Purchase the US Equities Researcher Dataset by Siblis Research and analyze the historical returns of almost ten thousand public US companies for the past 40 years. The dataset provides adjusted share prices (total returns) on a monthly level of 9,480 individual public companies. The dataset provides also market capitalizations & GICS sector classifications of the companies so you can analyze the market returns by size and sector.
The dataset includes all companies that have been part of the S&P 500 index at any point since 1979 and all companies part of the Russell 3000 index 1995, including delisted stocks. The data is available for large cap companies since 3/31/1979. For mid cap & small cap stocks the data is available since 3/31/1995. Examine a sample dataset from here.
Historical Returns of the US Stock Market: Indices vs Individual Stocks
If you would stop someone in the street and ask her what’s the return she can expect from buying stocks, the answer will be around 10% annually. Everyone who has ever invested in stocks know that average yearly return of the market (and market always meaning S&P 500 index without exception) is a few basis points below 10%.
But does this mean that if you randomly buy any stock you can be fairly sure to gain an annual return of 10%?
The returns of individual stocks versus the returns of stock indices have been discussed a lot after Hendrik Bessembinder from Arizona State University published his working paper “Do Stocks Outperform Treasury Bills?” in 2017. In his paper (with a very provocative title), Professor Bessembinder demonstrates that individual stock returns are very far away from the 10% annual return that every investor has been taught to consider the average market return.
Between 1926 to 2015, only 42.1% of the common stock maintained in the CRSP database are beating Treasury Bills holding period returns. If returns are examined on a monthly period, only 47.7% of all monthly returns of common stocks are larger than the one-month Treasury rate. The paper states 4% of the best performing stocks are responsible for the entire gain in the US stock market since 1926 and less than 0.4% of stocks account for over half of the total market returns.
Many researchers remind that the results of the paper are not really that novel but the paper has still been able to spark a lot of discussion around the topic. If the amount of winners is so small does active stock picking make any sense? Is passive indexing the only serious option that every investor should consider?
The results of Prof. Bessembinder’s paper have been confirmed and accepted by both academics and practitioner but many people claim that the passive indexing is not necessary the only takeaway from the study. For example Jack Vogel from Alpha Architect says that while the paper is a very strong argument that investors should definitely buy diversified portfolios, it doesn’t say which individual stock characteristics should be placed in these diversified portfolios. It has been demostrated that portfolios focusing on factors like value and momentum have outperform benchmark indices in the long-term. So while investing in a handful of individual companies might not make much sense in a long run, building factor based portfolios might be the key of beating the market.