A Wall Street sign is seen as a Christmas tree stands in front of the New York Stock Exchange (NYSE) on Wednesday, Dec. 21.Michael Nagle/Bloomberg
It's time again for Wall Street strategists to answer investors' favourite question: What's in store for the S&P 500 next year? (Never mind that U.S. large-cap stocks represent only a portion of most investors' portfolios.)
Good news, investors. As Bloomberg's Lu Wang noted this week, every strategist surveyed by Bloomberg is predicting that the S&P 500 will be higher next year. The average forecast is 2,356, which would be a 4.2- per-cent increase over the index's current price.
But before investors start counting their gains, it's worth asking whether Wall Street strategists are any good at forecasting the S&P 500's performance.
To find out, I looked at strategists' average year-end forecasts for the S&P 500 since 1999 - the earliest year for which numbers are available - and compared them with the index's actual year-end price.
At first glance, the strategists appear to be prescient. There's a high correlation of 0.76 between the average forecast and the year-end price. And the forecasts have been just 5.8 per cent higher on average than the year-end price. Not bad for a crystal ball.
Knowing that, investors might be tempted to discount the average 2017 forecast of 2,356 by 6 per cent and - simple math - come up with a year-end price of 2,215. Unfortunately, it's not that simple.
It turns out that while strategists' forecasts are remarkably close to the mark much of the time, they are woefully unreliable during inflection points. For example, strategists overestimated the S&P 500's year-end price by 26.2 per cent on average during the dot-com bust from 2000 to 2002. Then they underestimated the index's year-end price by 10.6 per cent during the early recovery in the first half of 2003.
It happened again during the financial crisis. Strategists overestimated the S&P 500's year-end price by 64.3 per cent in 2008 and then underestimated it by 10.9 per cent during the first half of 2009. In other words, the forecasts were least useful when they mattered most.
That's not an accident. There are two generally accepted market forecasting tools in polite finance society. The first is momentum, which commonly looks at the market's current price relative to a multi-day trailing average. A ratio greater than one implies that the market is headed higher, and vice versa. The further that ratio drifts from one, the stronger the ratio's predictive power, in theory. For example, divide the S&P 500's closing price of 2,261 as of Thursday by its 200-day moving average of 2,131. The resulting ratio of 1.06 would be a modestly bullish signal.
The second tool is valuation, which looks at how the market is priced relative to its long-term average, or compared with a strategist's estimate of fair value. A widely used valuation tool, for example, is Yale professor Robert Shiller's cyclically adjusted price-to-earnings, or CAPE, ratio. The current CAPE ratio is 28.3, which is well above its historical average of 16.7 since 1881. A high CAPE implies that the market is too rich, and vice versa.
The two methods aren't interchangeable, however. Valuation is useful only for medium-term forecasts, say five to 10 years. Momentum is useful only for near-term forecasts. So if you want to take a stab at where the market will be in a year, momentum is the go-to instrument.
The problem with momentum is that it isn't built for predicting turns in the market. By its nature, momentum extrapolates the recent past into the future. So it's not likely to warn that a correction is coming or when that correction is likely to turn the corner into recovery. That explains why strategists' forecasts failed them during the dot-com bust and the financial crisis and in the early months of the subsequent recoveries.
The stage may be set for strategists to miss yet again. The S&P 500 has enjoyed a multi-year bull run, and U.S. stocks look expensive by historical standards. That sounds a lot like the environments that preceded the previous two bear markets.
So as investors toast Wall Street forecasts for next year, they should keep in mind one thing: Those forecasts may be fairly reliable, unless the market has other plans.
--
Nir Kaissar is a Bloomberg Gadfly columnist covering the markets. He is the founder of Unison Advisors, an asset management firm. He has worked as a lawyer at Sullivan & Cromwell and a consultant at Ernst & Young.