Australian (ASX) Stock Market Forum

Equal weight investing

Joined
26 October 2008
Posts
2,931
Reactions
7
I am pretty excited today, because I realised that an equal weight ETF (MarketVectors Equal Weight) has come to Australia (finally).

Why does this excite me? Lots of reasons:

1. Diversification - the benchmark index S&P ASX 200 covers 80% of the market cap of the ASX. The top 10 holdings account for ~50% of that size. So when you buy an index fund like STW, you are essentially buying the top ten stocks (by market cap) with half your money and the remaining 190 stocks with the other half.

To give you an idea of how bad the issue of concentration is on the ASX, if you look at the S&P ASX 300, which adds an additional 100 stocks to the universe, this only covers an additional 1% (i.e. 81%) of the market cap of the ASX.

The equal weight ETF does not suffer from this issue, and in fact has the "opposite problem" (which I will get to in the next point).

If you research quantitative finance a lot, you will find that diversification is often referred to as one of the few "free lunches" that can increase returns and reduce volatility.

The other well recognised free lunch is:

2. Rebalancing -

Rebalancing in the benchmark (market cap weighted) index is a passive outcome, the portfolio manager doesn't actually have to do anything except replace stocks which are no longer in the top 200 or 300 based on market cap. This can be equated to "sell your losers and buy your winners".

Rebalancing quarterly to equal rather than market weights renders an outcome which is the opposite of that. i.e. "sell your winners and buy your losers". Maybe this sounds like a counterintuitive thing to do, but as research has shown, this sort of rebalancing actually generates its own returns (see http://gestaltu.blogspot.com.au/2012/02/volatility-harvesting-and-importance-of.html).

So now we are at two free lunches! Great, but actually it gets better. Thanks to the diversification and rebalancing we also get:

3. Small Minus Big (SMB) factor exposure -

Rebalancing quarterly to equal weight means that the largest stocks in the portfolio are (at the beginning of each rebalancing period) given the same allocation as the smallest stocks. This naturally increases the SMB factor exposure of the portfolio, which over the long run should provide extra returns from the factor.

If you don't know much about SMB factor, you can find out a little bit here http://blog.alphaarchitect.com/2014/07/02/does-the-size-effect-exist/

4. High Minus Low (HML) aka "Value" factor exposure -

Rebalancing quarterly to equal weight means that the most expensive stocks in the portfolio are (at the beginning of each rebalancing period) given the same allocation as the cheapest stocks. This naturally increases the HML factor exposure to the portfolio, which, again, should provide extra returns from the factor.

If you don't know much about the HML factor you find out some stuff here http://blog.alphaarchitect.com/2015...-tools-to-understand-what-drives-performance/

To give you an idea of how all of the above might add up, I just quickly screenshot the ETF holdings (based on 31/05/15)

mvw.png

and compare that with STW

stw.png

So, to me, that is awesome stuff to finally see the ASX offer investors what you can already get overseas in most other developed markets.

Other (less exciting but still tangible) benefits:
* The fund isn't actually a value or smallcap fund, it's "fundamentals agnostic" it just naturally heavier tilts towards small/value.
* Heavy exposure to market return factor (i.e. stocks return more than risk free assets)
* Some exposure to momentum factor (i.e. stocks in motion will continue to be in motion) between rebalancing periods.
* Management fee is not horrible (0.35).
* Buy/hold this ETF for factor exposure is much more tax, time and compounding friendly than manually running the portfolio yourself.

I personally look forward to seeing more ETFs like this come to the ASX, since it makes my life as an investor much easier. I would like to see a momentum ETF, and some value ETFs that aren't based on "high yield" stocks (there is the Russell Large Value, ETF which is good, but would be great to see a Small Value too). There is a minimum volatility ETF for US largecaps but they couldn't help themselves and had to make it "high yield" (ANZ ETFS S&P 500 High Yield Low Volatility ETF) which annoys me.
 
Agreed its a good concept.

That being said Sinner over the past 3-4 years I would imagine an ASX200 ETF would have outperformed an equal balanced ETF? I cant find the chart however there's been a big discrepancy in performance between the Big Ords and the Small Ords with the Big Ords grossly out performing. Does this development surprise you and I get the feeling you would expect them to 'return to the mean' over a long horizon?

Informative post however and one I would think many with large super balances should consider
 
I am pretty excited today, because I realised that an equal weight ETF (MarketVectors Equal Weight) has come to Australia (finally).

Why does this excite me? Lots of reasons:

1. Diversification - the benchmark index S&P ASX 200 covers 80% of the market cap of the ASX. The top 10 holdings account for ~50% of that size. So when you buy an index fund like STW, you are essentially buying the top ten stocks (by market cap) with half your money and the remaining 190 stocks with the other half.

To give you an idea of how bad the issue of concentration is on the ASX, if you look at the S&P ASX 300, which adds an additional 100 stocks to the universe, this only covers an additional 1% (i.e. 81%) of the market cap of the ASX.

The equal weight ETF does not suffer from this issue, and in fact has the "opposite problem" (which I will get to in the next point).

If you research quantitative finance a lot, you will find that diversification is often referred to as one of the few "free lunches" that can increase returns and reduce volatility.

The other well recognised free lunch is:

2. Rebalancing -

Rebalancing in the benchmark (market cap weighted) index is a passive outcome, the portfolio manager doesn't actually have to do anything except replace stocks which are no longer in the top 200 or 300 based on market cap. This can be equated to "sell your losers and buy your winners".

Rebalancing quarterly to equal rather than market weights renders an outcome which is the opposite of that. i.e. "sell your winners and buy your losers". Maybe this sounds like a counterintuitive thing to do, but as research has shown, this sort of rebalancing actually generates its own returns (see http://gestaltu.blogspot.com.au/2012/02/volatility-harvesting-and-importance-of.html).

So now we are at two free lunches! Great, but actually it gets better. Thanks to the diversification and rebalancing we also get:

3. Small Minus Big (SMB) factor exposure -

Rebalancing quarterly to equal weight means that the largest stocks in the portfolio are (at the beginning of each rebalancing period) given the same allocation as the smallest stocks. This naturally increases the SMB factor exposure of the portfolio, which over the long run should provide extra returns from the factor.

If you don't know much about SMB factor, you can find out a little bit here http://blog.alphaarchitect.com/2014/07/02/does-the-size-effect-exist/

4. High Minus Low (HML) aka "Value" factor exposure -

Rebalancing quarterly to equal weight means that the most expensive stocks in the portfolio are (at the beginning of each rebalancing period) given the same allocation as the cheapest stocks. This naturally increases the HML factor exposure to the portfolio, which, again, should provide extra returns from the factor.

If you don't know much about the HML factor you find out some stuff here http://blog.alphaarchitect.com/2015...-tools-to-understand-what-drives-performance/

To give you an idea of how all of the above might add up, I just quickly screenshot the ETF holdings (based on 31/05/15)

View attachment 63122

and compare that with STW

View attachment 63123

So, to me, that is awesome stuff to finally see the ASX offer investors what you can already get overseas in most other developed markets.

Other (less exciting but still tangible) benefits:
* Heavy exposure to market return factor (i.e. stocks return more than risk free assets)
* Some exposure to momentum factor (i.e. stocks in motion will continue to be in motion) between rebalancing periods.
* Management fee is not horrible (0.35).
* Buy/hold this ETF for factor exposure is much more tax, time and compounding friendly than manually running the portfolio yourself.

I personally look forward to seeing more ETFs like this come to the ASX, since it makes my life as an investor much easier. I would like to see a momentum ETF, and some value ETFs that aren't based on "high yield" stocks (there is the Russell Large Value, ETF which is good, but would be great to see a Small Value too). There is a minimum volatility ETF for US largecaps but they couldn't help themselves and had to make it "high yield" (ANZ ETFS S&P 500 High Yield Low Volatility ETF) which annoys me.

Only concern with some of these is very low FUM, and the turnover you get compared to market cap weighted ETFs.

Personally, I really like the Research Affiliates methodology, so QOZ and QUS.........should outperform over the very long term provided you can stick with it. When the alternative indexes underperform its hard for most investors to hold on.
 
Agreed its a good concept.

That being said Sinner over the past 3-4 years I would imagine an ASX200 ETF would have outperformed an equal balanced ETF? I cant find the chart however there's been a big discrepancy in performance between the Big Ords and the Small Ords with the Big Ords grossly out performing. Does this development surprise you and I get the feeling you would expect them to 'return to the mean' over a long horizon?

Informative post however and one I would think many with large super balances should consider

3-4 years is not really a long enough horizon to evaluate anything in terms of performance of various strategies.

Yes, ASX smalls have underperformed. I think mainly due to two separate situations:
* "Reach for Yield" by global investors who don't like 0% interest rates, when these global investors put money in the ASX, the top 50 stocks are pretty much the only ones which could take the liquidity, so this is where the global money is going. Lots of demand for "low vol", "high yield", and other "smart beta" products are also driving this demand for bigs.

* Industrial metals bear market, the ASX smalls is heavily weighted to miners and mining services, so we got a business cycle downturn in smalls along with macro cycle demand for bigs.

Yes, you would expect that over the long term, the performance of bigs to lag. If you are looking for examples of this behaviour, check out relative valuations between US bigs and smalls during the 2000 tech bubble peak and corresponding 10Y returns from then.
 
Only concern with some of these is very low FUM, and the turnover you get compared to market cap weighted ETFs.

Personally, I really like the Research Affiliates methodology, so QOZ and QUS.........should outperform over the very long term provided you can stick with it. When the alternative indexes underperform its hard for most investors to hold on.

I did look at QOZ, but from my understanding, "smart beta" products based on fundamental weights actually didn't outperform?

I did mention this in an edit to the original post, equal weight is "fundamentals agnostic" and I think that is worth something.

Do you really think this selection is going to meaningfully outperform after management fees?

Screenshot.png
 
I did look at QOZ, but from my understanding, "smart beta" products based on fundamental weights actually didn't outperform?

I did mention this in an edit to the original post, equal weight is "fundamentals agnostic" and I think that is worth something.

Do you really think this selection is going to meaningfully outperform after management fees?

View attachment 63124

Not sure where your understanding comes from. Refer attached, I've also seen index performance much further back than that. And yes I do believe that RAFI indexes will outperform market cap over the long term.

I don't believe the small cap premium exists in Australia, that end of the market is filled with cyclical rubbish.

My view; in Oz for mid/small I'd much prefer active management for that exposure, if you feel you want it.
 

Attachments

  • FRAU200_20150529.pdf
    567.2 KB · Views: 7
Not sure where your understanding comes from.

I don't have the data in front of me, but from what I recall, out of "Market Weight" CRSP, "Value Weight" CRSP and "Equal Weight" CRSP since 1970s, market weight and value weight had pretty much identical performance (value weight outperformed by a little but, but not statistically significant and not before fees). Meanwhile the equal weight outperformed consistently by a statistically significant amount.
 
I think (just a guess) the problem with value weight is most value ratios aren't systematic, the outpeformance of a given ratio is almost always best noticed in the spread between "value" and "glamour" deciles, deciles in the middle of the spread don't necessarily outperform the decile to their right or underperform the decile to their right.

Value weighting kind of ignores the spread, the weightings leave the value decile relatively underweight and the growth decile relatively overweight.
 
I don't have the data in front of me, but from what I recall, out of "Market Weight" CRSP, "Value Weight" CRSP and "Equal Weight" CRSP since 1970s, market weight and value weight had pretty much identical performance (value weight outperformed by a little but, but not statistically significant and not before fees). Meanwhile the equal weight outperformed consistently by a statistically significant amount.

Here we go, RAFI vs EW vs Cap Weight, 1985-2013 development markets

http://www.researchaffiliates.com/P...nd_Fundamental_Weight_Index_Investing_PDF.pdf

EW underperformed both RAFI and Cap weighted indexes in Australia over this period.

Take a look at the RAFI methodology, there is more to it than a typical value weight index.

Good discussion, this is an area of interest for me :)
 
Here we go, RAFI vs EW vs Cap Weight, 1985-2013 development markets

http://www.researchaffiliates.com/P...nd_Fundamental_Weight_Index_Investing_PDF.pdf

EW underperformed both RAFI and Cap weighted indexes in Australia over this period.

Take a look at the RAFI methodology, there is more to it than a typical value weight index.

Thanks that is interesting data, although I do usually like to see equity curves to get a good understanding of where/when the performance is coming from.

It is possible that I'm attributing general value weight concerns to the RAFI methodology which I'm not aware of but will look into today (thanks again).
 
I've always been a big fan of the RAFI Fundamental Methodologies as a way to get 'core' exposure to particular markets, while at the same time potentially generating outperformance. My biggest issue with equal weight is just how representative it is of the Australian or US economy. Assuming you want to invest in a particular market you are trying to get some form of representation rather than a random selection of stocks from a weighting perspective.

Also found this document which may be useful to some
View attachment 201405 Equal-Weight and Fundamental-Weight Memo.pdf
 
6 month % performance for:

* Agnostic equal weight constituents approx = MSCI Australia (MVW.AX)
* Market cap weight S&P ASX300 (VAS.AX)
* Bleeding edge fundamental weight RAFI magic constituents = S&P ASX 200 (QOZ.AX).

Screenshot_2016-02-10_12-00-42.png
 


Write your reply...
Top