Stocks and the Long Term - Some Solid Research to Consider

It is a cliché that one should only invest in stocks if one has a long term horizon but often this advice is dispensed with a definition of long term using a number pulled out of the air or even without any number at all! Is long term five years, 10, 15, 20, 25, 30?

Thanks to the fine IndependentInvestor.info website (you will need to register to see content but it's all free and unbiased info) for uncovering some credible answers. As one should expect, there isn't a single number but a sliding scale of declining risk with extension of years invested. How Long is a Long-Term Investment? The 1 in 9 Rule summarizes the paper by economist Pu Shen of the Kansas City Fed, available at How Long is a Long-Term Investment.

Some of Shen's Discoveries

• showing risk on the basis of a one-time investment at the start (the typical "if you had invested $10,000 in Fund X in 1970, it would be worth $ZZZZZ today") understates the chances of losing money; the more realistic scenario, where an investor puts in money gradually over time, which he calls repeated investments, took at least 24 years before a positive real return on stock investments was always achieved. Stocks = the Center for Research in Security Prices Index, an index for the entire U.S. stock market from 1926 to 2002. The one-time method always showed positive returns after only 19 years, a difference of 5 years. The reason is the net effect of two opposite forces - time diversification (which reduces risk) and shorter effective holding periods (which hurts). Check out Shen's chart 2 below
  

• stocks never under-performed bonds (US Government 20 year bonds) after at least 26 years holding period (repeated investment method used), not exactly a mere blink of an eye.
  
• though the risk of stocks declined progressively with longer holding periods, the odd time they did have poor results, and even after 20 years the worst stock vs bond under-performance was still quite a hefty difference - check out Shen's chart 5 below. Sobering data, I'd say.
  

• quote: "Worse than investing in stocks right before a market crash is liquidating stocks shortly after the crash." (He says this in the context of people needing to retire then but of course a retired person does not typically spend all his/her money, or cash everything out, the day of retirement.) The worst possible 20 year holding period for stocks was ending in 1974 but from then on, there was a bumpy but ever-upward recovery. Moral of the story: hang on, don't panic, don't sell everything, try to sell as little stocks as possible after a crash - viz 2008 crash and 2009 recovery to date.
• even after 25 years holding period bond investors only beat inflation 34% of the time!! Now that's what I call risky. Stocks always beat inflation over 25 years and beat bonds 99.8% of the time. Stocks for the long-term indeed.

She shows that it is very misleading and harmful to define the long term for stock investment as five or even ten years, such as this article does. One should also keep in mind that Shen's paper did not factor in the annual fund expense fees and tracking error that an actual investor would face. This would lower returns and extend necessary holding periods.