Keynes and the drunkard under the lamp post: Making sense of Palley

Abstract

In a recent article, Tom Palley begins his critique of Keynesian economics with the wellknown story of a drunkard who, when searching for his lost keys, looks not in the darkness of the nearby lawn where he misplaced them, but instead under the light cone of a lamp post because, when asked, he replies that’s where the light is. This story serves as a metaphor for the field of economics attempting to understand the workings of the capitalist economy solely through the lens of Keynesian economics. However, this endeavor is ultimately futile as comprehending capitalism requires a different analytical approach: acknowledging social conflict as an essential component of capitalism’s nature better addressed by Kaleckian macroeconomics. We attempt to illustrate that Palley is accurate in emphasizing the paradigmatic differences and even incommensurabilities between Keynes’ monetary production paradigm and the Marxian-Kaleckian social conflict paradigm. This suggests that any classification under the umbrella term “post-Keynesianism” is misleading. However, Palley is mistaken in his assertion that this distinction aligns Keynes’ economics with neoclassical (mainstream) economics, as the acceptance or rejection of social conflict is not the only fault line in terms of ontology. There are other ontological divisions that can exist. Or, to use our metaphor, the lamp post can be moved to different areas of the lawn, indicating different ontological perspectives.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Which is why these approaches are often dubbed as “bastard Keynesianism” using Joan Robinson’s slightly unceremonious label.

2 As Gregory Mankiw (1992, 560) bluntly formulated: “New Keynesians view their work as following in the broad tradition that evolved from Keynes, but their goal is to explain the world, not to clarify the views of one particular man. If new Keynesian economics is not a true representation of Keynes’s views, then so much the worse for Keynes.”

3 Just one statement should be singled out for annotation: “Some Post Keynesians hold that Keynes (1936) did not subscribe to either an aggregate production function (…) or the Neoclassical marginal product of capital. To this author, that is not correct. Keynes (1936, chapter 11) sought to emphasize that the MEK (marginal efficiency of capital, A.H.) is a future oriented construct that depends on expectations, which in turn are formed in a world of fundamental uncertainty. However, the MEK remains a fundamentally Neoclassical concept whereby it declines with capital-deepening” (TP, 30). This statement is disturbing because it confuses the neoclassical concept of marginal productivity based on capital deepening with the Keynesian concept of marginal efficiency based on capital widening (see Heise 1990).

4 One referee takes Palley’s contribution as rhetoric rather than scholarly. That is not my reading of it—otherwise it should probably not have been published in a scholarly journal such as the Japanese Political Economy—and I try to make sense of it beyond mere provocation.

5 Withdrawing from commenting on these views does not imply that I share them. However, I believe that the significance of assessing one’s political views in relation to their theoretical or paradigmatic approach is not convincing enough for me to pursue such an endeavor.

6 According to Heise (2023), it can be demonstrated that a capitalism built on “technocratic corporations” may exhibit greater volatility and lower growth dynamics compared to a capitalism driven by the entrepreneurial “animal spirits” of owners.

7 According to Robert Clower (1965, 111), an economic theory that aims to bring about a paradigmatic shift must be capable of challenging Walras’ law. In other words, any theory that relies on Walras’ law has “nothing fundamentally new to add to orthodox economic theory.”

8 While it is true that Marx explicitly rejected Say’s law, there is ongoing debate within the realm of Marxian economics regarding the extent to which this rejection aligns with the analytical foundations of Marxian theory. Scholars and economists hold different perspectives on this matter, and it remains an open discussion within the field; see e.g. Trigg (2020). Moreover, rather than refuting Walras’ law, Kaleckian economics offers a different analytical framework that complements and expands upon the traditional neoclassical approach questioning its status as “heterodox.” As one referee pointed out, this would imply that Kaleckian economics should be viewed more as a “critical” variant of neoclassical economics than as an opposing paradigmatic alternative to neoclassical economics. While it is not within the scope of this paper to deny Kaleckian economics the status of a heterodox paradigm, it is the responsibility of Kaleckian economists, such as Palley, not only to assert but also to analytically demonstrate how the social conflict ontology translates into an alternative epistemology that confirms this status.

9 Palley (2002, 24ff:) also refers to “fundamentalist Keynesian macroeconomics”—however, it appears that his notion of “fundamentalist Keynesianism” is not identical to the notion established in the literature; see Coddington (1976), Heise (2019), Davidson (2003/2004, 263).

10 Of course, variants of neoclassical (mainstream) economics have incorporated many kinds of imperfections in their models that can explain the effects on income distribution and the level of economic activity; see e.g. Shapiro and Stiglitz (1984), Lindbeck and Snower (1988), Ciminelli, Duval, and Furceri (2020), De Loecker, Eeckhout, and Unger (2020).

11 It is evident that, according to Marx, capital-labor relations in capitalist economies are not of a deliberate but coercive nature, rendering the distribution of their output not only antagonistic but also exposing the appropriation of profit as exploitation.

12 A positive-sum game setting rests on technical learning and specialisation effects on the one hand and social stakeholder effects on the other hand.

13 One referee contends that this notion, if attributed to Kalecki, is incorrect. This assertion may be valid if it implies that Kalecki never explicitly stated the profit rate as a residual, particularly if this profit rate pertains to the “normal rate of profit” in the context mentioned, as Kalecki rarely discussed a “normal rate of profit” in his work (see Nell 1989). However, it cannot be accepted if it denies that the rate of profit must be a mere residual in Kaleckian logic. This is because, in Kalecki’s approach, the primary determinant for the rate of profit is the “degree of monopoly,” which, in turn, can be causally influenced by the potential for social conflict among wage-earners (or their trade unions) (see Sawyer, 1985 108ff.; Bortz, 2017 569). Without this, Kalecki’s (and Palley’s) insistence on the importance of social conflict for income distribution would be rendered pointless.

14 This also seems to settle the dispute between so called “broad tent” (see e.g. Lavoie, 2005 2014: 42ff.; King 2012) and “narrow tent” strategies (see e.g. Davidson, 1994 2003/2004, 2005).

15 For a contrasting view see Heise (2020a).