Inflation in France Since the 1960s: A Post-Keynesian Interpretation Using the Conflict-Inflation Model

Abstract

This paper analyses inflation in France since the early 1960s based on a standard conflicting-claims approach. Applying an empirical version of the conflict-inflation framework, we adduce evidence suggesting the theory is sound and can explain variations in inflation over the long run. We provide a method for estimating indicators of workers’ and firms’ bargaining power as well as their respective distributional aspirations. Based on the literature we identify four periods: the Fordist regime, the Neoliberal regime, and two transitional periods. Our results cast light on institutional regime changes. It is shown that the evolution from the Great Inflation to the Great Moderation was the consequence of a collapse in the bargaining power of workers (and of firms to a lesser extent); but the narrowing of the aspirations’ gap because of workers’ renouncement was also significant. This analysis allows us to highlight differences between the stagflation observed during the 1970s and the inflationary surge in the post-pandemic period (2021–2023).

Keywords:

• Inflation
• conflicting claims
• institutional regime changes

JEL CLASSIFICATIONS:

• E31
• E65
• J52

Disclosure Statement

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

Notes

1 The empirical works of Sarantis (1991), Burdekin and Burkett (1996), and Fry and Webster (2006) remain exceptions over the last forty years.

2 Interested readers will find several extensions of this standard model explicitly incorporating productivity growth, exchange rate, or monetary policy considerations in Setterfield (2007), Vera (2010), and Zambrano (2014). Despite its apparent simplicity this basic model is still able to deal with financial and open economy issues as will be seen in section 3.

3 This assumption will be relaxed later.

4 Hereafter we assume that for any variable, $x,$ its rate of growth can be written as $\widehat{x}=\dot{x}/x.$

5 An alternative would have been to add a curve representing the price dynamics augmented by the productivity gains ($\widehat{p}+\widehat{\mu }$) instead of a wage curve reduced by productivity gains ($\widehat{w}-\widehat{\mu }$).

6 It could also be possible to use econometric techniques to identify structural breaks, but they should rely on strong assumptions about the values of key parameters, and they would offer fewer possibilities for the number of different periods.

7 For every year, a comparative normalization consists in subtracting the minimum of the variable over the whole period from the given year value, and dividing it by the difference between the maximum and the minimum value over the whole period. Consequently, for each year, the normalized value for the variable lies between 0 and 1.

8 For example, even though the analysis is focused on the United States, we can follow Setterfield (2007, 142) when he notes that “increases in non-standard employment arrangements have reduced the employment security”.

9 Nishi and Stockhammer (2020) also integrate an effect of the cycle on the aspirations of workers through the output gap (while it is through the variation of the unemployment rate in our model): in their model, its value depends on the demand and distribution regimes, while we posit a continuous decline along history in our approach. In their paper, Nah and Lavoie (2019) make the profit share targeted by workers depend negatively of the growth rate of the employment rate. For their calibration, they set the value of the parameter to 10. This difference in the level of our calibration stems from the fact that we use the variation of the unemployment rate to account for this cycle effect, while Nah and Lavoie (2019) use the growth rate of the employment rate.

10 This observation is fully consistent with the characterization of different inflation regimes (low inflation regime observed when wages are not index-linked, moderate inflation regime when indexation is generalized, and high inflation regime when indexation is made on an external index, generally the exchange rate) by Bastian, Charles, and Marie (2024); interested readers can refer to that paper for a discussion of how the distributive conflict and the course of inflation are affected by indexation.

11 If b₂ is positive, the other parameters take negative values.

12 A similar exercise could be conducted for other periods of oil shocks and imported inputs, as for example the period 2006–2008 and the immediate post-pandemic era including the war in Ukraine (2021–2023). For our purpose we want to preserve a single period from 1987 to 2018 because the upsurge in oil prices (and more generally in food prices during the so-called world food prices crises) of 2007–2008 did not cause a significant change in inflation observed, certainly because the global financial crisis and its macroeconomic consequences induced strong disinflationary effects. For the inflation that began in 2021, we do not yet have the historical hindsight to say whether or not the current disinflation will restore the neoliberal regime unchanged.

13 Our approach may be close to the one developed in Mazier, Reyes, and Chong (2023). See especially pp. 3-7 for an analysis of different periods of inflation in France. Moreover, in their estimation of the NAIRU for France, Sterdyniak et al. (1997) identify four different sub-periods, very similar to ours (1965-1973; 1974-1979; 1980-1986; 1987-1994).

14 Harkis are Algerians who fought on the French side during the Algerian War for Independence.

15 Adjustments at work are as follows for the period 1963–1967. On the one hand the imposition of price controls by the Government involves a downward revision of wage claims expressed by workers. On the other hand the slowing of workers’ claims pushes firms to accept a lower desired mark-up.

16 See the original agreement on the web site of the Ministry of Labour, Employment and Insertion: https://travail-emploi.gouv.fr/IMG/pdf/Constat_de_Grenelle.pdf.

17 Other measures include the nationalization of several corporate firms, the shortening of the working week, and the creation of more than 180,000 jobs in the public sector in a year and a half (on this last point see, for example, Fonteneau and Muet 1983).

18 Mazier, Reyes, and Chong (2023, 4) consider that: “the tournant de la rigueur was a success concerning the fight against inflation with only a limited part played by monetary policy.”

19 Interestingly the United Kingdom government chose the internal objective during the exchange rate crisis of October 1992.

20 The decline in unionism is intensified by the closing of industrial bastions.

21 While trying to explain the disappearance of the Phillips effect during the wage moderation period, Ratner and Sim (2022) show in a Federal Reserve working paper that “a nearly 90 percent reduction in inflation volatility is possible even without any changes in monetary policy when the economy transitions from equal shares of power between workers and firms to a new balance in which firms dominate.” They also add that “the decline of trade union power reduces the share of monopoly rents appropriated by workers, and thus helps explain the secular decline of labor share, and the rise of profit share.”

22 We have to recall here that the condition for a stable wage share is not that wage inflation must equal price inflation, but that wage inflation less productivity gains must equal price inflation. It should also be noticed that productivity gains slow down between the two periods, meaning that the gap between the wage curve and the wage-minus-productivity-gains curve narrows.

23 As noted by Mazier, Reyes, and Chong (2023, 3), “the shock is a net withdrawal in favor of the rest of the world which is undergone by domestic agents”, and each domestic agents are struggling for not being the ones who undergo the loss (reductions in the real wage, the profit margin, the fiscal balance).

24 For Celasun et al. (2022) more than 50% of the eurozone producers were obliged to reduce their productions because of shortages in intermediate inputs.

25 See Weber and Wasner (2023) or Rolim (2023) for a global picture of greedflation or Arquié and Thie (2023) for a specific study of France that estimates firms exhibit a more than 100% pass-through of costs increases in final prices.

26 AMECO Online, “Unemployment: percentage of active population (ZUTN series)”, data consulted on 20 December 2023.

Additional information

Notes on contributors

Sébastien Charles

Sébastien Charles is Associate Professor of Economics at the University of Paris 8-Saint-Denis (France). His topics of interest concern theoretical and empirical macroeconomics from a Post-Keynesian point of view.

Thomas Dallery

Thomas Dallery is Assistant Professor of Economics at the University of Littoral Opal Coast (France). He conducts his research at the Clersé Research Department. He is interested in the general field of political economy from a post-keynesian perspective.

Jonathan Marie

Jonathan Marie is Associate Professor in economics at the University Sorbonne Paris North (France), member of the CEPN. Adopting à Post-Keynesian framework, he focuses his research on effects of economic policies and on inflationary dynamics.