The purpose of the study is to determine why bid-ask spread anomalies occur in the quotations for American-style options of the S&P Index ETF (SPY). It is hypothesized that non-market-based constraints compel the emergence of risk-aversion price behaviors that force “abnormal” pricing. Anomalies were defined and identified as the independent variable; while moneyness, days-to-expiry, open interest, daily volume, and implied volatility were defined as dependent variables. Each variable was bin-range-defined, and frequency-probability analyzed. Variable incidence rates were tested against anomaly spreads for covariance and/or correlation, and regression was applied.
The R2 was small. However, all predictor variables were significant; the coefficients led to an endorsement of some expected behaviors, and rejection of others. Future research is suggested to determine possible causative forces of those behaviors inconsistent with the model. Overall, however, bid-ask anomalies appear to be the consequence of localized market-maker actions, such as bid-ask pricing suboptimal to the trader but optimal to the market-maker.
The study discusses the possible impact that algorithms may have on option bid-ask price discovery, and that algorithms may be a factor contributing to the decline in numbers of CBOE option market makers. A concentration of business to a few algorithmically-administered very large market makers could have a significant impact on option price implied volatilities and trade practices in the future, particularly for hedge or spread traders. This problem will likely be amplified as AI enters the trading arena. The impacts are yet unknown, but if the consequences of AI in other areas are a guide, the future suggests fewer trading and risk management professionals and a further concentration of market-maker activities.
Options, Spreads, Algorithms, Marker Makers, Market Microstructure, Efficient Market Hypothesis