Wednesday, April 30, 2014

Anti-Rebalancing

Investors find many ingenious ways to underperform stock indexes. Sadly, many have no idea that their brilliant moves actually work out badly over time because they don’t track their returns accurately. Among the ways that investors manage to harm their portfolios is what I call “anti-rebalancing”.

The idea of rebalancing your portfolio is that you start with fixed percentages of various asset classes, and when they drift away from the fixed percentages, you bring them back in line. For example, suppose Beth holds 4 funds in equal amounts: Canadian stocks, U.S. stocks, foreign stocks, and bonds. Every so often she checks which fund balance is higher than the others and which are lower, and she makes some trades to get the amounts back in balance.

This sounds easy enough in theory, but isn’t so easy in reality. After the explosion of U.S. stocks in 2013 and the relative underperformance of foreign stocks, Beth is supposed to sell some of her U.S. stock fund and buy more of her foreign stock fund and bond fund. It can be hard to make the decision to sell your best performer and buy the dogs.

This brings us to what I see some investors do: anti-rebalancing. They sell some of their dog funds and buy more of their top performers. This can actually work out well sometimes. Maybe U.S. stocks will shine again in 2014. Unfortunately, over time, anti-rebalancing works out worse than rebalancing. This is because rebalancing amounts to buy low and sell high, while anti-rebalancing amounts to buy high and sell low.

Few investors think of their strategy as anti-rebalancing, though. They offer intelligent-sounding reasons why their allocation percentages need to change. Maybe they decide the sentiment that the U.S. is losing its place of world dominance is overblown and they should keep a sizable portion of their savings in U.S. stocks. Whatever the reasoning offered, the bottom line is that they pour more money into whichever fund is up the most.

If you think of yourself as an index investor, but you make your own judgements about portfolio changes instead of simple rebalancing, I urge you to look over your history of trades and see what would have happened if you had rebalanced mechanically. The majority of investors who do this exercise honestly will find that their trades have actually hurt their returns.

8 comments:

  1. When is time to re-balance? Every month? When one of the categories deviates from its target share by xx% or by yy percentage points? Any rationale, rule of thumb, your personal practice or opinion?

    ReplyDelete
    Replies
    1. @AnatoliN: There are many approaches that can work just fine. I rebalance whenever my asset classes are out of line by certain amounts as explained here:

      http://www.michaeljamesonmoney.com/2012/03/portfolio-rebalancing-based-on-expected.html

      However, simple rules like checking once per year or once per quarter are fine too, as long as you act. Most people are able to rebalance with new contributions. Sadly, they often choose not to buy whatever is cheapest because it seems like such a dog and everyone else is buying the asset classes that have been flying high.

      Delete
    2. will check the recommended post. Thank you.
      Rebalancing with new contributions I do. I am concerned that I am not rebalancing often enough.

      If I rebalance with existing funds, I trigger deferrable taxes at non-registered account. Do you have a post with your considerations on which type of funds to use to avoid this?

      Delete
    3. @AnatoliN: I don't have a post about rebalancing in a non-registered account. However, it is sometimes possible to achieve rebalancing using only trades in registered accounts. I've never had to sell stock in a non-registered account to rebalance. Your mileage may vary.

      Delete
  2. Great reminder. I follow Doug Cronk's blog as well who recommends that individual investors pattern their asset allocation after pension plans and rebalance regularly to those allocations. The Pension Investment Association of Canada (PIAC) publishes the average asset allocation among all their member plans. I've used those averages as a baseline guide for my portfolio and plan to tweak my asset allocations as time goes along and the PIAC averages change. It's interesting to note, though, how little pension plan asset allocations change from year to year.

    ReplyDelete
    Replies
    1. @Juan: Investors tend to make poor choices when they change their allocations, so minimal changes sound good to me.

      Delete
  3. Great post as always!

    What would be the rebalancing situation if there are accounts in a TFSA and RRSP?

    Would the RRSPs be reduced by say, 30% (marginal tax rate) to account for the tax on withdrawals? And then the allocations compared across the two accounts?

    ReplyDelete
    Replies
    1. @SK: It's not the marginal tax rate that matters here, assuming that the RRSP withdrawals will be large. You need to estimate the average tax rate across your RRSP withdrawals. This may be below 30%, particularly if you plan to start drawing from your RRSP before beginning to collect OAS and CPP (and you have no other income).

      Delete