Asset Allocations by Country and Asset Type

InvesterJones

chiang_mai said:

A US allocation around 40% strikes me as appropriate, given the high valuations and Tech. concentration. Perhaps it's just me but it's becomming harder and harder to justify the 60% and 80% allocations of the global trackers.

You mean you're becoming more and more convinced that you know something other investors don't? I admire your confidence, though it is human nature to think we're above average Rotation away from the US will happen one day, but I'm not confident I can call the timing correctly - I'm addressing volatility concerns with asset type allocation instead - especially now I've discovered the joy of gilts (both conventional and index-linked).

chiang_mai

InvesterJones said:

chiang_mai said:

A US allocation around 40% strikes me as appropriate, given the high valuations and Tech. concentration. Perhaps it's just me but it's becomming harder and harder to justify the 60% and 80% allocations of the global trackers.

You mean you're becoming more and more convinced that you know something other investors don't? I admire your confidence, though it is human nature to think we're above average Rotation away from the US will happen one day, but I'm not confident I can call the timing correctly - I'm addressing volatility concerns with asset type allocation instead - especially now I've discovered the joy of gilts (both conventional and index-linked).

On the contrary, I think I'm now confirming in my own mind what others seem to have known for a long time, if the Blackrock data is to be believed. It was my sence throughout most of 2025 that a lower US allocation was warranted and now the BR data tells us that most British investors feel the same way and have for some time. One of the main hangups with this is the global tracker allocations to the US which are unchanged but still high at between 60% and 80%. How many people actually want that level of US investment, more importantly, how many people understand that's what they have. My US allocations are probably too low currently at a paltry 28% but since there are plenty of other opportunities available globally, I don't think that really matters much.

Cus

InvesterJones said:

You mean you're becoming more and more convinced that you know something other investors don't? 

Interesting point and generally accepted.  However imo, considering that the latest figures claim that 60% of equity inflows are meant to be passive investments, tracking indices, makes me think that those 60% of investors are not making any assessment of equity value in individual companies, but are relying on the 40% to make the price discovery, and they are trusting historical data and trending mantra.
Whether the 40% take this into account, whether the 60% overtly influence real equity values without direct research is another debate, but I would claim that someone like @chiang_mai, with their detailed analysis threads, probably does 'know' something 60% of investors don't.

chiang_mai

Cus said:

InvesterJones said:

You mean you're becoming more and more convinced that you know something other investors don't? 

Interesting point and generally accepted.  However imo, considering that the latest figures claim that 60% of equity inflows are meant to be passive investments, tracking indices, makes me think that those 60% of investors are not making any assessment of equity value in individual companies, but are relying on the 40% to make the price discovery, and they are trusting historical data and trending mantra.
Whether the 40% take this into account, whether the 60% overtly influence real equity values without direct research is another debate, but I would claim that someone like @chiang_mai, with their detailed analysis threads, probably does 'know' something 60% of investors don't.

Detailed analysis of single components is all well and good but this topic desparately needs the likes of @masonic and @OldScientist (and others) to put findings into context, because posters like them understand the 360 degrees, whereas I don't. 

masonic

While all this data is informative, I think it is important to recognise the limitations of any analysis performed on it. I am generally less concerned about what other people are investing in, but the data does tell an interesting story. If 60% of inflows are passive, yet the average UK investor still has a massive home bias and low US exposure, it suggests they aren't just buying global trackers. Instead, active allocation decisions are being made elsewhere - likely by pension default fund managers, robo-firms, and multi-asset fund houses. OTOH, the US has a large population, a greater proportion of whom invest in stocks, and probably larger sums on average, which could drive country weightings the other way.
Data from the Opinium/IA survey suggests that UK investors (particularly younger ones) actually want a degree of home bias, so we may just be seeing the market provide what is being asked for. Interestingly, this is happening while pension funds have reportedly spent the last decade reducing UK exposure, through a long period of UK underperformance. That shows that even professional asset managers struggle with timing.
What I find more interesting are valuation models, but these have serious limitations. We have the CAPE ratio, which is a fairly blunt instrument and cannot easily differentiate between irrational exuberance and a structural shift in the market. Today's capital-light/intangible-heavy tech companies have high profit margins, moats, and massive share buybacks that make them look expensive by historical standards almost all of the time. The Vanguard Capital Asset Pricing Model is more sophisticated, but it still only gives us probabilistic outputs with large uncertainties. It has been predicting low US growth for two years now, but we won't know if it was actually right for another eight years.
In my recent shift to a 60/40 allocation (and tilt away from the US), I've tried to keep these unknowns at the front of my mind. I don't have high conviction in any prediction. In adjusting my portfolio, I am not necessarily betting against AI or even predicting a crash. I simply view the current risks as outweighing the potential returns and I've reflected that uncertainty in the extent of my tilt. For me, overconfidence in the models is also a significant risk. But the ability to obtain a risk free ~2% real return outside of equities allows me to be wrong and still meet my objectives.

chiang_mai

The Goldman Sachs Research paper that is linked in the OP is seriously interesting. Those guys constructed a global portfolio model dating back to 1950 and then modelled the changes from then until now. The global portfolio in this context is, all the asset classes that are investable. The model considers two time frames, 1950 to 1990 and 1990 until today, peridocially little gems appear but it's slow reading so I've some way to go yet. One of those little gems is that, since 1990, equities have had on average the largest weight of the global portfolio at 48%, followed by global bonds (ex credit) at 37% and credit markets at 8%". Note, that's 48% equities, not 80%, 90% or 100%.

Another gem is, "Since the GFC US equities have materially outperformed non-US markets, helped by rising corporate profitability, especially for the large-cap US Tech sector but also a stronger Dollar". It appears that USD has weakened in recent times and that there is little reason to believe it will strengthen again, any time soon, not as long as governement debt keeps increasing and there remaions no viable plan to lower it.

.

OldScientist

Not a detailed analysis, but a comment on the source of the data for the original post (i.e., the Goldman Sachs paper, "Investing in Everything, Everywhere, All at Once" linked at https://elements.visualcapitalist.com/wp-content/uploads/2025/10/1760613583035.pdf ).

We can find the original graphs (exhibits 29 and 30) at the bottom of page 14 of the GS report. It would appear that the source of these data are "Based on national accounts, central bank and Fed data" and therefore, I think, includes all investors, i.e., retail, pension (DB and DC), and insurance.

Taking the UK as an example, - the 63% allocation to domestic bonds (I am not sure whether this includes corporate as well as government bonds - probably yes) may very well be driven by the strong, but declining, presence of DB pensions who represent a large customer for gilts (which then allow liabilities to be matched).

Home bias in equities may also be to some extent dependent on this since DB pensions and insurance companies invest in domestic private equity and infrastructure and some modern lifestyle funds (e.g., Vanguard Lifestrategy) also have a home bias.

I think other people have already said, but the global weightings are 'decided' by the global investor who will be made up of elements of the US, UK, Japan, etc. according to the weight of investments made in each of those countries. The well-known home bias of US investors (there are large threads at bogleheads) coupled with the US probably(?) being one of the largest sources of investments then results in a bias towards the US. Of course, speculative money will lead/follow trends, e.g., Japan in the 80s, dotcom in the 90s, tech in 20s(?), etc.

chiang_mai

masonic said:

While all this data is informative, I think it is important to recognise the limitations of any analysis performed on it. I am generally less concerned about what other people are investing in, but the data does tell an interesting story. If 60% of inflows are passive, yet the average UK investor still has a massive home bias and low US exposure, it suggests they aren't just buying global trackers. Instead, active allocation decisions are being made elsewhere - likely by pension default fund managers, robo-firms, and multi-asset fund houses. OTOH, the US has a large population, a greater proportion of whom invest in stocks, and probably larger sums on average, which could drive country weightings the other way.
Data from the Opinium/IA survey suggests that UK investors (particularly younger ones) actually want a degree of home bias, so we may just be seeing the market provide what is being asked for. Interestingly, this is happening while pension funds have reportedly spent the last decade reducing UK exposure, through a long period of UK underperformance. That shows that even professional asset managers struggle with timing.
What I find more interesting are valuation models, but these have serious limitations. We have the CAPE ratio, which is a fairly blunt instrument and cannot easily differentiate between irrational exuberance and a structural shift in the market. Today's capital-light/intangible-heavy tech companies have high profit margins, moats, and massive share buybacks that make them look expensive by historical standards almost all of the time. The Vanguard Capital Asset Pricing Model is more sophisticated, but it still only gives us probabilistic outputs with large uncertainties. It has been predicting low US growth for two years now, but we won't know if it was actually right for another eight years.
In my recent shift to a 60/40 allocation (and tilt away from the US), I've tried to keep these unknowns at the front of my mind. I don't have high conviction in any prediction. In adjusting my portfolio, I am not necessarily betting against AI or even predicting a crash. I simply view the current risks as outweighing the potential returns and I've reflected that uncertainty in the extent of my tilt. For me, overconfidence in the models is also a significant risk. But the ability to obtain a risk free ~2% real return outside of equities allows me to be wrong and still meet my objectives.

My major worry in these discussions is the novice investor who buys a global tracker and forgets about it, because they think it will do everything for them and they don't have to know or do anything. I think this is a throw back to a distorted Warren Buffet message that was taken out of context, combined with media messages to begin investing the simple low risk way.

 I don’t think it’s sufficient to simply decide on passive versus managed investment funds and then go ahead and buy something. Actually, I'm not sure many people conciously make that choice, I suspect with many it's a bandwagon they simply jump on, without really understanding all the pro's and con's. Whichever appropach a person decides on, they have to look under the hood and understand what they're buying. They might decide that buying a passive global index tracker will give them the world that is low risk but this is not always as true as it first appears. Global trackers can include anywhere from 60% to 82% US equties allocations, Emerging markets and small and medium, capitalised companies. If they don’t know which they want and why, I’d question whether they should be buying in the first place? 60%/82% US, when equities represent only 48% of investable global assets, does the investor really agree with that massive tilt, let alone understand it?  

 Investment portfolio’s contain so many variables that trying to maintain balance of all of them is very difficult.  An investor might balance for geography and the percentages invested in different markets but then there’s capitalisation balance as well as sector, risk and asset class. One small tweak in one particular area can easily cause other areas to become unbalanced, depleted or concentrated.  My experience is that it’s not enough to identify a good investment fund, you also have to see if it will fit within your portfolio and what the impact might be of acquiring the fund, on your other holdings.

 Of course, if an investor buys an off the shelf tracker, they  probably don’t need to worry too much about balancing its components, because the tracker will do some of that for them. But it wont balance geographies to the extent they might want them to be balanced, which mean they must check and make sure they're happy with them from the outset. It's slightly ironic that investors can have strong views on say the US and its future economy, yet not make the connection between that thinking and the US dominant tracker, in which they've tied up their financial future.  If they're not happy with 70% in the US and 3% in Japan, they either need a different tracker or will have to start adding other things, in order to achieve the desired balance (see the previous para).

 The same is true of bonds, it isn’t enough to simply buy bonds or a bonds fund, buyers have to understand what they're buying and why. Corporate vs Government, short duration vs intermediate and long, high credit rating, medium low or junk, etc. I'll stick my neck out here and say I suspect that a large portion of the markets growth we see is from uninformed investors who are following the crowd, without knowing where it's going or why.  

OldScientist

I think the advocacy for a global cap weighted portfolio in equities is largely derived from Jack Bogle's work as then applied to the US stock market and derived from the fact that the average investor in any given market will receive the average returns of the market less costs (e.g., see https://johncbogle.com/speeches/JCB_MS0205.pdf ).

If a retail investor is willing to modify their holdings in equities in response to events or 'sentiment', to consistently earn a return above a 'global' fund (and as you point out there are a number of different indices) or with lower volatility their predictions of future events would have to be largely correct. For example, if the US stock market consistently has a lower return than ex-US over the next few years, then those who have reduced their US exposure will win their bet while otherwise they will not.

I agree that bond funds in general do need some digging into since 'cap weighting' in bond indices (i.e., issue weighting) is a rather odd feature of global bond funds (IMV, it makes more sense for single countries) since the most heavily indebted countries have the largest weighting. The fund duration is of particular importance - short duration, generally lower return and lower volatility, long duration generally higher return but greater volatility with intermediate duration somewhere between the two.

badger09

Interesting thread. 

Which online tools (available to all) do posters use to analyse weighting of their portfolios?