I've done tons of study about HMC and Jack Meyer. The allocations make perfect sense. Multiple high beta assets with low correlations. Each of the different asset classes offer substantial returns over traditional bonds.
HMC doesn't care if EM's lose 10 b/c something else will be negatively correlated at hedge the position. HMC also does no tactical allocation. They don't believe they can add value over the short run. They set up trading parameters (asset allocation policy) which they adhere to. It's their asset allocation policy that is great, not just their secuity picking skills. If you took a standard asset allocation over 10 years and compared the results to their allocation (index returns only) you'd see 300 bps annualized outperformance. Some great reading on them can be bought at Harvard Business Online or at the Yale IMC site.
The comment expands on some of what I wrote back than. The concept is very important to proper portfolio construction. By blending together different stocks or assets you can achieve better returns with less volatility. While I believe that I can say it is easier to manage the volatility than ensure better returns. You should expect periods of being on both sides of the SPX in terms of returns. And that's ok.
If you manage your own money and you can accept that you will lag the market sometimes as well as beat it, your job gets much easier.
Here is the allocation as of December;
15% US equities
13% Commodities (about 3/4 of this part is in timber)
13% Private Equity
12% Hedge Funds
11% US Bonds
10% Foreign Equities
10% Real Estate
6% Inflation Index Bonds
5% Emerging Markets
5% High Yields
5% Foreign Bonds
-5% Borrowed Money





4 comments:
Is this your asset allocation or HMCs ?
sorry if I was not clear this is HMC's allocation.
This doesn't make an awful lot of sense to me or maybe it's a confusing usage of the word "beta". How can you have "high beta" assets with "low correlation" to one another? Beta means exposure to the common market factor; if these assets are uncorrelated with one another, then they must also be uncorrelated with the common factor, which would imply that they were low beta, not high.
-dsquared
Post a Comment