Academic research has proved that one of the most available pieces of business data—the growth in the book value of assets could be very valuable for investors. The data mentioned above can be utilized in a popular trading strategy based on the growth effect anomaly. There is a strong asset growth effect in U.S. stock returns, and this paper updates the previous results of Cooper, Gulen, and Schill. The idea is simple; on a risk-adjusted basis, low asset growth stocks significantly outperform high asset growth stocks. This effect is consistent over time with the returns of low asset growth stocks exceed those of high asset growth stocks in 90% (equal-weighted) or 72% (value-weighted) of the calendar years in the sample of the paper. Additionally, even though a return differential is larger among small-capitalization stocks, the authors have shown that the effect is still economically and statistically large also among large-capitalization stocks. Therefore, the strategy could be easily implemented with only small trading and slippage costs.
Interestingly, the authors state that a firm’s growth rate in assets is at least as powerful in explaining returns as other well-known effects such as size, book-to-market, return momentum, and reversals. Moreover, the proposed strategy, which consists of buying low asset growth stocks and selling the high asset growth stocks, has a low correlation to the equity market factor.
Similar results can be found, for example, in the work of Watanabe, Xu, Yao, and Yu: “The Asset Growth Effect: Insights from International Equity Markets“. The authors say that stocks with higher asset growth rates experience lower future returns in 40 international equity markets, consistent with the U.S. evidence documented by Cooper et al. (2008). This negative effect of asset growth on stock return is stronger in developed markets and in markets where stocks are more efficiently priced.
A variety of papers suggest that the return premium achieved by low asset growth stocks is consistent with compensation for risk (for example, Gomes, Kogan, and Zhang, 2003; and Li, Livdan, Zhang, 2008). Firms maintain a mix of growth options and assets in place, but growth options are inherently more risky than assets in place. As firms exercise growth options, the asset mix of the firm becomes less risky as assets in place displace growth options. The systematic reduction in risk following the exercise of growth options induces a negative correlation between investment and subsequent returns. However, empirical findings are all also consistent with systematic mispricing across asset growth as a firm characteristic. Therefore, the authors are unable to recognize whether the return premium for low growth stocks is due to systematic variation in risk or the return reversal caused by systematic overcapitalization of high growth stocks and undercapitalization of low growth stocks. Building on that, another past research has concluded that the asset growth effect is not fully explained by variations in risk.
However, there is a possibility that the effect is at least partially due to the systematic market mispricing of growing businesses. That source of mispricing could be caused by the extrapolation of past gains to growth for high asset growth companies. A good insight on the reasons for functionality could be found in the work of Kam and Wei: “Asset Growth Reversals and Investment Anomalies“. Quoting the authors: “We simultaneously test the prominent rational and behavioral explanations of the negative relations between corporate asset growth or investments and subsequent stock returns by extensively examining the effects of realized and predicted subsequent growth on the relations. We find: (i) returns on low growth firms with low subsequent growth are not higher than those on high growth firms with subsequent high growth; (ii) high growth firms that have subsequent high growth do not underperform, and the return spreads between low and high growth firms are lower when high growth firms have higher subsequent growth; (iii) the relations between growth and returns are weak or even in opposite direction when subsequent growth tends not to reverse but are significantly negative when subsequent growth tend to reverse and are stronger when the reversals are more extreme. Our findings are consistent with the hypothesis based on extrapolation and growth-based style investing but less consistent with the other explanations.”
Backtest period from source paper
Confidence in anomaly's validity
Notes to Confidence in Anomaly's Validity
Notes to Indicative Performance
per annum, annualized (geometrically) monthly return 1,59% from table 2
Period of Rebalancing
Notes to Period of Rebalancing
Notes to Estimated Volatility
estimated from t-statistic, data from table 2
Number of Traded Instruments
Notes to Number of Traded Instruments
more or less, it depends on investor’s need for diversification
Notes to Maximum drawdown
Moderately complex strategy
Notes to Complexity Evaluation
Simple trading strategy
The investment universe consists of all non-financial U.S. stocks listed on NYSE, AMEX, and NASDAQ. Stocks are then sorted each year at the end of June into ten equal groups based on the percentage change in total assets for the previous year. The investor goes long decile with low asset growth firms and short decile with high asset growth firms. The portfolio is weighted equally and rebalanced every year.
Hedge for stocks during bear markets
Not known - Source and related research papers don’t offer insight into the correlation structure of the proposed trading strategy to equity market risk; therefore, we do not know if this strategy can be used as a hedge/diversification during the time of market crisis. The strategy is built as a long-short, but it can be split into two parts. The long leg of the strategy is surely strongly correlated to the equity market; however, the short-only leg can be maybe used as a hedge during bad times. Rigorous backtest is, however, needed to determine return/risk characteristics and correlation.
Out-of-sample strategy's implementation/validation in QuantConnect's framework