Welcome! Please Log In
Essentials Popular Topics Favorite Forums & Blogs Join Discuss to setup a list of your favorite forums/blogs.
Discuss > Single Post
Emerging market equity inclusion
Robert T 08-21-2006, 10:11 PM | Post #2230112 | 
Back to EM. The 'Submerging Markets?' article raised a number of questions (at least for me) on what to use as the key criteria on deciding whether to include EM in a portfolio or not.

1. Should past returns be your EM equity guide? Investors have not been rewarded for additional EM risks over long-time periods. Over eighteen years from 1985-2003, emerging market equities produced an annualized return of 9.8% relative to 13.3% for the S&P500 and 11.0% for the EAFE (Swensen 2005). The same can be said for the S&P500 relative to T-bills for the 17 years from 1965-1981 when the S&P500 produced an annualized return of 6.3% relative to 6.7% for T-bills. This suggest that past EM returns may not be a useful guide to future EM returns (at least for the time period available).

2. Should past correlations be your EM equity guide? Similar to returns, correlations can vary significantly - as repeatedly emphasized by Rick Ferri. While useful, I don't think correlation should be the sole guide to a decision for exclusion or inclusion. Its not surprising in Jeff's article that EM (large cap) has a slightly higher correlation to S&P500 (large cap), than Intl small value. An alternative interpretation could be that in the absence of intl. small value choices (which is the case for many), EM large cap maybe a useful substitute (although not perfect), based on correlations.

3. Should past efficient frontier analysis be your EM equity guide? Chart 4 in the article raised my interest enough to run some numbers of my own - using a similar allocation to my own portfolio rather than only with the S&P500. I used two portfolios each with a 37.5:37.5:25 - US:Non-US:5yr T-notes, with data from the same time period 1988-2005. The first had a 0.2:0.4 size:value factor loading for both US and Non-US equities with no-EM. The second had a 0.2:0.4 size:value loading for US and Non-US - Non-EM equities. A 10% EM equity allocation was added (10% of total portfolio). The annualized return %:SD results were 12.3:12.4 without EM and 13.3:12.7 with EM. So adding EM added 1% in annualized return with a marginal increase in SD.

Whenever, I do these analyses I always recall the quote from Bill Bernstein in The Intelligent Asset Allocator on the use of efficient frontier analysis "It's like trying to generate electrical power by placing a battery and a lightening rod at the last place you saw lightning stike. It isn't likely to strike there again. In other words, next years efficient frontier will be nowhere near last year's". However he adds - "if you are trying to capture lightening in a jar, you are better off in Texas than in Alaska. There are certain asset combinations and portfolios which are likely (but nor certain) to do reasonably well". What the above results seem to say is that slightly over-weighting EM in a global small and value tilted portfolio increased returns with moderate increases in risk. I place a bit more weight on this analysis than on 1 and 2 above. This says nothing about future returns - both hopefully similar allocations will do 'reasonably well'.

4. Should cost of capital and expected return be your EM equity guide? Given the limited time period of data, I think this probably carries more weight (at least for us) than 1-3 above. As Jeff points our in his article "since companies in emerging markets have a higher cost-of-capital than companies in the developed markets, returns should be higher as well". Given some of the uncertainty about what the 18 years of EM data is actually telling us - the higher cost-of-capital and higher expected return criteria maybe the most useful.

As a result of 4 (and to a lesser extent 3 above

Originally posted in thread: 52623
View Complete Thread
Reply Quote
  • Favorites
  • Flag
  • Contact
    © Copyright 2019 Morningstar, Inc. All rights reserved. Please read our Terms of Use and Privacy Policy.