Issues in assessing trade execution costs

This study assesses the sensitivity of trading cost estimates derived from publicly-available trade and quote data to two methodological issues: The time adjustment made before comparing trades to quotes, and the procedure used to designate trades as buyer or seller-initiated. The results indicate that making no allowance for trade reporting lags is optimal when assessing whether trades are buyer or seller initiated, for both Nasdaq and NYSE stocks. However, trade prices are best compared to earlier quotations when assessing trade execution costs, in order to capture the effect of systematic quotation revisions in the seconds before trades are reported. A technique for inferring trade direction recommended by Ellis et al. (J. Financial Quant. Anal. 35 (2000) 529) leads to significantly smaller estimates of trading costs than the well-known Lee and Ready (J. Finance 46 (1991) 733) algorithm. Despite the sensitivity of trading cost measures to these methodological issues, inference as to whether the Nasdaq dealer market or the NYSE auction market provides lower trade execution costs is not sensitive.