Post New

Here are the global valuations as of March 31, 2026.

Quarterly CAPE Ratio Updates

Median CAPE Ratio: 19
25% cheapest: 14
25% most expensive: 28

Average of Foreign Developed: 22
Average of Foreign Emerging: 18

https://theideafarm.com/quarterly-global-valuation...

Brazil is the second cheapest in the world, and India is the second most expensive. Brazil is going to do well if oil prices stay high and India is going to suffer. Japan, which is often touted as being very cheap, is among the pricier ones. The rap against Brazil is that the largest companies are all mining, energy or finance related, very cyclicaland deserving of low valuations.

As an aside, not disagreeing at all, here are two good rules of thumb for this sort of comparison--

* Ignore the CAPE of a small country whose market is totally dominated by just a couple/few huge firms. In that case, there is no particular reason to expect mean reversion on the theory that market-cap-to-GDP-ratios should have a normal level, it's just a bet on the fortunes of those specific companies. Consider instead the CAPE of the region in which the company lies.

* Ignore the CAPE of a country compared to other countries (mostly). Instead, compare the CAPE of a country or region to the historic average of that area's CAPE. Several years ago there was a poster quite tempted by the CAPE of 8 in the Russian market, and several folks pointed out that that's close to the usual figure (historical median 9.15 if you're wondering), not a cheap figure. Time has passed, and the reason has become clearer to all.

In an older table of mine, the historical median for Brazil was 17.3, so if both that figure and the figure in the "idea farm" table are correct, then the current 11.8 does seem pretty cheap.
This is a table that only goes up to mid 2012, but might still be of some use for a sense of scale

                     Min          Max        Median
Australia            7.65        31.60        17.15
Austria              6.04        59.11        26.80
Belgium              4.88        29.49        14.89
Brazil              11.04        29.73        17.34
Canada               5.83        63.33        19.85
Chile                9.71        32.95        20.93
China               14.30        61.98        24.40
France               6.20        57.16        19.92
Germany              7.83        56.86        17.90
Greece               1.95        39.81        15.91
Hong Kong            8.55        34.55        18.16
India               12.69        47.80        24.56
Indonesia            5.05        34.96        16.37
Ireland              3.02        18.25        10.94
Italy                5.92        52.91        21.66
Japan               13.27        94.25        43.89
Malaysia             7.77        26.50        18.49
Mexico              11.69        35.33        19.62
Netherlands          4.62        38.52        11.95
Portugal             7.02        39.36        16.43
Russia               5.13        22.88         9.15
Singapore            9.40        37.80        21.96
South Africa        10.16        24.12        16.22
South Korea          4.74        27.65        17.84
Spain                7.25        40.06        17.33
Sweden               4.82        74.18        19.54
Switzerland          7.12        57.94        18.16
Taiwan               9.18        42.28        19.43
Thailand             3.00        17.97        11.48
Turkey               8.24        42.96        17.02
UK                   4.43        28.69        11.84
USA                  4.72        45.08        14.63

It might be fun to compare the current levels in your linked table to the median levels in this table.

Jim

                    Min          Max        Median     Current 

Brazil              11.04        29.73        17.34    11.8
China               14.30        61.98        24.40    13.7
Hong Kong            8.55        34.55        18.16    15.8
India               12.69        47.80        24.56    32.2

Japan               13.27        94.25        43.89    25.3  
France               6.20        57.16        19.92    20.3
Germany              7.83        56.86        17.90    18.0
UK                   4.43        28.69        11.84    16.8
USA                  4.72        45.08        14.63    35.8

Brazil, China, Japan are significantly lower than median. India and USA are significantly higher than median. Europe very close to median.

Japan is an odd one. It had such extreme valuations, then a multi-decade slump when valuations declined, that median maybe overstated. Its current valuation seems high given its vulnerability to high energy prices, increased competition from China and USA in auto, batteries and AI.

Jim,

Thanks for sharing the historical data — this is very helpful. Could you clarify the starting date for the CAPE figures?

Also, if you were incorporating this data into an investment strategy, how would you use it? For example:

a) Broad allocation via country-level ETFs focused on the most attractive valuations

b) Using it as a filter — e.g., applying an existing stock selection strategy but only investing in stocks within countries with the most attractive valuations

Or would you approach it differently altogether?

Appreciate your perspective.

Thanks for sharing the historical data — this is very helpful. Could you clarify the starting date for the CAPE figures?

The paper I was quoting from had varying amounts of history for each country.
Their comment was: "While only two had century long data (US and UK), most of the other countries go back to the 1970s and 1980s.". And as mentioned, the study covered history ending mid 2012. I'm sure there are better comparable tables out there now.

Also, if you were incorporating this data into an investment strategy, how would you use it? For example:
a) Broad allocation via country-level ETFs focused on the most attractive valuations
b) Using it as a filter — e.g., applying an existing stock selection strategy but only investing in stocks within countries with the most attractive valuations
Or would you approach it differently altogether?

If using this, which I wouldn't, I'd probably prefer use slightly broader regions rather than smaller countries if I could. I'd prefer a Scandinavian fund, if it existed, to a Danish or Finnish fund, because of my comments on market concentration. Some years ago you wouldn't want to have thought Finland was about to soar just because Nokia had a bad year and the CAPE looked cheap: they had a lot more bad years after that.

But I'm an individual stock guy, mostly, so I go to the FT.com free global equity screener. https://markets.ft.com/data/equities/results
I usually eliminate industries with high risk or no control over their pricing (banks, insurers, basic materials, energy, funds, real estate). A couple due to personal taste (tobacco).
I require a meaningful dividend. Bad for me tax wise, but a lot of firms in "secondary" markets are controlled by insiders and/or are not run for shareholder returns, so this at least ensures that SOME of the profits get to the shareholders.
I then screen mainly on sane leverage, not crazy P/E, and highest five-year-average ROE.
My favourite leverage screen is net income not more than 5 times typical annual profit, but that particular screener doesn't allow that, so there I use a more conventional cap on debt-to-equity.

If I'm not doing individual stocks, I essentially never use a cap-weight fund. They tend to underperform almost any other weighting method you can think of.

As an aside: what goes for the pitfalls of comparing country CAPEs also goes for comparisons among industries. I wouldn't buy an industry because its CAPE was cheaper than the broad market CAPE. You'd want to buy into an industry (if at all) if its CAPE were lower than that sector's own historical average CAPE. [in that region]. I have a table of those, too. For example, the historical average CAPE for US industrials, materials and health care are all about 30% higher than the hsitorical averages for the equivalent sectors in Europe, while European utilities and financials have had average CAPEs quite a bit higher than their US counterparts.
(based on data from Thomson Reuters’ Datastream Market Indices starting in 1973, ending around 2013)

Jim

Hello Jim,

> "I usually eliminate industries with high risk or no control over their pricing ... real estate)."

This statement sounds strange to me. Real estate isn't an industry but rather a diverse variety of industries.

In terms of 'high risk', there are REITs you can choose where you have excellent insight into cashflow for years ahead and macro risks are hedged.

In terms of 'pricing power', again it depends on the REIT, and it depends how you perceive pricing power.

---

Take NVIDIA GPUs. If you wanted to work with CUDA-based engineering applications, you had three choices for most of the last decade:

- buy a new NVIDIA GPU

- buy a used NVIDIA GPU

- rent a NVIDIA GPU in someone else's machine.

Pricing power can be demonstrated by stagnant offerings that people relentlessly buy - NVIDIA GPUs today have 8GB and NVIDIA GPUS in 2016 had 8GB.

It can be seen in the absence of special offers and deals to incentivise bringing purchases forward.

It can be seen by continual supply shortages and buyer desperation.

---

Now, suppose NVIDIA announced a new '20 year subscription' GPU package. You would no longer be able to buy GPUs as you need them.

Instead, you'd have to commit to an annual subscription for 20 years, and every few years you will get a GPU shipped to you.

The price each year would continually rise contractually, and the quality of the GPU would hardly improve much over time.

Further, they insist you must secure your subscription against assets you own, in case you try to back out.

I don't believe most businesses or consumers would rush to embrace that deal.

NVIDIA simply doesn't have the pricing power to force that on consumers.

---

Similarly, imagine a 'Disneyland for 20 years' package where you can only go to Disneyland if you commit up front to 20 years of annual holidays there.

With the price set so that it will rise if there is inflation, but will not go back down if there is deflation.

And if there's sudden huge demand for theme park holidays, the contract lets them push the prices up *even more*, every few years, if they want.

I don't believe consumers would rush to embrace that deal.

Disney simply doesn't have the pricing power to force that on consumers.

---

But there are REITs that have their contracts set that way and businesses /government desperately rush to take up the offer!

Because there is no choice. To operate the business, they *must* have warehouses or hospitals or student flats or whatever in a particular place.

Or because demand is so outrageously strong, that if a 20 year contract is what it takes to secure the site, that's what you sign.

Unite is pivoting to this model - looking for places where they get 100% occupancy year after year, and universities will guarantee the cash flows without fear.

These are the kinds of REIT I am quite interested in.

When the customers have to sign up for long term contracts at prices that keep going up for decades, and can't easily get out.

When you can see who their customers are, and whether those customers are going to be able to pay for years to come.

---

"You will pay what I ask, and more each year, for 20 years, or you are screwed" is pricing power to a degree you hardly see outside of real estate.

Perhaps in defense? I'm not sure what other sectors can force customers into 20 year contracts that always go up in price for the same dull product each year.

If you have ideas, I'd be interested to hear them.

Anyway this is why I would disagree with you about real estate lacking pricing power.

---

In cases where real estate / REITs operate on a short-term basis, and compete to attract customers regularly, those don't have pricing power.

Offices and high street shops, which are quite 'substitutable', often require incentives to bring in business, have fierce price competition.

Personal storage facilities like Safestore and Big Yellow, are always competing with attic space, garages, and 'your friend's spare room'.

Residential properties can be substituted; they lack pricing power except where there are more people needing homes in an area, than there are homes.

But there are real estate sectors - logistics, health, elderly care, some student accomodation - where you may see good pricing power and relatively low risk.

---

Pricing power is also driven by the ability for someone else to set up shop in competition, by the substitutability of the thing being bought.

In some sectors, some places, and some times, this happens a lot in real estate, but in other times, places and sectors, it doesn't seem to.

TRS

> "I usually eliminate industries with high risk or no control over their pricing ... real estate)."
...
This statement sounds strange to me. Real estate isn't an industry but rather a diverse variety of industries.

I eliminate only "pure" real estate firms. i.e., firms whose business is primarily owning property. Related industries, like real estate brokerages, are often very good businesses to be in. Further afield, sellers of building suppliers can be very good.

That's because, when screening, I'm only interested in "value added" product or service businesses. That's where the money is--the rest of the business world makes decent money only when luck favours them, which isn't all the time. In the case of a firm whose core is owning real estate, the value added is relatively minor and the total is very much at the mercy of the prevailing prices of real estate and rental rates.

Coming at it another way, the long run return of almost any firm eventually approximates their long run returns on equity and assets, which is generally a very low cap rate for firms whose primary business is holding property. In a handy database of (mostly) US stocks, REITs rank 80th out of 94 industries for average ROE, and 82nd for average ROA. I'd rather be in (say) beverages with ranks 12 and 10 on those metrics.

(developers are in the middle: they do add value and do better than real estate holders on average, but their average return hides way too much risk and cyclicality to make them worth considering. It's easy, almost normal, to get almost wiped out in the wrong part of the cycle)

I eliminate property holders from my screening for the same reason I eliminate gold miners. A gold miner can be creative and efficient in the ways they extract, refine, and sell gold...but in the end, the investment results are dominated by the market price of gold, over which they have no control, and which they don't even produce. A real estate holding firm which does very high quality buying, selling, and some development will still have results dominated by the prevailing prices of real estate (capital and rental), which in most areas don't rise much in real terms over the long term, and can fall precipitously depending on the market cycle. Just look at prime central London housing, often cited as the world's best example of a "never lose" real estate market...prices are ~10% lower than they were 12 years ago even before you count inflation.

That's not to say there aren't occasional great deals in property firms, but those are exceptional, and require selecting the right security at the right moment, meaning deep analysis--not the investing style related to broad global screening. The average return, which is what you're expecting for if you're screening for a slate of stocks, is neither particularly attractive nor predictable. The world has many industries. There are lots of better fish in the seas, so you might as well stack the odds in your favour by eliminating the known laggards. If you're going to have a broad portfolio with less informed analysis per security, you need to choose from a pool which high average returns so the inevitable duds will be more than compensated for.

As a random example, a simple slate of US-listed medical device firms, equally weighted, has beat the S&P 500 by 3.9%/year in the last 29 years, despite lagging hugely in the last 5 years while giant tech ruled the roost. With "good tailwind" industries like that out there, I'm happy to skip the ones with laggard averages and leave them in the "too hard" pile.

Jim

> "US stocks, REITs rank 80th out of 94 industries for average ROE, and 82nd for average ROA."

Don't get me wrong. I wouldn't invest in US REITs either, separate from the boycotting issue, they're just too expensive.

But you have a good chance to get a humble 10-11% with relatively little volatility in your income, and very little stress, from UK REITs.

Assuming you are in the UK and investing optimally with ISAs/SIPPs etc.

For people in other countries, it's less appealling. That 10-11% becomes 9% after withholding tax.

At which point you might well say 'unless some REIT offers a chance of additional gains, I'd rather have the index at 11% than UK REITs at 9%'.

Alternatively, you might say 'well, 9% sucks, but it behaves a bit differently to bonds and non-REIT stocks, it's some useful diversification'.

> "Just look at prime central London housing, often cited as the world's best example of a "never lose" real estate market...prices are ~10% lower than they were 12 years ago even before you count inflation."

Prime? Central? 12 years? That's an odd combination.

Especially since I just made the argument for *not* viewing residential as having pricing power, in my previous post.

And I thought you had a stance against things that resemble cherry picked timeframes/datasets?

Let's look at the last 25 years, every 5 years, government data, all of London, and see what result we get.

https://landregistry.data.gov.uk/app/ukhpi/print?i...

Jan 2000: 140k
Jan 2005: 250k
Jan 2010: 300k
Jan 2015: 430k
Jan 2020: 510k
Jan 2025: 563k

*And* generating substantial rental income that whole time. *And* with tax relief as a UK investor that you wouldn't have had on shares & dividends.

Those numbers above are hard facts - it's government land registry data for achieved prices at city level.

I wouldn't invest in regular London residential property myself, but the fact is, it has *printed* money.

More generally, using such a niche of real estate and a VERY niche timeframe as a criticism of REITs/real estate, presented without any source to verify it, is hardly fair to the argument that I actually made in the post you replied to.

Since I had pointed out, directly, that *residential is in general not a good example of REITs with sustainable pricing power*.

Would it not be fairer to respond to the argument that was made, than the opposite of it?

As far as long term returns go, I would not advocate holding REITs (UK or US) blindly regardless of price for the long term, even though the total return from REITs is roughly the same as stocks over very long periods of time.

But as long as they are much cheaper, and less vulnerable to Trump's insanity (tariffs, etc), and efficient (for me) to hold in savings wrappers, then why not?

TRS

> "US stocks, REITs rank 80th out of 94 industries for average ROE, and 82nd for average ROA."

Don't get me wrong. I wouldn't invest in US REITs either, separate from the boycotting issue, they're just too expensive.

But you have a good chance to get a humble 10-11% with relatively little volatility in your income, and very little stress, from UK REITs.


I'd agree that 8% seems pretty likely...at the moment, maybe your 10-11% for a while. But that's because of the specific opportunities and price levels at the moment. At the moment anybody who bought UTG at any time in the last 11 years has lost money, even counting dividends, for example, as there have been two separate big down cycles. I think it's possible that those who bought not long before the pandemic may never see a positive real total return.

My point is that REITs (those that are really in real estate) are, on average, poor investments, no matter what country they're in, because they don't generate value at a very high rate as a direct inevitable consequence of the nature of their business. They have dreary long run *average* returns, they have very long bad cycles, and they occasionally run into debt crises when management isn't as good as you thought.

A well informed investor picking the right security at the right point in the cycle can do very well, for sure. I've bought a couple in the past. But if one is not a subject matter expert, it's an industry that the average person will do better avoiding. There will be some winners, for sure. But if you're going to throw 40-60 darts and a dartboard to get the highest average score, and somebody lets you remove some of the below average numbers first, it makes sense to do it. There is no reason to start with a needless handicap.


Prime? Central? 12 years? That's an odd combination.

Not at all. It was chosen because it was considered one of the safest real estate bets in the world, and because that result emphasizes that even the safest bets in real estate have truly terrible cycles. Other things being equal, I'd rather skip industries that hurt you that badly for such long stretches with such regularity. Given an easy choice, I'd rather wander through a regular field than a minefield, even with a map : )

Would it not be fairer to respond to the argument that was made, than the opposite of it?
As far as long term returns go, I would not advocate holding REITs (UK or US) blindly regardless of price for the long term, even though the total return from REITs is roughly the same as stocks over very long periods of time.

You were responding to my comment that I filter out real estate from my global screening--which is a "shotgun" approach to portfolio construction. As that is the context, I absolutely stand by it. It's a poor sector to include if you aren't a subject matter expert, specifically BECAUSE you'd be doing it relatively blindly. It's a below average business sector in terms of both risk and return on average over time, no matter where it is, so the non expert will do better by simply excluding it.

Same with gold, basic materials, banks, insurers, and number of other sectors known for frequent pitfalls that are so easy to avoid just by avoiding the industry.

Jim

Bit old but for a lt overview of the brazilian equity markets valuations as well as the Real (important for all non Brazilian based investors)

And yes, combining many MI models also work to select stocks in Brazil...

https://quantasticworld.substack.com/p/unlocking-v...

Sorry wrong link...

https://quantasticworld.substack.com/p/brazils-equ...

Interesting discussion,

AI suggests Reits have outperformed equities in the US. The US market is also best for Reits vs UK.

Reits in the UK on average have performed similarly to general equities but are far more cyclical as you say and doing well is timing dependent.

AI suggests Reits have outperformed equities in the US. The US market is also best for Reits vs UK.

The problem there might be "AI suggests..."?

As a random demonstration, the largest US REIT ETF is VNQ.
It has pretty steadily underperformed the average S&P 500 firm since inception in 2004. It has lagged RSP by a pretty remarkable 2.93%/year over 21.5 years with dividends reinvested. (RSP tracks the S&P 500 equal weight index, so it's a proxy for the average big firm)

Some investors are attracted to certain REITs for very high coupons. (I'm not saying anyone is suggesting that). But stocks with very high yield tend to have poor total returns. For example, the top 5% of yielders among reasonably sized US stocks lagged the S&P 500 by 4.6%/year since then, counting dividends for both.

Jim

Yep AI hallucinations!.

It occured to me well after the post that it sounded like bollocks 😂

Can anyone explain to me how AI hallucinations work?.

I subscribe to Chat GPT put in it phd mode and ask it a bunch of questions about companies and specific markets.

Can anyone explain to me how AI hallucinations work?

I can share my personal view/overview.

The main principal underlying the way they work is to pick the next most likely word in a sentence. To almost everyone's surprise, given some local context and enough probability table entries, this ends up sounding like a rational person.

But it isn't. The end result is more like a very well spoken and confident person who (a) has read everything on the internet, but (b) unfortunately believed it all. Also (c) an imperfect memory, and (d) is very good at plausible bullshitting.

The reason for the last one is that, given that it's just picking probable word combinations, it's inevitably picking whatever is most plausible sounding, not necessarily with any connection to a fact. That's what hallucinations are: strings of words that go together in a way that is statistically likely given the context, but don't actually have a connection to the real world because...why would they? Since by construction it is so darned plausible sounding, after a few useful answers we all have a HUGE temptation to start skipping the necessary task of validating whatever it says.

Some fancy models being developed then take what they've written and actually try and go and validate it, but that requires good and locatable source for the validation of any possible answer. They tend to search the web.

To get an idea of the scale of the bullshit problem, a variety of state of the art models were tested on examples from a medical diagnosis test. Each round starts with a few symptoms, and one has to work through a series of steps of test results to do the differential diagnosis correctly. The AIs were wrong, and confident, at early stages in 80-90% of the test cases. They jumped to plausible sounding conclusions too early with high confidence, and couldn't even give complete lists of the conditions that met the facts known so far. If you hear hoofbeats you should suspect a horse before you suspect a zebra, but you should actually have a look before asserting that it's definitely a horse.

Jim

Thankyou Jim, fascinating.

What with all the hundreds of hours of podcasts / interviews with "experts" online,why have I nrver hrard this discussed in any detail?

It sounds like a fundamental flaw.

Never heard* fat fingers.....

The main principal underlying the way they work is to pick the next most likely word in a sentence.

I would like to apologise for that mistaken post.

Obviously it's principle, not principal. Now I have to give up complaining about other people's mistakes for a few weeks. Dang, that's my principal joy.

Jim

Post New

US Policy FAQ

---

Contact Shrewd'm
Contact the developer of these message boards.