Earlier I explored reasons why value investing beats growth investing over a long term. The article was based on the data from the US stock market going back many decades. In this day and age when investors often invest in global stocks either through ADRs or via global funds or ETFs, we still need to ask if the investors are better off preferring value vs. growth stocks.
Fortunately for us, Brandes Institute has been tracking the global performance of growth versus value stocks. They call this study value versus glamour*, where glamour includes stocks that exhibit the quintessential growth characteristics: high multiples, high expected earnings growth rates as well as significant popularity among investors. You can download the 2012 edition of this study directly here. Keep reading for the key findings of the study.
Definition of the value and growth stocks in this study
We all understand the basic concepts of growth and value stocks. However, when it comes to a study like this that aims to serve as a benchmark, the definition of growth and value stocks need to be precise and consistent so appropriate baskets of stocks can be chosen for every year over the time period of the study and the results are dependable and actionable.
For this study, authors divided the spectrum of stocks in deciles based on fundamentals such as Price to Book, Price to Cash Flow and Price to Earnings ratios. Decile 1 included stocks with highest multiples (i.e glamour or growth) and Decile 2 included stocks with lowest multiples (value). While this method may end up miscategorizing certain stocks, when done over entire markets it should provide a solid representative sample. This has the added benefit of consistency and repeatability.
The study used this methodology to build baskets for geographies, market capitalization and valuation metrics.
Other Adjustments in the Study
The authors removed 50% of the smallest stocks from the sample. This eliminated micro caps from the study, arguing micro caps are not truly part of a large investor’s investable universe. The smallest stock in the study had the market capitalization of $740 million. Of the remaining stocks, the top 30% by market cap were classified as large cap, and the bottom 70% were classified as small cap. While the definition of small cap and large cap stocks vary, these are likely to be at variance to most investors understanding of these asset classes. Still, it does not affect the conclusions from the study.
Key Results from the Study
Returns for US stocks between 1968 and 2012 and stocks outside US from 1980-2012 were examined. Significant value premium was observed across geography, market capitalization and valuation metrics over long term.
- in US, long term returns for low P/B Decile 10 stocks (value) averaged 12.8% annually while that for high P/B Decile 1 stocks (glamour/growth) averaged 4.3%
- value stocks in emerging markets outperformed glamour stocks in emerging markets as well as value stocks in US and elsewhere
- in the short term during market declines, value stocks decline further than growth stocks, while during bull markets they appreciate more than growth stocks
- P/B valuation metric delivered the smallest value outperformance globally at 7.1% for average 5 year annualized return (this is interesting and unexpected)
To study the shorter term performances of growth stock vs value stock, the authors divided the 32 year period between 1980 and 2012 into 28 rolling 5 year periods and then compared the returns for each period. They found
- value beat growth in nearly every period. 1995-2000 was the only rolling 5 year period when growth stocks led by internet/tech companies outperformed value stocks by greater than 5% margin, but only in the small cap segment.
- outperformance averaged 5.1% for large cap stocks and 6.3% for small cap stocks
- value deciles showed smaller standard deviation in returns compared to growth deciles and corresponding Sharpe ratio (risk-adjusted return) was significantly higher for value stocks
(please note that the volatility as measured by standard deviation can be high in the very short term. In this case the returns are being measured over a period of 5 years)
Other interesting results
- emerging markets delivered an average value premium of 15.7% annualized, highest among the 23 countries surveyed
- of the 16 out 23 countries with statistically robust data, 13 countries’ returns suggested a value premium while results were inconclusive for 3 countries
- no country returns indicated presence of a “growth premium”
Conclusion: Value Premium is Widespread and Persistent
In 1998, Fama and French concluded that “the higher average returns on value stocks in United States are local manifestation of a global phenomena”. Brandes Institute’s study lends additional support to this conclusion. Market efficiency in the developed markets work to reduce the outperformance margin (compared to say, emerging markets), but value stocks remain the asset class to focus on for medium to long term investors.
* Please note that I have used the terms glamour and growth interchangeably based on the valuation metrics used to define these terms