An open issue in the microstructure of equity markets is the role of common cross-firm variation in short horizon returns, order flows and liquidity. Since order flows are generally held to contain informed components, does common covariation in stocks' orders account for the covariance structure of short-term returns? Furthermore, is liquidity driven by strong common factors? The equity market breaks of 1987 and 1989, as well as the debt market crisis of 1998, for example, are widely perceived as systematic breakdowns in liquidity.
These issues are important for both microstructure theory and for institutional trading practice. Subrahmanyam (1991), Chowdhry and Nanda (1991), Kumar and Seppi (1994), Caballe and Krishnan (1994) have all extended the work of Kyle (1984, 1985) to multiasset markets by adding investors who are informed about macroeconomic factors and/or who have portfolio-wide liquidity shocks (e.g., portfolio substitution). In such environments inter market price discovery and order flow dynamics are obviously more subtle than when private information and/or trading noise is purely idiosyncratic.
Until recently, however, little direct empirical research has been conducted on the magnitudes of cross-stock interactions at the microstructure level. Given the sheer size of the Fitch, ISSM, TAQ and TORQ databases, it is perhaps not surprising that previous work has tended to focus on individual stocks in isolation from each other. This focus on stocks in isolation has, however, left us ignorant of even the most basic facts about cross- sectional interactions between stocks.
This paper answers these questions in two ways. First, we use principal components analysis to show that common factors exist in the order flows and returns of the 30 stocks in the Dow Jones Industrial Average (DJIA). In addition, canonical correlation analysis documents that the common factor in returns is highly correlated with the common factor in order flows. Second, we find some evidence of a common factor in quote-based proxies for liquidity, and to a lesser degree, in inferred price impact coefficients, after controlling for previously documented time-of-day seasonalities.
Download
Common Factors in Prices, Order Flows and Liquidity
