Good morning. Wednesday brought some calm to markets (outside of South Korea). Stocks inched higher, oil prices stabilised a bit and even bitcoin staged a mini-rally. Have investors already started to look past the war? Email us your thoughts: [email protected].
Geopolitics, bubbles and gold
The day the UBS Global Investment Returns Yearbook appears is something like Christmas Day for sad markets nerds. Its tabulation of returns going back more than a century and covering all the big asset classes gives the feeling of holding all the secrets of the market in one hand (and yes, if you are a proper sad market nerd, you get the paper version, all 300-odd pages of it).
This year’s version has some interesting new stuff in it that got us thinking.
Ignore geopolitics (almost always). That is hard to do at the moment, for several obvious reasons, but data on market performance over the past 126 years suggests in almost all cases, long-term investors should block it all out and hang on.
In a call to launch the latest yearbook this week, one of the three authors, Paul Marsh, noted that the biggest geopolitical shocks, such as the world wars and the 1970s oil shocks, did leave a mark. But “the really big events that impact stock prices tend to be pretty infrequent”, Marsh said. And peacetime bear markets have, in the past, done as much or more damage than even big wars and the like. The Wall Street crash in the 1930s, the dotcom crash at the turn of this century and the great financial crisis 18 years ago (yes, you are that old) were all on the scale of the world wars as market events. Irish stocks lost 81 per cent in 2008, for instance.
From the yearbook:
One hard lesson of this chart is the great financial benefit the US has enjoyed by fighting its wars (since the civil war) on other people’s territory (and being on the winning side). Compare US returns in the two world wars with those of Germany and France.
Keep calm and carry on, then? History says so, but history is retrospective, whereas investing is prospective. Modern warfare technology means that Iran just might be able to keep the Strait of Hormuz on lockdown for a long time, pushing global growth down and inflation up. We now know that the two world wars were big ones. But that was harder to know before they began.
Stick with tech (if you’re young). “People associate new technology with bubbles because of the dotcom crash,” said Marsh. And so they get worried about the high current concentration of the US market in technology generally and AI technology specifically. But if you think the market is concentrated now, have a look at the market in 1900, when it was basically a forum for owning bits of railroads. Chart from the yearbook:
Railroads do not play anything like the role in American life as they did 125 years ago. Parts of their functionality have been taken away by cars, highways, trucking and air travel. But despite this, $1 invested in US rail in the year 1900 would now be worth a stonking $176,000, Marsh and his colleagues calculate. By contrast, the same $1 invested in rail’s supposed replacement, road transport (trucking, buses and so forth) would be worth $27,000, and airlines would be worth $8,000.
Of course, the story of rail is not the story of every technology. The yearbook points out that a major technological innovation and stock market component of the late 18th and early 19th centuries was canals. But if you had stuck with your canal shares after the appearance of the railroads you would have lost everything. Rail’s story is different because it has remained the most cost-efficient way to move very heavy cargo, and the industry has become steadily more productive through process improvements and consolidation. High costs of entry and network effects mean that the rail industry’s competitive moat (unlike, say, airlines’) is very wide.
What about information technology? The dotcom bubble looms large here. According to the yearbook, in the late 1990s, “too many dotcom companies had no viable business model. Many subsequently failed — just as, more than a century earlier, too many railroads were built with little prospect of profitability and with most subsequently going bankrupt.” But, as with railroads, investors who stuck with the tech industry (as opposed to any particular company within it) have done well over the long term. Below is a performance chart comparing the tech sector to the broad market, starting on the day the dotcom bubble peaked in March of 2000:
The unlucky investor who first bought tech in March of 2000 would have to wait 17 years to break even, 23 years to catch up with the wider market and 25 to start outperforming. But that is the perfectly unlucky investor. As the yearbook points out:
An investor who put money to work in US technology at the end of 1996, when Fed chair [Alan] Greenspan famously questioned if the market was “irrationally exuberant” would still have enjoyed an annualised return of 14.1% versus 10% for the US market.
The key thing here may be that technology sector indices cast a wide enough net to benefit from various technologies over time, from PC operating systems, to internet search to AI. The lesson: stick with technology, even when it looks expensive, but diversify within it.
Use gold, don’t hoard it. Finally, this year’s yearbook has a new and informative section on gold, which paints a balanced and informed picture of the yellow metal’s investing characteristics.
Since 1900, the real annualised return on gold has been 1.3 per cent, worse than global bonds (1.7 per cent) and stocks (6-7 per cent). But the long history is not really relevant, because gold’s return profile changed dramatically after the US went off the gold standard in 1971. Since then, the annual real return has been a respectable 4.7 per cent.
But there are several asterisks next to that number. It is flattered by the amazing 2025 run-up in the gold price, and it has come with 40 per cent more volatility than US stocks. And finally, gold has been a poor inflation hedge: the yearbook finds that of the 28 years since 1971 in which inflation exceeded 3 per cent, gold returns were negative in 13 of them. And on average, gold’s performance is not meaningfully negatively correlated with stocks’ (though the two are not correlated positively, either). All of this makes gold look like a poor portfolio component. But it stands out in one key respect: it tends to do well when equities are performing terribly, and in recessions.
According to the yearbook:
Gold prices rose in eight of the eleven [post-1971] drawdowns and fell less than the S&P 500 in the other three. Bonds also gave protection, with positive returns in eight drawdowns, and smaller losses than on stocks in the other three. Bonds beat gold in five drawdowns, but across all eleven, the average gold return was 2% higher than on bonds.
The fact that, historically, gold only consistently outperforms in very bad moments suggests that investors should not simply hold it. If they do, all they are getting over the long term is lower returns than stocks, with higher volatility. Instead, investors who like gold should buy it when times are good, enjoy its stability when a storm comes, and then sell it when the weather calms, investing the proceeds elsewhere. With this sort of rebalancing, gold can be useful. If you simply hoard it like Scrooge McDuck, all gold does is make you poorer over the long term than you would have been otherwise.
One good read
Insider picks.
One good read
Chef’s pick
Read the full article here