Authored by Nicholas Colas via DataTrekResearch.com,
We don’t look at technicals much in these notes, but we do admire what good technicians offer: a clear sense of where to buy and sell. Their assessments are generally quite precise: buy a stock when it gets to $50.33, but sell if it breaks $47.10. There is simply no direct analog in the squishy world of fundamental analysis.
Today we will, however, take a page from the chartists’ pinpoint process to answer one question: where do you back up the truck on US large cap stocks? There’s no sense in denying the obvious – US equity markets feel shaky. Some of this may just be quarter-end positioning, but we’re also about to walk into the most anticipated/hyped earnings season since the Financial Crisis. So it makes sense to plan, if only to understand the downside through the lens of preparation rather than the rear view mirror of regret.
Here are three scenarios to consider:
Scenario #1: a short (1-5 day) serious drop in US stock prices akin to 1987’s Black Monday. That may sound scary, but it shouldn’t:
The S&P 500 posted a total return of 5.8% in 1987 and ended the year 9.8% above the close on October 19th.
If you swore off US stocks for the following 5 years, you missed the next double (yes, double) in the S&P 500.
So what was the valuation on the S&P 500 on the close of Black Monday, and what would an equivalent number look like today? Here’s the math:
At the close on October 19th 1987 the S&P 500 traded for 9.3x the following year’s earnings (224 index close, $24.12 in 1988 earnings).
Keep in mind that 10-year Treasuries yielded 8.9% in October 1987, so PEs plus Treasuries equaled 18.2. That’s close enough to the old trader’s “Rule of 20” for this math to make sense, in other words.
Using the same calculation on today’s S&P 500 yields a target “disaster” PE of 15.4x forward earnings (18.2 minus the current 10 Year yield of 2.8%).
The current consensus earnings estimate for the S&P 500 is $158/share. We will be conservative and haircut that by 10%, down to $142/share.
The conclusion: Put the numbers together and you get a “1987-style low” of 2187 on the S&P 500, or 16.0% below current levels and roughly 25% below the all time highs from earlier this year. Not great, but hardly awful either.
Scenario #2: we can do a similar exercise with long-term valuation measures like the Shiller Cyclically Adjusted Price Earnings ratio, or CAPE.This approach has gotten a lot of press lately thanks to a recent Research Affiliates paper. The basics are simple enough: the S&P should trade for some multiple of average earnings over the prior 10 years, essentially a measure of corporate earnings power through a cycle.
The most likely scenario where the CAPE multiple becomes relevant is a slow motion train wreck for US stocks – something akin to a “Sloth market” (i.e. a slow moving furry intruder rather than a speedy bear) that takes 12-18 months to play out. We don’t even need higher interest rates to see that happen – just a creeping move out of equities from a mindset change among investors. Not likely, but we’re deep in the 100 acre “what if” woods anyway….
Here’s the CAPE math and S&P downside target:
The mean Shiller PE going back to the 1880s is 16.8x earnings.
The current multiple is 31.4x, which includes the Financial Crisis when S&P earnings went to zero in Q4 2008 and therefore troughed at $40/share on an annualized basis.
Average S&P 500 annual earnings from 2009 to 2017 are $98/share on an operating basis. That includes almost all the aftermath of the Financial Crisis and the Great Recession that followed. It is, therefore, a reasonable estimate of long-term earnings power.
The upshot here (and its not pretty): an S&P 500 at 1646 ($98/share times 16.8x). This amounts to a decline of 36.8% from today’s close. If it is any solace, the last time the index was at these levels was September 2013. We’d be giving up several years of gains, but at least the S&P 500 would still be above the old 1500 double top of 2000 and 2007.
Scenario #3: in between the 1987 case study (a bolt out of the blue) and the CAPE revaluation (death by 1000 cuts), we have a catalyst-driven price reset: the need to discount a near term recession, caused by reason or reasons as yet unknown. This one feels the most logical right now, if only because the US Treasury yield curve continues to flatten. This may spook investors into thinking an economic slowdown is at hand, even if one never appears. The old saw that “stock markets have predicted 9 of the last 5 recessions” is valid – and those 4 other times are serious buying opportunities.
The math for discounting a recession in the next 12-18 months into the S&P 500:
Garden-variety recessions hit corporate earnings by an average of 20-30% from prior peak to trough. We’re leaving out the 2008 experience from that math.
Current estimates call for S&P 500 earnings of $158/share. Worth noting: without tax reform, those would be more like $146/share.
If recession fears mount, markets will start to discount earnings of $111 - $126/share, or 20-30% current recession-free expectations based on earnings performance in prior cycles.
Assuming the equity market’s current 16.0x forward multiple remain the same (rather than expand due to lower interest rates), this implies an S&P 500 fair value of 1776 to 2106 with a mid point of 1,941. That is 25% below today’s close. If you want to give some credit for higher recessionary earnings multiples, perhaps 17x is fair. In that case, the S&P 500 would trade for 2023 (23% below today’s close).
We should emphasize that none of these scenarios are anything close to our base case. The purpose of this exercise is to consider what levels on the S&P 500 correspond to historically valid and profitable entry points. To sum up our levels:
Market crash scenario: buy the S&P at 2187, down 16% from today.
Protracted bear/revaluation market: buy the S&P at 1646, down 37% from today.
Recession scare: buy the S&P anywhere from 1776 to 2023, down 22-32% from today.
No comments:
Post a Comment