Forecasting The S&P 500's Returns

| About: SPDR S&P (SPY)

I come from the hedge fund industry and one of the biggest myths in that world is the one about "absolute return strategies." Have you heard it? The one about the hedge fund manager whose strategy makes money whether the market goes up or down?

Easier said than done.

The reality of the hedge fund industry is that most of these guys generally correlate with the stock market. Not everybody, of course. But a disturbing majority sees their fortunes rise with a bull market and fall with the bear. Alpha is a tricky creature to trap. And after spending a decade in that wilderness, I am convinced that there are very few managers actually capable of doing it, at least over windows long enough to matter.

So even if you're a hedge fund investor happily paying 2/20 to your supposedly-awesome manager who's got a lot more beta hidden in his portfolio than you and possibly even he is aware of, the question of "how's the market going to do?" remains incredibly relevant.

And for the average investor at home, this is perhaps the most important question.

What sorts of returns should we expect from the market over the next decade or so?

Only by properly aligning our expectations will we be able to avoid the suffering that's inevitable to follow -- whether that's the suffering of losing money in a correction or missing out during a rally.

Suffering is what happens when reality fails to conform to our expectations, you know. The Buddha said so!

Getting Our Expectations In Line

Absolute return strategies or no, ultimately, all portfolio management decisions are a relative value proposition. Investors need to select a discrete lineup of positions, and sadly, the way that most of them make decisions is by picking whatever one has performed the best lately. A large number of hedge fund managers, I regret to inform you, make decisions the exact same way.

Performance chasing is a disturbingly emotional endeavor, at least as I see it. The good news is that there are ways to remove a lot of that destructive emotion and better ways to make decisions than to simply chase whatever sector was hottest last year.

To start, we must first decompose the stock market into the most basic elements of its return. Over the long run, this is equal to GDP + Inflation + Dividends. This true by spiritual definition and also by exhibition.

It works for shorter windows of time as well, but we have to make an adjustment for psychology. Depending on the risk appetite of the moment, investors are either willing to pay a lot for one dollar of corporate earnings or they're willing to pay very little. Over the long run, this psychology tends to oscillate around a historical mean. When psychology is grim and market valuations are cheap, people who invest typically reap larger long-term rewards. When psychology is hyper-bullish, investors entering the market anew tend to struggle over the long-run.

Knowing this, we can feed our model three basic variables (inflation will be baked in automatically):

  • Macro-Economic Growth
  • Dividends
  • Current market valuation

All of those data points are easy to obtain and are fortunately quite reliable. We'll start by using ~6% as our nominal growth rate and the S&P's dividend yield. For the valuation component, we'll use a cyclically-adjusted price/earnings ratio, in this case one that normalizes 10 years of earnings.

Then we simply map that model all the way back to the beginning of the 20th century and relate its predictions to what the actual returns wound up being.

Here's what it looks like:

Pretty cool, huh?

I've seen John Hussman use a similar model to project stock market returns. I'm not exactly sure how he does it and which precise inputs he uses. But mine is crafted from the same basic idea. My gut tells me that my way is probably easier and less accurate while his calculation is probably more complex and works more effectively as a real model.

Two things are immediately apparent with our model, even to the layman. It does an exceptionally good job at approximating future returns but it cannot project them with reliable precision.

This is OK. Even if we can't derive an exact projection of how the market will perform, a model like this matters a great deal because it helps us frame our expectations and make more prudent decisions in the present.

Currently, our model projects future S&P 500 returns to accrue at a rate of roughly 3.35% per year. Those will not come peacefully, however. Volatility for the S&P has historically averaged 16% per year. Our model says that, during the next decade, the market will have some 20% years, some -15% years, and a bunch in between, and when the dust has finally settled, investors will have earned something close to 3.35% per year.

One could argue that this model is a bit optimistic. Will the U.S. economy really grow at a 6.3% nominal rate in the decades to come? Also, was that a fair number to use in the first few decades of the century when growth was generally lower and lumpier than the growth explosion in the second half of the 20th century? And what about the post-crisis years? Wouldn't you rather use something like 5% (2.5% growth 2.5% inflation) for the next decade or so?

Since our model is fairly flexible, we can make some broad adjustments to make it a little more accurate. We can separate the data into pre-WWII, post-WWII, and "post-crisis" periods.

As you can see, this one fits a bit better:

This revised model features a correlation of 0.77 and an R-squared of 0.60, which is statistically significant given the massive size of our data set.

Under this revised model, the stock market is projected to grow at an annual rate of 2.12% -- 2.10% of which, it's worth mentioning, is expected to come from dividends. In real terms, ex-dividends, our model is projecting the stock market to go essentially nowhere for the next decade.

Don't worry, I'm completely aware that this sort of prediction will get me laughed out of the room today. But keep in mind this model was projecting annualized returns of -1.74% at the end of 1998 (they wound up being -1.75%) and 20.55% in February 1982 (they wound up being 19.55%). I have no doubt you all would have laughed at those predictions at those moments in time as well. Such is life as a rational investor in the world of fierce emotion and misleading narratives.

What To Do About It

If this truly is the long-term reality with which investors are faced, how should we respond?

After the crash of 2008 we all were reminded of the lesson that it's not such a great idea to be simply long a passive S&P mutual fund or the (NYSEARCA:SPY). But in the years since, many of us have forgotten that lesson and we're building new myths about how it maybe isn't such a bad idea after all to just always be long the market.

The first thing that investors should be doing to work around these types of market expectations is employing some kind of basic asset allocation strategy. Whether you're using a tactical approach, rebalancing into certain places and out of others depending on the relative expectations or employing a "perpetual" asset allocation approach, any strategy here is better than none. Add some bonds, real estate, and commodity strategies and you'll be better prepared to deal with whatever the future holds.

The next thing is to make sure your psychology is prepared. A few months ago I wrote about the "10 Qualities for Successful Investing" and those are even more important if the future of the market will be more difficult to handle than it was in the past. When the market was going up and up and up during the long-boom, investors didn't really need to worry about traits like Patience, Objectivity, or Fortitude.

The last easy thing we can do to deal with this type of market projection is take a closer look at where our returns are coming from. Depending on what you use as your future economic growth input, most or all of the market's future returns will come from dividends. This is not like 1984 where future returns are projected at 15% per year, only 4% of which will come from dividends.

In a 2-4% total return world, it is suddenly all about the 3% dividends. Perhaps this is why "high quality dividend payers" is one of the most popular strategies around right now. Seeking Alpha now has an entire section devoted to it, and based on some of the limited statistics I've seen, it's one of the most popular sections. A lot of investors in this community understand that this is one of the strategies that's going to get them through the next decade in one piece.

As I see it, there is no more reliable, easy-to-execute way to beat the market over the next decade than to overweight high-quality dividend stocks. I run a model portfolio of this exact strategy in my Alpine Advisor newsletter. You can get that one for free if you want to see all the exact positions. Since its inception in 2010 that portfolio has outperformed the market by a few percent and with less volatility as well. Past performance is never indicative of future results, and its live track record so far does not include any major market corrections. I'm curious to see what will happen. But with names like Intel (NASDAQ:INTC), ConEd (NYSE:ED), and McDonald's (NYSE:MCD), I'm sure it'll be OK relative to the rest of the market.

What Lies Ahead

Maybe the market returns the 2% projected by this model or maybe it won't. As I said, this model is far from perfect and at various times in the past has over- or under-estimated eventual market returns.

The point is that with a 2% projected return, it's probably unreasonable to think the market averages 10 or 15% over the next decade. It's probably even unreasonable to think the market averages 7-8% over the next decade, unless we get another "productivity miracle" to carry us to a higher economic plateau. For what it's worth, I'm more comfortable betting on that outcome than its opposite (a total deflationary breakdown) and so I'd rather have a portfolio tilted towards being long equities with sufficient caution than avoiding equities altogether.

After all, we've gotta go somewhere for return, don't we?

Disclosure: I am long INTC, ED, MCD. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Additional disclosure: For additional disclosure, see:

About this article:

Author payment: $35 + $0.01/page view. Authors of PRO articles receive a minimum guaranteed payment of $150-500. Become a contributor »
Problem with this article? Please tell us. Disagree with this article? .