Nicholas Kaldor and Critical Realism

Abstract

Kaldor was among the first Post Keynesians to be identified by Tony Lawson as a forerunner of critical realism, and there are some insights to be gained by considering why this was so. Kaldor's arguments about stylised facts and theory building also provide an opportunity to consider issues of research method. We suggest some criteria that might be used to assess whether a research program is broadly consistent with a critical realist agenda. We then examine Kaldor's contributions to see whether his theoretical and empirical insights fit these criteria. In doing so, we identify several key areas of interest that extend methodological discussion in useful directions. We conclude that Lawson was broadly correct: Kaldor was indeed a forerunner of critical realism.

Notes

¹This does not mean that that Kaldor's work cannot be consistent with other ontological traditions, and we make no comment here on the extent to which this might be the case (Boylan & O'Gorman, 1999).

²A notable exception being Boylan & O'Gorman (1999).

³There is one exception to this rule. In 1955 Kaldor published a two-page ‘Rejoinder’ in response to David McCord Wright's ‘Methodological footnote’ to Kaldor's Economic Journal paper on growth and fluctuations (Kaldor, 1954, 1955; Wright, 1954). Kaldor's defence of scientific realism is discussed in Section 3.1 of the present paper.

⁴Lawson argues that, in both the natural and social sciences, the aim is to understand the causes of observed phenomena. However, the social realm requires different research methods from the natural sciences because, unlike the ‘natural’ realm, social reality can be transformed by human agency (Lawson, 2003, p. 35).

⁵Lawson's use of the term ‘internal relations’ is linked with his arguments about the open nature of social systems. Internal relationality, or ‘other orientation’ of social practices, means that individual actions cannot be treated in isolation from the actions of others (Lawson, 2003, p. 39).

⁶See, for example, Kaldor (1960, p. 181; 1961, pp. 3–4; 1966b, pp. 309–310; 1972, pp. 1238–1240; 1981, p. 433; 1996, pp. 101, 132).

⁷And it was criticised as such by econometricians in the UK (Stone & Jackson, 1946) and the US (Hart, 1945).

⁸On the implications of non-ergodicity, see Davidson (1996).

⁹To give just one example, in his Okun lectures he noted that ‘our ability to predict what can happen or what is likely to happen becomes progressively less as we consider the more distant future as against the nearer future’ (Kaldor, 1985, p. 62; emphasis in original).

¹⁰For some early examples, see Kaldor (1950–51; 1955, p. 157; 1957b; 1961, p. 3; 1962).

¹¹A construct that has proven useful in developing economic theory and providing insights into relevant human behaviour is that of ‘labour supply’. A researcher observes a school leaver looking for work, a part-time café worker wanting a full-time position, two grandparents planning to retire, and a friend going to work every day. These disparate activities are interpreted as all related to the amount of time these people wish to spend doing paid work, and are described by the label ‘labour supply’. This construct can then be used in combination with other constructs to develop a theoretical model of the labour market. The concept of ‘labour supply’ has been uncontroversial, and there is little debate over whether ‘labour supply’ is a real event.