Returns for Savers

Can A Saver Estimate Their Future Market Returns?

This website extols the idea that the markets are unpredictable. However, it would be foolish to invest in a risky marketplace without any idea of the potential outcomes. Unfortunately, if you are still saving for retirement, those potential outcomes are very tough to gauge.

Consider somebody who is closing in on retirement. Even over a 10 year period, it is very difficult to predict what the market might do. There are all sorts of ways to try to predict the market’s direction, but even the best methods can explain only about 40% of the results over a decade. That is nothing to hang your hat on.

If you are 5, 10, or 15 years from when you hope to retire, your returns from this point onward will matter a great deal. You have hopefully been saving for a decade or two already, and you want the shares you have accumulated to rise in value and build up your nest egg. But having any certainty what will come over a time period this short is next to impossible.

What about on a longer timescale? Once we move beyond a few decades, we can in fact make a reasonable prediction of future market returns. Some sources will suggest that we can make such predictions based on historic returns, but this approach is wrong. It is much better to base future estimates on current market valuations. One such method is to use the current dividend yield and knowledge of average long-term dividend growth. If the dividend yield of the S+P 500 is 3%, for instance, you can add that to the dividend growth rate of 1.3%, for a total expected real (inflation adjusted) return of 4.3%. (For more on this method of calculating long term returns, I recommend digging into some reading from William Bernstein.)

Is it a perfect guess? No, but it usually comes reasonably close. Keep in mind the markets can be a wild ride from year to year, but over time they should fall back onto this steady pathway. Below is a graph of this method put to the test. The blue bars are expected real returns and the red bars are actual real returns, with dividends reinvested. They cover 50 year forward-looking periods starting in 1871, all the way until 1962 (the last year that we can get 50 years of data from). The chart is based on the S+P 500 using data from Robert Schiller.

Image and video hosting by TinyPic

So if a young investor has 30 or 40 years to go until they retire, shouldn’t they be able to make an educated guess of their long term returns? Again, the answer is no. If the young investor had a big chunk of money to invest all at once, then it would be a different story. Our long term expected returns can work for lump-sum investing, when there are no additions to or withdrawals from the account. But that doesn’t apply to most people. Most young investors are starting out on a savings plan, making periodic additions to an account with the hopes to retire someday. Consider that over 30 or 40 years the market will be rising and falling, and savers will be making contributions during many different periods with many different expected returns. What is much more important to the saver is the order of returns. They must hope to start out their savings program with a period of poor returns, that way they can buy up lots of shares cheaply. Only once they approach retirement and have accumulated plenty of shares do they want the market to rise.

For the lump-sum investor, the order of returns does not matter. If you know you are headed to Rome, then all roads do in fact lead there. For the saver, the order of returns is paramount. Great returns early followed by poor ones later will not bring good results. So back to our original question, can a saver estimate their future returns? In effect, no. When you are starting out a savings program estimating returns is of little use because you don’t have much invested yet. As you approach retirement, you have more invested, but your estimates are less and less likely to be correct. An uncertain order of returns and shorter-term market fluctuations seem to overwhelm any meaningful predictions.

Consider the following chart. Again, using Robert Schillers data back to 1871, I have plotted the total accumulation of a person who invests a real $5000 each year for 30 years in the S+P 500. It is compared to the expected real returns of the index predicted at the start of the savings program using the dividend method described above. Just by looking at the cloud it is clear there is effectively no correlation. Calculating your expected returns when you start out your savings program will tell you nothing useful.

Image and video hosting by TinyPic

So what is a young saver to do? I still feel it is wise to keep appraised of current market valuations throughout your savings program. Even more importantly, you should understand past trends and the range of possible outcomes you can expect. The order of returns that takes place over your accumulation phase is merely a roll of the dice, so focus instead on what you can control. Limit the impact of a bad order of returns by starting as early as possible and saving as much as you can. You won’t regret it.

Malcolm Lewis, November 2012

See also The Ages of the Investor: A Critical Look at Life-Cycle Investing, by William Bernstein


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s