Can we predict stock returns over the long term?


(MoneyWatch) There's so much confusion in the debate about the future long-term performance of stocks that it's impossible to know who you should believe:

On one side are the naysayers who say stocks won't return what they have in the past due to major problems such as slow economic growth caused by our large national debt. On the other are proponents who point to the wide spread between the yields of stock earnings and Treasuries.

    The research team at Vanguard recently tackled this issue and found that, basically, neither view was right. In a recent study, the investment giant found that "many commonly cited signals have had very weak and erratic correlations with actual subsequent returns, even at long investment horizons." Its findings include measurements used by both sides of the debate, though one measurement stood out above the rest.


    However, Vanguard did find that valuations do matter. The researchers found that metrics such as price-to-earnings ratios have a little value over the long term, but "even then, P/E ratios have 'explained' only about 40 percent of the time variation in net-of-inflation returns."

    The important lesson for investors is that while valuation metrics, such as P/E ratios, are the strongest indicators of future returns, they still leave a large portion of returns unexplained. An important reason is that the risk premium investors demand to own stocks isn't constant. Instead, it varies based on market conditions. And since the research shows that a large percent of returns are explained by surprises, it's important that you think of expected stock returns in terms of probabilities, not certainties. Consider that for the period 1926-2011, real 10-year U.S. stock returns have ranged from approximately -5 percent to 20 percent!

    Vanguard also stated that stock returns should definitely not be forecast on a short-term basis. The reason is that stock returns are essentially unpredictable at short horizons. Vanguard's research found that the estimated historical correlations of the various valuation metrics they looked at with the 1-year-ahead return were close to zero. They added: "Quite frankly, this lack of predictability is not surprising given the poor track record of market-timing and related tactical asset allocation strategies."

    Monte Carlo simulation

    So what can you do? One way to look at future returns is through the use of Monte Carlo simulation, which uses a set of assumptions -- such as investing time horizon, inflation rates and asset allocations -- to give you probabilities of your portfolio achieving a specific result.

    While it won't tell you how your portfolio will perform in the future (since that's impossible), it will help you understand the need to prepare for the various possible outcomes. That includes putting in place a contingency plan detailing what actions should be taken if financial assets fall to such a degree that you might not reach your goals. Those actions might include remaining in or returning to the work force, reducing current spending, reducing the financial goal, selling a home and/or moving to a location with a lower cost of living.

    Image courtesy of Flickr user 401(K) 2012

    • Larry Swedroe On Twitter»

      Larry Swedroe is director of research for The BAM Alliance. He has authored or co-authored 13 books, including his most recent, Think, Act, and Invest Like Warren Buffett. His opinions and comments expressed on this site are his own and may not accurately reflect those of the firm.