Thoughts On Geoff Considine's Model ETF Portfolio

Roger Nusbaum submits: Posted earlier today on ETF Investor was a portfolio put together by Geoff Considine using Monte Carlo Simulation. Geoff's idea is a spin on David Jackson's portfolio.

I'll toss up my usual disclaimer that all portfolios like this have flaws. Anything along these lines that I might try would have flaws too. Flaws themselves don't have to be a problem but not knowing what they are could be. I realize the portfolio adds up to more than 100%. [Editor: Thanks for catching my error in transcribing Geoff's allocation, Roger. It's now been corrected -- with a credit to you! -- David.]

The energy weight is colossal at 26.96% (all of these numbers are from Morningstar) compared to 10.29% in the S+P 500, utilities have a 13.76 weight compared to 3.6% in the S+P 500. Tech is very underweight vs the market. The yield is 2.41% which is above the market yield but not that high when you consider that 30% is in bonds and 10% is in REITs.

I am as bullish on energy as anyone, I think, but there is no way I would overweight it by 16 percentage points, not even close. I think there are fundamental drivers for energy so I am overweight. I don't think there are necessarily any fundamental drivers for utilities to outperform the market again, tripling up on the benchmark weight is very extreme. I am very close to equal weight (just a hair over) utilities because I want the yield but I do not think the sector will beat the market by the roughly 8% it did in 2005.

I would also note that the median cap size of the portfolio is $14 billion. That is fairly small. Small could continue to lead but it is worth knowing that the run of large cap lag is very long in the tooth by historical standards.

The other thing is no emerging market exposure, I don't think EFA really have enough to matter. I have not paired back my exposure for clients but it is possible that pairing back is the right trade ( to be clear I don't think it is but I could be wrong). Almost ignoring an asset class could be a big mistake, at least it is today.

I am quite sure that all of the things I pointed out are easily justified by Mr. Considine but this is just how I see the portfolio.

Comments

[RogerNusbaum]

Geoff, thank you for the reply. I should take umbridge with the <em> skilled analyst </em> remark (insert smile). If the weightings are different that what was originally spelled out that would probably change things. I do this type of picking apart of any ETF portfolio I come across in an attempt to offer readers a second opinion on what they are reading. I think this helps people be better do-it-yourselfers.

The portfolio might turn out to be fantastic but understanding potential drawbacks will make for better allocation decisions. Thanks again.

[witchdoctor]

Roger, You now have the benefit of hindsight, both energy and utilities outperformed in 2006. Your concerns were unfounded and you should recognize how this portfolio actually performed vs. the one you recommended in a future article. Give credit where credit is due.

[RogerNusbaum]

I don't know how this portfolio did, I don't know which portfolio that I wrote about that you are referring to or how it did but the focus of this post was the risk taken to get the return.

100% in Garmin would have been even better.

26% in energy is a big bet, you either agree with that or not; I guess not.

[GeoffConsidine]

Roger:

I just ran across the comment from Witchdoctor--but don't know if its from Sep 2007--thats my guess. Nonetheless, to compare my portfolio analysis to a random stab on a single stock (like Garmin) seems fairly dismissive. Two years since I wrote that article and the portfolio that I analyzed has done very well on a risk-adjusted basis, despite the concentration in energy and utilities. Yes, it could be luck--but I think that I made a well-reasoned case back then--hardly the same as a naked bet on a stock.

A lot of articles down the road from that one (it was my first portfolio contribution to SA), the application of portfolio theory to develop portfolios with the most risk-adjusted return has a lot of support in operation. The point is that concentrated positions (as compared to index concentrations) can make for a much better portfolio overall---and Arnott's work supports that too. The relative weight in a portfolio to a sector as compared to a market-cap-weighted index is, based on my research and that of many other people, close to irrelevant.

Best,

Geoff

[RogerNusbaum]

The Garmin comment was about concentration in general.

You might be able to out-statistic me all day, don't care, 26% in energy is way, way bigger than I would ever go.

[GeoffConsidine]

Right--I get that you feel in your heart that this portfolio is 'too risky' because it is 'over concentrated.' You don't really know if this portfolio is risky relative to the original David Jackson ETF model portfolio though--you simply assume that. This is my point. Many investors fall prey to the idea that concentration = risk and, worse yet, that a portfolio is concentrated if it holds more of something than is in some arbitrary index. There is zero evidence that the weights represented in the S&P500 or any other index are optimal or even rational.

This was a model portfolio--I have not endorsed it for any actual person. The point was to get people to think about portfolio risk in a more meaningful language.

In the concrete numbers of portfolio theory, my model portfolio was no more risky than David Jackson's--sure, there is uncertainty in all this, but this is a meaningful result. Let's frame this another way. Do you have any concrete evidence other than gut that this portfolio is riskier than the portfolio proposed by David?

[RogerNusbaum]

this was a year and a half ago, let it go.