Comments on: Comparing Market Cap ETF vs Dividend ETF – How Much Duplication?

By: Louis

Louis — Tue, 29 Mar 2011 02:23:12 +0000

Mike,
What do you mean when you say that you like to be diversified over the whole globe, so the XIU large cap 60 is good enough?
The XIU is only Canada– what else are you adding?

By: Dividend Growth Investor

Dividend Growth Investor — Fri, 05 Dec 2008 21:33:42 +0000

Actually in the US there also are stocks that are way overweighted in investors portfolio. I did a research for http://www.TheDiv-Net.com today, and found out that the top 10 stocks in S&P 500 account for over 22% of the weight in the index. Not surprisingly, most of those are solid dividend stocks that certain income investors like me hold in their non-fund portfolios..

By: Four Pillars

Four Pillars — Fri, 05 Dec 2008 18:16:46 +0000

Jordan – that definitely isn’t crazy!

I’m not really saying that buying more securities for small amounts of diversification is a waste of time – I’m a little biased because I get tired of having too many securities so I would rather own less investment products if I can get almost as much diversification.

One thing about XIC is that the capped composite index limits any stock to 10% of the index so you can’t have another Nortel. Mind you 10% is still quite a bit and if you had the odd situation of 2 or 3 Nortels at the same time, that could still influence the index in a big way.

By: Jordan

Jordan — Fri, 05 Dec 2008 18:02:06 +0000

I mean to say if there were pricing errors the correlation would be lower, not greater.

By: Jordan

Jordan — Fri, 05 Dec 2008 17:50:58 +0000

I’m actually thinking of doing something similar in my investments where I am splitting a standard asset class like Canadian equities with a traditional cap weighted index (XIC) but then also invest in a fundamentally weighted index (CRQ) (which takes dividends as a fundamental factor).

It might only add a bit of negative correlation, and most of the holdings would be duplicate. But when there are mispricing errors (like Nortel) then the correlation would be even greater so the impact probably wouldn’t be as significant to the asset class as a whole. I don’t have any proof of this but I’m thinking maybe together they would offer a lower standard deviation then either one individually without any more risk. Plus I’m not fully convinced which index is better, this way I get the better real history, liquidity and lower MER of XIC but also the potential that over the long term CRQ will outperform as the back testing has shown.

What do you think? Is it a crazy idea?