Last summer, Howard Marks raised some eyebrows by variously suggesting that tech investors cannot truly see the future and by reiterating the idea that passive investing may have unintended consequences.
That wouldn’t be the first time anyone (including Marks) has expressed those ideas, but as is usually the case when one challenges conventional wisdom or otherwise pokes the groupthink hornet’s nest, it brought out the tomato throwers.
As a reminder, here are some of the key passages from Howard’s July 2017 note:
Bull markets are often marked by the anointment of a single group of stocks as “the greatest,” and the attractive legend surrounding this group is among the factors that support the bull move. When taken to the extreme — as it invariably is — this phenomenon satisfies some of the elements in a boom listed on page four, including:
- trust in a virtuous circle incapable of being interrupted;
- conviction that, given the companies’ fundamental merit, there’s no price too high for their stocks; and
- the willing suspension of disbelief that allows investors to extrapolate these positive views to infinity.
In the current iteration, these attributes are being applied to a small group of tech-based companies, which are typified by “the FAANGs”: Facebook, Amazon, Apple, Netflix and Google (now renamed Alphabet). They all sport great business models and unchallenged leadership in their markets. Most importantly, they’re viewed as having captured the future and thus as sure to be winners in the years to come.
True as far as it goes . . . just as it appeared to be true of the Nifty-Fifty in the 1960s, oil stocks in the ’70s, disk drive companies in the ’80s, and tech/media/telecom in the late ’90s. But in each of those cases:
- the environment changed in unforeseen ways,
- it turned out that the newness of the business model had hidden its flaws,
- competition arose,
- excellence in the concept gave rise to weaknesses in execution, and/or
- it was shown that even great fundamentals can become overpriced and thus give way to massive losses.
The FAANGs are truly great companies, growing rapidly and trouncing the competition (where it exists). But some are doing so without much profitability, and for others profits are growing slower than revenues. Some of them doubtless will be the great companies of tomorrow. But will they all? Are they invincible, and is their success truly inevitable?
Around the same time Marks wrote those words, Goldman inadvertently triggered a mini-tech rout with a note that asked if FANG was overvalued. The key takeaway from that note (to me anyway) was the discussion of factor crowding and how smart beta and other factor-based strategies can end up perpetuating a self-feeding loop. Recall this from the infamous Goldman note released in early June of last year:
While not exactly a Fields Medal worthy observation, we note that FAAMG is positively correlated with Growth and Momentum and this relationship has strengthened in recent months. The bigger anomaly, however, is that FAAMG is almost as highly correlated with Low Vol (as measured by standard deviation of 1Y daily price returns), which is not a characteristic typically associated with cyclically driven names.
Now consider that with the following from the same Howard Marks piece mentioned above:
Organizers wanting their “smart” products to reach commercial scale are likely to rely heavily on the largest-capitalization, most-liquid stocks. For example, having Apple in your ETF allows it to get really big. Thus Apple is included today in ETFs emphasizing tech, growth, value, momentum, large-caps, high quality, low volatility, dividends, and leverage.
If you’re interested in a lengthy discussion of this in the smart beta context, you should maybe check out something funny I wrote for Dealbreaker called “One Wall Streeter Discovers It’s Ok To Be ‘DUMB’“.
Ok, anyway, Goldman is out with a sweeping new piece called “Why Tech Is Not A Bubble”, which is worth a read regardless of whether you agree with the premise as communicated by the rather straightforward title.
First, Goldman notes the obvious, which is that the biggest names are getting, well, bigger.
“The biggest 20 technology stocks globally have a combined market cap of over $6trn, but the top 5 companies comprise 60% of this total,” the bank writes, adding that “these super – large companies are concentrated in two geographies: 1) US – FAAMG: Facebook, Amazon, Apple, Microsoft, and Alphabet’s Google. 2) Asia – STTAB: Samsung, Tencent, Taiwan Semiconductor, Alibaba, Baidu.”
Here’s a look at their weight in regional benchmarks:
One saving grace is that market diversification means dominant sectors and companies have tended to comprise a smaller percentage of the aggregate over time as markets have evolved. Here’s that at the sector level:
And here it is at the single-stock level:
Goldman goes on to list a variety of reasons why tech’s success is “justified by the fundamentals” including, obviously, topline growth and margins. And then there’s this:
Of course regulatory risk comes up, and that’s something I’ve spent more time talking about than I ever imagined I would. Goldman recently dedicated one of their “Top Of Mind” posts to the subject and the overarching point I’ve tried to make is that it was always ridiculous to assume that regulatory risk wouldn’t come calling at one time or another, even if the risk of regulation emanates from something as constantly misguided as Donald Trump’s mouth.
Recall this assessment from a piece we did back in late March when tech was looking shaky:
Buying opportunity or not, the important point is the bit about “we already know this.” It’s been clear since he became President that Trump was likely to try and “clip Bezos’ wings” (as Axios puts it) and the more negative WaPo coverage Trump gets, the more angry he becomes.
We also “already know” that there are potential problems with Tesla’s autopilot and that Facebook and other social media platforms were likely hijacked by global bad actors in the service of promoting a populist agenda in the U.S., the U.K., Germany, and France (among other locales). Because we knew that, we should have known that it was just a matter of time before this or that “new” bombshell finally lit a fire under regulators’ asses.
And see this gets me back to the point made here at the outset. It was (and is) ridiculous to assume that we’re going to simply transition seamlessly to a world where Jeff Bezos serves your healthcare needs, provides you with retail banking services, gets you a mortgage, sells you drugs covered by the health insurance he also sold you, and delivers those drugs to your home that he owns the mortgage on via a drone that was activated by his female alter ego “who” now giggles at you for no reason.
In the same vein, it was always absurd to think that Facebook and Twitter were going to continue to operate unfettered as the preferred medium of (un)civil discourse once it became abundantly clear that those platforms were being used (in some cases by bots operating at the apparent behest of foreign intelligence services) to manipulate the public.
And on and on. This is problematic because as Doug Schwartz, a partner at CGCN Group, former Chief of Staff to the Senate Republican Conference, working on legislative and political strategy, and former senior advisor to Senator John Thune of South Dakota, told Goldman in an interview out in April, “tech seems to be politically ‘homeless’ right now–facing tough questions from critics on both sides of the aisle.”
Getting back to the tech bubble piece, here’s what Goldman has to say about comparisons to previous bubbles, some of which are mentioned by Howard Marks in the excerpts above and for now, we’ll just present this without further comment other than to note that there’s a ton more in the full Goldman piece:
Two previous periods when a group of stocks dominated the equity markets were the in 1960s to early 1970s in the so-called ‘Nifty Fifty’ era and the rise in technology in the late 1990s. The ‘Nifty Fifty’ period saw the dominance of a group of 50 companies that, unlike the 1990s, were not focused on a particular sector but rather a concept. There was significant optimism that US economic dominance would allow a new breed of US corporations to become truly global market leaders – multinationals. Many of the companies that were favoured did enjoy very high returns (rather different from the tech bubble of the late 1990s when the market was dominated by new companies with no returns) and a belief that these could be maintained into the long term future. For that reason they were often referred to as ‘one-decision’ stocks. You bought and held them irrespective of the price. There was a popular shift away from value investing towards growth investing. As a result the valuations increased hugely. By 1972 when the S&P 500 had a P/E of 19, the average across the Nifty Fifty was over twice this level. Polaroid traded at a P/E of over 90 and Walt Disney and McDonald’s over 80x forward expected earnings. Interesting, despite these very lofty valuations, Professor Jeremy Siegel argued (see Valuing Growth Stocks, Revisiting the Nifty Fifty, American Association of Individual Investors, October 1998) that most of the stocks did actually grow into their valuations and achieved very strong returns.
A similar narrative later drove the focus on the ‘New economy’ of the late 1990s. Then, as in the 1960s, Value (or ‘old economy’ ) stocks became very unloved.
The current rise in technology companies that followed the financial crisis is rather different from the frenzy that drove the bubble in the late 1990s. In the years before the crisis banks dominated the sector weights in many equity markets (benefiting from a cocktail of strong growth, high leverage and product innovation). With the demise of the banks leadership in markets, technology has quickly become the major leader of market returns and a dominant sector once again. Since 2008 technology in the global stock market has increased from 7% to 12% – at the same time it has nearly doubled in the US from 13% to 21% in the S&P. In the late 1990s the technology share of global market capitalisation went from just 10% of the S&P in 1996 to a peak of 33% in 2000.
Most importantly, however, the valuation of the companies in the earlier periods was much higher than for those of most technology companies today. As Exhibit 21 shows the largest tech stocks in the tech bubble traded at an average of over 50x PE (although many stocks were far more expensive than that). The largest Nifty Fifty stocks traded at an average 35x.Today, the largest tech stocks trade at a little above 20x expected earnings despite the very low level of interest rates today (particularly relative to the early 1970s)
Is Goldman comparing trailing earnings to expected? That’s fun.