The Little Book that Beats the Market has published annual returns of over 30% from a 17 year period from 1988 to 2004. But is it realistic to expect those kinds of returns going forward over the next 17 years? While doing some research, I came across some interesting technical data that suggests that it may not be realistic.
The Dow in the 20th century returned about 5.3% per year, before dividends (adding dividends in, the return was slightly over 8%). However, these returns were not smooth. You could have had 2 long, 30-year investing careers in the 20th century and produced very different results depending on the time period. Have a look at the chart below from StockMarketTiming.com:
What you see here are a few distinct patterns over 100 years. For the first 40 years or so, the market traded in a relatively narrow range, with one major spike (the 1929 boom) and one major pit (the height of the depression in 1932). Overall, however, Dow investors would have only about 2.5% per year, or just over 5% with dividends reinvested. The post-war period from 1945-1966 was the first "boom" period, with the Dow on a steady upward march that returned investors an impressive 8.2% per year (over 11% with dividends). This was followed by another 20-year flat period which included the difficult 1970's, with inflation and gas shortages, as well as 2 major recessions in 1974 and 1982. However, after that followed one of the greatest 17-year bull markets in history, with nearly 13% annual returns in the Dow, 15% with dividends, as major new developments including the personal computer, the Internet, and major medical advances drove productivity forward.
It was also during this last bull market that Greenblatt performed his Magic Formula trials. While most long-term technical analysts agree that the last bull market ended in mid-2000, the majority of the 1988-2004 period fell within it. The S&P 500 and Dow both returned 12% annually during this period, while the Magic Formula strategy's annual return was over 30%. However, the Dow and S&P have both exhibited the characteristics of previous bear markets since the bubble burst in 2000. In these kind of markets (lasting 20 years or more), market returns have been much lower. Therefore, it seems reasonable that 30% annual returns are unlikely from the Magic Formula for the foreseeable future. It's purely hypothetical, but if the market delivers the "bear market" 2.5-3% annual returns over the next 10 years, the Magic Formula might be expected to do 2-3 times better, which would be a still solid 5-6% annually. Focusing on dividend stocks may improve this result a few percentage points. Focusing on only the best companies in the Magic Formula should improve results even more.
Now, a few things should be clear. First, MagicDiligence does not necessarily believe that trends exhibited in the past will necessarily manifest themselves in the same way going forward. Secondly, the Magic Formula is a fundamentally sound strategy that should outperform the market significantly. Numerous studies and books have confirmed that buying low P/E (or in our case, high earnings yield) stocks outperforms the market. And lastly, stocks have been confirmed over the long term to be the best returning class of investments. Wharton professor Jeremy Siegel, in his book Stocks for the Long Run, shows that, since 1800, stocks have far outperformed any other investment class, including gold, bonds, and real estate. There is no reason to believe this won't continue in the future.