Factor investing is an investment approach that involves targeting quantifiable firm characteristics or “factors” that can explain differences in stock returns. Security characteristics that may be included in a factor-based approach include size, low-volatility, value, momentum, asset growth, profitability, leverage, term and carry.[1] [2] [3]
A factor-based investment strategy involves tilting investment portfolios towards or away from specific factors in an attempt to generate long-term investment returns in excess of benchmarks. Proponents claim this approach is quantitative and based on observable data, such as stock prices and financial information, rather than on opinion or speculation.[4] [5] Factor premiums are also documented in corporate bonds[6] and across all major asset classes including currencies, government bonds, equity indices, and commodities.[7]
Critics of factor investing argue the concept has flaws, such as relying heavily on data mining that does not necessarily translate to real-world scenarios.[8]
The earliest theory of factor investing originated with a research paper by Stephen A. Ross in 1976 on arbitrage pricing theory, which argued that security returns are best explained by multiple factors.[9] Prior to this, the Capital Asset Pricing Model (CAPM), theorized by academics in the 1960s, held sway. CAPM held that there was one factor that was the driver of stock returns and that a stock's expected return is a function of its equity market risk or volatility, quantified as beta. The first tests of the Capital Asset Pricing Model (CAPM) showed that the risk-return relation was too flat.[10]
Basu was the first academic to document a value premium in 1977.[11] The roots of value investing date to decades earlier, and were formalized by Benjamin Graham and David Dodd as outlined in their 1934 book Security Analysis. Their student Warren Buffett outlined their findings and application in his 1984 article "The Superinvestors of Graham-and-Doddsville". In 1981 a paper by Rolf Banz established a size premium in stocks: smaller company stocks outperform larger companies over long time periods, and had done so for at least the previous 40 years.[12]
In 1992 and 1993, Eugene F. Fama and Kenneth French published their seminal three-factor papers that introduce size and value as additional factors next to the market factor.[13] In the early 1990s, Sheridan Titman and Narasimhan Jegadeesh showed that there was a premium for investing in high momentum stocks.[14] [15] In 2015 Fama and French added profitability and investment as two additional factors in their five-factor asset pricing model.[16] Profitability is also referred to as the quality factor.[17] Other significant factors that have been identified are leverage, liquidity and volatility.[18] [19] [20]
The most well-known factor is value investing, which can be defined primarily as the difference between intrinsic or fundamental value and the market value. The opportunity to capitalize on the value factor arises from the fact that when stocks suffer weakness in their fundamentals, leading the market to overreact and undervalue them significantly relative to their current earnings. A systematic quantitative value factor investing strategy strategically purchases these undervalued stocks and maintains the position until the market adjusts its pessimistic outlook.[21] Value can be assessed using various metrics, including the P/E ratio, P/B ratio, P/S ratio, and dividend yield.
Low-volatility investing is a strategy that involves acquiring stocks or securities with low volatility while avoiding those with high volatility, exploiting the low-volatility anomaly. The low-volatility anomaly was identified in the early 1970s but gained popularity after the 2008 global financial crisis. Different studies demonstrate its effectiveness over extended periods. Despite widespread practical use, academic enthusiasm varies, and notably, the factor is not incorporated into the Fama-French five-factor model. Low-volatility tends to reduce losses in bear markets, while often lagging during bull markets, necessitating a full business cycle for comprehensive evaluation.
Momentum investing involves buying stocks or securities with high returns over the past three to twelve months and selling those with poor returns over the same period. Despite its established phenomenon, there's no consensus on its explanation, posing challenges to the efficient market hypothesis and random walk hypothesis. Due to the higher turnover and no clear risk-based explanation the factor is not incorporated into the Fama-French five-factor model. Seasonal effects, like the January effect, may contribute to the success of momentum investing.
In a 2019 paper, Rob Arnott, Campbell Harvey and colleagues identify several problems with factor investing.[8] They assert that due to data mining, very few factors have statistical significance in real-world scenarios. They also argue factors may not offer promised diversification under all market conditions, as factors may change in their level of correlation over time.
In a 2016 paper, Arnott and colleagues noted that many factors become popular among investors, leading to high valuations among such stocks and subsequent expected poor returns.[22]