apex84 wrote:Random Musings wrote:Even over long time periods, would an average investor, let alone a reasonably astute Boglehead really "feel" the difference between a std of 9.9 and 9.6 besides the fact that it came out mathematically (let's assume annualized returns are exactly the same)? I wonder at what level of volatility change would most investors start to "feel" it?
I agree that the benefits as described in the white paper are small. I still think that the availability of a low cost fund from Vanguard that tracks the largest asset class in the world is a useful addition to their fund offerings. Even if the benefits of the fund are slim in the current bond environment, it is possible that this diversification will be useful in the future. Of course, like all diversifying assets, it has a chance of lowering portfolio return, but at least with bonds, one knows the yield at the outset.
Well 75% of the bond market is outside the US. (Proportionally the US is a much larger chunk of the global stock market than of the bond market.) So from the standpoint of a market-cap weighted passive investor, this fund fills in a huge gap.
As for the diversification benefits, I don't think the Vanguard paper did the right analysis. We aren't really interested in how this impacts the volatility of a balanced portfolio because we've already reduced volatility by about as much as we can without altering our exposure to the basic asset classes.
What we'd really like to know is how this reduces your exposure to "peso problem" events that only occur with very small probability (e.g. 1-2%, or even less). Unfortunately, because of the low probability, it is unlikely that the historic record will be sufficiently illuminating. So what we really need to do is an extreme value analysis that extrapolates out to the tails of the distribution and then estimates the reduction in tail-value at risk. (Which is, confusingly, very different from "value at risk".) Common sense tells us that reducing our exposure to one large issuer (the US Federal Government) reduces our losses in the event of extreme events, but we really don't know how big of an impact this is or what the best allocation to international bonds would be to achieve this.
Similarly, looking at recent-ish history in the US (since the early 80s) isn't sufficient to get a good idea of the diversification benefits of doing this. US bonds were in a very long, very strong bull market for the entire time period. We'd like to know whether or not the global bond market is correlated enough that all bonds will be in bull or bear markets together or whether it is possible for the US bond market to be in a bear market while international ones are still bullish. Again, it seems to me that, in a rising interest rate environment in the US, having the international exposure would leave you better off than otherwise, but I don't know how big of an effect this would be. More data and more analysis is needed.
abuss368 wrote:What happened to the emerging markets bond fund that was to be originally offered with this one?
The Vanguard paper linked in the very first post still has the section talking about the benefits of emerging market bonds. I'd assume that if they didn't plan to launch one that the new version of the bond paper would have taken that section out. So it would seem to be still in the works.