Some comments on the Sraffian Supermultiplier approach to growth and distribution

Abstract

The article discusses the Sraffian Supermultiplier (SSM) approach to growth and distribution. It makes 5 points. First, in the short run, the role of autonomous expenditure can be appreciated within a standard post Keynesian framework (Kaleckian, Kaldorian, Robinsonian, etc.). Second, and related to the first, the SSM model is a model of the long run and has to be evaluated as such. Third, in the long run, one way that capacity adjusts to demand is through an endogenous adjustment of the rate of utilization. Fourth, the SSM model is a peculiar way to reach what Garegnani called the “second Keynesian position.” Although, it respects the letter of the “Keynesian hypothesis,” it makes investment quasi-endogenous and subjects it to the growth of autonomous expenditure. Fifth, in the long run it is unlikely that “autonomous expenditure” is really autonomous. From a stock-flow consistent point of view this implies unrealistic adjustments after periods of changes in stock-flow ratios. Moreover, if we were to take this kind of adjustment at face value, there would be no space for Minskyan financial cycles. This also creates serious problems for the empirical validation of the model.

Keywords:

• Sraffian Supermultiplier
• long run
• utilization
• autonomous demand

JEL Classification Codes:

• B22
• B5
• E12
• E32
• O4

Notes

1 Skott (2017: 2) adds to this definition a second criterion that “its movements must not be offset automatically by changes in other components of demand.”

2 This means that we could rewrite equation (1) as $g^i=g^i[{\gamma }_w, \left(u-u_d\right)]$. If profitability did not have any effect on investment, the warranted rate, thus defined, would be equal to γ.

3 It is worth mentioning that not all Sraffians share the view that in the long run the actual utilization rate needs to be equal to the desired rate. Early discussions of why actual utilization—even on average—deviated from the normal rate in the long run include Garegnani (1992), Vianello (1985), and Ciccone (1986). More recently, some Sraffians have criticized the SSM model on these grounds (Trezzini 1995, 1998; Palumbo and Trezzini 2003).

4 This is important because some Kaleckians had argued in favor of an endogenous rate of utilization on the basis that the desired rate of utilization is merely a convention, which changes as the actual rate changes (Lavoie 1995, 1996; Lavoie, Rodríguez, and Seccareccia 2004; Hein, Lavoie, and van Treeck 2012). This is not a convincing justification of the endogeneity of utilization. The choice of the level of utilization is analogous to the choice of the technique of production.

5 The endogenous adjustment of desired utilization as a long-run adjustment mechanism of the Kaleckian model has been recently proposed by in a note by Dávila-Fernández, Oreiro, and Punzo (2017). However, they proposed this as an alternative to the adjustment of the warranted growth rate. Lavoie (2018) replied with some critical remarks.

6 The term “Keynesian Hypothesis” was coined by Kaldor (1955, 95).

7 This theory of investment goes back to Garegnani (1962, 1978, 1979).

8 It is also worth mentioning that the first one to express doubts about “a hard-and-fast line between induced and autonomous investment” was Harrod (1959, 461) with reference to the original supermultiplier model of Hicks (1950). Harrod’s concern was to what extent the presence of autonomous investment invalidated his “instability principle.” He argues that certain components of investment that are exogenous in some phases of the cycle might be revised in some other phases.

9 See Pariboni (2016) for a more detailed discussion.

10 Obviously, a big part of all the phases of a cycle like that involves changes in product and asset prices; these changes further magnify the elasticity of expenditure to the state of the economy.

11 The ability of the government to increase its debt is obviously much greater than that of the private sector. However, from an SFC point of view, assuming that autonomous government expenditure in itself can be the long-run driving force of the economy means that public deficits and public debt are accepted politically and do not matter economically (Godley and Lavoie 2007b).

12 The saving rate of the household sector started increasing in 2006. This was the trigger of the crisis that followed.

13 For a recent survey of the SFC modeling principles and literature, see Nikiforos and Zezza (2017). Minskyan models of financial cycles and crises are discussed in section 3 of that paper.

14 For an analysis of the increase of the indebtedness of the households at the bottom of the distribution in the United States before the crisis along these lines, see Nikiforos (2016a).

15 In a recent paper, which they explicitly locate within the SSM approach, Fiebiger and Lavoie (2017) argued that debt-financed household semi-autonomous expenditure contributed to the growth of the U.S. economy before the crisis and the lack of it explains the slow recovery that has followed (the prefix “semi” makes a world of difference here). I am in agreement with their argument, but I disagree that it provides support to the SSM approach. As argued here, semi-autonomous expenditure can be a useful analytical concept in general, but it is not satisfactory in the case of the SSM model because of its long-run nature.