ABSTRACT
Using three datasets of French manufacturing firms, this article studies the role of trade openness, in relation with the cycle, as a determinant of company margin rate. Margin rates increase as capacity utilization tightens (and vice versa), reflecting the procyclicality of margin rates. However, high import rates are limiting this procyclicality: when capacities are tight, domestic producers may not be able to serve demand, but foreign producers may substitute for them if they are already present on the market as reflected by the level of import rates.
Acknowledgements
The views expressed in this article do not necessarily reflect Banque de France’s or Eurosystem’s opinion. We thank anonymous referees from the journal for their contribution to the improvement of this article.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 In some specific activities, the margin rate depends on specific aspects. Barnea and Kim (Citation2007) show, for example that in the oligopolistic banking industry, the changes of the margin rate depends both on dynamic oligopolistic conduct and dynamics of market fundamentals.
2 Trade liberalization changes competitive pressures, by itself and also through complementary policies which may be needed (see Moore Citation2010, for example on capital account liberalization and the banking industry). And the trade structure depends itself on various factors, as regulations but also as shown by Xing and Xu (Citation2014), on saving rate level.
3 Haskel and Martin (Citation1994), for example, aggregate concentration ratios and import rates in ‘adjusted concentration ratios’. Interestingly, the interaction term between their cyclical term and this ‘adjusted concentration ratio’ is not any longer significant, which is consistent with our results: import rates in interaction with the cycle have an impact on margin rates, that is of opposite sign to the one of concentration ratios, which may explain why mixing both indicators yields non-significant results.
4 The SMI sample is available only from 1996.
5 We also tested the robustness to the use of system GMM estimators. Results are consistent in sign and significance, but the Hansen tests do not support the use of this estimation technique.