Informational externalities

Short-term traders, Learning and Informational externalities
Sahar Parsa Massachusetts Institute of Technology June 2009
Preliminary version; Comments Welcome; Check for updates at: http://econ-www.mit.edu/grad/sparsa/research

Abstract This paper examines how short term trading impacts the aggregation of information in …nancial markets. I develop a model where short-term traders, in anattempt to learn about the average beliefs of future market participants, make the price relatively more noisy. This typically introduces a negative informational externality on long-term investors. I show that (i) as the horizon of the informed traders decreases, the price becomes relatively less precise; (ii) an in‡ of informed ow traders in the market can decrease the informativeness of the pricewhen the traders have a relatively short horizon or the market is expected to be thin in the future; (iii) …nally, as rational informed short-term traders have access to an extra source of information about the future price, they end up creating more noise and a decrease in the informativeness of the price might result. Thus, paradoxically, more informed trading could lead to a less informativeprice.

I am deeply indebted to George-Marios Angeletos for its invaluable guidance. I also thank Fernando Duarte, Pablo Kurlat, Jean Paul L’ Huillier Bowles, Pablo Querubin, Johannes Spinnewijn, as well as the seminar participants at the MIT Macroeconomics Lunch, and International Breakfast. All remaining errors are my own. Correspondence: Department of Economics, MIT, Cambridge, MA 02142. Email:sparsa@mit.edu

1

1

Introduction

Financial markets are populated by traders with di¤erent horizons that meet at a given time. Some traders have short horizons. They enter and exit the market within a day or a week. Other traders have a longer horizon. Typically, they buy an asset from a …rm and hold on to the asset for much longer periods. Hence, long term traders are likely to carerelatively more about the underlying fundamental value of the stock they invest in than about predicting short-term movements. In contrast, short term traders who decide to buy a risky asset today to sell it tomorrow, will care about forecasting the price in the market tomorrow. Nevertheless, the price tomorrow will depend on the forecast of the price the day after tomorrow, and so on. Using backwardinduction, short term traders should also care about forecasting the fundamental value of the asset they will buy. In a world with symmetric information, by applying the law of iterated expectation, today’ forecast s of tomorrow’ forecast of the day after tomorrow’ price is given by today’ forecast of the day after s s s tomorrow’ price. Generalizing, the intertemporal higher order forecast boilsdown to today’ forecast. s s Hence, even though a trader cares about the future price, ultimately the only object of interest reduces to the fundamental. Short term traders’forecast of the price at which they will sell the asset comes down to the forecast of the fundamental value just as with long term traders. On the other hand, in the presence of asymmetric information, traders with di¤erenthorizons often behave di¤erently and try to predict very di¤erent objects. In particular, short term traders seem to gather and use di¤erent information than long term traders who mainly use information about the underlying value of the company. In an economy with asymmetric information, traders’ inference of the price at which they will sell the asset di¤ers from their inference of the fundamentalvalue.1 Rational short term traders will want to learn about the average forecast of future traders in order to forecast the future price. Typically, this will give an important role to the horizon of traders in …nancial markets. In this paper, I study the incidence of the horizon of traders on the informational e¢ ciency of …nancial markets. One of the major roles of …nancial markets appears to...