Portfolio Rebalancing: Get It Right

What is the best thing you can do to improve your clients’ portfolio performance? For a long time, the accepted answer was simple: asset allocation. Now, however, it appears that there is a second factor that might just be more important: consistency in that asset allocation, as maintained by steady rebalancing.

Get access to this article and thousands more...

All On Wall Street articles are archived after 7 days. REGISTER NOW for unlimited access to all recently archived articles, as well as thousands of searchable stories. Registered Members also gain access to exclusive industry white paper downloads, web seminars, blog discussions, the iPad App, CE Exams, and conference discounts. Qualified members may also choose to receive our free monthly magazine and any of our daily or weekly e-newsletters covering the latest breaking news, opinions from industry leaders, developing trends and growth strategies.

Already Registered?

Comments (8)
Allan, Nice article and incredibly simple to implement if one chooses such a course! I see much of your other work here too. I can tell you are incredibly well respected by your peers. Keep up the great work and thank you!
Posted by Bruce B | Monday, June 02 2014 at 1:34PM ET
We published an article "Let it Ride" in this magazine (Dec.8, 1988) where we studied and established the results of quarterly rebalancing over the period of Dec. 1968 to Dec. 1987. We chose 3 month treasuries, high grade long term corporate bonds, gold, Dow Utility, Dow Industrial, Dow Transportation indexes. The non-rebalanced portfolio slightly beat the rebalanced one, but surprisingly, with less volatility.

At that time we were about to market a pension investing process to businesses, and thought that the idea of methodical rebalancing would endear us to that market and give us a competitive edge. Of course we were surprised (and disappointed) at the results of out study.

I believe that managing portfolios requires a lot more than just blindly rebalancing, which sort of flies in the face of that other parcel of investment wisdom dictating that you should cut your losses and let your winners run.

Of course, financial planners would rather keep it simple, buy some number of inexpensive index funds or etf's and distinguish themselves as more valuable from their competition by promoting the idea of rebalancing on some fixed schedule. This will remain true until "rebalancing" becomes the can't do without criteria for investment management, and than maybe someone will come forth with this study or a newer one, finally shattering the hook so many investment advisors have hung their hats on.

Bill Kaufman (Money Watch Consultants Inc.)
Posted by william k | Tuesday, June 03 2014 at 10:20AM ET
Interesting article. However, we should be careful not to draw general conclusions from one 15-year time period and one that included 2 significant recessions. If the authors were to perform an empirical analysis incorporating a large number of historical time periods, including longer ones, the conclusions would likely demonstrate that rebalancing, especially as frequently as annually, is suboptimal assuming the goal is to maximize portfolio value. The reason is that rebalancing is essentially locking in a static asset allocation by selling better performing investments to buy more of the poorer performing investments.
Posted by Eric S | Tuesday, June 03 2014 at 12:07PM ET
In thinking through the results that we found, I have come back to the idea that rebalancing misses long term trends in financial markets. As a consequence, rebalancing forces continual buy ins to asset classes that could be going through long term down trends or cycles (think long term bonds purchased over rebalancing intervals of years' long rising interest rates, ugh. Or conversely, bonds sold off to rebalance during long periods of decreasing interest rates.

Moreover, for rebalancing to make sense, one has to choose "asset classes" to work with. Ideally these are chosen based on correlation histories.

No one would choose multiple asset classes for a portfolio that were wholly correlated (what would be the point?). But the truth is that asset class correlations change. For example, an asset class can be inversely correlated with another asset class until it's not inversely correlated. Correlations themselves can change. They can be arbitrary and/or temporary.

So we concluded that rebalancing over simplifies the portfolio management process. It survives because it is a "good story" to tell naive prospects, enabling many investment advisors to sell their wares.
Posted by william k | Tuesday, June 03 2014 at 12:23PM ET
It is unfortunate that you continue to propagate the wrong conclusion, so skilfully exploited by the mutual fund industry and other purveyors of retail investments, as a marketing tool, that the "notable research paper" you referred to showed that asset allocation was the single most important determinant of portfolio returns. It did not and it never claimed to do so. What the authors of that paper argued was that it explained about 94% of the portfolio's behaviour; the ups and downs of the portfolio. In fact, the same authors wrote a subsequent paper in which they argued that asset allocation only explained, if memory serves me right, about 40% of the portfolio's performance.
Posted by Victor M | Tuesday, June 03 2014 at 3:06PM ET
Add Your Comments:
Not Registered?
You must be registered to post a comment. Click here to register.
Already registered? Log in here
Please note you must now log in with your email address and password.

Already a subscriber? Log in here