A Framework for Assessing Factors and Implementing Smart Beta Strategies

Abstract

Twenty years ago there were only five equity factors (market, value, small-cap, momentum, and low beta). Today the literature contains research papers on hundreds of supposed factors, most of which will not produce a reliable positive premium in the future. Rather than adopting a statistical approach, we propose a three-step heuristic to help investors discriminate between genuine premium-bearing factors and the spurious products of data mining. A robust factor is one whose persistence and economic meaning have been studied in numerous papers published in top-tier journals; whose premium is statistically significant across time periods and in most countries; and whose effect survives reasonable changes in the definition. We also address practical aspects of factor allocation. Appropriate factor allocations depend on the investors’ definition of risk (volatility versus tracking error), risk tolerance, and ability to implement tactical or dynamic allocations, as well as the investment organization’s governance structure and politics.