Country Funds En Masse Could Be the Answer

Country funds often get a bad rap from mainstream media even though any S&P 500 index fund is also a single-country product. The utility of country funds is that they allow access into an investment destination without having to pick a stock. The downside to them, for example, could be that a weighting in banks might hold back what might generally be a technology exporting-based country.

A potential issue with using a bunch of country funds is ending up too lopsided in some sectors and zeroed out in others. I was curious to see if a portfolio of just country funds could be assembled that captured some regional and economic diversification without ending up 35% in financials. So I plugged in the following ETFs into Morningstar to see what I could see.

iShares Canada (EWC)
S&P 500 (SPY)
iShares Australia (EWA)--personal holding
iShares UK (EWU)
iShares France (EWQ)
Market Vectors Russia (RSX)
iShares Taiwan (EWT)--a few clients own this one
iShares Malaysia (EWM)
iShares Israel (EIS)
iShares Brazil (EWZ)

A quick disclaimer before proceeding: I gave very little thought to the ten chosen and I'm not disclosing the simplistic weighting on the off chance someone tries to implement this.

Without looking under the hood just yet, it is clear that the mix captures regional and economic diversification. You don't need to know too much about stock picking or portfolio construction to realize that, but this was probably never in doubt.

As for the sector breakdown, shockingly the weight to financials (all sector info according to Morningstar) was only 19.95% versus 17.43 for the S&P 500. Tech was a meaningful underweight at 8.42 versus 16.73. The other overweights were telecom, materials, energy and industrials. The underweights were healthcare, staples, discretionary and utilities by a whisker.

According to portfolio science the stats are mediocre. The overall beta was 0.95, the correlation to the S&P 500 was 0.88, and the standard deviation was 23.91 versus 21.08 for six months. The average market cap was about half of the S&P 500 and the yield was a little better at 1.96%.

Morningstar was not able to calculate the performance for 1 year because EIS is too new. So backing that one out the portfolio was up 7.57% for one year (actually it only went back to May 31, 2007 for some reason) versus a 7.46% loss for the S&P 500 as measured by SPY. The Tel Aviv 100 Index was down 8.9% in the period studied so if EIS had existed it would have take a small bite out of the return.

The portfolio is mostly foreign and foreign did much better, so the result is not a surprise--again foreign has done better. The next time the US is the best performing market, a portfolio like this that is mostly foreign would obviously lag.

The bigger macro of this exercise is that it probably removes home bias (well, I think it does anyway). I also think that by drawing various types of countries it reduces, not eliminates, currency risk, for example lately the British pound has generally struggled against USD while the Aussie has been strong.

I think anyone putting some thought into country selection and weighting, and again I did not, could create a diversified portfolio without having to pick stocks or sectors or manage things like cap size or style.

The downside includes ignoring the things listed in the above paragraph which at times will hurt performance, although that might not be fair; if you don't manage those things now, you would not miss them in the future.

Comments

[rajtrades]

I regularly look at alternative ETFs to further diversify my portfolio. I often wonder whether country funds or regional funds offer better results when viewed from a practical sense. The 1 year daily correlation is over 80% between most country indices and their respective regional and/or type indices. Shorter term correlations can diverge significantly however, during brief spurts of country specific event driven volatility. These correlation characteristics imply that a short term rebalancing regime would be needed to capture relative price dislocations among similar country funds found within a broader index. Longer term, these price disparities should be arbitraged away thus providing little if any additional return. Thus, I believe country specific funds are only beneficial to portfolio methodologies that are short term oriented.

As an aside, what should be the goal when constructing a diversified portfolio; should it be a weighted net correlation of 0% (based desired investment timeframe), or maybe + or – 10% or something completely different?

[kurt walter]

I would like to see country funds that go short the index, too.