Rational data driven investment advice - the all world stock index

[steveo73]

My take is that an all world stock index (the most diversified stock index you can buy) will outperform any other investment strategy from a risk adjusted perspective over the longer term.

Does anyone have any data or facts to refute this statement ?

Who has the king of all investment vehicles in their portfolio ?

I do - it makes up about 50% of my portfolio. I have about 30% in my home country stock indexes and 20% in bonds and cash.

What I love about this is that once you've purchased this investment vehicle you know that you are going to beat the pants off the average investor. You are automatically in the top end of the bell curve. It feels good.

[Bonde]

"Talk about inverting expectations! Accounting for both investing uncertainty and the effects of compound returns, betting on the higher risk of the Total Stock Market rewarding you with superior long-term returns is no sure thing. In fact, even over long 15-year periods that would stretch the patience of even the most stoic investor, the highest-risk portfolios like the Total Stock Market have sometimes lagged even the lowest risk options such as the Permanent Portfolio. Once you think beyond over-simplified mathematical assumptions, the relationship between risk and return gets a lot more complicated."
https://portfoliocharts.com/2019/02/25/ ... you-think/

Due to taxation in Denmark I cannot profitable invest in gold or other commodities. We invest mostly in a World ESG index and around 20% in bond index. We also own a world index but we bought it before we changed our strategy to ESG for ethical reasons.

[zbigi]

I share the worry of Buffet and Munger that the accounting practices in most countries in the world are basically not up to very high standards, so the numbers that companies are reporting are often likely to be bogus. Essentially, you don't know what you're buying. Doesn't matter in a bull market (people were buying much riskier/crappier securities after all), but might start to matter when the money gets tight.

[prudentelo]

I held world (not US or residence countries) exclusively since 2012.

Whatever those may say who argue US-only, for most it is only a momentum play. They would not stick to such theories if world had outperformed US for past 10 years.

but 10 year underperformance in stock categories signifies nothing

[prudentelo]

One can also be careful that there are several "world" indexes of which none (to my knowledge) actually represent whole world by market cap.

Some other represent "developed world" (with definition from like 1985, so Korea is "not developed"). I dont think any index represent China at markert cap (due to accounting etc. issues)

[Tyler9000]

My take is that an all world stock index (the most diversified stock index you can buy) will outperform any other investment strategy from a risk adjusted perspective over the longer term.

Does anyone have any data or facts to refute this statement ?

The key qualifying term here is "risk adjusted". How do you define that?

As an example, here's a bookmarked chart showing the risk and return of 19 different portfolios. The portfolios with the best risk adjusted returns are at the top-left. Copy the code and paste it into the cell with the Z, and it will also map a portfolio of 100% developed world stocks on the list. Then you can tweak the risk and return measures to update the chart in real time. One's preferred definitions clearly influence the answer to your question.

From perusing the data, my initial impression is that an all world stock index is indeed a fine option but I wouldn't necessarily put it on a pedestal. And I suspect we agree more than your opening statement might indicate, since you only use it for 50% of your portfolio.

[WFJ]

Several textbooks on Finance and advanced stats required to answer question, but here's a single bite-sized appetizer.

All stocks = market risk, but a lot of trust is being placed on the manager of fund in their definition of "all world". I've reviewed portfolios created under Modern Portfolio Theory that contained weird oil leases/MLPs in Nigeria/Bharain, credit default swaps on exotic futures contracts, oil hedges, shipping rates, weather and similar that were all nearly worthless in "conservative" MPT portfolios for Mom and Pop retail investors. Broker had no idea there were 10,000+ positions in these investments and couldn't explain why the exotic positions were all nearly worthless. I suspect these vehicles were dumped in retail accounts after trades blew up in accounts of HNW individuals but can't prove it.

A better practice is to buy an investment vehicle that is clearly defined "S&P 500" or some other widely accepted index. Better is to self-index and avoid letting someone else vote your shares and avoid radicals being able to pollute corporate boards.

An issue with blind indexes is the weighting of a few companies that defeat the diversification benefits indexes had in the past. There is no solution to this risk beyond self-indexing and reducing size of these positions (APPL, MSFT, AMZN, GOOG, META, NVDA, TSLA).

[steveo73]

The key qualifying term here is "risk adjusted". How do you define that?

Good question. In my opinion the world stock index is the king of all investment options due to the following factors:-

1. Stocks have the best long term growth options out of all investment options. So straight away if you want your assets to beat inflation which is your key long term risk this is the best asset class to do that.
2. The all world stock index is the most diversified option out there.
3. It's self-cleansing longer term. The poor performers drop out and the good ones come in. This is the same as all indexes but the all world stock index is the most diversified option so it's probably even better,
4. There is an added benefit of diversifying your currency risk. This sounds funny because it's an additional risk but it also means it your home country currency goes down you are somewhat protected.
5. Simplicity. You get it all in a single investment vehicle.

From perusing the data, my initial impression is that an all world stock index is indeed a fine option but I wouldn't necessarily put it on a pedestal. And I suspect we agree more than your opening statement might indicate, since you only use it for 50% of your portfolio.

I love your stuff but it suffers from historical data bias. So what has performed well in the past looks good which is fine but it can suffer from data mining bias. The best way around this in my opinion it to take out the principles of good investing namely indexes to give great diversification and diversification across asset classes and taking into account the lifecycle of investing.

You shouldn't put anything on a pedestal but I've yet to see a more rational approach than utilizing an all world stock index as the core component of your long term investing strategy.

Have you watched these investment videos. They are fantastic. https://www.youtube.com/watch?v=_chiIIxMGl0&t=4s

You are right though in that so long as you are following modern portfolio theory there are other options that may suit you. For instance I don't just use an all world stock index but mainly because there are tax benefits in my country from home domiciled stocks.

[steveo73]

but 10 year underperformance in stock categories signifies nothing

Correct. It's noise. If you are investing for a 10 year timeframe I suppose it matters. When it comes to early retirement it's noise.

[steveo73]

One can also be careful that there are several "world" indexes of which none (to my knowledge) actually represent whole world by market cap.

Some other represent "developed world" (with definition from like 1985, so Korea is "not developed"). I dont think any index represent China at markert cap (due to accounting etc. issues)

It's not perfect but it's pretty darn good. There isn't a better single option out there. I wouldn't be 100% stocks unless I had a huge stockpile of money. I probably still wouldn't do that. I'm 80% stocks now though and I'm really confident on that.

[steveo73]

A better practice is to buy an investment vehicle that is clearly defined "S&P 500" or some other widely accepted index. Better is to self-index and avoid letting someone else vote your shares and avoid radicals being able to pollute corporate boards.

An issue with blind indexes is the weighting of a few companies that defeat the diversification benefits indexes had in the past. There is no solution to this risk beyond self-indexing and reducing size of these positions (APPL, MSFT, AMZN, GOOG, META, NVDA, TSLA).

I use Vanguard. They show you the stocks they invest in.

I disagree completely on self indexing. There is no way you can get the same diversification for the same cost. It's unrealistic.

I agree that it would be good to reduce the size of those positions but I don't see that as a big thing. Those stocks will either stay high or go down. A good index will simply adjust it's holdings and longer term you will be fine. That is a key advantage of an index.

There has actually been lots written about the scenario that you are talking about.

[prudentelo]

It's not perfect but it's pretty darn good. There isn't a better single option out there. I wouldn't be 100% stocks unless I had a huge stockpile of money. I probably still wouldn't do that. I'm 80% stocks now though and I'm really confident on that.

I am not saying world indexes are bad. I hold them of course (as I wrote)

Just saying there is not "the world index" like there is "the SP500 index"

Some "world index" exclude MANY countries including many you wouldnt think e.g. because they were considered "poor" in 1985

And no world index yet incorporates all Mainland China at market capital weight, as far as I know, but there is hope to do so.

(also no index in the West has Russia stocks any more ...)




To put it other way, WFJ is exaggerating that just because "World" rather than "SP500" is on the label it is going to contain dead CDOs and other crud, but he is right that buyer must look more closely at what exactly is being bought than just a label. There are reputable world indexes e.g. by MSCI, but they are not "commodity" like SP500 index.

[MBBboy]

What fund are you looking at for your backtesting? The oldest all world index I'm seeing is VT, which only goes back to 2009 and badly lags VTSAX, which is US only.

Asked another way, what data do you have that says all-world has actually outperformed over any sustained time period?

[steveo73]

I am not saying world indexes are bad. I hold them of course (as I wrote)

Just saying there is not "the world index" like there is "the SP500 index"

I get it. It's not perfect. It would be great to have something like an S&P Index for the world as a whole.

It's still a great investment vehicle.

Can you think of a better alternative ?

I think the best argument against a world index is as per Tyler's point regarding creating a portfolio that works best for you that may not include a world index.

[steveo73]

Asked another way, what data do you have that says all-world has actually outperformed over any sustained time period?

Good question. I don't view it like this. I view it as looking at the data (the best I can see is from Tyler's site) and then taking broad principles from that.

As I stated above the world index has numerous advantages over other investment options:-

1. Stocks have the best long term growth options out of all investment options. So straight away if you want your assets to beat inflation which is your key long term risk this is the best asset class to do that.
2. The all world stock index is the most diversified option out there.
3. It's self-cleansing longer term. The poor performers drop out and the good ones come in. This is the same as all indexes but the all world stock index is the most diversified option so it's probably even better,
4. There is an added benefit of diversifying your currency risk. This sounds funny because it's an additional risk but it also means it your home country currency goes down you are somewhat protected.
5. Simplicity. You get it all in a single investment vehicle.

In my home country houses have outperformed stocks. I wouldn't invest any more into housing other than owning my own house. My house is worth a little more than my investment portfolio and I think that is nuts.

It's really hard to pick a specific index and have historical data going back for over a century.

This is the same issue I have with Tyler's site in that you can get caught up in the details rather than the principles derived from data that relates to good investing. Tyler's site is the best. It's awesome. It's just that you need to understand that you are investing for the future 20-50 years rather than picking an specific index and trying to replicate that performance ongoing especially when the data isn't that good.

I view that as being caught up in the trees rather than looking at the forest.

[ertyu]

Asked another way, what data do you have that says all-world has actually outperformed over any sustained time period?

If we are indeed experiencing a structural change in the global economy, what has actually outperformed in the past might not be relevant. Many argue that while US has outperformed over the past 20 years, this will not hold going forward, for instance.

[Tyler9000]

This is the same issue I have with Tyler's site in that you can get caught up in the details rather than the principles derived from data that relates to good investing. Tyler's site is the best. It's awesome. It's just that you need to understand that you are investing for the future 20-50 years rather than picking an specific index and trying to replicate that performance ongoing especially when the data isn't that good.

I don't really have a problem with that viewpoint other than the statement that the data isn't that good. One is free to disagree about how to interpret the numbers, but I'm confident that the data I've collected (very carefully verifying sources and matching methodologies) is as solid as you'll find. I do share some of your skepticism of very old data (100+ years), though. So I agree that quality matters. I'm picky about it, too.

Going back to your original question, I think I was most hung up on your use of the term "risk-adjusted" when talking about portfolio performance. That usually implies very specific things involving historical data (like Sharpe or Treynor ratios), which is why I offered risk & returns data. But now I understand that you are really more interested in investing principles than traditional risk-adjusted returns. That's perfectly fine, but I also think the terms being used might cause a bit of confusion.

[Qazwer]

Theoretically in a open market all assets will have the same risk return or else capital would flow to them. The problem is that the stock market is not the only available assets and it varies country to country as to what businesses and assets are tradeable and capital flow has limits. Therefore in indexing you can only do so much in representing the ideal and possibly trying to replicate the world ideal you might be further off from the correlation of all assets than if you just used a US based undex. Probably though when you get to that level, taxation, costs, luck and personal circumstances might dominate over any strategy. S&P 500 is pretty good if you want to simply index. I personally try to include the world but I am not sure how smart that is or whether it even matters.

[prudentelo]

Back testing is not the strongest argument for world allocation

[WFJ]

I use Vanguard. They show you the stocks they invest in.

I disagree completely on self indexing. There is no way you can get the same diversification for the same cost. It's unrealistic.

I agree that it would be good to reduce the size of those positions but I don't see that as a big thing. Those stocks will either stay high or go down. A good index will simply adjust it's holdings and longer term you will be fine. That is a key advantage of an index.

There has actually been lots written about the scenario that you are talking about.

The firms have to disclose holdings but can hide and use turnover to hide odd holdings. Most exotic investments were in a Wells Fargo retirement plan TDMFs that was loaded with hundreds of busted exotic investments. The other was a managed account pushed by Merrill Lynch or JP Morgan (don't remember) had oil leases and MLP all over the Middle East that were nearly worthless for an investor that specifically said they only wanted US only investments. I've also seen mutual funds pitched to Mom and Pop conservative investors in a taxable account with over 100% turnover that made absolutely no sense. There are no easy solutions to the OPs question that will hold over time.

Self-indexing is easier now with services like Slices where one can easily follow an index removing companies they don't want to invest in and also avoid firms from voting shares to elect board members based on traits rather than competence. Avoiding META, NFLX, TSLA has more than made up for extra costs in small market orders. For example, COST, CSX, DLTR, Dr. Pepper, LULU, MAR, O'reily auto parts, Old Dominion Railroad, PEP, ROSS. SBUX, Kraft-Heinz, WAG and Xcel energy are in Nasdaq 100. Makes no sense for these to be in tech index, but they are for some reason. Unfortunately, the prospectus are long, detailed and flat out painful to read, but the only document that states what an investment has to follow, regardless of what the investor assumes.

The main point is "Buyer Beware" and read the documents if you want to understand what you are buying.