To rebalance or not to rebalance?

Rebalancing a portfolio (bringing the asset class weights to or close to their neutral weights) serves two essential purposes:
1. Preserving the defined risk profile of the portfolio over time
2. Increasing the probability of locking in market related performance gains by cashing in on outperforming investments (reducing the impact of a subsequent selloff in the investment) and buying more of underperforming investments (enabling the portfolio to capture greater returns from a subsequent rebound).
Rebalancing, together with diversification are amongst the core pillars of modern portfolio theory.
There is a caveat, however, that should be regarded carefully by those that employ rebalancing techniques when managing portfolios. It is that the effectiveness of rebalancing techniques depend largely on market conditions. In periods of market turmoil, rebalancing can prove to be counterproductive. Just like in the event of an earthquake, we are less likely to go about our daily chores, rebalancing is not necessarily the right thing to when market conditions are extraordinary.
The best way to illustrate this would be by taking an example of a "balanced" risk profile portfolio with an even split exposure to equities and bonds. Say, for example, that $1000 dollars is invested into equities and the remaining $1000 into bonds for a total portfolio value of
$2000 at the beginning of the year. Suppose that a financial crisis occurs and that by mid year, equities are down 30% whilst bonds remain flat (I know, its not very difficult to imagine such things these days). The portfolio value will be as follows:
Equities: $700
Bonds: $1000
Total: $1700
Suppose that right after the drop, the portfolio manager decides to rebalance with a final result of:
Equities: $850
Bonds: $850
Total: $1700
Suppose, also, that for the rest of the year, equity markets drop another 30% and bonds remain at the same level. Our portfolio at the end of the year will look as follows:
Equities: $595
Bonds: $850
Total: $1445
Now suppose that the manager did not rebalance at any time during the year. Our portfolio would end up as follows:
Equities: $490
Bonds: $1000
Total: $1490

We find ourselves with a portfolio that has a smaller exposure into equities but also with a better performance relative to the rebalanced portfolio. The non rebalanced portfolio total value stands at $1490 versus $1445 for the rebalanced portfolio. The difference (3.11%) may not seem like much but that is mainly becuase the total portfolio value is small!

Although this example illustrates the performance advantage of not rebalancing when markets are in turmoil (such as what we are experiencing at present), what it doesn’t show us as clearly is the increased risk of shifting the risk profile that not rebalancing results in. It is easy to imagine what would happen to the performances of the two portfolio examples in the event of a subsequent market rebound of similar scale to that of the drop.

Moral of the story: Rebalancing doesnt work well in times of extreme volatility but you have to weigh in the increased probability of a shift in risk profile if you decide to ignore rebalancing.