There has been a long history of analyzing the workability of markets regarding concentration ratios as indicators of workability. In this paper, we discuss the Coordination Failure Diagnostics (CFD) concept introduced by Grossekettler (1982, 1999, 2005), which has been empirically tested in a lot of studies and is constantly reviewed (e.g. Blanckenburg (2007), Kubani (2007), Lorenz (2005)). The CFD-concept analyses real market processes with tools of time series analysis and investigates whether they operate efficiently or not. Furthermore the concept can be used as a tool for detecting cartels.

The CFD postulates the efficient operation of five processes. In this paper we use the analysis of the market clearing process and the rate of return normalisation process as examples to demonstrate the general methology. A working market clearing process aims to eradicate discrepancies between supply and demand quantities in the market under consideration in order to prevent a surplus of goods resulting from overproduction as well as consumers queuing due to inadequate consumer demand satiation. The rate of return normalisation process aims to ensure that production factors are guided towards their best economic use. This is achieved by variations in capacities and should result in a rate of return equalization. Workability is given with a significant long run mean of zero of both processes: the difference quantity between supply and demand (market clearing process) and of the rate of return difference (rate of return normalisation process). To specify the model, we implement a test of structural breakpoints, followed by a test based on the assumption of stationarity.

For our analysis we use data from the ifo Institute for Economic Research (Munich), the Bundesbank and Federal Statistical Office of the selected German industries from 1965-2007. We find workability of the German Chemical-, Electrical- and Automobile Industry and an efficient absorption of exogenous shocks. As an example for a non-workable market we analyze the German Cement Market, where a cartel has been detected and prosecuted by the German antitrust authority. We show that this would have been indicated by the CFD-Concept, even before the cartel was detected: As can be expected in the case of a cartel we see a stationary surplus quantity, a stepwise raising price, an excess rate of return and decreasing capacities. The CFD-Concept is capable to check for the existence of a cartel before starting an official proceeding.

The paper is structured as follows: Firstly, we introduce the CFD-concept and give a short overview of empirical studies that have used the concept. Secondly, we discuss the statistical methods and data we use for our analysis. Thirdly, we present the empirical results of selected markets. The paper ends with a short summary and conclusion.