Long-term shifts in demand and distribution in neo-Kaleckian
and neo-Goodwinian models
Robert A. Blecker*
November 2017 [revised, June 2018]
Draft chapter for Hassan Bougrine and Louis-Philippe Rochon (eds), Money and Crises in Post-
Keynesian Economics: Essays in Honour of Marc Lavoie and Mario Seccareccia (Edward Elgar,
forthcoming). [6026 words total]
*Professor of Economics, American University, Washington, DC, USA, blecker@american.edu.
The author would like to thank participants at conferences and seminars in New York (Eastern
Economic Association), Brasilia (Brazilian Keynesian Association), and Buenos Aires
(Universidad Nacional de San Martín) for comments on earlier versions of material in this
chapter.
American University Working Paper Series
https://doi.org/10.17606/c8vj-q531
1
INTRODUCTION
The neo-Kaleckian branch of post-Keynesian economics emphasizes the multi-faceted
relationships between the functional distribution of income and various indicators of
macroeconomic performance, such as the rates of capital accumulation, capacity utilization, and
economic growth (Hein 2014; Lavoie 2014). Empirical studies in this tradition have, however,
reached diametrically opposed conclusions about whether aggregate demand is normally wage-
led or profit-led in the short run. As surveyed by Blecker (2016) and Stockhammer (2017),
1
studies using a “structural” (multi-equation) approach have (with some exceptions) tended to
find that most countries have wage-led demand, except for very small or highly open economies,
while studies using an “aggregative” approach have typically found that demand is profit-led
(often for the same countries or similar panels of countries). The latter studies have frequently
relied upon the so-called “Goodwin cycle” variant of the neo-Kaleckian model pioneered by
Barbosa-Filho and Taylor (2006), in which it is also found that there is a “profit squeeze” (that
is, higher utilization leads to an increased wage share or reduced profit share).
2
Following
Stockhammer (2017), this variant will be referred to as the “neo-Goodwinian” approach.
In spite of these varied empirical findings, one thing is clear: over the past few decades,
most advanced capitalist countries (and some emerging market nations as well) have suffered
from slower growth along with rising inequality. In particular, wage shares – which in the past
usually exhibited the “stylized fact” of mostly cyclical fluctuations around roughly constant
trends – have fallen notably in many countries since roughly the 1980s or 1990s, depending on
the country (Storm and Naastepad 2012; Karabarbounis and Neiman 2014; Kiefer and Rada
2015; Wolff 2015; Stockhammer and Wildauer 2016). The wage-led demand view suggests a
2
ready-made explanation for this coincidence of increasing inequality and worsening
performance: various structural changes and policy shifts (globalization, financialization,
deregulation, deunionization, and neo-liberal policies generally) have combined to weaken labor
relative to capital; the resulting redistribution of income toward profits inevitably causes
aggregate demand to be depressed and growth to falter in economies where demand is wage-led.
It is less apparent how the coincidence of greater inequality and slower growth in the
longer term can be explained by a model in which demand is profit-led. As two skeptics of the
profit-led view have written,
despite the fact that in almost all OECD countries real wage growth was
significantly restrained after 1980, allowing profitability to recover to its golden-
age level, post-1980 macroeconomic performance is in general characterized by
lower output growth, lower rates of investment, and higher rates of
unemployment than witnessed during the period from 1960 to 1980.... The
disappointing performance raises the question of why the redistribution of income
from wages to profits in a supposedly profit-led demand regime has so far failed
to bring about a more adequate long-run economic performance. (Storm and
Naastepad 2012, p. 113)
This chapter will argue that there is a logically coherent explanation for the paradox that
an increased profit share has failed to bring about improved macro performance at medium-term
horizons (across a few decades) in economies that have profit-led demand in the short run (that
is, at business-cycle frequencies). The key to this explanation lies in the other finding of the neo-
Goodwinian literature: that the distributive relationship slopes upward in utilization-wage share
space, or in other words, the economy exhibits a “profit-squeeze” in distribution. If the
distributive relationship slopes upward while the aggregate demand relationship slopes
downward, then any medium-run shifts in a southwesterly direction (toward lower
utilization/growth rates and a lower wage share) should be driven primarily by downward shifts
in the aggregate demand relationship. Thus, any factors that tend to depress aggregate demand
will also move the economy toward lower points on the distributive relationship – lower
3
combinations of the wage share and utilization rate (or some other measure of output or demand)
– provided that this relationship is characterized by profit-squeeze behavior.
Essentially, chronically depressed aggregate demand relieves the profit-squeeze on
capital by holding down output and employment, thereby preventing workers from reaping wage
gains both during cyclical upturns and over longer periods. Thus, this explanation reverses the
causality in the standard long-run story: in this alternative view, it is chronically depressed
aggregate demand that prevents workers from winning wage gains in line with their productivity
growth, rather than the falling wage share being an exogenous cause of depressed demand.
Before proceeding further, several qualifications are in order. First, many economists
have rejected the entire idea of characterizing economic systems as uniquely either wage-led or
profit-led (for example, Nikiforos and Foley 2012; Dos Santos 2015; Palley 2017; Skott 2017). I
am sympathetic with that critique, especially since my own previous work has emphasized that
the impact of a redistribution of income between wages and profits on aggregate demand varies
according to the source of the shift in distribution in an open economy (Blecker 1989, 2011). In
this chapter, I apply the wage-led versus profit-led distinction only to the slope of the aggregate
demand relationship in the short run; I do not use it to characterize the behavior of the overall
economic system or longer-term outcomes.
Second, some economists have argued that the findings of profit-led demand are likely to
be biased by factors that are not controlled for in the neo-Goodwinian empirical studies. Lavoie
(2014, 2017) argues that those findings could be driven by the procyclical behavior of labor
productivity in the presence of overhead labor. Assuming that firms do not layoff overhead labor
in proportion to the decline in output during a cyclical downturn, labor productivity will fall in a
recession and rise in the recovery, and since the wage share equals the ratio of the real wage to
4
labor productivity, the wage share will naturally move in the opposite direction. This endogenous
movement of the wage share could create a misleading impression that demand is profit-led,
since the profit share would be falling as utilization falls and conversely. Also, Stockhammer and
Michell (2017) demonstrate theoretically that a debt-driven “Minsky cycle” can mimic a profit-
driven Goodwin cycle even when aggregate demand is not truly profit-led.
3
Third, and most importantly, the argument in this chapter is intended only to establish a
logical possibility, not to claim that it is the only or exclusive explanation for the coincidence of
worsened economic performance and higher profit shares. It is clear that the massive structural
changes brought about by the globalization of production and finance and shifts toward neo-
liberal policies have directly weakened labor’s bargaining position and inhibited workers from
winning wage increases commensurate with their rising productivity, and the resulting fall in
wage shares (especially for production workers) certainly could contribute to depressed
economic activity in the long term. Indeed, I have argued elsewhere that the impact of
distribution on demand and growth is likely to vary over different time horizons (Blecker 2016):
the positive effects of profitability on investment and negative effects of unit labor costs on net
exports are likely to dominate in the short run, implying profit-led demand at business-cycle
frequencies, whereas the positive effects of high wages on consumption are likely to dominate in
the longer term, implying wage-led demand over longer time horizons.
4
But what this chapter
emphasizes is not the demand relationship per se, but rather how it interacts with the
distributional relationship in such a way that depressed demand and growth could be as much
causes of a more unequal distribution of income as they are consequences of the latter.