A Difficult Balancing Act

June 28th, 2011

How best to invest is always a difficult decision to make, but once you have your strategy, the next decision to make is how best to maintain it.  This article will look at this maintenance and in particular the relative merits of rebalancing in conjunction with the asset allocation method of investing.  It should be remembered that this is merely one type of investment and maintenance strategy and is not necessarily the one most suited to your needs/desire.

Allocating assets

The key point of an asset allocation strategy is to achieve the best possible relationship between risk and return by investing in several different asset classes such as cash, fixed interest securities, property and equities. This exploits the fact that, due to their divergent behaviour patterns then there is usually a low correlation between the returns produced by these different asset classes (i.e. the behaviour of one should not be overly copied by another).  By combining them selectively, it should be possible to either increase your return based on your risk appetite or to reduce the risk you are taking to achieve a specific return when compared to investing in a single asset.  This is sometimes referred to as the ‘efficient frontier.’

Consequently, this strategy allows you to manage risk – which is why it is extremely important that you do not stray too far from the recommended asset allocation. If that happens, your portfolio will change in shape over time and will move away from your original risk appetite (either increasing or decreasing in risk dependent upon how the various asset classes perform) that laid the foundation for the recommended allocation in the first place.

Beneficial rebalancing?

Rebalancing can help to maximise the returns received from your portfolio based on your specific appetite for risk, but it is very important to remember that if you rebalance too often, the trading costs (the cost of buying and selling investments) will reduce the positive returns (or increase losses) and potentially damage your long-term returns. This leads to the almost impossible to answer question of once the asset allocation strategy has been determined, when and how often should a portfolio be rebalanced?

The technical answer is: rebalance when the expected benefit (i.e. the increase in returns for the risk adjusted portfolio) is greater than the trading costs involved.  In other words, the more often it is done the more costly it is but does this produce sufficient extra growth to offset the costs?

Rebalancing can be quite a complex matter which has been the subject of many academic studies. However, assuming that rebalancing is suitable and fits in with your investment profile and that your asset allocation has been implemented, you can basically choose between the following rebalancing strategies:

1) No rebalancing
2) Dynamic programming with cost function minimisation
3) Fixed frequency (monthly, quarterly or annually)
4) Fixed tolerance band for maximum deviations

Most professional investors opt either for a ‘tolerance’ strategy (which means that when the portfolio has changed beyond a set percentage it is then rebalanced back to its original position) or a ‘frequency’ strategy, where this rebalancing is performed over a set period such as annually. Despite the fact that both are quite close in terms of benefit, however, neither of these strategies offers the optimal solution.

According to leading academic researchers, the best solution is optimal dynamic programming. This is a very complex strategy and not many – even institutional investors – use this method. Luckily though, a much simpler ‘tolerance’ or ‘frequency’ strategy will produce comparably good results.

Many economists have examined which rebalancing strategies produce the best results and in 2006, in the Journal of Portfolio Management, Walter Sun and his colleagues at M.I.T. in Boston concluded that significant excess returns can be generated by rebalancing your portfolio.

Amongst other things, they tested a portfolio comprising five different asset classes and examined the performance generated after 20 years when adopting four different rebalancing techniques. The trading costs (of buying and selling assets) were assumed to be between 0.4% and 0.6% for the different asset classes – although an important aspect is that the costs did not necessarily include the cost of the individual who had to instigate the changes.

In the table below, from the research, several rebalancing strategies were considered and the cost of each compared.   The tolerance band strategy works on the basis of rebalancing once the portfolio has moved more than 5% from the original allocation.

Summary of the empirical findings of Sun et al. (2006)

Source: Sun, W., A. Fan, L. Chen, T. Schouwenaars and M. Albota, 2006, “Optimal Rebalancing for Institutional Portfolios”, Journal of Portfolio Management, 32, 33-43

Rebalancing strategy

Trading costs (bps per year)

Reduction in growth compared to optimal strategy (bps per year)

Total costs (bps per year)

No rebalancing

0.00 30.18 30.18
5% tolerance band

7.29

0.70

7.99
Monthly

23.67

0.00

23.67

Quarterly

13.69

0.28

13.96

Yearly

6.84

1.55

8.39

* BPS are basis points and represent 100th of 1% so 30 bps = 0.3%

Whilst this suggests that rebalancing is advantageous it is still dependent upon the costs involved in the whole process, although as mentioned previously, it also helps to maintain the risk of your portfolio within its original intended level.

A key benefit of asset allocation and automatic rebalancing is that it creates a disciplined approach at all times (including when rebalancing), although it is also possible to counter a common and unfortunate weakness of human nature – the desire to invest more in an asset class performing well and less in an asset class that is underperforming. Asset allocation and rebalancing strategies create a non- emotional buying and selling process and therefore the removal of this human emotion which can create the ‘buy high and sell low’ failure of many.

It should be remembered, however, that these are not the only investment methods that can be utilised. And, despite the benefits, they do counter the investment expertise of a financial adviser/investment manager and the potentially difficult to ascertain aspect of market timing, such as disinvesting before the credit crunch where asset allocation and most other strategies failed to avoid falls in values due to the market wide nature of the issues.

The value of your investment can go down as well as up and you may not get back the full amount invested.