Re-Evaluating Rebalancing 

Whilst enjoying Parmenion’s Fundamentals of Investing course, a brief diversion around Capital Gains Tax planning in portfolio management led me to do some investigation on the current thinking about rebalancing.

This is a word much more in use these days, probably as a result of the increased availability of the appropriate functionality through platforms and the explosion in model portfolio offerings from an increasing number of discretionary portfolio managers.

Of course, rebalancing for advisers, meaning those without discretionary powers is, and will always be, a nightmare.  The advice process to recommend the sale and repurchase on each occasion, the requirement to obtain client consent, and the actual administration process makes it all rather cumbersome. 

So it’s no wonder that DFMs have been keen to emphasise this aspect of their service, making consistency and simplicity core elements of investment portfolio planning.

An analyst explained at a PFS conference some years ago that they had found no benefit in rebalancing more frequently than once every 18 months, but accepting that most clients were on some form of annual review cycle there was no real disadvantage in rebalancing annually. 

Vanguard published a paper on this subject (http://tinyurl.com/l7y5987 ) which concludes there is no optimal time period or threshold which delivers enhanced returns.  However, semi-annual or annual rebalancing with 5 per cent thresholds is likely to produce an appropriate balance between risk control and cost minimisation for most investors.  Annual rebalancing is recommended where the portfolio is outside a tax-wrapper. 

A second article by William J Bernstein (http://www.efficientfrontier.com/ef/100/rebal100.htm ) looks at returns for portfolios rebalanced monthly, quarterly, annually, biannually, and every 4 years, looking at twenty-four twenty-eight year periods, staggered by one month.  The results are illuminating: over this relatively long time frame there is little difference between quarterly and annual rebalancing.  He notes that the average difference in annualised returns between quarterly and 4-yearly rebalancing is only 18 basis points.

He concludes that the more volatile the portfolio the more reason there is for rebalancing.  He also suggests that portfolios where the constituents have low correlations might lead to higher rebalancing returns.  But the most interesting conclusion is that taxable portfolios should not be rebalanced at all, and that rebalancing really only works as a risk control mechanism.

Investors rebalancing to seek enhanced investment returns are likely to be disappointed, even without the consideration of the impact of costs and taxes.  Are the DFMs that you are using ready to explain and justify their rebalancing strategies, especially where they contradict this research?  And are you explaining the rationale behind this strategy (i.e. to control risk, not to enhance returns) properly to your clients?  
 

Gill Cardy, 21/11/2014