How To Forecast Future Stock Prices . . . Usually

Wall Street research can be extraordinarily useful in helping to gain a better understanding of the inner workings of companies.  Having been on the research side of the investment world for many years, I’ve had extensive experience in exploring the nuts and bolts of dozens of companies, both through analysis of numerous financial reports and regular contact with corporate executives.  Some of it was useful; some was not.

In all cases, the goal was to come up with estimates of where company earnings were likely to go in the year ahead and the years beyond. As analysts, we looked for consistency of progress and the potential to grow at above average rates.  The task of coming up with these kinds of conclusions was a matter of translating discussions of ongoing operations and developments into numbers that made sense.

Once that was done, it was a matter of viewing the current analysis against the backdrop of the past.  The range of past valuations would often provide an indication of what might lie ahead.  So if the price-earnings ratio (p/e) had previously run from 15 to 18 times earnings, one could project future prices by multiplying future earnings by the historical p/e.

This, of course, is not a flawless equation.  For one thing, overall stock valuations are heavily influenced by the current level of interest rates.  When rates are low, as they are now, valuations tend to be high.  And vice-versa.  For another, there are many instances when interim developments disrupt a lengthy record of progress, which invariably has a significant negative impact on stock prices.  When that happens, you can throw the historic range out the window.

Still, the equation is quite useful from a macro view.  As an illustration, I studied the S&P 500 from 1960 to the present and compared the actual levels of the index with levels calculated by multiplying S&P 500 earnings by 18 times.  Over the 57-year period, there has been a significant correlation between the two, though there have been a few times when it was clear that valuations were stretched in the extreme.

The outstanding example of excess was during the dot.com era of 1999-2000, when the S&P turned out to be overvalued by almost 50%.  The debacle that followed corrected that aberration, but it took more than a decade before the S&P 500 was able to move to new high ground.

The flipside was back in 2008-9, when the S&P plunged, virtually in line with the change in underlying earnings.

Where are we today?  That depends on where S&P earnings are likely to end for the full year.  At the moment, Wall Street analysts are estimating earnings at about $134 a share.  Based on that estimate, the index is priced about where it should be.  The estimate, however, represents a gain of better than 20% from 2016 results.  If it turns out that earnings come in at $125 a share or even less, we’d be looking at overvaluation on the order of about 10%.

The problem is that we won’t know the actual earnings for the current year until sometime in early 2018 and by then anything can happen.

Share this article