There are three ways to generate investment performance. The first and the most important one is by using asset allocation to shape future investment return. The other ways are market timing and improved execution. Portfolio rebalancing is a management operation that creates an opportunity to use the two latter sources of performance. It can be a clever form of market timing, selling high and buying low. Additionally, a well-designed rebalancing schedule can reduce trading costs. Every portfolio needs to be rebalanced, so why not make the most of it?
Recent literature indicates that this performance source is exploited in the noncapitalization-weighted equity indexes, known as “alternative beta.” Beta investments that, in theory, give only “allocational” performance, can deliver extra returns if the portfolio is rebalanced periodically. The alternative beta indexes have weight targets that are not proportional to market capitalizations. This automatically imposes regular rebalancing to keep the portfolio in line with the target. Consequently, these indexes are natural candidates to garner the rebalancing bonus—if it exists.
The effect of rebalancing on portfolio performance has been relatively well studied in the case of diversified portfolios. If one sets a strategic asset allocation target at, say, 50 percent equities and 50 percent bonds, market forces make the allocation drift as time passes. Thus, a rebalancing is needed to bring the allocation back to the initial target. Since the seminal work of Perold and Sharpe ,1 we know that periodic rebalancing of a portfolio to its target allocation can bring an additional performance benefit in the presence of strong mean-reverting behavior. Further research concluded that certain other conditions need to be met to obtain the rebalancing payoff: low return dispersion and low average correlation among assets. But the studies indicate that there is no crystal ball to predict the direction of this rebalancing effect, and no “magic” rebalancing rule to follow.
Do alternative beta strategies really benefit from rebalancing? How much does it really contribute to overall performance? What are the conditions that determine any bonuses? The answers to these questions are quite straightforward when one applies the theoretical findings regarding diversified portfolios to equity indexes. It turns out that the rebalancing effect is a combination of two forces that always act in opposite directions. The first one is the contrarian effect that allows investors to benefit from short-term price reversals. It is always positive, and grows when portfolio components are increasingly volatile and less correlated. The second one is the dispersion effect that favors nonrebalanced portfolios, and is a result of long-term trends. It is always negative for a rebalanced portfolio. The sum of the two effects can be positive or negative, depending on the investment universe and on market conditions.
In what follows, we provide a brief review of the theoretical considerations behind the rebalancing bonus. Next, we show the structure of the rebalancing bonus using U.S. and European equal-weight portfolios derived from the S&P 500 and the Stoxx Europe 600 indexes for the 1999-2013 period. As is the case in multi-asset portfolios, a rebalanced stock portfolio does not guarantee a rebalancing bonus. The U.S. equal-weight strategy has indeed delivered a positive rebalancing bonus, while the European equal-weight portfolio has not. Further analysis shows that the difference in rebalancing payoff is mostly explained by different levels of average correlation in the two markets. European stocks were more correlated than U.S. stocks, which explains the absence of a positive rebalancing effect.
Rebalancing Bonus In Theory And In Practice
A portfolio with constant weight targets cannot remain passive, since price movements imply an allocation drift. At the end of a given period, the assets that performed better than average end up having higher weights, and the assets that underperformed the average end up underweighted with respect to the initial allocation. Consequently, one must periodically reduce the investment in relative overperformers and increase positions in relative underperformers to restore target weights.
Though rebalancing may seem a purely maintenance-related operation, a simple example shows that it might generate a positive return. Imagine a portfolio composed of two assets, A and B, with equal amounts of $100 invested in each asset. Suppose that after one period, the investment in asset A fell to $90, and that the investment in asset B rose to $110. Further, during the next period, the investment in A climbed back to $100, and it fell to $100 in B. In this experiment, a buy-and-hold investor had zero overall return. A constant-mix investor targeting a 50/50 mix at the end of the first period discovered that his allocation moved to 45/55 and performed a rebalancing, selling $10 worth of asset B and investing $10 more in asset A. At the end of the second period, the rebalanced allocation became worth $111.11 and $90.91, for a total of $202.02. Thus, the constant-mix investor achieved a 1.01 percent return when the buy-and-hold investor had none.