Asset Allocation

Thanks for a very enlightening post. Does anyone see a case for buying sector specific funds while keeping region specific ones (Aus-US-Rest of the world) as a bigger percentage of the portfolio?
When i put in a test portfolio of region specific funds in portfoliovisualizer for example and compare it to another portfolio with some sectors thrown in (tech, health, consumer staples and discretionary) i get a better historic return with the sectors thrown in than without.
Does anyone have a view on this?

 

You may just be picking sectors that happened to have outperformed over an arbitrary time period (likely tech and health). Why not go all in on NDQ?

If you have a well diversified portfolio, you already own the stocks from those sectors. By betting on specific sectors you'd be increasing idiosyncratic risk and the complexity of your portfolio. So extra funds to worry about for lower/no extra benefit.

 

@Gypsyblood

The 2019 S&P 500 Sector Quilt

 

Given the inherent urge for humans to fiddle / optimise / think they’re smarter than the market (generally equates to stuff it up) the above statement equates to:

 

This article was excellent! Helped visualize it so much better!

If you look at S&P 500 index in the mix of all the sector based indexes in this article, would you argue that buying 100% of S&P 500 would have averaged a "lower" return than if someone had bought all the Sector indexes as well as S&P 500 in equal weights?

 

Yes, I do feel like I am complicating it. Also not quite clear if the complication will make things better or worse yet.. trying to understand it better. I do appreciate your insight as someone having gone through the journey!

 

Looking to construct a portfolio from VGE, VAS, VGS and VAE.

This is sort of where I'm at with respect to the minimum and maximum ranges in the composition.

30-50% VGS
20-30% VAS
20-30% VAE
10-20% VGE

There seems to be significantly more allocation to VAE/VGE than those that have been through multiple cycles. Why might 30-50% in VAE/VGE be too much?

Another issue is that US makes up such a large chunk of VGS, and is at extreme valuations. Would it make sense to underweight VGS (ie: lower end of the desired range) at this point in time?

 

Overlap, why?

 

What I would do:

70% VGS / VGAD
20% VAS (maybe even lower if you own other assets like property in Australia)
10% VGE (no need to go above its world market weight)
0% VAE (overlaps with VGE)

I would decide on the allocation and not worry about valuations.

 

In a vacuum, no idea what will do better.

In reality, I reckon 99% will be better off with just a total market fund.

 

Happy with having more exposure to the overlapped countries especially China, India. VGE + VAE for exposure to all the smaller weighted countries in each.

Perhaps a bit of a non-consensus view, but have conviction either China or US will be the dominant country in my investing lifetime (currently early 30s). At the same time, don't have conviction on who will emerge the winner, so happy to allocate roughly equally to each.

 

Can I please have the names of the 1.3% companies that produce outsize returns. Happy to skip the 98.7% that just match US 1 month treasuries return

FYI

1) VAS (S&P/ASX 300) as of 31st Aug 2020
Price/Earnings Ratio = 19.17
Price/Book Ratio = 1.93
Return on Equity = 13.16%

2) IVV (S&P 500) as of 21st Sep 2020
Price/Earnings Ratio = 22.48
Price/Book Ratio = 3.55

3) VAE (FTSE Asia ex Japan Shares Index) as of 31st Aug 2020
Price/Earnings Ratio = 16.84
Price/Book Ratio = 1.69
Return on Equity = 15.25%

4) IJH (S&P US Mid Cap) as of 21st Sep 2020
Price/Earnings Ratio = 18.59
Price/Book Ratio = 2.03

5) IJR (S&P 500 Small Cap) as of 21st Sep 2020
Price/Earnings Ratio = 15.38
Price/Book Ratio = 1.52

US small caps and VAE are the cheapest based on fundamentals like P/E and P/B.

Surprisingly, VAS even though it is 16.5% below its all time high is not cheap by historical standards.

(PS: The above P/E values are all trailing 12 months. Stock markets are forward looking and don't really care about the past. So depending on what the forward P/E values are S&P500 might look expensive based on trailing P/E but could actually be cheap compared to say VAS when looking at forward P/E values)

Cheers,
Oracle.

 

Do Global Stocks Outperform US Treasury Bills? by Hendrik Bessembinder, Te-Feng Chen, Goeun Choi, K.C. John Wei :: SSRN

Page 44 will give you the top 50.


VEU:
P/E = 17.91
P/B = 1.58

IJS:
P/E = 11.64
P/B = 1.07

 

Nice

Disappointed to find out although we are all overweight ASX none of the ASX companies make it in the top 50. Cut off is $152billion USD. CSL misses out.

Can we invest in IJS here. Not sure if iShares offers that ETF on ASX.

Cheers,
Oracle.

 

O.K. grump time……


I keep seeing this 1.3% figure, and although @orcle hasn’t explicitly said it and neither did Ben Felix, how it normally gets interpreted is that 1.3% of stocks outperformed bond returns.


What the paper reported was that approx. 40% outperform bonds. But the aggregate out performance of 38.7% of them is required to offset the 60% that under performed hence leaving 1.3% of companies equaling total out-performance.

But there are more issues with the paper than just people mis-interpreting the results as displaying extreme skewness of 1.3% out performers.


The methodology was to calculate assuming dividend re-investment and that’s not how the real-world works. Take something like General Motors for example that went into chapter 11 bankruptcy – it would show up as total wipe-out and hence an underperformer against bonds yet over its time it added lots of economic value to the economy. Everybody didn’t re-invest their dividends back into GM and if they did the company would be fat with cash and not insolvent post GFC. Those dividends had economic value and got spent or re-invested somewhere else, maybe even into Amazon….. and Amazon is one they put in the 1.3% outcome but how does that story end? Will all the current mark to market value be realisable in actual economically usable future cash flow to investors?


Academic methodology to make the math possible, that doesn’t accord with the real world on this one….. tread carefully with what you conclude from it.

2c

 

Hmmm, I see where the confusion can come from.

How I interpret the study: for each stock you pick that out-performs, the next you pick will most likely cancel out that return entirely. It reiterates the point of owning the entire market, and not relying on managers to "pick out the bad stocks" (good luck with that).

Good point on General Motors. If someone was drawing down on a portfolio that included GM at the time, they would be benefiting from that wealth creation which is not captured by this study. As always we tunnel vision by looking at specific periods in time.

 

Yes am slowly unwinding one of my RE trusts which gives us an option to look at it early 2021 - It has done well but am interested in seeing what my options are then plus will be offloading a property once another one is completed in 2022

In terms of capital gains not sure if it fortunate or not but we have some losses that are carrying forward that we can assist in offsetting some of the gains - its a hangover of GFC and investing with a 'superstar who had a foolproof strategy involving options calling" - wish I knew about passive index investing then

 

An Australian domiciled low cost total market fund would be nice. VDHG may also progressively suit better approaching and through retirement with reduced risk including that through hedging (selling down will be forced but currency can be volatile). Satellite holdings can be used for high conviction.

 

No. I emailed iShares about it and they have no plans to introduce it here. I don't either bother asking Vanguard Australia about VIOV (their equivalent)