Factor Investing: Gross to Net Returns

From Theory to Reality

January 2018. Reading Time: 10 Minutes. Author: Nicolas Rabener.

SUMMARY

  • Long-short multi-factor portfolios generate attractive returns before fees
  • Returns are much less attractive post fees charged historically
  • However, some fees in the long-short space are likely justified given higher complexity

INTRODUCTION

Reality is the murder of a beautiful theory by a gang of ugly facts (Robert Glass, 2002). Factor investing can be considered one of the beautiful theories of the investment world as it is backed by a significant amount of empirical research and can be implemented by investors across markets and asset classes. The gang of ugly facts is that most of the research is based on backtesting, which is full of inherent biases, and investors get charged a variety of fees for accessing the returns from factor investing. In this short research note we will analyse the path from gross to net returns for a long-short multi-factor portfolio.

METHODOLOGY

We focus on four factors namely Value, Momentum, Low Volatility and Quality in the US stock market. The multi-factor portfolio is created via the intersectional model, i.e. by ranking the stock universe by the four factors simultaneously (please see our white paper Multi-factor Models 101 for further details). The portfolio is constructed from the top and bottom 10% of the ranked stocks, is adjusted to achieve beta-neutrality, only includes companies with a market capitalisation above $1 billion and is rebalanced monthly.

MULTI-FACTOR PORTFOLIO PERFORMANCE POST TRANSACTION COSTS

The chart below shows the performance of the long-short multi-factor portfolio before and after transaction costs, which are assumed to be 10 basis points per transaction. Institutional investors can trade for less than 1 basis point per transaction, but the impact costs of executing orders are often multiples of that and should be reflected as well. The portfolio rebalances monthly, but given that the stock selection is based on the ranking of several factors, the turnover is relatively low.