Abstract

Liquidity Measurement in Experimental Stock Markets

The liquidity that goes along with various organisational forms of the trade in securities and its measurement are key issues in the shaping of security markets. This contribution starts with a discussion of the appropriateness of several liquidity measures proposed in the specialised technical literature. On the basis of a series of market experiments, a comparison is made of the liquidity of the three basic forms of security trading - total price determination, continuous auction selling and the market-maker system. Both the volume of trade and the measure proposed by Roll (1984) are apparently inappropriate for comparing the liquidity of differently organised security markets. There is no support by empirical evidence of overestimated transaction costs prognosticated for total price determination on the basis of theoretical arguments. The explicit bid/ask price spread, however, is basically an appropriate measure of liquidity. Such a spread can be ascertained also for total price determination on the basis of experimental market data. This spread has turned out to be smaller than the ones existing in continuous auction selling and in the market-maker system. However, the bid/ask price spread must be subjected to careful interpretation as well. This contribution shows that there may be systematic distortions in certain circumstances. These could be identified in the experimental markets examined by this contribution. But this is not necessarily possible with investigations of field date, which suggests that there is a real risk of distortions.