Yesterday we had a pre-planning conference call for the Inside ETFs Conference next month in Hollywood, Florida. The panel I’m participating in is on sector ETFs. I’m a believer in building portfolios at the sector level but not necessarily just with ETFs. My approach is to be wrapper agnostic, simply trying to find the best tool available, as I see it, for each exposure.
Although I did not mention it on the call, as important as I think sector decisions are I think they will be far less important in the new decade, sorry guys. The idea here, as mentioned before, is that what made sector decisions so important in the last decade were the extreme sector distortions in tech and then financials. To repeat, a sector that grows to greater than 30% is a serious problem waiting to happen and a sector greater than 20% is a threat that I would tell anyone to heed; certainly I will heed it the next time it happens but I don’t think will be for a while after happening twice so close together.
Value can still be added at the sector level of course. In 2010 as the S&P 500 was going up almost 13% (plus dividends) health care and utilities were flat, tech, financials and staples were pretty close to SPX, materials, energy, discretionary, telecom and industrials were noticeably ahead of the index. For this info I used the nine sector SPDRs as benchmarks and the Vanguard Telecom ETF (VOX) which clients own. Noticeable does not mean dramatic as materials was the best performing sector but it was only up a little more than 20%, so not that dramatic.
As opposed to sector decisions I think country selection will be far more important going forward. Country selections were very important from 2000-2009 as I’ve discussed quite a few times but the massive implosions of tech and financials made for a great and easy way to add value for people who paid attention to sector weightings and if I am correct about there not being a similar sector distortion for a while then performance dispersion between the sectors should be smaller as was the case in 2010.
Compare the numbers above to the iShares Chile ETF (ECH) which is how some clients own Chile. Long time readers will know I’ve been investing in Chile for many years, I write about it all the time. ECH was up 46% in 2010 and the iShares Peru ETF (EPU) was up 58%, we own EPU for one client as part of a specific mandate. This is just a microcosm of course but in thinking about the next ten years country selection and diversification will be far more important.
This raises a theoretical question about portfolio construction using country funds. I’ve made a big deal out of looking under the hood of any ETFs you might use for all sorts of things including sector weights and more importantly how the attributes of one ETF blend in with the rest of the portfolio. For example I would not be too concerned that the Market Vectors Egypt ETF (EGPT) is 46% weighted in financial stocks as opposed to every fund you have being extremely lopsided in financials taking the sector to 40% of the entire portfolio.
But if there is going to be less sector dispersion then can it be suitable to worry less about the sector weightings of funds in terms of blending with the entire portfolio? You still have to like the country to buy the fund of course. While I have zero interest in buying the France ETF should we be concerned that a portfolio of Global X Colombia (GXG), Market Vectors Egypt ETF (EGPT) and the iShares Poland ETF (EPOL) would be grossly overweight financials?
While 40% in any sector is a rotten idea, 20% may not be unthinkable if, and this is an important if, the attributes of the countries do provide reasonable diversification for the portfolio. As a simplified example Sweden is a much different country than Colombia and so sector overlap there would be less of a threat to the portfolio than a bunch of country funds from countries using the euro as their currency.