by Jeffrey Miller, Partner, Eight Bridges Capital Management
September 5th, 2017
“I mistrust all frank and simple people, especially when their stories hold together”
? Ernest Hemingway, The Sun Also Rises
In case you’ve been out, quite a few market experts are warning about the inflated valuations in financial markets lately. Everyone from Jeffrey Gundlach to Ray Dalio to Paul Tudor Jones to Mark Yusko is saying stocks are due for a pullback. Even Warren Buffett, who seemingly hasn’t said a negative word about the stock market in decades, on CNBC this week gave a nuanced response to the question about whether or not stocks are expensive. He said he didn’t think stocks were expensive relative to bonds. But he added (I’m paraphrasing, since I heard it on TV) “Well, you see, the 10 year is yielding something like 2.15%, so its trading at 45 times earnings for no growth, so stocks with some growth trading at less than that are relatively ok.” He went on to add that if rates went up, then this would change his answer on stocks.
Bulls on bonds say that with weak growth and low inflation, bonds yields should be low. Bulls on stocks will point to the low bond yields and say that they justify higher than normal P/E ratios. These are very simple, seemingly logical stories to tell about markets, and they make for simple sound bites on TV. Again, Warren Buffett himself just did it. Stated differently, stock market bulls are playing the relative value game. But this is a very dangerous game to play. It’s effectively musical chairs with your money. Everyone is hoping they can get a seat when the music stops, and we all know that there aren’t enough chairs for everyone in this game. I’d mistrust these frank and simple stories, because the reality is a lot more complex.
Restated, justifying high P/E ratios with low interest rates doesn’t make sense, unless you’re also willing to concede that future stock market returns are likely to be significantly lower than they have been in the past, and lower than the average market participant is expecting to earn. It’s just math. You can’t turn a 4-5% current earnings yield coupled with a 1.9% dividend yield into anything approaching a 10-12% total return without some serious multiple expansion from here. And while moving from a 10 p/e to an 11 p/e gives you a nice 10% return bump, moving from a 20 p/e to a 21 p/e only gets you 5%. Still like the simple story that low rates justify higher P/Es? Then check out this chart from GMO – there is no correlation between actual 10 year real interest rates and the Shiller P/E:
“You're not a moron. You're only a case of arrested development.”
? Ernest Hemingway, The Sun Also Rises
Conservative investors who think about valuations and prospective future returns (I consider myself to fall into this category) are having a very hard time in this market. We look like morons. Usually in markets like this, value investors start to look like time has passed them by. Their investing acumen is questioned, and they lose assets under management. It’s at these times that I like to start over, asking basic questions about risk, return, growth rates, and valuation. I pull out my handy chart of implied growth rates and discount rates for various P/E ratios, trying to triangulate what the market is discounting at the current prices. I’ve been wracking my brain trying to understand what I am missing about this market, trying to place myself in the shoes of the marginal buyer of stocks today, at today’s prices.
This started with an examination of the banking sector, my specialty. I was trying to come up with a model that justified the sector’s current valuations. I’ve done this exercise periodically over the past 23 years, usually when I’ve been wrong on the market for longer than I, and my investors, feel comfortable with, and I’m trying to come up with a reason to “get involved.” And usually, this is right before being patient and cautious and trying to be prudent turns out to have been the right call. Will it once again this time? Beats me. Check back in a year. But I spent some late nights this past week deep in spreadsheets and databases trying to construct a bull case. I kinda failed. I do think there are a few sectors and companies that are fairly cheap right now. The issue I’m having is that if the overall market tanks, it won’t matter – everything will go down. In a correction, correlations have a habit of going to one fast.
Here are the facts: The S&P 500 is trading at about 22 times trailing earnings, which isn’t so good. That’s a 4.5% earnings yield. But wait, I’m using last year’s earnings you say? Ok, let’s look at forward estimates instead. On forward earnings, the market is cheaper at about 17 times, but forward earnings have a bad habit of never actually happening. That’s just reality. As a whole, we analysts are an optimistic bunch (or, more likely, the sell-side doesn’t like making management teams mad by posting estimates that are materially lower than company guidance – that’s a good way to lose access to management and your ability to get paid for it). So consensus estimates always start off too high then come down during the year. A more accurate method would probably be to just take last year’s number and add a few percent to it. But again, that’s not what happens. Check out the chart at the top of the next page. Will this year be different?
So what’s the “right” Price/Earnings ratio for stocks? After a lot of time spent digging into spreadsheets, I came to the same conclusion I always do: I don’t know. There are too many moving parts to be certain. Peter Lynch in his classic book One Up on Wall Street always used “the earnings line” in the Value Line tables as his “right” number. Well, if you look at the charts he put in his books, the earnings line is a P/E of…15x earnings. If it was good enough for Peter Lynch, it’s good enough for me. We can try to justify a higher number by doing all sorts of math, but frankly, I spent about 20 hours this week trying to reprove everything I learned in business school and in the years since, and then results are so susceptible to small changes in any of your inputs that I’ve come back to where I started. A 15 P/E for a non-cyclical company gives you a 6.7% earnings yield, so with 3% growth you can get close to a 10% “cash-on-cash” return. I’d argue that banks and cyclicals should trade for less, because in recessions they have a nasty habit of losing money, so you need to be compensated for the “lost years” of earnings, and, in a bank, potential dilutive capital raise if things get dire enough. It turns out, banks are a good way to examine the market, because their balance sheets tie well to their income statements.