Stocks-for-the-long-run type charts commonly plot the marvellous growth story of the US market. Sometimes you’ll also get the UK thrown in for good measure.
However you rarely see much mention of our great European frenemies: Germany and France.
Partly that’s because our cultural conversation is dominated by the US.
But it’s also because the continentals’ stock market history isn’t such a wonderful advert for investing. In fact if long-term US stock returns were similar to theirs, I suspect investing wouldn’t be anywhere near as popular as it is in the Anglosphere.
So let’s turn to our near neighbours to discover what a torrid equities experience looks like.
(All charts show inflation-adjusted total returns, reported in local currency.)
German stock market returns
Data from JST Macrohistory1 and MSCI. March 2024.
- Average real annualised return = 4.0%
- Cumulative growth of 1DM/euro = 426.7
- Best annual return = 149.7%, 1923
- Worst annual return = -90.0%, 1948
- Volatility = 31.4%
The German graph looks remarkably similar to the UK experience, with three main exceptions. Namely 1920s’ hyperinflation, the aftermath of World War 2, and Germany’s comparatively smooth sailing through the 1970s.
You can’t help but stare in wonder at the priapic spike driven by the stock market frenzy that accompanied hyperinflation from 1921 to 1923.
We’ve all heard of the wheelbarrows full of worthless money in Germany back then. In that climate, the stock market was a rare place you could protect your wealth – at least for a time.
Even in after-inflation terms, the market rose 722% between 1921 and 1923. It then imploded – falling by 92% over the next two years.
By 1931, in the midst of the Great Depression, the index had been set back 50 years, to levels last seen in 1881.
War hammered
From that nadir, equities rose by double digits for five years in a row. By which time the Nazis were firmly in power.
After a slight wobble in 1938, markets advanced again from 1939 to 1940 in lockstep with German tanks. Stocks were largely domestically-owned and the 30% increase in 1940 speaks to the string of victories scored on the battlefield.
The market continued to rise, even as the Germans were stopped outside Moscow. But then the Nazi government imposed a stock price floor from 1943 as its fortunes deteriorated. This move essentially froze prices for the remainder of the war. Traders declined to buy stocks that were kept aloft by artificial gravity.
1948’s vertiginous 90% drop accompanied the revaluation of the German currency to 10% of its former value.
At that point, German stocks were worth 33% less than they had been in 1871.
So much for ‘stocks for the long run’.
The only way is up
However this uncompressed calamity was followed by a 121% rebound the following year, as the post-war Wirtschaftswunder2 began to take hold.
By 1958 your stocks would have made 2021% if you’d bought into the German market in 1948.
How many people could or would have done that? Vanishingly few, I suspect.
Elsewhere the UK’s worst stock market crash still lay ahead. Our home market tombstoned -72% from 1973 to 1974.
But in contrast the German market only declined 24% during the same period.
And now, if you look back 50 years, German returns average 5.9% annualised. That compares to 6.2% annualised for the UK and 7.1% for the US.
Nevertheless, the catastrophic German war experience has left its imprint in the country’s relatively subdued overall market return of 4% annualised over the very long-term.
French stock market returns
Alas, as the French chart shows, there are other roads besides defeat in war that lead to stock market perdition:
- Average real annualised return = 1.2%
- Cumulative growth of 1F/euro = 6.58
- Best annual return = 115.9%, 1954
- Worst annual return = -46.0%, 1945
- Volatility = 21.8%
Japan is the cautionary tale commonly used by seasoned investors to scare the younglings – but it should be France.
Unlike Japan, the French market is still 33% below its World War 2 peak some 80 years later.
French equities lost 96% of their value from 1942 to 1950. But the slide didn’t stop there. The market continued to crumble for another 27 years, until 98% had been lost peak-to-trough.
Paradoxically, the French economy and people enjoyed a 30-year boom after World War 2 – a period that came to be known as Les Trente Glorieuses.
But the benefits weren’t felt by French investors.
Returns were undermined by industrial nationalisation and high inflation. It wasn’t until 1983 that the market was defibrillated back into life by Mitterand’s tournant de la rigeur economic reforms.
By then, the stock market had been a disaster area since 1914. That long era of investor sorrow has saddled French equities with a bond-like 1.24% long-run annualised return.
Yes, the past 50 years have seen French shares recover to a perfectly respectable 5.3% annualised. Even so I still believe the gallic experience is the best riposte to home bias imaginable.
The German and Japanese downturns are clearer illustrations of investing risk.
But France’s lost years demonstrate that equity rewards do not necessarily flow from economic success (something we’ve seen again more recently with certain emerging markets).
UK and US stock market returns
By way of contrast, here’s the growth charts for UK and US equities:
Real total return data from JST Macrohistory3 and FTSE Russell. March 2024.
- Average real annualised return = 5.3%
- Cumulative growth of £1 = 2,521.55
- Best annual return = 103.4%, 1975
- Worst annual return = -57.0%, 1974
- Volatility = 17.5%
Data from JST Macrohistory 4 and Aswath Damodaran. March 2024.
- Average real annualised return = 6.8%
- Cumulative growth of $1 = 24,640.33
- Best annual return = 60.9%, 1933
- Worst annual return = -41.0%, 2008
- Volatility = 18.4%
International long-term returns
And for completeness here’s how our foursome compare when you plot them all on the same chart:
I wonder how many people look at the blistering US performance and decide to go all-in on an S&P 500 ETF?
Especially after US stocks’ recent stunning results.
Or how about a bet on nordic tigers Sweden and Denmark? They’ve enjoyed US-level returns over the past 150 years.
Me? I don’t think any regime can last forever so I’m sticking with my global tracker fund.
Take it steady,
The Accumulator
- Òscar Jordà, Katharina Knoll, Dmitry Kuvshinov, Moritz Schularick, and Alan M. Taylor. 2019. “The Rate of Return on Everything, 1870–2015.” Quarterly Journal of Economics, 134(3), 1225-1298. [↩]
- ‘Economic miracle’. [↩]
- Òscar Jordà, Katharina Knoll, Dmitry Kuvshinov, Moritz Schularick, and Alan M. Taylor. 2019. “The Rate of Return on Everything, 1870–2015.” Quarterly Journal of Economics, 134(3), 1225-1298. [↩]
- Òscar Jordà, Katharina Knoll, Dmitry Kuvshinov, Moritz Schularick, and Alan M. Taylor. 2019. “The Rate of Return on Everything, 1870–2015.” Quarterly Journal of Economics, 134(3), 1225-1298. [↩]
@TA:
Nice post.
Re: (All charts show inflation-adjusted total returns, reported in local currency.)
Would it not have been instructive to include a Table/Chart showing comparative exchange rates (or the effect of them) across the period too?
Interesting article.
I think https://rationalreminder.ca/podcast/224 is worth a listen for those interested on this topic, as it digs into these questions not only from the perspective of single country, but also whole world, using a lot of historical data, including expropriations.
@all — Apologies if you saw the version of this article (delivered over email) that made an excessively generous claim about the performance of Japanese equities.
Gremlins! Line since deleted.
It definitely feels like the conclusion is to diversify globally, but it doesn’t necessarily flow from that to do so using a market-cap weighted fund. Maybe we should consider other ways to choose allocations?
@Caligula — I’m inclined to agree, but then I’m a naughty active investor.
For a long-time equally-weighted indices outperformed market cap weighted, but the remarkable strength of US tech/growth in recent years has undone that. (And it was mostly just a small cap effect anyway, IIRC, with smaller companies having a bigger share of an equal-weighted index).
For a view from the other side, here’s Lars saying market cap weighted is the rationale choice:
https://monevator.com/why-invest-in-alternatively-weighted-index-tracker-funds/
Anything else is a claim for edge / data mining, in his view.
All hail the global market cap fund. It’s all you need.
Plus a little extra tilt to EM… and Global Small too… oh, lets add a bit of Gold… not forgetting those Dividend funds for passive income…
Definitely global market cap fund all the way though!
@TA — Great article! Any thoughts on why Sweden and Denmark have performed so well over the last 150 years? Does the size of an economy correlate with stock market performance at all? I would guess that the two measures are not intrinsically linked. That said, the huge size of the US economy and the fact that many of the world’s most prominent companies are listed there — even ones that were originally started overseas — makes me worry less about the outsize contribution of US equities to a global tracker.
@TI – I don’t think equal weighing is active investing in any way. It’s not about trying to identify which country will perform better – in fact it’s doing the opposite. I’m specifically not talking about saying things like “The US market is too large, I’ll allocate more to Europe” – I’m saying something like “I don’t know anything about which will perform better, therefore I’ll go 50/50”.
Market-cap weighting is more active: it assumes that larger markets will perform better. This is only reasonable if we assume that:
a) current market caps are purely due to past performance and
b) past performance predicts future performance
Both those assumptions are manifestly false.
All of that being said – I’m not actually doing equal weighting. I have nearly everything in a global market cap weighted fund, with a slight tilt to non-US and small cap. Full equal weighting *feels* naughty, like you say…
Very interesting! I often wonder what would happen to stocks if we had a proper war going on. The German experience was a rollercoaster, but … perhaps also not too bad, especially compared to *not* being in the stock market where you would have lost everything during hyperinflation and after the war was over.
@Rich #9 “I often wonder what would happen to stocks if we had a proper war going on”: Mitsubishi UFJ Financial Group’s recent Capital Markets Strategy Policy Note states: “Bloomberg’s Economics team has modeled the potential cost of a military conflict in Taiwan across two scenarios. In both cases – a Taiwan blockade and a Taiwan war – they have projected a cost to the global economy of $5 and $10 trillion, respectively. The corresponding 5% and 10% contractions in global GDP would precipitate a global recession, and dwarf the cost of other recent conflicts including Ukraine, Gaza, the Gulf War and 9/11. The economic impact assessment takes into account the disruption to semiconductor supply chains, shipping, trade sanctions and tariffs, cyber-security and financial markets.” Not sure that would be worse for shareholders as compared to being invested in French equities from 1940 to 1983. That experience makes Japan’s ‘lost decades’ from December 1989 to March 2024 look benign.
Is it just me or were those Y axes logarithmic? I don’t understand how a scale that goes 100,1000,10,000, 1,000,000 makes sense.
If they were linear scales – 1000,2000,3000 etc would those lines be a helluva lot steeper.
@ Colin – yes, logarithmic graphs. The correct way to show long-term returns otherwise later fluctuations are out of all proportion with earlier ones. See:
https://freefincal.com/why-are-stock-market-graphs-plotted-in-logarithmic-scale/
@ Rich – I guess what we’re not feeling is the desperate panic of losing 90% and not knowing the market will bounce back.
@ Wodger – I would guess that Sweden’s neutrality helped it avoid the worst of both World Wars. Denmark was non-combatant in WW1 but invaded in WW2. Even then the returns aren’t terrible. We’d have to dig into the archives to find out what happened there. Inflation looks benign in both countries. Both markets recovered strongly after financial crisis, came through Stagflation relatively unscathed, Great Depression not awful. As far as I’m aware labour relations are good in both countries, but I don’t know enough about the composition of their stock market to even hazard a guess.
Last time I checked, Australia and South Africa both had superb long-term returns while Switzerland was fairly mediocre.
This is a total guess, but the number of times the banks are involved in stock market panics, perhaps their markets have a relatively low weighting to financials? Perhaps some combination of that plus didn’t suffer much damage in World Wars, plus good at containing inflation, plus stock market not dominated by companies or industries that have gone into irreversible decline?
That’s fascinating. What strikes me is how relatively inconspicuous the big stock-market “events” (I mean the infamous Black Mondays, Tuesdays, etc.) are, compared with the longer term depressions, wars etc. on this timescale. I guess looking at total returns rather than bare stock prices helps with that. On the other hand, unless you’re a college endowment fund, an aristocratic dynasty or a vampire, I suppose a 100-year timescale isn’t terribly helpful.
@ColinThames – Yes, logarithmic. My only concession to chart-watching/”technical-analysis” has been to take the view that when a chart goes exponential, and you switch it to a logarithmic scale and it’s _still_ exponential, then you’re probably in a bubble. I reckon you can see that in the US and French graphs prior to the “Great Depression” and the Dot-Com bust, but it’s easy to say that with the wisdom of hindsight and I may well be deluding myself.
@caligula — I think all investing is active, it’s all a matter of degree, but anyway equal-weighting is certainly more active versus the common definition of passive as used by those who describe themselves as such.
Let’s say Microsoft is 7% of the S&P 500. That is how the weight of international capital has ‘voted’ — it’s derived from the valuation that the market has put on the earnings of Microsoft, taking into account all other uses of money, etc etc. (All in theory, obviously).
As an equal weighter, however, you do not put 7% into Microsoft. You put 1/500th into it, and you put the same amount into the company that the market-cap weighted tracker has at position 500 in the index.
This is an active decision. You are voting your money differently from the market.
It’s basically the same with countries.
Obviously it’s all more convoluted than this, various frictions and biases come in, markets aren’t fully rationale, etc etc. But that’s the gist.
Equal-weighted was thought to have done better because the smaller companies had until recently a tendency to outperform larger ones (the small cap effect). If you follow the link to Lars’ piece, he’s sceptical of most such factors, and one would have to concede it’s been a good few years to be sceptical to be fair. 🙂
How about the graph for the bond component of ones portfolio ? The German bond investor would be wiped out by hyperinflation to virtually nothing in the 1920s – it was much more deadly to bonds than to shares (the book “When Money Dies” has been mentioned on these pages before, certainly worth a read). And then after regime changes, did the new states in France and in Germany honour the bonds of their predecessors to any extent ?
Very interesting read.
@TI (#14) – this is a really convincing argument for market-cap weighting.
When I started out, I had the too-clever idea to over- and underweight countries by CAPE. This has made lower returns for more faff, and seven years in I think this was probably a mistake, at least from the return point-of-view.
For diversification I’d still want to keep any single country well below 50% of the total, so a global tracker is too heavy in US stocks.
@LateGenXer has posted the link to an excellent interview with Prof Cederburg above (#2). In a more recent interview he said about geographical diversification: “At the monthly level, domestic stocks and international stocks have much higher correlation than that [stocks and bonds, correlation 0.15]. But if we look over the long horizon, like a 30-year horizon, the correlation between stocks and bonds is almost 0.5. There’s actually much higher correlation at long horizons for bonds and stocks, and then domestic stocks and international stocks have a lower correlation at a long horizon compared with stocks and bonds, so the geographical diversification is actually more beneficial than the diversification across the two asset classes.”
from https://rationalreminder.ca/podcast/284
@Brod I feel attacked.
Although you missed out EM Small Cap Value and Physical Uranium too 😉
Interesting article. Are you able to do a similar one on India ?
@ Mark – I have the data for bonds. You’re quite right German bonds are a wipeout. I haven’t looked at French but I know the French state has defaulted on debt and their post-war inflation record makes the UK look like Switzerland.
@ flyer123 – I don’t have total return data for India. Do you know of a good public domain source?
Another argument in favour of Brexit?
Mainland Europe has historically always operated a vigorous “protectionist “ economic policy-rightly or wrongly
U.K. and America historically are equally forcibly for “free trade” -rightly or wrongly
This salient fact shows in their relative stockmarket outcomes
Of course there are many other ways of making money than the stockmarket
Those investors however who ignore history and cultural norms pay the price of performance for their investment portfolios -if the stockmarket is their vehicle of investment choice
Economics come very second to a country’s deep seated cultural beliefs and ideas
xxd09
@xxd09 — A more appropriate phrase would be “the first economic argument for Brexit”…but I’d still disagree.
Through your proposed historical lens, countries that pursue petty nationalistic agendas or worse combined with jingoistic populist manifestos that set their face against evidence – and led by leaders that undermine institutional norms – do not have a good record of delivering strong and durable shareholder returns, in the main.
I’m thinking any number of South American banana republics, most recently Venezuela and arguably perhaps Brazil or even China of late.
Protectionism can have a role in the development of an emerging economy, particularly when it’s in its high growth / low wage / heavy industry / exporting goods phase. (E.g. South Korea back in the day).
That was hardly the UK at the time of the Referendum, and even with the slow bleed of a less-efficient economy since Brexit I don’t think anyone is forecasting we degrade back to that.
One thing to remember when thinking of going all in on a Global tracker is equity confiscations. Russia froze western owned shares in its companies, and vice versa recently. During the IIWW, US forced a split off the US subsidiaries from their german parent companies. There are probably way more examples, but the important part is that it’s no use if the US/China/etc. market is going up, if your shares are confiscated as your country is considered hostile.
A perverse incentive for home bias perhaps
I wouldn’t say it’s an argument for home bias but more an argument for investment in countries with clearly defined property rights. English Common Law and its US derivative have always been the best at that… Hence London’s outsized role as a financial centre. You saw what happened with Baillie Gifford’s heavy weighting in China recently where private property essentially doesn’t exist.
Does this have to do with a different business culture, in terms of how businesses raise finance?
My impression is that German businesses – their famed “Mittelstand” – are often small/medium-sized enterprises, possibly family businesses, that rely on bank loans rather than equity. But I’m really not an expert, just a thought.
@TA, an interesting comparison, especially with the log plot.
How different the countries performed depends critically on the time frame considered. They all looked similar for the first 40 years of the comparison, and have been broadly similar from different starting points for the last 40 years (well, except for one country where growth has levelled off somewhat since … looks like 2016).
You have to ask whether the divergence in the mid-20th century represent fundamentals that are relevant to investment decisions in 2023 or were specific to a period with two world wars and their aftermath.
And yet, back in 2002 we invested £3.5k in both an L&G UK Index Trust ACC and an L&G Euro Index Trust ACC. Today, they stand at £11.5k for the UK and over £15k for the Euro. So the Euro index performed far better, for us at least. Had we been reading Monevator back then we would of course put the money in a Global tracker (if such things existed back then). I wish 😉
@Haphazard
You touch upon a salient point. Given the size of the French & German economies their indices encompass a small amount of companies. The corporate structures and cultures are different with many companies privately held and many companies holding cross holdings in each other. Workers in those countries have also had traditionally larger claims on the state for savings and social insurance than the “Anglo Saxon” economies, with their performance only of tangential interest to the average citizen.
@Nimbus (#26),
For info: £ vs € at start of 2002 c. 1.6; today 1.16
@ Jonathan B – Excellent point! I’d say the two bookend eras represent high watermarks for globalisation while the mid (2oth) century period marks a retreat into varying experiments with managed economies, capital controls etc.
The fact that capital couldn’t flow freely could be one reason why outcomes diverge in the post-WW2 period.
The main relevance now: globalisation may be going into reverse and financial regimes change.
Finally, it’d be interesting to throw some more countries into the mix. Japan is the obvious example of a stock market that took a different path over the last forty years.
@Accumulator – Is it Website like this ?
https://primeinvestor.in/nifty-50-returns/
Underlying source data links in the page above.
Another argument for market cap weighting is that it is the cheapest to run, as it doesn’t need much rebalancing (price and market cap move together), or trading in smaller less liquid names.
@flyer123 – Ah, nice. It looks like a price return index and I think the data source requires a paid sub for total return data. I could be wrong but first foray suggests that’s the case.
Intriguingly, the chart you link to shows why linear graphs are misleading re: stock prices. The 72% gain in 2003 barely registers whereas the much smaller 2020 gain looks like an ascent up El Capitan.
Yes, logarithmic scale makes so much more sense.
Try this link although not sure if it includes dividend info for total returns – Also may need to download to CSV and calculate/graph the data of the report downloaded between 2 dates. https://niftyindices.com/reports/historical-data
Accumulator – Not a complete but maybe this ?
25 Years Journey of Nifty 50_V2 – NSE https://nsearchives.nseindia.com/content/indices/25_Years_Journey_of_Nifty50_2022-01.pdf
Very interesting. Really amazed at France. Particularly between end of WW2 and early 80s. Not least because France’s industry was not bombed to smithereens and its population was not killed in such large numbers/proportions. You conclude this was about “home bias”, which I can’t really judge. But we thought WE were far and away the sick man of Europe by the late 70s: as a teenager then I remember that all too well. Puts it in some perspective at least.
@ Mike – I think we need to distinguish between shareholder returns and standard of life gains for the general population. France did very well by that light during those lost decades for its stock market. The economic gains accrued to labour not capital. It was policy choice and the will of the electorate to take that path, and who can blame them?
We were overtaken by France and Germany in GDP-per capita terms in those post-war years. Effectively surrendering a lead bestowed by the Industrial Revolution. We started to reel them in again in the 80s and 90s but have rather blown it again recently.
France to me is a great example of how economic returns don’t necessarily flow to shareholders. China is a better example these days. Home bias is only a concern from a diversification pov.
@flyer123 – yes, it’s hard to tell. Usually an index provider doesn’t give total returns data away for free, or no more than a few years worth. I’ve had more success sourcing total returns from academics. I can’t open that pdf you sent.
Do you know how far the Indian stock market goes back historically?
@Accumulator – I have the downloaded pdf, I can email, let me know the mailid.
NSE was established by 1992/1994, so data maybe available all the way through.
BSE sensex is more older, established in 1875, so unlikely for older data.
@flyer123 – contact me at: taccumulator ‘the-at-symbol’ gmail.com
@The Accumulator — Eek! I’m wary of putting emails on the Internet, you’ll be harvested by a robot. I’ve edited accordingly. 🙂
Latest from Charlie Bilello (The Week in Charts (9/25/24)) on the asset class return ‘tapestry’ from 1st Jan. 2011 to 24th Sep. 2024:
US large caps: 13.8% p.a.
US growth: 16% p.a
NASDAQ 100: 18.4% p.a.
US small cap.: 9.4% p.a.
US value: 10.6% p.a.
All very respectable. But:
EM stocks: Just 1.8% p.a.
Commodities: A disastrous minus 0.8% p.a.
And for good measure 😉
BTC/XBT: 144.8% p.a.
Underneath every generalisation though there’s another narrative. From the ‘Off the Charts’ substack (‘Buy Everything’), reporting data from Refinitiv today (using MSCI India and China Total Returns indices):
$1 invested in India’s market on 31st Dec. 1993 returns $13.58 by 18th Sep. 2024 but $1 invested in China on 31st Dec. 1993 returns just $0.89 by 18th Sep. 2024.