This article can be read by selected Monevator members. Please see our membership plans and consider joining! Already a member? Sign in here.
US stocks vs the World: how often does the lead change hands? [Members]
Next post: Reasons not to downsize in retirement
Previous post: Weekend reading: rightsizing to a richer old age
Comments on this entry are closed.
The interesting bit about ‘lead switching’ is you can see that when US retakes the lead, all the gains of World are reversed quickly.
The bigger problem for World, apart from the lower growth and currency issues, is that many good companies are kept private. The culture of going public and sharing wealth growth is lacking in Asia and Europe. Many businesses are family run and the greed in keeping profits in-house leaves very few great companies listed on markets. On top of that, pressure from governments to reduce job cuts and forced price levels are huge disadvantages.
Will anyone invest in French energy companies after Macron forced them to take the hit during the start of Ukraine war? Good government policy, but terrible for capital markets.
The lesson is simple. Live in Europe, invest in US and travel in Asia.
Truth be told, we’re all active investors, even if we think that we’re being passive and benefitting from the wisdom of the crowd by buying VWRD ETF and then accepting that 71% goes into the S&P 500 and Nasdaq.
There is no such a thing as a truly passive investor.
We just make our active choices a bit differently to someone choosing funds, companies or countries to invest in based of some other metric than their weight in an index.
Choosing to invest in an “All World”, periodically rebalanced, market cap weighted, index tracking product is still an “active” choice.
Country/region, industry sector, & Fama French factor allocations for equities specifically, and for asset class allocations more generally, are all active choices.
And IMHO if you own a portfolio that mirrors the cap weights of every global publicly traded asset then that’s an active choice in and of itself albeit that the rebalancing between the asset mix and different companies is rule based/ pre-determined/ automated.
Even if you call that approach “passive”, then it still isn’t a complete mirroring of the true “global market portfolio” (if there is such a thing), as it doesn’t include weighted allocations to unlisted shares (and more businesses are held this way via PE) and nor does it cover ‘alternative’ assets like forestry etc.
Once we accept that nothing is truly passive, and that “passive” is a terribly non-descriptive and low information way to label a systematically executed cap weight strategy for certain listed and liquid assets, then we’re liberated from the sense that we have to try and follow those weights at all times and in every and all circumstances.
The obvious alternative to a 62% to 71% weighting to North American large caps is not to go 100% into the high fee UK equity ‘active’ fund manager du jour.
Rather, you can adjust your US weightings without sinning (too much) against the Gods of the Efficient Market Hypothesis.
After all, there’s no agreement what the EMH means, which is the right version of it, and which, if any, versions of it are (or might be) right.
Personally I use 95% index trackers but I’m always open to considering something else and I have dialled down my own US exposure from around a 6 to 7 out of 10 to about a 5.
Even some index trackers have a maximum single country, single sector or single company cap to try and preserve diversification.
And it’s harder to argue that you have a good global diversification if 71% of the portfolio is in just the one country, however well that country has performed, and however promising it is relative to companies in the other 200 countries in the world.
One can only scratch the surface but I think one of the most succinct rebuttals to the US bear arguments is the chart of Nvidia’s market cap versus those of the entire UK, French and German stock markets. Granted you may scoff that Nvidia is the poster child of the bubble in US stocks but still, when you put it in that context, that the US has the framework in place to spawn businesses that become larger than the main European markets in a short space of time, I’m just not that convinced that the whole complicated concept of US exceptionalism should fade away anytime soon or that the equity market is delusional in the way it is allocating capital.
I’m staying humble personally and assuming the market has this weighting somewhere in the region of fair.
@ AoI – Your point about US economic dynamism is well made!
I think Nvidia is super interesting. As per the old cliche: it took 30 years to become an overnight success. Moreover, it was a fluke. It just so happened that Nvidia’s GPUs were the ideal workhorse for LLMs. Nvidia didn’t see it coming. Nobody seems to know whether the current tech pathway for AI will continue to pay dividends or whether it will run into the sand. Nvidia’s future may well not be in its control.
I’d be interested to know whether Nvidia has a defensible moat in the way that TSMC seems to. Nvidia reminds me more of Nokia or Blackberry: right place, right time but not necessarily durable.
All the same, we don’t have to predict the downfall of the US tech titans. All that need happen is that the market overpays for future S&P500 cashflows relative to the rest, then realises its mistake.
@ Delta Hedge – originally passive investing just meant accepting the market return i.e. not attempting to beat the market through stock picking or timing. Somehow – and I blame the Internet – the label was misinterpreted as meaning you somehow shouldn’t make any personal decisions at all about your portfolio. Including, for example, not choosing your individualised bond allocation in line with your risk tolerance and time of life. The idea that “passive investing” doesn’t involve decision-making is bogus and I sense largely floated by naughty active types seeking to portray passive investors as hypocrites or idiots or… Anyway, vive tribalism.
@TA: “I’d be interested to know whether Nvidia has a defensible moat in the way that TSMC seems to. Nvidia reminds me more of Nokia or Blackberry: right place, right time but not necessarily durable”
Not necessarily the right thread for my comment here (as this is the ‘passive’ aisle, and not Moguls) but: Nvidia could have a moat but IMHO there’s some better candidates in this space.
They did actually, AFAIK, deliberately go down the bleeding edge GPUs and state of the art nanometer chips in order to try (as they have successfully done) capture 3 different markets: high end gaming, crypto PoW mining, and AI (neural nets and latterly LLMs).
Got to give them credit for that. Not just down to luck.
But could Nividea be replicated or beat? Sure they could.
How far ahead are they (and how quickly could they be overtaken, if at all)? Who knows?
I think 2 things are fairly clear though:
a) they’re very reliant on TSMC and a lot of the semiconductor fabrication plant infrastructure for TSMC is in Taiwan, which is not exactly the safest place in the World for it TBH, and;
b) a sizeable and growing part of Nvidia revenues depend on LLMs panning out as a commercially valuable General Purpose Technology.
So there’s plenty of risk with the stock to balance out the opportunity.
I would personally have thought though that a company like ASML would offer a better trade off here long term.
They are literally the only entity in the World that can do the very highest end of the Extreme Ultra Violet Lithography needed for the latest TSMC foundries and the best of breed Nvidia chips which they make. Economics Explained on YouTube has an excellent recent piece on why the Netherlands is the most overpowered country compared to its geographical and demographical size which goes into the frankly incredible technology of, and the vital role played by, ASML. It’s worth a watch.
Senior moment – it’s was Real Life Lore who covered ASML in their Netherlands piece.
Economics Explained is also excellent, but I was misremembering there.
With 70% US in the World Index, its getting to be a moot point, but I’m happy to take a high level of ‘single country risk’ when it comes to the US for many reasons – accumulated over a couple of decades mainly working for you guessed it – US Semiconductor companies. Second guessing the next US exclusive Black Swan has been costly insurance. Asset allocation and time diversification are for me more important parameters.
@ Delta Hedge – interesting! I’ll check out that piece on the Netherlands. As far as I can tell, it’s fair to say that Nvidia realised GPUs had applications beyond gaming and enabled those capabilities in their chips. The AI community adopts GPUs and at some point, as you say, Nvidia bet on that opportunity. I’m reminded of an interview I heard years ago with a Toyota exec. He was explaining their approach to developing non-ICE cars. He said they didn’t know which technology was going to win – hydrogen, electric, hybrid etc – so they bet on them all. Diversification wins again (he says in a belated attempt to get back on topic 😉
@ Calculus – excellent point! You’ve reminded me that another way of solving the dilemma on the equity side is to diversify beyond cap weight holdings using a global multi-factor ETF.
@Delta, On ASML – yes great tech but how many fabs are there to sell these high end printers to! Maybe more the railway to the mine rather than the picks and shovels. Plus its always cyclical.
Agreed ‘fluke’ is harsh on NVidia – new markets are always sought out, if not exactly the plan. Where are the other GPU makers btw? There are lots of hurdles to competing at this level – not least the US export restriction rules which cover these chips specifically and pretty much anything else with US content they want to restrict!
@Calculus. Yes. ASML is vulnerable and it’s expensive on any normal metrics. Trailing Twelve Month P/E just hit 50. Price to TTM Sales is over 12. Forward Twelve Month P/E is very nearly 40. Price to 12 month Forward Earnings Growth of forecast 25% (conventional PEG ratio) now at 2. PEG on the 5 year analysts’ consensus earnings growth (of 21% p.a.) now hitting 2.4. PEGs usually need to be below 1 to be considered good value. But ASML are the only ones with the tech. Some moats based on brands (looking at you Diageo plc) have not proved as durable as expected. Some moats involve financial difficulties of replication (why Buffett brought into Burlington Northern Santa Fe Railway for $44 bn, as he estimated it would cost a competitor $100 bn to build an equivalent rival line). But in principle it would be possible for a competitor to build a rival, just as TSMC fabs in Taiwan could be replicated by semiconductor firms in the US with the time and the money. The tech isn’t unique to TSMC per se. But for ASML the tech is unique. Noone else knows how to do it for sure regardless of the cost involved in trying to. That’s quite a moat in principle.
@TA: global multi factor takes one away from cap weighting, but it still has a major look through weighting to the US.
Depending upon the factor in question, the last time that I looked into this it was roughly from around 50% (size and value factors) to over 70% (momentum factor).
Across the ~10% of my own portfolio which I’ve got equal weight between the 3 global multifactor ETFs: JPLG, HWWA and (your own favourite) FSWD I’m guessing that the look through exposure to the US is about 60% (averaged over the 3 of them). It’s not so easy to tilt away from the USA 😉
There is, of course, the rebalancing bonus between factors though.
FWIW as the Russell 2000 has so lagged the S&P 500 since the recovery began in October 2022 that I’m personally looking at shifting a fifth of my exposure to the S&P 500 (mostly via the Source S&P 500 UCITS ETF, SPX) – so 10% of overall portfolio – into the Legal & General Russell 2000 US Small Cap UCITS ETF (RTWP) which has (to quote the fact sheet) “a quality tilt by adjusting the market cap weight of those constituents with better quality characteristics upwards and adjusting the market cap weights of those with poorer quality characteristics downwards such that the target active quality factor exposure, set at 0.4, is achieved…..Quality factor is defined as a composite measure of..profitability and leverage”.
IMO this is important for the Russell 2000 as the index has lots of unprofitable biotechnology firms, which increases its volatility, reduces its returns and generally make it less attractive (with the biotechs the Russell 2000 has a P/E of 17, but without them one of just 10).
In effect the quality tilt has the added benefit of aligning with the value factor so that you can get size, value and quality at the moment in one tracker. Still US country specific risk though, but at least it’s a risk that is no longer concentrated in just the large/mega caps.
@Delta makes sense on the quality bias in US small caps, also mitigates the issue of private equity arguably having hollowed out some of the quality at the broad index level
I think your final point highlights the key question on this broader topic in terms of what one is seeking to address. Is this topic fundamentally a risk management conversation about diversifying idiosyncratic country specific risks associated with all US listed equities, is it about managing position sizes / sector exposure in mega cap tech or more about return expectations in the US versus elsewhere. Naturally they’re closely linked and I guess coming at it from the perspective of an index fund portfolio the practical response is the same either way. I feel they’re also very distinct topics though.
From my own humble perspective I’m more concerned by the latter than the former. If Trump 2.0 involved Liz Trussing the Treasury market I would question to what extent non US equities will be a safe haven, when the US catches a cold the rest of the world catches a fever and all that.
For what its worth, in my own portfolio I have reduced holdings in the mega cap tech names and sold puts on them. My rationale simply being I have no concern on them from the perspective of business quality only valuation and maybe crowded positioning hence I like the risk reward in being paid decent premiums for the risk of buying them back at a lower price. August 5th presented an (entirely lucky) window to short the puts
@ Delta Hedge – Yes, multi-factor has strong US allocation but my suggestion was in response to Calculus’ comment: “I’m happy to take a high level of ‘single country risk’ when it comes to the US for many reasons.”
As AoI mentions, much depends on the risks you’re seeking to address. So if we reframe “potential over-exposure to the US” as “potential over-exposure to megacap growth stocks” then a multi-factor holding is a simple solution.
But, as you say, if your concern is US exposure then multi-factor is no solution at all.
I’m reading your ASML discussion with Calculus with great interest. For one it’s fascinating that a single firm in the Netherlands should prove to be a key link in the semi-conductor supply chain. I’ve no idea how that happened but it’s a great illustration of the complex interdependencies upon which our modern world relies.
Secondly, even in the case of a firm that seemingly has a wonderful moat, it’s extremely difficult to assess the risk of ploughing substantial capital into one stock. Fun to think about though.
This article suggests to me that there may be something worthwhile in Gary Antonacci’s GEM strategy, described in his book Dual Momentum. Here is an article about the performance of the GEM strategy applied by non-US investors:
https://www.optimalmomentum.com/dual-momentum-for-non-us-investors/
@Peter Kelen #13. It’s an excellent website, as is Gary’s 2014 book Dual Momentum Investing. You might be interested in the Dual Momentum Systems website, which was inspired by Gary’s original idea and tries to improve on it. It covers several strategies from very conservative through to super aggressive. Yesterday I linked to a review of it by the 7circles PF site and to one of its strategies in my comment #41 on the thread under @TI’s 15th August 2024 piece “Now could be a better time to retire”.
@Delta Hedge #14. I would strongly recommend not investing in a ‘super aggressive’ strategy if one wants to sleep soundly at night!
Indeed @pka #15. But to quote Soros, it’s not whether you’re right or wrong that’s important, but how much money you make when you’re right and how much you lose when you’re wrong. If you pursue an aggressive strategy with a minimal amount of starting capital (<1% of total) then you in effect cut your losses in advance of any failure. If the strategy outperforms by enough annually in the longest term (multi-decades) the higher compounding returns outweighs the minimum initial capital allocation. Bottom line, put in <1% of starting capital and the most you can/will lose is <1% of starting capital. It's trying to find a suitable long term asymmetry between the impact of permanent loss of capital and uncapped possible future returns.
https://www.apolloacademy.com/foreign-companies-migrating-from-european-stock-exchanges-to-us-stock-exchanges/
Another complicating factor on the US vs the World topic… an increasing share of US listed companies aren’t US companies
Fun facts: US listed companies now make up 72% of the MSCI World Index (i.e. Developed Markets). But US GDP is only 26% of Gross Global Product (sum all countries’ US $ GDPs). Emerging Markets are still only 11% of the wider MSCI All World Index (covering DM and EM). Yet EMs’ GDPs are now around 40% of Gross Global Product. In December 1989 Japan hit a 45% share in the MSCI World Index. Now it’s just 2.8% of the MSCI World tracking HMWO ETF. Will the US go same way? This question has been asked before, but since crossing above the 50% mark in its share of the MSCI World Index back in 2013, US stocks have outperformed European stocks by over 200%.
@Delta Hedge — Your Japan weightings look low there. For example I’m seeing 5.62% in HMWO (link below) and about 6% in VWRL.
https://www.fidelity.co.uk/factsheet-data/factsheet/IE00B4X9L533-hsbc-etfs-plc/portfolio
This doesn’t change the main point of course, in terms of the dramatic re-weighting of a global portfolio versus the late 1980s. 🙂
@TI: makes you wonder about the accuracy of the rest of HL’s info. See here (Top 10 countries) for the 2.8% figure:
https://www.hl.co.uk/shares/shares-search-results/h/hsbc-msci-world-ucits-etf-gbp
Getting info. on the Japanese MSCI World Index share stumped Perplexity AI by the way, which said this:
“The specific weighting of Japan in the FTSE and MSCI country indices is not directly provided in the search results. However, Japan remains a significant component of these indices due to its developed market status and economic resurgence. Historically, Japan’s share in such indices has decreased from its peak in the late 1980s but continues to be a key player.”
And it gave this less than impressive response on a second attempt at an answer:
“As of 2024, Japan’s share in the MSCI World Index has significantly decreased, though exact figures for 2024 are not explicitly provided in the search results.”
Not exactly impressive bearing in mind the God only knows how much energy and cost went into providing those answers as compared to one of Google’s conventional search results. Makes you wonder if the air is eventually going to come out of the AI balloon. That’ll have an effect on US share of the index to be sure! 🙁
And on subject of US mega cap tech concentration, this Dutch poster on Twitter has US retail portfolios holding 20% in Nvidia and Apple – what could possibly go wrong???
https://x.com/beursanalist/status/1838489253426135094
Also note the 3% in 3x NASDAQ or S&P 500 LETFs. I hope these people know what they’re doing and have a pre determined plan for rotating out of leverage.
Misquoting Mark Twain, history doesn’t repeat but it may rhyme. Based upon the US record, 3 historical scenarios haunt the dreams of those of us Brits with large allocations to US cap weighted indices:
1. The US equity market underperforming cash from 1966 to 1982 during inflationary times.
2. The 0% average return in US shares from 1929 to 1949
3. And the ‘lost decade’ prior to the 2009 commencing, 15 years duration so far, bull market.
The biggest relative valuation gap is between the US equity market and the rest of the developed world; but large cap is richly valued compared to small cap, growth style compared to value, and developed markets compared to emerging and frontier ones.
Whilst it’s hard to imagine all types of equities won’t fall if US mega cap growth falters sooner or later; the risks of big falls would seem least in emerging market small cap value, save that the dollar tends to appreciate when equities go risk off which then screws up emerging markets.
Perhaps US Treasuries and USD denominated commodities are the best hedge against (respectively) either a deflationary or an inflationary risk off regime and/or take down of US large cap equity dominance.
A very clever 10 years’ look back from a fictional 2035 perspective from AQR, with an all too plausible US loses global equity leadership scenario:
https://www.aqr.com/Insights/Perspectives/2035-An-Allocator-Looks-Back-Over-the-Last-10-Years
@TA #12: ASML monopoly intact:
https://open.substack.com/pub/multibaggernuggets/p/overview-of-the-week-13-is-asml-doomed
Rotation US/RoW?:
https://open.substack.com/pub/topdowncharts/p/10-charts-to-watch-in-2025-q1-update
Schiller CAPE 10 now @38.7x, in its 96th percentile (only higher 1929 + 2000) and double 1871 to date average of all CAPE 10 readings (17.6x). Is it Soros (“When I see a bubble forming, I rush in to buy, adding fuel to the fire”) or Rob Arnott (reversion to mean)?
Well, although finding myself in the position of arguing for the Inelastic Market Hypothesis (and that US Large Caps are getting pushed ever higher by passive flows), and against my gut instincts that this is what’s happening; the relative much more attractive valuations of ex-US have tipped it for me now.
By next week I’ll have full portfolio look through US equity exposure down to 40% from 50% following the publication of this Maven piece last year, and as compared to a US Large Cap Share of the Global All Cap totals (including Mid and Small Caps and both EM and FM) of roughly 60%-65% (depending on how and what’s counted).
As I mentioned in the most recent W/e reading comment thread, deep down I fear that this decision to intentionally underweight the US, and, therefore, go to in effect short US Large Growth, will end up as a backfire, but the valuation gap is just too huge now.
Ex US SCV is now >5x cheaper than US Large Growth.
This is worse than even March 2000, and there was a time, long ago now, when ex US SCV was even at a premium to US Large Growth (and people thought nothing of it!)
Perhaps the gap between the US and the rest just keeps getting bigger from now onwards (without limits), but I’m seeing arguably high quality Small Cap deep value plays in Europe, Japan, Emerging Asia, Latin America and Africa with consistent profitability, growth potential, moats, solid dividends well covered by FCF, and trading at single figure PEs.
It’s well understood that stock market returns do not consistently align with economic indicators. Despite decades of research examining factors like GDP growth, profit margins, and interest rates, these variables show no significant correlation with stock market performance over a 5 year period. Even an unrelated factor like national rainfall has as much explanatory power (0.06 correlation) as GDP growth — i.e. effectively none.
Instead, 5 year market movements are more heavily influenced by starting PE (0.38), the countries’ debt/GDP ratios (0.23) and starting dividend yield (0.18) – with investor mood, psychology, and sentiment driving the shorter term fluctuations in valuations.
The ex US and especially ex US SCV heavily is heavily favoured over the next 5 years on that basis as compared to US Large Growth.
We have stretched US valuations, minimum dividends and worsening debt/GDP.
But…I still fear I’m making a mistake here and the American rocketship will just keep rising – if only for changed market structural reasons.
Time will tell. If in the next 5 years US Large tech crashes hard in a repeat of sorts of 2000-02/03 then I’ll be backing up the truck, selling everything else, and using thr proceeds to buy America again no matter how grim it looks (2009/10 being a much worse feeling time to invest in America than 1999/2000 even though it was in fact a so much better one).
The most successful sovereign wealth fund of the past decade — New Zealand’s, who knew? — is overweighting Europe apparently, though the tilts are pretty tiny (a few percent off US, a few more on Europe):
https://www.ft.com/content/ba559f94-9e95-49f3-b9bc-831482c8bc79
Interesting. Muchas Gracias @TI 🙂
The future long duration interest rate regime is the massive imponderable.
What will 30 year US, Euro and GBP rates be in 2034/35?
That single data point would tell us so very much if only we could know it beforehand.
I can’t go lower than a 40% US allocation because the risk and consequences of being wrong here are both too high.
It seems absurd to think about it, but, given that US indices are up so hugely since March 2009 (S&P 500 from 670 to nearly 6,500, Nasdaq 100 from 1,200 to 23,600, and even nearly 24,000 earlier this year); one can’t really rule out entirely a continued relentless grind higher, and another stellar outperformance of the US (never bet against America, as Buffett put it), and then finding oneself in consequence in 2034/5 with the cheap European, UK, Japanese and ex US SCV stuff having gotten even cheaper, and up only say 1.5x since 2025, whilst the S&P 500 somehow (market inelasticity?) manages to go up another 4x, to something ridiculous like 25,000 on a PE of 50 and CAPE of 70. The Nikkei 225 hit a CAPE of 90, IIRC, in December 1989.
I just can’t run that risk of such a crushing underperformance, so I’ll have to keep to at least 40% US exposure, even if that means then exposing myself somewhat to the possibility of a huge crash in US mega cap tech. Swings and roundabouts.
Overall, for me, there’s now just too much cheap stuff outside the US to maintain Global Market cap weights though.
But if that changes and the US becomes cheap again, as it was briefly after 2009, my allocation will change too. A song for each season.
Ted calls the top on the US. Time for a rotation?
https://open.substack.com/pub/tedgioia/p/is-the-bubble-bursting
Gioia’s latest post nails the stark divide between the frothy AI economy and the robust non-AI sector. AI’s $3 trillion infrastructure tab looms large, with tech giants scrambling for cash and even eyeing nuclear power to feed data centers’ insatiable energy demands.
Yet, enthusiasm is fading.
ChatGPT’s user base leans on free riders, developer sentiment’s down 10%, and 75% of AI projects flop, and then there’s *that* MIT study showing 95% of firms surveyed report no RoI.
High valuations, like OpenAI’s, lack revenue to match, smelling like a classic bubble driven by hype, not fundamentals.
Meanwhile, out IRL, the non-AI economy is struggling bigly as the Donald would say.
But people still crave and pay up for ‘real’ experiences.
Taylor Swift’s $2 billion Eras Tour and Beyoncé’s $600 million haul dwarf some nations’ GDPs. Live music, sports, and theme parks are cashing i. $37 billion projected for concerts alone in 2025.
Substack’s steady subscription growth underscores demand for tangible value over digital pipe dreams like AI or the metaverse. Consumers are voting with their wallets for real experiences, not speculative tech.
Gioia’s right: bubbles can defy gravity, but AI’s cracks with unmet expectations, unsustainable costs are glaring.
Maybe the non-US and non-AI economy, grounded in what people actually want, looks like the smarter bet.
Time to rethink those US focussed, and AI heavy, portfolios?