Andrew Sheets: What Will Markets Return in the Long Run?

One of the great conundrums of finance is predicting what markets will return over the long run. But with some historical research and the power of math, the future can become a bit clearer. ----- Transcript ----- Welcome to Thoughts on the Market. I'm Andrew Sheets, chief cross strategist for Morgan Stanley. Along with my colleagues bringing you a variety of perspectives, I'll be talking about trends across the global investment landscape and how we put those ideas together. It's Thursday, October 28th at 2 p.m. in London. The question of what markets will return over the next decade is a conundrum. It's complicated because of just how much can change in a given year, let alone a decade, but also simple because over longer horizons, valuation measures such as bond yields or stock price-to-earnings ratios tend to matter a lot more for how well a market does. A 10-year horizon really matters to investors saving for the future. But most investors, and also this podcast tend to focus on events happening in the much more immediate future. So what do we think this return picture holds? When estimating what a market will return over the long run, there are really two basic approaches. The first, sometimes called the demand approach, assumes that markets are efficient, and that investors will always demand that the market is priced to deliver an average historical return. In this approach, future returns for the market are simply assumed to be the long run average. We don't use this approach, but others do, and it is appealing for being relatively straightforward. An alternative, which we favor, could be called the supply approach. This attempts to quantify just how much return a given asset can supply. So, for bonds with a fixed yield, this approach is attractively simple. On a 10-year horizon, the return for a broad bond index should be pretty similar to its yield today, regardless of the path that interest rates take between now and then. That might sound somewhat counterintuitive, but there's some pretty good math, we think, to back it up. After making a few minor adjustments, we think the U.S. Aggregate Bond Index may be able to supply a return of about 2% per year, over the next decade. For stocks... now there are more moving parts and more assumptions that can ultimately be proven right or wrong. The long run return of a stock market can be broken down into three parts: the dividends of the stock market pace, the growth in the market's earnings, and the change in the valuation that's applied to those earnings. The dividend yield is relatively easy to estimate, but earnings and valuations create a lot more debate. For earnings, our starting point is to assume that they grow, at least with the rate of inflation. We see a good argument for this, if prices everywhere are rising, companies should book higher sales and profits. This is one reason why equities tend to be a better asset class in higher inflation because they can grow their cash flows much more easily than, say, a bond can. So how much do earnings grow over and above the rate of inflation? We average two trend lines: a very long run trend of historical earnings growth and one that only focuses on more recent history. There are pros and cons of each. For example, only using the recent historical trend may better reflect current conditions in the market, but it also might overstate what's been an unusually favorable environment for companies. By taking the average, we split the difference. Now, with stock market earnings are above trend. We assume that there is some convergence down. And if earnings are depressed, we assume some normalization up. We think there's some good historical arguments for this, as earnings do tend to oscillate around these trend lines over time. Finally, what about those valuations? Well, we assume that valuations move back to long run averages, but do so only gradually, as we believe history says this gravitational pull takes time. Putting all of this together, we think the U.S. stock market could return about 5.2% per year over the next decade. The bad news is that's roughly half the long run average. The good news? It's still two and a half times higher than the return from that broad bond index. So where can investors find higher returns, especially relative to inflation? For equities, our framework suggests the highest so-called real returns are in Europe, where we think stocks could beat inflation by about six percent per year. In fixed income markets, we see the highest inflation adjusted returns in emerging market bonds. Finally, where could our assumptions be wrong? The return for bonds should be pretty well anchored by their yields, but for stock markets, there are several swing factors. Higher corporate taxes, for example, or higher interest rates could mean we're too optimistic about our assumptions for earnings growth and valuations. On the other hand, a stronger economy and importantly, a more permanent shift higher in market profitability could mean that our assumptions for mean reversion back to historical averages are simply too pessimistic. Either way, time will tell. Thanks for listening. Subscribe to Thoughts on the Market on Apple Podcasts or wherever you listen and leave us a review. We'd love to hear from you.