.png)
I have immense respect for Bob Shiller. We’ve never met, but he is a forefather of behavioural finance along with Richard Thaler (and he is from Detroit, which isn’t uncool). Moreover, the CAPE is actually a decent framework (more or less advocated by Ben Graham back in the day), but there is an argument that its relevance not only time-varies, but that it does so due to omitted variables (like opportunity costs) or because of it being occasionally tainted by poor (and rare) post-recession data points.
Additionally, I am much more comfortable thinking that stock prices, and the indices which consist of them, are trading at value based on fundamental expectations about tomorrow, rather than what happened yesterday – or ten years ago. [1]
And despite the fact that many – perhaps rightly – are arguing that interest rates are artificially low, they are what they are; and it is extremely illuminating to actually take a step back and see some actual figures. Today, the yield on US 10Y government debt is 2.47%. That’s an effective bond P/E of 36x. In the autumn of 1981, the yield on the same debt reached 15.85%. That’s a P/E of 6.3x. Can we really ignore interest rates, and use the P/E of equities in 1981 as a relevant historical benchmark on the P/E for equities in 2018? [2]
I don’t think we can. Even when adjusting for the “real vs nominal” issue outlined in footnote 2 below, the opportunity cost of investing in equities and the market multiple is inexorably linked.
In the above example, on September 30, 1981, the Shiller CAPE was 7.6x, a 20% premium to bond yields. Today, the Shiller CAPE is at 33.0x, a 8% discount to bond yields.
As Warren Buffett and a few others have suggested, the market isn’t necessarily expensive; and I am sure there are others who could use this data to proclaim it may actually be a screaming buy. We truly do not have a view here (about the market itself), but are just highlighting that at least some of the Shiller CAPE doomsayers may be comparing apples and oranges.[3],[4] Specifically, today’s “CAPE” includes the historical look back into 2008 and 2009, which was a multi-deviation negative event, literally.
“Real Earnings” in Shiller’s dataset had averaged 66 over the previous 120 months (from 1998 through 2007) preceding the Global Financial Crisis and fell to less than 8 by March of 2009. [5] As we march forward in time [6], those GFC datapoints are going to fall off, and the Shiller CAPE is, very simply, going to mathematically drop (unless the world ends again).
Even at the current level of real earnings, assuming no improvement as we march forward to December of 2019, the Shiller CAPE will see multiple compression of 15%, automatically. In other words, the market is going to start looking a lot cheaper to the Shiller CAPE shorts, but they probably don’t realize it.
FOOTNOTES
[1] But smarter people than me disagree, like Cliff Asness, who suggests that P/E’s don’t do a great job predicting earnings growth, which he suggests counters the argument that tomorrow matters more than yesterday.
[2] And yes I know we are looking at nominal yields and not real yields, so we’re committing a bit of apples and oranges sin ourselves, but our allusion still holds if we make the correction. The World Bank estimates a “United States Real Interest Rate” adjusting by inflation as measured by the GDP deflator. It currently stands at 2.21%, so implies a “bond P/E” of 45.3x. In 1981, it was 8.72%, so a P/E of 11.5x. So, in 1981, the Shiller P/E traded four multiple points below the real yield on bonds, and today it trades a full 12 points below. Still cheap. Moreover, in both of these periods (amazingly) the Shiller P/E traded at a ~30% discount to the bond P/E. So tell me interest rates don’t matter…
[3] The smarter “CAPE doomsayers” will point out that there are duration arguments across the spectrum of securities that comprise an index, and that there are scenarios where individual stocks (and perhaps the market if dominated by these stocks) may be “cheap” vs. interest rates, but actually expensive on their own.
[4] The other thing worth thinking about is that the FAAMGs currently represent a significant weight in the S&P 500. Not only does this distort things a little (maybe), but basically, these FAAMG names, especially Google, Amazon, and Facebook have introduced game-changing business models unseen before by Mr. Market. And they got big quick (I mean, they are all bigger than Exxon now, which has been around since John D Rockefeller started Standard Oil of New Jersey in 1870), basically in the blink of an eye. So, when someone so new and special grows so rapidly, looking back on a ten-year horizon really overstates how expensive the market is on a CAPE (especially when said offenders represent a mid-teen weight in the index).
You think Facebook is expensive today with trailing adjusted earnings of $5.89? Well, nine years ago they printed EPS of $0.10, so they are on a historical 2009 P/E of 1,850x! The S&P doesn’t go back and recreate ten years of history for any new entrants, they just delete someone and add someone new – so when the likes of Facebook are introduced to the index (as a reminder, it just happened in 2012), they come in with whopping market caps against some deleted loser that probably was by definition on its earnings’ knees, also inflating the Shiller CAPE. Again, apples-to-oranges, or should I say Apples-to-Exxons.
[5] The standard deviation of trailing twelve month averages of monthly observations over the previous ten years to March of 2008 was 18.3. Over the next twelve months, the average dropped by 48.4, and the actual monthly figure dropped by over 60 points (from real earnings in March of 2008 of 69.26 to 7.90 in March of 2009).
[6] I have no idea why “ten years” stuck as the Shiller CAPE lookback period, and why it wasn’t five years or 15 years (or 9 or 11), but it stuck nonetheless. My concern is similar to my argument about the silly 200 day moving average adored by noise traders.
DISCLAIMER
The views and opinions expressed in this post are those of the post’s author and do not necessarily reflect the views of Albert Bridge Capital, or its affiliates. This post has been provided solely for information purposes and does not constitute an offer or solicitation of an offer or any advice or recommendation to purchase any securities or other financial instruments and may not be construed as such. The author makes no representations as to the accuracy or completeness of any information in this post or found by following any link in this post.
YOU MIGHT ALSO LIKE
Stock Market History, Illuminated, 2022 Style
I’ve been keeping a graphic of the long-term performance of the US stock market for many years now, and I’ve been sharing this information in the histogram below. I think it does a reasonable job of revealing how long-term returns are manufactured and the tilts over time. I’ve color-chunked the data by decade to highlight decades of historical weakness as well as periods of strength. I’ve also added additional tables and charts which I think are interesting, and in some cases, stunning.
Read More
