of what Keynes allegedly brought to the table was already there: price rigidities, monetary disequilibrium, et cetera. So, it was not so much even that Friedman accepted Keynesian models, but that Keynesian models were already predicated on concepts long accepted by the economics profession (also, it should be noted that Hicks’ model of Keynes is based on an interpretation, which itself is influenced by other non-Keynesian economists).

    I think, though, that in ways Krugman is being disingenuous. Friedman may have accepted Keynesian models, such as IS/LM, but IS/LM was not really the crux of Keynes’ argument in The General Theory. IS/LM, as developed by Hicks, is based on the assumptions Keynes makes (i.e. liquidity preference theory of interest, marginal efficiency of capital, et cetera), not on the conclusions. Keynes’ central argument, which has to do with underemployment and the disconnect between investments and savings is not something, I don’t think, that Friedman agreed with. So, in the most important sense, Friedman was not a Keynesian.

    • “A lot of what Keynes allegedly brought to the table was already there: price rigidities, monetary disequilibrium, et cetera. ”

      He also brought a theory of demand for money, which was ultimately fallacious. But, it was more than the neoclassicists and many Austrians have to offer (e.g., the monetary equilibrium theory that the Austrian free bankers base their system on is Walrasian to the core).

    • Jonathan:

      In that snese Friedman is not in line with Keynes but:

      While Keynes’s verbal analysis in the General Theory continued to emphasis the role of investment, interest, and money in determining output and employment, his abandonment of the natural rate concept masked the intertemporal coordination issues at the heart of fundamental economic problem, made it easier to ignore the important capital theory issues involved in the original Hayek-Keynes debate, and facilitated the morphing of the economics of Keynes into the IS-LM single macroeconomic output aggregate Keynesianism.
      Relative to most quantity theorists, old or new, and most modern macroeconomics which model the economy with a single aggregate production measure, Keynes, even in the General Theory, continued to stress the importance of the distribution of production and resources between present uses, consumption, and the future oriented uses, investment. The single aggregate approach makes it nearly impossible to even recognize intertemporal coordination problems. Keynes does recognize potential problems. But a major factor differentiating Keynes from the Austrians is Keynes’s lack of any well defined capital theory compared to the Austrian use of structure of production capital theory, a capital structure -based macroeconomics (Cochran and Glahe 1999, pp. 103-118 and Horwitz 2011). Hence, “In the judgment of the Austrians, Keynes disaggregated enough to reveal potential problems in the macro economy but not enough to allow for the identification of the nature and source of the problems and the prescription of suitable remedies” (Garrison 2001, 226).

Comments are closed.