Imperfect Competition
Macroeconomic effects

The last decade has seen the emergence of a substantial literature concerning the macroeconomic implications of imperfect competition in output and labour markets. In Discussion Paper No. 636, Research Fellows Huw Dixon and Neil Rankin consider the low and inefficient levels of activity and the `coordination failure' exhibited by imperfectly competitive macroeconomic equilibrium models. If all economic agents reduce prices simultaneously, output and employment rise, and utility increases; but firms and unions acting independently have no incentives to cut prices or wages. Partial equilibrium models of imperfect competition do not capture this `coordination failure', since they omit the feedback effects of decisions on incomes and thus on demands, and they take no account of agents' different roles as consumers, workers and shareholders.

Dixon and Rankin then consider the impact of macroeconomic policy on output and welfare. Monetary policy has real effects only if imperfect competition is accompanied by a further `distortion', such as the `menu costs' incurred by firms in changing prices, expectations of future prices that do not respond proportionately to changes in current prices, or a nominal rigidity in some sector. Fiscal policy, in contrast, nearly always has real effects on output, although there can be no general presumption concerning its sign, since imperfect competition allows government spending to operate through various channels outside traditional Keynesian macroeconomics: altering the overall price elasticity of demand for goods, enhancing the income effect on labour supply, exploiting asymmetries in government spending between sectors, and inducing firms' entry to and exit from imperfectly competitive industries.

In such imperfectly competitive economies, output and employment are too low, price generally exceeds marginal cost and the wage exceeds the marginal disutility of labour. Dixon and Rankin find that monetary or fiscal policies that increase output are then probably welfare improving, and very possibly Pareto-improving. Multiple equilibria are also likely to occur and to be Pareto- rankable; thus the economy could be `flipped' from a low- to a high-output equilibrium, in which all are better off, by coordinating monopolistic agents' expectations about each others' actions. The underlying cause of multiple equilibria in most of these models, and of their imperfectly competitive market structure, is increasing returns to scale in production.

In conclusion, Dixon and Rankin suggest possible extensions of this research to the open economy, possibly related to the literature on imperfect competition in international trade, and to dynamic models: modern theories of industrial organization focus heavily on strategic issues in dynamic game-theoretic models of imperfect competition.

Imperfect Competition and Macroeconomics: A Survey
Huw Dixon and Neil Rankin

Discussion Paper No. 636, May 1992 (IM)