GDP growth? US non-farm payrolls? Fed funds rate ? Oil price? VIX, change in the Baltic Dry index?
All of those might make good candidates, but when it comes to overuse one number trumps them all thanks to its seductive power.
The Price/Earnings ratio. Let me explain why.
There’s quite a lot of psychology bound up in this that we need to unpack. Firstly, our natural human need to simplify. Brain cycles are expensive, so we do all we can to avoid them and seek out simplifications, rules of thumb to avoid having to do the work. P/E ratio fits perfectly. Second, groupthink. When we hear enough people reference the P/E ratio it assumes greater and greater significance. Which finally turns to overconfidence and false certainty. The markets are uncertain so we yearn for clear indicators. Something to tell us whether we should buy or sell, something that makes a complex world black or white, that gives us a story. We rarely stop to interrogate the P/E ratio from first principles, instead taking its conclusions at face value and making decisions based on it while lacking any real understanding of the underlying. Why do the work when one little ratio contains all you need?
A final psychological truth that explains a lot here: Fear sells. The dot-com crash looms particularly large in the psychology of a certain cohort of investors, and the PE ratio effectively “sounded the alarm” so it has become imbued with much power as a gauge of exuberance and bubble-risk, which goes well beyond its initial use as a tool to compare two stocks. Robert Shiller pinned this down in a series of seminal papers in the late-1990s in coming up with the CAPE (cyclically adjusted price earnings) which used an average of earnings to avoid sensitivity to a single year, and was used to generate forward looking returns forecasts (of questionable predictive power). The CAPE took the win in the late 90’s in calling the dot com bust but it has consistently sent incorrectly bearish signals of returns and overvaluation in the decades since then. A lot of brain cells now routinely get burnt looking for the next dot-com crash indicator, Barry Ritholtz wrote on this recently. Sounding the alarm can make you sound awfully intelligent – everyone wants to be that person who called it.
If you’re open to it – allow me to chip away a little bit at your faith in the P/E ratio.
I see at least five key issues with the PE ratio that are rarely scrutinised:
1. Earnings. Most people probably think that accounting is a bit boring but at least it’s factual. We assume Earnings are some kind of objective truth, that tells us something useful about the economics of a company when in reality they are purely an accounting construction – a mirage some would say. As Generally Accepted Accounting Principles (GAAP) have changed over the years so earnings numbers move. There is huge latitude in terms of what management estimate and how they can present the numbers. Some skeptics such as NYU professor Baruch Lev (author of the book The End of Accounting ) argue that earnings are more or less useless for equity investors today, thanks in part to the skewed way they account for intangibles ( investment in brand, patents and research) which dominate in today’s world (more on this below). GAAP principles have also changed through time – allowance for stock options, depreciation, dividends and tax policy are just some examples that have all impacted earnings differently in different time periods but historical comparisons are frequently made without recognising any of this. This podcast is a great summary of Prof Lev’s work.
Companies do issue reams of data on non-GAAP numbers these days to try and address this issue as well as holding earnings calls Q&As which are informative to the real drivers of value in the business. Sure, some of these stats are rubbish and rightly mocked (remember “community adjusted ebitda”) but that doesn’t give you an excuse to hide behind a worn-out ratio, you need to do the work to sort the useful from the nonsense. There’s never been more high quality analysis freely available to the casual investor on Twitter and blogs (professor Aswath Damodaran’s work is a great example). But of course, if you want a simple ratio and a neat story none of this fits, and the vast majority of people in investing won’t do the work to understand the extra info.
In 2013 Jeremy Siegel published work showing that the CAPE-based equity return forecasts had been persistently too pessimistic for decades, due to changes in the way earnings were reported for the S&P500 post 1999, this paper provides another good overview of the sort of GAAP changes that do matter for accounting earnings but tend not to get discussed. The CAPE has consistently been above its long term average for a long time now.
2. Intangibles. At a minimum Michael Maubousin’s paper on valuing intangibles ought to be required reading for all investors to really understand this big issue . Because the FASB board that set GAAP principles have decreed that investments in research ought to be treated like expenses, these hugely depress earnings for companies making large intangible investments today (by also not appearing in book value it distorts the balance sheet). Sceptics love to poke fun at tech firms that are “yet to make any money”, they might think twice if they realised that these losses – perhaps unfairly – reflect investment in research and development being made now that could underpin revenues years into the future. If they were investing in factories or acquisitions it would look very different.
This isn’t just about tech here – Amazon and Shopify are in consumer discretionary, Netflix, Google and Facebook in communications, Tesla in autos and Airbnb in hospitality are all examples of this. Today’s leading firms in almost all sectors are marked out by their investment in intangibles. Not to mention firms with huge brand like Nike and healthcare companies who have had big R&D for decades.
Critics of accounting methods say that the powerful FASB board that set the rules for how earnings are calculated is an un-accountable committee mired in all the conformity, groupthink and status quo bias you’d expect, who stand guard over an approach rooted in the early twentieth century economy of plant, factories and widgets, which has not adapted to a world of subscription services and research.
I’m seeing EBITDARD (where RD stands for R&D) being used more and more to try and adjust for this – as one attempt at standardising these comparisons outside GAAP.
3. Interest Rates. Other things being equal you’d expect to pay more for the same earnings at lower levels of interest rates. Michael Mauboissin has made some good illustrations of this. Robert Shiller, the architect of the CAPE ratio has recently released work updating his classic thinking on CAPE for a low rate world and inverted the ratio to create an earnings yield, and translated this to an excess over the 10 year treasury yield to try and adjust for this effect. That measure gives a very different picture of the valuation of equities today than the raw CAPE ratio.
4. Growth. I think most investors do realise that stocks with high growth potential do merit a higher PE ratio but lack a framework for actually making that adjustment and it is not that intuitive. Again , Michael Mauboussin comes to the rescue here with some very simple examples which show that a growth rate just 5 percentage points higher can justify a PE ratio twenty points higher. As growth rates increase, PE ratios ought to increase exponentially, which is always bound to be unintuitive.
I think this relationship is poorly understood which means investors generally fail to account for how growth ought to change the PE ratio. In fact the very construction of the ratio itself as a snapshot in time frames any subsequent conversation in a way that works against any realistic appraisal of growth effects and more-or-less anchors investors to a zero growth assumption.
5. Sectors, standards and debt. I didn’t want to have an excessively long list so I’ve grouped a final few points together here. Among the more egregious uses of the PE ratio in my book is comparisons among different international equity markets , although these are frequently made. The differences in accounting standards (eg GAAP vs IFRS) themselves can result in huge differences in reported earnings for the same stock, and vastly different sector compositions in different markets make this even more of an apples-to-oranges comparison. Finally, the PE ratio doesn’t take into account how much debt a company has – other things being equal you might expect to pay more for earnings from a less-indebted company.
I hope I’ve given you something to think about here and a pause for thought the next time you hear the Price Earnings ratio being cited. Look I’m not saying it has no value whatsoever, in a complex world we do need some simplifications and price ought to matter in investing. Using the PE ratio to compare two similar firms at a point in time is probably among the least-bad uses, and I know that many in investing will be aware of some of the shortcomings described about. But framing really matters, and once you’ve framed a conversation with the PE ratio it is hard to make the appropriate adjustments away from it that are warranted. Of course, the market could crash tomorrow, making me look a little stupid and entrenching the power of CAPE for another generation. Even so I do think that more investors would benefit from better understanding the shortcomings and tempering their use of this ratio accordingly.
If you liked this you might also like :
4 thoughts on “The most overused number in investing”