The European Science Foundation Network on Financial Markets, which was founded in 1988 in order to promote interaction and debate among European researchers in financial economics, is administered by the Centre for Economic Policy Research and chaired by Colin Mayer, Co-Director of CEPR's Applied Microeconomics programme. The theme of the third workshop of the Network, held at the Université de Lausanne between 31 January and 2 February 1990, was `Market Microstructure', and it was attended by thirty participants from all over Europe. The conference was organized by Jean-Pierre Danthine (Université de Lausanne and CEPR).

Price Volatility

Riccardo Rovelli (Università Bocconi, Milano) presented a paper on `Measuring the Value of Hedging Opportunities: The Price of French Government Bonds before and after the MATIF', written jointly with Alessandro Citenna. They sought to determine the effect of the introduction of a futures contract on the spot price of the underlying asset. If the introduction of a futures market allows investors to hedge risks, then there should be a reduction in the required rates of return on the underlying securities. In order to test this they modelled the excess return on the French Treasury bond market of the premia attached to a number of specific risk factors in the tradition of the `arbitrage pricing theory. Only the interest rate risk was found to be significant and even the premium on this was not significantly different from zero from April 1986 onwards.

Heinz Zimmermann (Hochschule St-Gallen) and Patrick Artus (Caisse des Dépôts et Consignations, Paris) both thought that there was a need for more formal modelling in this area. Destabilization may result either from asymmetries of information or from increased levels of speculation. Markets in futures contracts may be too large in relation to those in the underlying assets and this may compromise liquidity. Questions were raised about the suitability of using a single factor instead of estimating a more general multi-factor model.

Over the last decade the Paris Bourse has undergone fundamental reforms. Continuous auctions have replaced batch auctions; agents de change have been replaced by intermediaries who can trade on their own accounts; and commissions have been liberalized. In `A Comparison between the London and Paris Markets for French Cross- Listed Shares' Marco Pagano (Università di Napoli and CEPR) and Ailsa Roell (LSE) examined the effects of these reforms. They found that both share price volatility and trading volumes have declined, and that the Paris Bourse currently offers greater liquidity than the London market to investors, but that there are no remaining unexploited arbitrage opportunities between the two markets at present.

Isabelle Bajeux (ESSEC) warned against making comparisons of volatilities before and after reform on the grounds that the fundamentals might have changed and that controlling for these would not be easy, while Ragnar Lindgren (Stockholm School of Economics) attributed the declines in volatility to increased activity by market makers and thought that empirical evidence should be sought on this possibility.

Insider Information

The `efficient market hypothesis' asserts that in markets with asymmetries in information between different investors, equilibrium prices aggregate information efficiently and investors can infer all relevant information from traded prices. In a paper on `The Efficient Market Hypothesis and Insider Trading in the Stock Market' (written jointly with Eric Maskin), Jean-Jacques Laffont (Université de Toulouse) demonstrated, however, that large traders have incentives to restrict the amount of information that they convey to the market, which causes the efficient market hypothesis to break down and leads to the emergence of pooling rather than separating equilibria.

Rafael Repullo (Banco de España and CEPR) and Jürgen Dennert (Universität Basel) agreed that the best informed traders are not generally small and that an analysis of equilibria with large traders was necessary. It was pointed out that separating equilibria are informationally efficient but do not lead to efficient risk sharing. When the variances of returns are small, the inadequacy of risk sharing dominates the provision of information, and pooling is preferred by all investors.

In `Using Privileged Information to Manipulate Markets: Insiders, Gurus and Credibility', written jointly with Roland Benabou, Guy Laroque (Institut National de la Statistique et des Etudes Economiques (INSEE), Paris) demonstrated that it may be possible for insiders to manipulate asset prices by means of strategically distorted announcements and predictions. They also examined the extent to which insiders' influence on the market is limited by public awareness of deliberate manipulation. Their model provides an economic rationale for restrictions on insider trading to preserve the integrity of information where truthfulness is either not readily verifiable or unenforceable.

Jean-Charles Rochet (Université de Toulouse) argued that while the paper offered a convincing and elaborate model of how opportunistic journalists (or `gurus') can manipulate financial markets and how their own reputations vary over time, such journalists would always be detected by the public in the long run, since the preservation of their own reputations would not be a sufficient incentive for them to report honestly all the information available to them. Giovanna Nicodana (Università Bocconi, Milano) questioned whether the model can explain persistent mispricings in the market, since in its current form it describes agents whose credibility has been reduced to zero.

Stock Markets

Anthony Neuberger (London Business School) and Ailsa Roell (LSE) studied `Components of the Bid-Ask Spread: A Glosten-Harris Approach' using transactions data from the London Stock Exchange. They wished to establish the extent to which the spread is gross profit as against compensation for trading with better informed investors. The two can be distinguished because the gross profit component is transient while the asymmetric information component is permanent. The analysis of the data suggested that the gross profit component comprised by far the greater part of the spread.

Ron Anderson (Columbia University and Université Catholique de Louvain) thought that the study added significantly to the literature on bid/ask spreads. Bruno Biais (HEC-Paris) thought that the theoretical basis of their analysis was unclear. He presented a simple inventory model whose implications were consistent with the results presented by the authors.

Walter Wasserfallen (Foundation of the Swiss National Bank, Gerzensee) reported an analysis of `Pricing Initial Public Offerings - Evidence from Germany', written jointly with Christian Wittleder. This study covered a data set of 92 initial public offerings (IPOs) over the period 1961-87. He showed that the initial excess returns were substantial and that only six of the IPOs were overpriced. Several possible reasons for overpricing were examined, and the results suggested that underpricing is positively related to proxies for ex ante uncertainty about the value of the newly issued shares.

Tim Jenkinson (Keble College, Oxford, and CEPR) argued that no satisfactory explanation for the underpricing of IPOs has yet been found. In particular, he noted that since virtually all IPOs are underpriced the `winner's curse' theory, which suggests that outside investors receive overpriced issues, cannot be applicable to Germany. Pekka Hietala (INSEAD) argued that the only variable that seemed to provide an explanation for underpricings was the standard deviation of stock returns after the IPOs, but was doubtful as to whether this could be interpreted as a proxy for ex ante uncertainty as opposed to an ex post response of market trading volumes to the levels of discounts on new issues.

The Foreign Exchange Market

The unbiasedness of forward rates as predictors of future spot rates has received considerable attention in finance, and most results reject the unbiasedness hypothesis. In `Market Microstructure Effects of Government Intervention in Foreign Exchange Markets', written jointly with Peter Bossaerts, Pierre Hillion (INSEAD) examined whether the explicit modelling of the bid-ask spread explains this rejection. Their paper notes that if government intervention is awaited, this introduces skewness into the expectations of future spot rates and hence biases tests of efficiency that use the average bid-ask quotes of forward and spot rates.

Ailsa Roell was not sure whether devaluations or revaluations lead to asymmetries of information, and suggested that this will depend on the distribution of insiders' informational advantage. She also thought that risk aversion, and inventory holding costs might be more important than the asymmetric informational factors emphasized in the paper. Bruno Biais suggested that skewness in distributions may reflect differences in the informational content of sales and purchases, and that more attention should be given to the information contained in trades about the beliefs of insiders.

Christian Wolff (Universiteit van Limburg and CEPR), in a paper on `EMS Exchange Rates', written jointly with Frederik Nieuwland and Willem Verschoor, examined the alternative hypotheses that exchange rates in the European Monetary System (EMS) follow random walks and that exchange rate coordination introduces mean reversion into the distributions. He concluded that EMS exchange rates are most adequately described by a combined jump-diffusion- ARCH model, which is consistent with the patterns of exchange rate movements observed elsewhere.

Both Mark Taylor (City University Business School and CEPR) and Hermann Garbers (Universität Zürich) thought that this was a valuable contribution to the literature. Mark Taylor made a number of suggestions of how the analysis could be extended and further tests performed. Hermann Garbers suggested that it would have been useful to search for common trends in the vector of all exchange rates instead of looking at them one by one.