Introduction

Agreements between undertakings aimed at fixing prices, limiting output or sharing markets without any expectation of offsetting benefits to consumers are prohibited under Article 81 of the EC Treaty. Cartels are obviously included in this. A violation of this prohibition can for each violator attract a fine up to 10 per cent of its worldwide turnover. This fine will in the first instance be levied by the Directorate General for Competition following their investigation and is formally an administrative rather than a criminal fine. Their decision can be appealed to the Court of First Instance (CFI) and subsequently to the European Court of Justice (ECJ). In order to overcome the information problems in detecting and prosecuting cartel agreements, the EU has introduced a leniency programme which rewards cooperation by cartel members through a reduction in the fine.

The law on cartels, like other areas of competition policy, is founded on principles of economics, but economists are rarely involved in its application against explicit hard-core collusive agreements. In part, this is due to the absence of evidentiary problems – direct evidence may be found from a whistleblower or in dawn raids (e.g. e-mails) – and thanks to leniency programmes, cartel members may either reveal details of their infringement in return for immunity from fines or start cooperating and providing evidence in return for a leniency discount once an investigation into their industry has been opened.

Introduction

The failure of prices to adjust plays a crucial role in Keynesian macroeconomics. In particular the failure of nominal prices to adjust instantaneously to nominal shocks is important for the role of demand shocks as a source of business cycle fluctuations. Empirical evidence shows that nominal shocks contribute to business cycle fluctuations (see, for example, Andersen, 1994). Insufficient price adjustment may at a general level be caused by either adjustment being costly in terms of explicit or implicit costs or by pricesetters lacking sufficient information to make the proper adjustment.

The most widespread model of price inflexibility is the so-called menu cost model, assuming price adjustment to be costly. This model has recently been extensively analysed (see Ball, Mankiw and Romer, 1988 and Andersen, 1994 for introductions and references) and has provided a number of important insights on price adjustment. Still, the empirical relevance of price adjustment costs remains an open question, and it is not obvious whether price adjustment costs are more important than costs of adjusting quantities.

An alternative approach to the explanation of price rigidities is to consider informational problems arising in decentralized economies. Small departures from the benchmark case of full information are sufficient to cause adjustment failures, especially if firms are differently informed or if there is confusion between permanent and transitory changes (see Andersen, 1994). In this chapter we extend the analysis of informational problems by making information acquisition by actors in the economy explicit.