The Magic of Hindsight

Abstract

The crucial assumption in most pre-trade transaction cost models is market neutrality. Consequently, expected costs for these models are entirely based on one's own trading strategy and direct market impact. Market effects due to other market participants are generally completely ignored. This article describes how to improve transaction cost estimates (benchmarks) for post-trade analysis. After deriving the expected transaction costs from a specific pre-trade model, we incorporate the general market effects at the time when the trades actually took place. The proposed model incorporates market returns, spread variables, and trade imbalances. The model also allows for the decomposition of the cost of a transaction into two components: the cost due to one's own trading and the cost due to general market effects. The framework is applied to a proprietary pre-trade transaction cost model.