Asset Allocation by Geography

My point is that your objective should be to minimise risk across your portfolio.

Geographical diversity is one criterion to consider, but not the only one.

I don't think anyone did say it was the only one but I am also not sure how a property owner in Ireland with a pension fund invested in global equities is reducing his portfolio risk by buying brazillian forestry just because apparently it is not correlated.
 
@NoRegretsCoyote
With all due respect, you might re-read your post, particularly the bit I quoted.

Or you might expand: how could investment in a single equity be less risky than investment in a weighted global basket of equities? Which is what you said - even if you meant something completely different.
 
@NoRegretsCoyote
With all due respect, you might re-read your post, particularly the bit I quoted.

Or you might expand: how could investment in a single equity be less risky than investment in a weighted global basket of equities? Which is what you said - even if you meant something completely different.

Geographical diversity is a necessary, but not sufficient, condition for risk minimisation.

@Sarenco argued that he/she didn't know what the future held, therefore for investment success (his/her objective) it made sense to diversify geographically. My (facetious) point was that if you don't have any idea what patterns hold at all, ever, then you don't need to bother diversifying.

Obviously a basket is less risky than a single stock. I never said it wasn't. I just don't think geographical diversity should be an objective purely for its own sake.
 
@Sarenco argued that he/she didn't know what the future held, therefore for investment success (his/her objective) it made sense to diversify geographically. .
That's not what I argued.

I aim to be a diversified as possible across global stocks because I've no idea what companies, sectors or regions will succeed in the future. So, I simply hold the entire market - regardless of where companies are headquartered or listed.

I don't know what is going to happen in the future so I diversify both within and across the major asset classes. In other words, I'm hedging my bets.
 
I've no idea what this means or what patterns you are talking about.

Could you elaborate?

I am talking about correlations (negative and positive) between asset classes. If I wanted to minimise risk, and I had lots of shares in a Spanish builder then I would not buy lots of shares in a Spanish bank, because the returns tend to be correlated.

I don't know what is going to happen in the future so I diversify both within and across the major asset classes. In other words, I'm hedging my bets.

I just see a contradiction here. You claim you have zero idea about what will happen in future. But you are adopting an investment strategy which hinges on the assumption that certain correlations between asset classes will hold in future.

I think your investment strategy makes sense by the way. I just don't think that geographical diversity is an end of itself.
 
geographical diversity

Is Alphabet (aka Google) an American company? Is BP a British/Dutch company? a lot of the world's biggest companies by stock valuation are already geographically diversified if you look at their sales and profits by region - so spreading your capital around different markets doesn't make a lot of difference as to where the earnings are coming from imho
 
You claim you have zero idea about what will happen in future. But you are adopting an investment strategy which hinges on the assumption that certain correlations between asset classes will hold in future.
No, that's not what I'm assuming.

Correlations vary considerably through time – they are not remotely constant. I have no idea what correlations will prevail in the future – nor does anybody else.

I diversify between and across asset classes simply because I have no idea what assets will outperform over my investment horizon. I might have an expectation that certain asset classes will outperform others but the markets are under no obligation to meet my expectations. It's simply a case of not wanting to put all my eggs in the one basket.

If it turns out that there is a low correlation between some of my investments that provides a reduction in the return variance of my portfolio, well, that's just a bonus.
I just don't think that geographical diversity is an end of itself.
I don't think anybody said that geographic diversity is an end in itself.

Again, I invest in the entire global equity market because I don't know what companies, sectors or regions will outperform in the future.
 
I am talking about correlations (negative and positive) between asset classes. If I wanted to minimise risk, and I had lots of shares in a Spanish builder then I would not buy lots of shares in a Spanish bank, because the returns tend to be correlated.



I just see a contradiction here. You claim you have zero idea about what will happen in future. But you are adopting an investment strategy which hinges on the assumption that certain correlations between asset classes will hold in future.

I think your investment strategy makes sense by the way. I just don't think that geographical diversity is an end of itself.

I would very interested in how you manage the correlation of your portfolio. There are professional fund managers who spend their days trying to design the most efficient non-correlated portfolio and struggle. As Sarenco says, it is not static over time because it is based on historical returns. It has also shown not to hold up as expected in periods of volatility like the financial crash where nearly all asset classes that were thought to be negatively correlated fell when market volatility spiked.
 
@Sarenco
@Sunny

Suppose there are only three stocks in the world:

Bank of Ireland: an Irish bank
Cairn Homes: an Irish builder
Petrobras: a Brazilian oil producer

You want a balanced portfolio, but can only afford to buy two of the stocks. Which two do you pick and why?
 
Can I say none?:cool:

I've no interest in investing any material part of my retirement savings in just two stocks.

Nobody has suggested that correlations don't exist. But I don't invest on the basis of any projected future correlation between assets, which you keep insisting for some reason.
 
Nobody has suggested that correlations don't exist. But I don't invest on the basis of any projected future correlation between assets, which you keep insisting for some reason.

I don't know which way equity markets will go in general (up or down).

I do think that there are good rules of thumb about which assets generally have correlated returns. They don't hold all the time, but they do to some extent. Personally I would use these rules of thumb to construct a balanced portfolio.
 
Asset allocation by geography. How much does it matter?

If you invest in the MSCI World Index you are not asset allocating by geography, you are investing in 23 developed equity markets. https://www.msci.com/world and https://www.msci.com/documents/1296102/1362201/MSCI-MIS-USA-Dec2018-Brochure.pdf/7cf87d48-fccf-372e-0ac4-f2fa9f208fb7.

Obviously keeping an anchor to the wind in terms of home currency is important. However, I come across critiques of portfolio construction in terms of how under or over weight they are in comparison to the MSCI World Index.?

Constructing a portfolio underweight / overweight to the MSCI World Index presumably means adding other asset classes, e.g. emerging market equities, property, commodities, timber, debt, etc. that are not captured by this index. You could do this, but it would not be wise.

How you allocate your investments between different asset classes is the most important decision you can make in determining if you will meet your investment goals. This entails uniquely investing in a set of assets to develop an optimal portfolio, i.e. the one that cannot generate a higher rate of return relative to your personal risk profile.

It's difficult to see how being underweight / overweight relative to the MSCI World Index (i.e. in essence underweight / overweight relative to developed market equities) could assist you in developing an optimal portfolio. As best you could invest in a financial instrument that tracks this index as your means of investing in developed market equities. But your allocation to other asset classes should be determined by your optimal portfolio.
 
Last edited:
If you invest in the MSCI World Index you are not asset allocating by geography, you are investing in 23 developed equity markets. https://www.msci.com/world and https://www.msci.com/documents/1296102/1362201/MSCI-MIS-USA-Dec2018-Brochure.pdf/7cf87d48-fccf-372e-0ac4-f2fa9f208fb7.



Constructing a portfolio underweight / overweight to the MSCI World Index presumably means adding other asset classes, e.g. emerging market equities, property, commodities, timber, debt, etc. that are not captured by this index. You could do this, but it would not be wise.

How you allocate your investments between different asset classes is the most important decision you can make in determining if you will meet your investment goals. This entails uniquely investing in a set of assets to develop an optimal portfolio, i.e. the one that cannot generate a higher rate of return relative to your personal risk profile.

It's difficult to see how being underweight / overweight relative to the MSCI World Index (i.e. in essence underweight / overweight relative to developed market equities) could assist you in developing an optimal portfolio. As best you could invest in a financial instrument that tracks this index as your means of investing in developed market equities. But your allocation to other asset classes should be determined by your optimal portfolio.


I think you need to re-read the original post.

Also, you’ve misrepresented a statement by inserting a question mark where there was a full stop.
 
Back
Top