This article notes that dealers' bid/ask spreads should vary directly with their costs of adjusting to inventory imbalances. Thus, well-diversified dealers are expected to quote lower bid/ask spreads on stocks with substantial total risk caused by undiversifiable risk. Furthermore, the effect of systematic risk on bid/ask spreads should be negligible if dealers are compensated for systematic risk by market returns. This article shows that, empirically, bid/ask spreads of OTC stocks are insensitive to the systematic risk of individual stocks--even when only stocks with stable betas are considered. Furthermore, bid/ask spreads are not sensitive to changes in market variance, as would be expected if systematic risk affected spreads. While unsystematic risk affects bid/ask spreads, its effect is pronounced for stocks traded by small, undiversified dealers. If stocks are only traded by large dealers with low diversification costs, unsystematic risk does not affect bid/ask spreads.
Spreads between dealers' "bid" and "ask" prices comprise a major component of total trading costs in securities markets. Consequently, current microstructure research is concerned with the relative importance of informed trading, inventory holding, and order processing costs as they are reflected in the dealer's bid/ask spreads (Glosten & Harris, 1988;Stoll, 1989). This study contributes to this literature by examining the effects of portfolio risk which is an important aspect of security dealers' inventory holding costs.
In the microstructure literature, previous investigators have tried to explain how bid/ ask spreads vary with respect to security dealers' portfolio risks. Some of those papers (Stoll, 1978a;Benston & Hagerman, 1974) develop contradictory conclusions regarding the issue of whether a stock's systematic risk affects its bid/ask spread.
From a practical viewpoint it is important to determine whether systematic risk affects stock's bid/ask spreads. If it does, spreads and trading costs will increase whenever market risk rises. Elevated trading costs, in turn, could cause trading volume to diminish and stock volatility to rise further in a vicious circle effect. In addition, if systematic risk affects
The authors would like to thank Hans Stoll, Jess Chua, and finance faculty members at the