Introduction to Factor-Based Investing
Posted on 02/15/2015
This article is sponsored by S&P Dow Jones Indices.
For decades investment portfolios have been constructed from a combination of cap-weighted index funds and active funds. Cap-weighted index funds allow investors to acquire the market portfolio in a simple, transparent, and cost effective manner. By contrast, active funds promise higher returns at the cost of greater complexity and higher fees.
In recent years institutional investors have employed a new approach to portfolio construction: factor-based investing. This increasingly popular approach lies between the passive and the active, allowing investors to target specific risk factors (return drivers) as well as market beta. These strategies use a transparent, systematic rules based approach, at relatively low costs.
The origin of factor-based investing can be tracked back to the linear Capital Asset Pricing Model (CAPM), one of the first financial theories to model asset returns as a function of factor risks. Formulated in the 1960s, it stated that there is only one factor, the market factor, driving the returns of assets.
Moreover, the CAPM framework highlights two sources of risk within any portfolio, one systematic, the other specific. This has important implications for portfolio construction. First, the specific component can be diversified away by holding many assets. Second, the systematic risk is a function of the portfolio beta and market risk. CAPM reveals valuable insights behind the mechanics of investment performance, namely that expected returns of assets are proportional to their systematic risks as measured by their betas. On the other hand, specific risks can be diversified away and are not rewarded with excess returns.
By the 1970s additional factors were introduced to improve CAPM as a risk tool. The first multifactor model was developed by Stephen Ross in 1976. Many of today’s commercial risk models are based on his Arbitrage Pricing Theory (APT), including macroeconomic factor models, fundamental factor models, and statistical factor models.
Pricing anomalies were soon discovered that contradicted CAPM and its use as a pricing model. The Fama and French three-factor equity model, incorporating the size and value effects in addition to the market, was widely regarded as an improvement . An extension of this three-factor model is the Carhart four-factor model, where the momentum effect is included . From a practitioner’s point of view, this highlights that there may be other priced factors, in addition to the market, that will reward investors over time. These factors drive the performance of investment portfolios. They underpin many of the factor-based products currently available in the market.
Pricing factors such as value, momentum, and quality have provided excess returns within the equity domain. The same principles are increasingly applied to commodities where factors such as roll yield and momentum are popular. Awareness is growing within the fixed income sphere too. Portfolios that target factors such as the term, credit, and high yield spreads are likely to follow. Within asset classes, factors can be combined to target multiple exposures – a multifactor approach. When combining factors, cross correlations can reveal diversification benefits improving portfolio risk return characteristics.
It seems inevitable that practitioners will continue to utilise and develop factor-based products, due to their transparent and systematic rules and relatively low costs. The next few years will be interesting.
The Story of Factor-Based Investing research paper is available at: Research Paper.
2 Carhart, M.M., (2012). On Persistence in Mutual Fund Performance. Journal of Finance. 52 (1), 57-82.