Note: This is a copy of a post I made in another thread:
"Bonds: Ballast for your portfolio".
I think that it is relevant to the topic of this thread.
------------
aj76er wrote: ↑Sun Aug 13, 2017 11:57 pm
I recently found this passage in Sharpe's RISMAT-7 to be pretty compelling:
To see this, consider a simple policy calling for a mix of 60% stocks and 40% bonds. Now assume that stocks outperform bonds so the portfolio proportions change to 65% stocks and 35% bonds. To rebalance to a 60/40 mix, one would have to sell some stock holdings and buy bonds with the proceeds. But not everyone can sell stocks and buy bonds, since for every buyer there must be a seller.
Prof. Sharpe is simplifying things a little too much. What he is saying is not exactly right. Let me explain.
What would happen, if all investors wanted to rebalance their portfolio to 60/40, is that prices would adjust accordingly; that's how markets work.
In other words, the price of stocks for which there's no buyer would drop to attract buyers and the price of bonds for which there's no seller would go up to attract sellers. This would continue until either the aggregate market prices of stocks and bonds settles to 60/40 (removing the need for anybody to rebalance), or until some investors decide to switch off 60/40 investing.
Assuming that stocks and bonds were initially
fairly priced, affecting prices to match a target 60/40 allocation (in the chosen 65/35 example) would naturally increase the future returns of stocks (because their price would drop below fair value) and decrease the future returns of bonds (because their price would go up above fair value).
What is
more likely to happen is for
arbitrageurs to take advantage of the situation and beat the market, leading 60/40 investors to underperform the market.
In other words, the real problem Sharpe is trying to avoid, is for a passive investor to be putting excessive pressure on market prices and get taken advantage of, in the process.
Personally, I don't care.
I don't think that stock markets and bond markets are so closely integrated. Money taken off the stock market does not necessarily move into the bond market; it can go elsewhere (bank account, private real estate, spent, etc.).
Also, Sharpe's
Global Portfolio doesn't include all traded securities; it doesn't include things such as junk bonds and commodities, for example.
Furthermore, Sharpe's argument fails when it comes to international bonds, because he is recommending the use of funds which hedge currency. There are two problems with this. For one thing, currency hedging requires the use of derivative investments. This, in itself, causes a departure from passive indexing; it's not even clear if there's a good argument to explain how hedging counterparties would naturally be in the proper proportion so as not to cause excessive pressure on hedging costs. For a second thing, the "hedge return" will cause total returns to differ from those of the underlying
markets. This will force investors willing to keep their international bond holdings in proportion of international bond markets to make transactions to rebalance their holdings (based on currency fluctuations), putting excessive pressure on market prices and opening the way for arbitrageurs to take advantage of the situation. This is actually pretty bad, as short-term currency fluctuations can be quite brutal.
Finally, Sharpe's overall proposal contains a
significant departure from his Global Portfolio concept; that is, he lets investors choose the amount of TIPS they want to include in their portfolio, as if this wouldn't put excessive pressure on prices and open the way for arbitrageurs to take advantage of it. Considering TIPS as riskless is an oversimplified concept. There is simply no such thing as a
riskless marketable financial security; TIPS investors who thought otherwise learned as much in 2008.
The closest thing to a
riskless security is an I-Bond, which is not a tradable security, and thus, for which there is
no market. Unfortunately, there are rather low annual limits to how much one can buy.
(In Canada, where I live, there is nothing equivalent to I-Bonds, unfortunately).
As a result, I've decided to keep things simple:
- I use total-market index funds to invest into four selected types of investment-grade securities: domestic stocks, international stocks, domestic nominal bonds, and domestic inflation-indexed bonds.
- I do not include international bonds (currency-hedged or not) into my portfolio.
- I do use a fixed target allocation (25/25/25/25) which is 50/50 stocks/bonds.
- I rebalance my portfolio very lazily:
- I mostly rebalance by diverting fund distributions and new portfolio contributions into assets below their target allocation. In retirement, I will take withdrawals from assets above their target allocation.
- Infrequently, when it gets too much off target, I actively rebalance my portfolio using a rather sloppy two-step rebalancing process, which conveniently sidesteps tax problems (such as wash sales) and keeps my trades difficult to anticipate by arbitrageurs.
I know that what I'm doing is not perfect, but there is simply
no perfect solution. As I've explained above, even Sharpe's proposal has significant problems.
I'm not saying that Sharpe's proposal is
bad; far from it. I'm just saying that, given its own problems, it doesn't provide any
foreseeable improvement over my
simpler approach.
Note that I used the name Global Portfolio to refer to his recommended portfolio, in my text; Sharpe actually uses the name World Bond/Stock Portfolio, which is more accurate.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)