The Beauty Contest And Short Term Trading
Cass Business School; Centre for Economic Policy Research (CEPR)
University of Navarra – IESE Business School; Universitat Pompeu Fabra (UPF); Centre for Economic Policy Research (CEPR); CESifo (Center for Economic Studies and Ifo Institute for Economic Research)
January 2014
IESE Business School Working Paper No. WP-915-E
Abstract:
We consider a two-period market with persistent liquidity trading and risk averse privately informed investors who have a one period horizon. With persistence, prices reflect average expectations about fundamentals and liquidity trading. Informed investors engage in “retrospective” learning to reassess the inference about fundamentals made at the early stage of the trading game. This introduces strategic complementarities in the use of information and can yield two stable equilibria which can be ranked in terms of liquidity, volatility, and informational efficiency. We establish the limits of the beauty contest analogy for financial markets and derive a rich set of implications to explain market anomalies, and empirical regularities.
The Beauty Contest and Short Term Trading – Introduction
We study the drivers of asset prices in a two-period market where short-term, informed, competitive, risk-averse agents trade on account of private information and to accommodate liquidity supply, facing a persistent demand from liquidity traders.
Traders’ “myopia” ranks high on the regulatory agenda testifying policy makers’ concern with the possibly detrimental impact it has on the market.1 The issue has a long tradition in economic analysis. Indeed, short term trading is at the base of Keynes’ dismal view of financial markets. According to Keynes’ ‘Beauty Contest,” traders’ investment decisions are driven by the anticipation of their peers’ changing whims, rather than by the actual knowledge of the companies they trade. In this context it has been claimed that this type of behavior introduces a particular form of informational inefficiency, whereby traders tend to put a disproportionately high weight on public information in their forecast of asset prices (see Allen, Morris, and Shin (2006)).
In this paper we present a two-period model of short term trading with asymmetric information in the tradition of dynamic noisy rational expectations models (see, e.g., Singleton (1987), Brown and Jennings (1989)). We find that when liquidity trading is persistent there is strategic complementarity in the use of private information and provide sufficient conditions for it to be strong enough to generate multiple and stable equilibria which can be ranked in terms of price informativeness, liquidity, and volatility; this allows us to establish the limits of the beauty contest analogy for financial markets, and deliver sharp predictions on asset pricing which are consistent with the received empirical evidence (including noted anomalies).
Suppose a risk-averse, short term trader has a private signal on the firm’s fundamentals. His willingness to speculate on such signal is directly related to how much of his information will be re
ected in the price at which he liquidates. However, and importantly, it is also inversely related to his uncertainty about such a price. Indeed, the more volatile the price at which he unwinds, the riskier his strategy, and the less willing to exploit his private signal the trader becomes. But in a market with asymmetric information, a reduced response to private information can feed back in the volatility of the price. This is because, the less information is transmitted to the price, the larger is the contribution of non-fundamentals trades to price volatility. Therefore, the willingness to act on private information depends on the trader’s uncertainty over the liquidation price, and at the same time affects such uncertainty. In this paper we show that this two-sided loop can be responsible for the existence of multiple, stable equilibria that can be ranked in terms of informational efficiency, liquidity, and volatility.
The crux of our argument revolves around a particular type of inference effect from the information reflected by prices that arises when liquidity traders’ positions are correlated across trading dates. Indeed, with persistence second period investors can retrospectively reassess the first period inference about the fundamentals, based on the new evidence gathered in the second period. We thus term this effect “retrospective inference.” In a market with risk averse, asymmetrically informed investors, it is well-known that the price impact of trades arises from the sum of an inventory” component, and an “inference” component.2 While in a static market the two terms are positive, in a dynamic market retrospective inference can make the inference component turn negative. This diminishes the price impact of trades, reducing the volatility of the asset price, and boosting traders’ response to private information. The intuition is as follows.
See full PDF below.
