Prior to Brexit, I kept getting asked about how and when the stock market would break out to new highs. Now that the post-Brexit volatility has subsided, I’m getting the same question again. My answer is the same now as it was then—realistically, only one of two things will drive stocks to new highs: (1) earnings growth improves and we get an earnings-led rally, or (2) failing that, we have a multiple-expansion-driven rally, meaning an increase in the price-to-earnings (P/E) ratio, or multiple, of the stock market.
An earnings-driven rally would provide the healthiest and most sustainable kind of upturn. Stock prices tend to follow earnings, after all, at least over the long run. Unfortunately, earnings growth has been slowing for the past 18 months. But given the sustained improvement in financial conditions since February,1 there’s hope that improved earnings results will follow. If not, any rally will by definition have to be multiple-driven (i.e., a willingness to pay a higher P/E).
Discounted earnings model
While the market’s P/E is a useful snapshot of current valuation, I tend to look at valuation from the perspective of a discounted earnings model, which I consider to be a superior valuation measure. Equity investors are no different from business owners in this regard: They all need to make an assessment of what a future earnings or cash flow stream will look like, and then they need to discount that earnings stream by a discount rate, which is the sum of the risk-free rate (RFR)2 and the equity risk premium (ERP).3
Therefore, stock market valuation is a function of three moving parts: expected earnings growth, the 10-year Treasury yield (the RFR proxy), and the ERP. The latter two comprise the discount rate against which to calculate the present value of this expected future earnings stream. The projected earnings stream is like the numerator in an equation, and the discount rate is the denominator. This is what makes the valuation debate so tricky and interesting, in my opinion. It’s dynamic, not static—always changing and evolving.
The market is very efficient in terms of pricing in what is currently known. So when people debate whether the market is rich or cheap, it’s not so much a matter of looking at a current snapshot of price divided by earnings, but rather a debate over what kind of regime we are heading into. I guess you could say that valuation is in the eye of the beholder.
Therefore, making a blanket statement that says the market is in a bubble because the market’s P/E is at a certain percentile of its historical range is insufficient; one needs to make the argument that one or more of the three variables are going to change in the future. Based on this discounted valuation matrix, I can tell whether the market deserves a P/E ratio of 12x or 15x (or 30x, for that matter), depending on the assumptions about changes to the earnings (numerator) and the discount rate (denominator).
Market valuation is sensitive to both, because changes in the numerator (expected earnings growth) can be significant. However, in a low-growth environment, changes in the discount rate can really move the needle; in a high-growth environment, the discount rate is less important. This is why secular growth stocks sometimes trade at sky-high P/E ratios.
Is a breakout in the cards?
The question at hand is whether a breakout is in the cards, and if so, which of the three variables will be responsible. Let’s assess earnings first, as they are by far the most important. The year-over-year change in trailing operating earnings has fallen from its cycle peak of +7.4% in 2014 to about 0.0% currently (as of May). The best that can be said for recent earnings growth is that it has been flat-lining for the past few months. So, at least the downside momentum has waned.
Until recently, the same has true for forward earnings estimates. Here, the year-over-year growth of operating earnings is +0.2%, and this too has been flat-lining for several months. The weekly progression of earnings growth estimates for 2016 started the year at 6.8% and have been stuck at around +1.0% to +1.3% for several months now4 (see chart). Not including the weak-performing energy sector, earnings growth estimates for the S&P 500® Index are +4.0% for 2016, also with little movement in recent weeks.
Looking at revisions, we do have a very subtle improvement here. The earnings revision index for the S&P 500® Index (three-month upgrades less downgrades divided by the number of estimates) has improved from –15% to –7%. That’s still negative, of course, but inflection points are often signaled by changes.
So there are your “green shoots” in earnings. It’s not much, but it’s an improvement over the steady declines from last year. Things are getting less bad, or at least not getting any worse. This shouldn’t be a surprise, considering that crude oil is up 100% from its February double bottom, the broad dollar index is down some 6%, and financial conditions have greatly improved.
These three factors have driven most of the estimate progression since 2014, and they may even suggest that earnings estimates are too low for 2016, unless there’s something more broad-based going on that’s not explained by these variables. Assuming for now that there isn’t, a stabilization and eventual turn in the earnings cycle would almost certainly send the market higher. And with the S&P 500® Index still within spitting distance of its all-time high, it will take very little to propel things forward to new highs.
Think about it: If earnings growth reverts back to its trend growth rate of +5% per year, and we add +2% for dividends, then compounding a +7% compound annual growth rate (CAGR) off a 2,100-ish base for the S&P 500® Index could lead to meaningful new highs even this year. But it needs to happen first, because it’s also possible that earnings growth is in a holding pattern on its way to a further contraction, which is why the market is so hesitant to discount it. Historically, the earnings cycle tends to last five years from peak to peak or trough to trough, and we are only 18 months on the downside. So maybe it’s too early to be talking about an upward swing in the cycle.
If earnings aren’t going to do the heavy lifting, any breakout will have to be P/E-driven. As mentioned, the market’s P/E can expand only if expected earnings growth picks up or if the RFR and/or the ERP contract. Assuming in this scenario that earnings growth is not rebounding, then it will have to be the discount rate that comes down. With the 10-year Treasury already yielding 1.69% (as of June 21) and unlikely to fall much further unless earnings growth also contracts (which could force the P/E ratio lower), we’d have to see a meaningful decline in the ERP. In other words, a stock rally without actual earnings growth would have to be based on a possibly unjustified pickup in sentiment and a willingness to own stocks at lower returns.
A purely sentiment-driven rally could pack a punch, given that the ERP (at 3.8%) is not at all stretched by historical standards. But such a development seems unlikely without some fundamental catalyst like a pickup in nominal GDP growth (which should lead to better earnings growth). That doesn’t mean it can’t or won’t happen. But if it does, it would likely be a temporary “sugar rally” based on false hopes (because if it were justified, then earnings would improve as well).
As I ponder this riddle of what can drive stocks out of their bull market purgatory, I keep coming back to earnings. With a trailing P/E of 18x, an RFR of 1.63%, and an ERP of 3.8% (in line with its long-term average), I believe it will have to come from earnings. The good news is that there are a few very subtle green shoots emerging. Whether it’s the start of a cyclical upturn or just a temporary reprieve remains to be seen.
While the global repercussions of Brexit are still a wild card, I would encourage investors to keep their eye on the ball on their long-term investment objective and focus on what really drives stock prices over time. Historically, earnings growth has been at the top of that list.
Investing involves risk, including risk of loss.
Past performance is no guarantee of future results.
Fidelity Brokerage Services LLC, Member NYSE, SIPC, 900 Salem Street, Smithfield, RI 02917