Inflation stabilization and normal utilization

Abstract

This paper presents a model of inflation and distribution that examines the relationship between the employment of labor and the utilization of capital. It features a structural difference between the wage Phillips curve and the price Phillips curve that gives rise to persistent changes in the real wage whenever the inflation-neutral level of activity fails to utilize the existing capital stock at its normal level. Assuming an inflation-targeting central bank that is obliged to run the system around its inflation-neutral level, these changes will reduce the gap between the inflation-neutral level and normal utilization by moving the system along a stable wage curve. In the end this implies that the inflation-neutral level of employment and full or normal utilization of capital will tendentially coincide, lending some support to the Duménil-Lévy thesis that monetary policy makes normal utilization a long-run center of gravity.

Keywords:

• Capacity utilization
• Phillips curve
• wage curve
• Kaleckian and classical growth models

JEL:

• E11
• E12
• E24
• E31
• E52

Disclosure statement

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

Notes

1 As do most of the classical models in Foley, Michl, and Tavani (2019) and hybrid Keynesian-classical models such as Michl (2017). For a recent survey of literature on the role of capacity utilization in the long run, see Blecker and Setterfield (2019, Ch. 6).

2 Michl (2008a) extends the Duménil-Lévy argument to the inflation rate. Examples of other mechanisms that restore normal utilization are the Harrodian investment equations proffered by Skott (1989) and the variable rate of business saving suggested by Shaikh (2009). Normal utilization also plays a central role in Sraffian supermultiplier growth models where capacity adapts to demand in fully adjusted states as demonstrated by Freitas and Serrano (2015).

3 It deserves to be said that Hein et al. (2011) and other critics of the Dumenil-Levy formulation do not confine their criticism to this central point. A comprehensive reply to the full bill of particulars they raise is beyond the scope of this paper.

4 There are several models of normal utilization available. Some define it by cost minimization. Petach and Tavani (2019) treat normal capacity as the solution to a social coordination problem in which firms optimize the user cost of capital. Skott (1989) appeals to the use of excess capacity as an entry deterrent. The current paper can be interpreted as an explanation of how demand management will achieve normal utilization, at least tendentially.

5 Relaxing the assumption that k = 1 leads to $e=(u/k)\kappa .$

6 Skott (2015) has argued from a neo-Harrodian perspective that prices and profit margins are actually quite flexible in practice, which is consistent with the treatment here. Another possible influence is the idea of “real competition” advocated by Shaikh (2016). Finally, a reader commented that since the profit margin adjusts (see below) to generate full utilization, the approach bears a family resemblance to neo-Keynesian writing, such as Robinson (1962).

7 The pre-set real wage is basically a state variable that is predetermined by conditions in the temporal past.

8 If we included a mechanism like wage aspirations in the wage curve, this would probably attenuate the changes in the real wage in the thought experiments we run through in the rest of the paper, and affect the value of $\overline{w}.$

9 Something like this theoretical foundation for the wage Phillips curve was originally proposed by Rowthorn (1977) but it is not inconsistent with much of Kalecki’s writing.

10 These weights might also be conceptualized as the proportion of firms oriented toward their competitors’ prices (flex-price firms), and the proportion of firms following strict mark-up pricing oriented toward their labor costs (fixed-price firms), but our model makes no allowance for heterogeneity among firms.

11 There is some empirical support for the idea that the inflation-neutral level of activity cannot be easily tied to one measure of slack, which is one way of interpreting e_n. Forbes (2019) estimates national and global Phillips curves using the principal component of seven separate measures of slack to model the output gap.

12 This specifies that the central bank should raise the nominal interest rate by more than the positive deviation of the inflation rate from its target, thus creating a negative feedback that will create disinflation (and vice versa for a negative deviation). The assumption is that it is the real, not the nominal, interest rate that regulates spending.

13 Some staff economists at the Federal Reserve System consider the utilization rate a valid measure of economic slack, and refer to the “non-accelerating inflation rate of capacity utilization” based on personal conversations with the author.

14 The step across the second equal sign uses the trick of adding $e^{*}{\alpha }_p/\kappa -e^{*}{\alpha }_p/\kappa$ (i.e., 0) to the term in parenthesis and then rearranging.

15 This case bears resemblance to an economy under a gold standard in which the long-term price level evolves according to the relative costs of producing gold. Absent a complete model, one might speculate that if the gold standard operates like a central bank with a price level target, it could create gravitational forces that stabilize inflation, albeit around some slowly moving value, making normal utilization a center of gravity.

16 I am indebted to an anonymous referee of this journal for suggesting the need for supporting evidence.

17 Keynes did not actually support Kalecki’s emphasis on underutilized capital, so the term Keynesian is being used here in its modern context. Also, just to be clear, long-run normal utilization does not necessarily mean that aggregate demand has no role in the long run.

Additional information

Notes on contributors

Thomas R. Michl

Thomas R. Michl is Emeritus Professor of Economics at Colgate University. The author thanks two referees of this journal and Mark Setterfield for their helpful comments.