In this paper, we investigate the relation between stock returns and aggregate liquidity. The market microstructure literature establishes that liquidity, reflecting asymmetric information and trading cost, is an important factor in price discovery. Recent empirical studies suggest that aggregate liquidity may be a systematic factor that impacts asset pricing. According to financial theory, investors require a higher expected return if systematic and non-diversifiable risks affect financial assets.
Hence, a market-wide liquidity shock should affect returns of all assets and the average returns of the aggregate market. If individual assets exhibit different sensitivities to liquidity, a market wide liquidity shock should be reflected in the cross-sectional variation of asset returns. Moreover, given risk-averse traders, asset returns should be positively related to the second moment of liquidity.
We investigate whether market-wide liquidity is priced. First, we ask whether the average returns of the aggregate market are related to aggregate liquidity, and whether cross-sectional differences in expected stock returns are related to the sensitivities of returns to fluctuations in aggregate liquidity. Second, we ask whether stock returns are related to the variability of market-wide, systematic liquidity.
Pástor and Stambaugh (2003) construct an order-flow induced “stock return reversal” to proxy for market-wide liquidity risk, investigate the cross-sectional relation between expected returns and systematic liquidity risk, and find that stocks that are more sensitive to aggregate liquidity have substantially higher expected returns. In this paper, we examine the time series relation between market returns and aggregate liquidity. We use the aggregate effective commission rate as the liquidity measure. The commission rate is an intuitive measure of trading costs that impact investor decision-making. Investors are generally concerned with expected returns net of transaction costs. When investors anticipate higher transaction costs or lower liquidity, stocks will be priced in such a way that they generate higher expected returns before deduction for transaction costs. Hence, we predict a positive relation between the aggregate effective commission rate (illiquidity) and market returns.
From a unique dataset of aggregate commissions for the U.S. securities industry and aggregate trading volume, we recover the quarterly effective commission rate for the aggregate market. We use this commission rate to proxy for market liquidity and examine the time series relation between aggregate liquidity and market returns for the period 1980-2001. Our results show that, over time, there is a significant and negative relation between market returns and liquidity, and a strong positive relation between market returns and the variability of liquidity. The impact of aggregate liquidity on market returns survives a number of robustness checks and is significant both statistically and economically after controlling for the well-known interest rate risk factors and trading volume.
We also examine the cross-sectional relation between liquidity and returns. The liquidity factor exhibits significant explanatory power on the cross-sectional variation of average returns, after controlling for the interest-rate risk factors and Fama-French factors, respectively. We find that the sensitivities of returns of size-based portfolios to aggregate liquidity are statistically significant and decrease monotonically with market capitalization. Our findings are consistent with those of Pástor and Stambaugh (2003) and further establish the role of liquidity as a risk factor in expected returns. This paper complements the cross-sectional negative return liquidity relation documented by recent studies.
