“Stocks aren’t cheap and popular at the same time.”
Robert Shiller is a Nobel laureate and well-known professor in economics at Yale, where he is developing his heterodox theory of “narrative economics”. He largely built his reputation by presciently predicting the bursting of the internet bubble in 2000 and the financial crisis in 2008. In both cases he identified overvaluation as a key contributing factor.
Shiller is also famous for developing the Cyclically Adjusted Price/Earnings ratio, popularly known as CAPE, a valuation tool for equities. Calculated as the price divided by the 10-year average earnings adjusted for inflation, it now shows high readings. It thus came as a bit of a shock when Shiller announced a few weeks ago that “sky high stock prices” actually “make sense” if using a further adjustment to his CAPE indicator. Renamed into “Excess Yield CAPE”, the adjustment consists of inverting the CAPE, which turns it into a yield, and subsequently deducting the 10-year bond yield.
This justification of high stock prices was met, to put it mildly, by surprise from many in the investment community, including Albert Edwards and Jeremy Grantham. The general criticism varied, from “Shiller is changing his narrative” (pun intended) to “Shiller is moving the goal posts”.
So, how should we view this question of valuation? Are stocks fairly valued or in a bubble?
Stocks are popular . . .
I like the above opening quote (even if we don’t know who said it). It combines the issue of valuation with that of crowd psychology. Regarding the latter, seminal works, such as Kindleberger (2011) and Mackay (1841), have identified common characteristics of bubbles, hypes, and manias. Those traits, updated for today’s conditions, include:
- Easily available credit, in our case via central banks, which allows leveraged bets. Today that includes record volumes of traded call options, often on highly speculative stocks.
- Growing retail investor participation, now facilitated by user-friendly mobile apps created by zero-commission brokers.
- Lack of breadth characterised by a concentration in a few stocks or sectors that move broader indices. “FANG” (Facebook, Amazon, Netflix, and Google) and its variations are the current reincarnations of this phenomenon.
- A frenzy in IPOs, nowadays involving loss-making companies (“unicorns”). Today it is accompanied by a craze in Special Purpose Acquisition Companies (SPACs), which are corporate vehicles established to make acquisitions.
- Narratives and stories whereby popularity is correlated with the level of implied “disruption” or “transformation”, not its probability. In other words, the more outrageous the claim, the more popular the theme and the stocks playing it.
- Exponential price patterns without significant corrections. Examples include semiconductor stocks or Tesla.
Still, the bubbly nature of the market is not the main topic of this article. Instead, I would like to return to the question of valuation.
. . . but are they expensive?
Reviewing numerous articles and comments since Shiller’s flip-flop, it has become clear where the disagreement on valuation hinges.
Let me start by removing one technical misunderstanding. Some commentators incorrectly equated the Excess CAPE Yield to the infamous Fed-model. A key critique of the Fed-model is that it is flawed because it deducts a nominal bond yield from a real earnings yield. However, the original CAPE already corrects for this error by adjusting the earnings for inflation. In short, the Excess CAPE Yield is not the same as the Fed-model.
Other commentators make a more valid point. They argue that bonds are in a bubble and massively overvalued owing to extraordinary monetary policies. With yields at, or in some cases below, zero, they have a strong argument. Accordingly, the Excess CAPE Yield remains high―and thus seemingly attractive―only because bond yields are absurdly low. The Excess CAPE Yield would be much lower, making stocks more expensive, if a ‘fair-value’ or historic average bond yield were employed.
But that does not mean that artificial valuations cannot endure for longer, which is a common argument from equity bulls. It goes something like this: “based on policy makers setting the rules, monetary policy and regulatory financial repression guarantee that bond valuations will remain way above what we consider ‘fair value’ for many years to come. This keeps equity valuations attractive.”
That claim, however, requires qualification involving its underlying assumption of ceteris paribus. Any “guarantee” depends on policymakers not making any mistake, not losing (e.g., yield curve) control, etc. Moreover, and more relevant here, is the duration of valuation regimes. After all, nothing lasts forever, so at some point some kind of mean reversion in bond valuation will occur. Importantly, this is implicitly also acknowledged in the above claim.
Which brings us to the crux of the debate.
While there is little disagreement on valuation―everything is overvalued―there is considerable difference on timing. Specifically, those reducing equity exposures today based on valuations do so because they (implicitly) admit that they cannot time mean reversion. Those who accept the risk of current valuations by staying fully invested do so because they (implicitly) believe they can time it and get out ahead of everyone else.
For what it is worth, traditional valuation approaches take into account risk, exemplified by the principle of “margin of safety” popularised by Benjamin Graham and Seth Klarman. We term it an equity risk premium for a reason. From that perspective, valuation is not intended for timing purposes.
To conclude, the new CAPE is appropriate as a valuation tool provided its advocates make clear that bonds are assumed to be fairly valued. Once this assumption is relaxed, the valuation question turns into one of timing mean reversion.
Today, a fair value assumption is foolhardy. Mean reversion is more probable, if even with uncertain timing. Moreover, as I have previously expressed here, mechanical economics, and especially policy makers’ overconfidence in exerting control over markets and the economy, is flawed and could lead to disruptive future adjustments. In short, economic, financial and heuristic risks don’t justify prevailing valuations. More likely, they convey a dangerous ‘priced for perfection’ situation.
For now, the valuation debate is likely to continue. It relates to that other familiar question which is always looming in the background: will history rhyme or is it different this time?