People keep going on about how important asset allocation is. I left the party when they claimed that a stock & bond mix would perform better over the long-haul then an all stock portfolio. They acknowledged that the stocks would do better on average, but that the mix would somehow magically outperform the all stock.
This didn’t make any sense to me. If you leave two investments for 30 years and get the average stock return on the all-stock portfolio and the average bond return on the all-bond portfolio, how is the return greater if you mix them together instead of keeping them separate?
The only way I thought it could possibly make sense is the idea that you’d have funds available in the bond portion of your portfolio to use to buy stock after a crash. The primary advantage from this perspective would be that the bonds aren’t correlated with the stocks, so you could use one to buy the other. Asset evangelists don’t talk about this though, and since you’d have to actively capitalize on this, it seems like something they’d mention as a necessary step to get the good returns promised by asset allocation.
The realization finally hit me when reading “Four Pillars” this morning: The benefit comes from using the assets within your portfolio to determine when the other parts of your portfolio are cheap, then buying them. E.g. if you have a 90% stock, 10% bond portfolio, you’d expect that the stocks will outperform the bonds. If you’re re-balancing (say by adding money) and your portfolio has 85% stock, then clearly the stocks are selling cheap (relative to your bonds) and its worth buying more of them (which you’ll do to re-balance). When you add money, you’d normally expect to be buying more than 10% bonds (since on average the stocks will outperform the bonds and push them to a lower proportion of your portfolio), but if the stocks REALLY outperform the bonds, then you’ll buy even more then normal (and take advantage of the bonds being cheap relative to the stock).
Basically, asset allocation is just an easy way to determine when the investments you want to buy are cheap (relative to your other investments), then buy more of them. You could accomplish the exact same “benefit” from tracking when your investments are cheap or expensive and buying them directly, however this would be a lot more work. You could also just track the different assets within your portfolio and have the return for the last year and the return for the lifetime of the portfolio. When the last year’s return is lower then the lifetime return you’d put more of the money you’re adding to that portion – again not quite as easy but uses the same idea.
Year 1 $90,000 stock, $10,000 bonds
Year 2 $96,300 stock (7% return), $10,300 (3% return)
Say we’re adding $1,000, in order to rebalance the asset allocation, we’d be putting $540 into stocks and $460 into bonds (bringing them back to 90/10).
Year 2 $96,840 stock, $10,760 bonds (after adding $1000 and re balancing)
Year 3 $96,840 stock (0% return), $11,082.8 bonds (3% return)
Say we’re adding $1000 again in order to re balance, in order to get back to the 90/10 split, we need to sell $192.52 of bonds and buy $1,190.52 of stock (we’re buying lots of stock since it had such an awful year that we now feel that its cheap).
Year 3 $98,030.52 stock, $10,892.28 bonds (after adding $1000 and re-balancing).
Part of me feels like “supporting” under-performing bond returns just for the information of when stocks are cheap doesn’t seem like the best approach, but maybe there’s more to it then I’m seeing. I’d be tempted to construct a portfolio and assign it the expected return for each asset class. When an asset doesn’t live up to expections, add money to it to bring it in line with where it “should be”. For example, in the above portfolio with 90% and 10% with an expected stock return of 7% and an expected bond return of 3%, after 10 years you’d expect the stock portion to make up 93.4% of the portfolio. *THIS* is what you’d use to re-balance, not 90%. This way you’d get the information that an asset class was under-performing and have the opportunity to buy it cheap, without unduly subsidizing the weaker portions of your portfolio.
Please correct me if I’ve missed the point. Given my current understanding, I’m far more convinced of the value of asset allocation.