Thu, 22 Aug 2013
It's unlikely we'll see the 6.5% long-run real returns of the past decades, but 4.0%-5.5% returns going forward shouldn't disappoint investors, says StockInvestor editor Matt Coffina.
Jeremy Glaser: For Morningstar, I'm Jeremy Glaser. I'm joined today by Matt Coffina; he is the editor of Morningstar StockInvestor. He has recently completed some research into what's driven stock returns over time. We're going to talk to him about what some of those factors are, and also his expectations for returns going forward.
Matt, thanks for joining me today.
Matthew Coffina: Thanks for having me, Jeremy.
Glaser: Benjamin Graham famously said in the short term the market is a voting machine, but in the long term it's a weighing machine. When you look across the last 100 some years of stock market performance, do you find that to be the case?
Coffina: I think very much so that that's the case. If we look at the drivers of stock market performance over the very long run, there is a very clear connection between dividend returns, dividend yields, increases in real earnings, and then changes in the price/earnings multiple being the big swing factors in the short run. But over the very long run, it's really those first two factors, dividend yields and real earnings growth, that drive real afterinflation returns.
Glaser: Eventually the stock market really does follow those fundamentals. Looking across those factors, which ones did you find were the most important, and have those factors changed over time?
Coffina: What's interesting is that dividends used to be a much larger contributor to the markets real afterinflation returns than they are today. Payout ratios and dividend yields have been falling steadily over the last 50 years or so, such that now the yield on the S&P 500 index is only about 2%, whereas it used to routinely be in the midsingle digits.
You could ask what does this mean for returns going forward. What's interesting is you would think from finance theory that the lower dividend payout ratios go, the higher real earnings growth would be. And that hasn't really been the case. Real earnings growth has picked up some of the slack for lower dividend yields, but not all of the slack that's left there, which to me indicates that companies on average are not necessarily making the best investments with the capital that they're retaining. Maybe they're repurchasing stock at inopportune times, they're overpaying for acquisitions, they're making bad internal reinvestment decisions, and so on.
Glaser: If there is shift of emphasis away from dividends and toward companies keeping more of their earnings, does that mean you expect returns to be lower over the next 20 years than they were over, say, the last 20 years?
Coffina: Over the last, let's say, 100 years, the real returns--again, we're talking in afterinflation terms on the S&P 500 or predecessor indexes--has been about 6.5% a year. However, if you strip out the P/E changes over the last 30 years or so, even during the bull market of, call it, 1980 to 1999, the market only had a sustainable return--which again I would define as dividend yields plus real earnings growth--of about 5.5%, so below that 6.5% long-run average.
Over the past decade or so since 2000, we've seen that fall further to about 4% dividend returns plus real earnings growth. To me this indicates that the 6.5% real return that investors had become accustomed to over the very long run is unlikely to be repeated going forward. But you could still expect a very reasonable 4.0% to 5.5% real return on equities, which really is nothing to be depressed about. With that growth rate you would expect the purchasing power of an investment in the S&P 500 index fund to increase or double in about 13 to 18 years, which is at least much better than you're going to do from long-term bonds, given the current interest-rate environment.
Glaser: How do you think about the price/earnings ratio changes then? How important is it when thinking about future return to look at where the current P/E ratios are?
Coffina: Again, the price/earnings ratio has really been the key swing factor between good decades and bad. This could literally take more than a decade where we've started sometimes with P/E multiple in the high teens, and then at the end of the decade it ends in the midsingle digits. Returns during that period could be very poor, even if you have real earnings growth and dividend yields.
On the other hand, the great bull markets have happened when you started with a very low P/E multiple, say, in the midsingle digits, and then ended at a very high level. For example, the 1980-99 bull market started with a P/E somewhere in the midsingle digits, and by the end of it the P/E was somewhere around 30 almost. Without those huge swings in P/E multiples, you're very unlikely to see another really strong bull market like we saw at the end of the 20th century.
Glaser: Where do we stand today with the P/E ratio, and what does it mean for the next decade?
Coffina: It's very hard to say what's a fair P/E for the market, not least of all because we never really know if earnings are at a normalized sustainable level. The P/E multiple during the last bear market actually peaked in 2009 at the market lows, not in late 2007 at the market highs. The reason being that earnings were very much depressed in 2009 and somewhat in 2007. Now, so what we can say about the P/E multiple is that it does have a very strong tendency toward mean reversion. If you go all the way back to 1871, the average P/E multiple has been 15.5, and 14.5 if you look at just the median. It's been higher than that in the last 50 years, more in the high teens, and that's about where we are now in the high teens on the S&P 500, which is consistent with what we've seen over the last 50 years.
What does that mean for the next decade? I think, for one thing it means that we're probably not set up for very strong bull market, or if we were to see a very strong bull market, it would mean that the market would probably be very overvalued 10 years from now. Alternatively, we could see a weak market, in which case the market might end up very undervalued 10 years from now and we'd be primed for a very strong bull market after that. I would say, looking at Morningstar analysts' fair value estimates that the market is more or less fairly valued, which is what we've been saying for some time now, in which case, I would expect the market to generate again that 4.0%-5.5% real afterinflation return over the very long run. But again, for any given decade, we can't really say one way or another because it depends on what's going to happen to that P/E multiple.
Glaser: How about some of the alternative measures of P/E, things like the Shiller P/E, that try to come up with that normalized earnings number? Some people have said that looks somewhat elevated. What are your thoughts about that?
Coffina: There's all sorts of attempts to try to normalize earnings. Probably one of the most prominent would be the Shiller 10-year average earnings [also known as the P/E 10], using 10-year average earnings in the denominator, what's called cyclically adjusted price/earnings. That also does look relatively inflated by historical standards. The challenge here, I think, is that the Shiller P/E is very much backward-looking. So you're going to include the depressed earnings of 2008 and 2009 for another five years in that measure. Unless you expect another financial crisis or another great recession in the next 10 years, say, those numbers might not be as relevant as current earnings levels. I think current earnings are probably neither inflated, nor depressed. I would say that looking at that multiple on current earnings is probably reasonable. Again, the S&P 500's P/E in the high teens is basically in line with where it's been over the last 50 years, indicating a market that's more or less fairly valued.
Glaser: Do you think at today's levels, it would still make sense to invest in stocks versus moving into cash or looking at fixed-income products?
Coffina: That's the challenge for investors, is that there are really no better alternatives out there. I still very much like stocks over the long run as probably the best vehicles to create long-run wealth. When you have long-run interest rates, even after the recent runup, still below 3% on the 10-year Treasury, that's pretty much the most you're going to get in terms of returns. If we get inflation of 2% or 3%, especially after taxes, you're very unlikely to see a significant real growth in purchasing power with fixed income. So I still think stocks are the best place to be. The market's certainly not as attractive as it was in 2008 or 2009, but we have to play the hand that we're dealt by the market, and I still think stocks are a great place to be over the long run.
Glaser: Matt, thanks for joining me today.
Coffina: Thanks for having me, Jeremy.
Glaser: For Morningstar, I'm Jeremy Glaser.