Geographic Re-Balancing of Stock ETFs

Various channels claim that rebalancing your stock portfolio increases your return on investment. In this article, I will present the analysis of historical data on geographic allocations of the popular indices MSCI World and MSCI Emerging Markets.

2 remarks before we start:

• This article was translated from German to English in collaboration between man and machine *To be honest, I was a little disappointed by the translation performance of translate.google.com.
• Even though I am talking finance in this article, I’m not a finance expert at all. I’m just a private individual who is interested in their own pension scheme and other financial topics. So I accumulate solid half-knowledge and rely on my more or less common sense and program a calculator^1,2 every now and then.

Let a world portfolio be given

Suppose a person, let’s call her Elisa, wants to benefit from the global capital market for her retirement. Her effort should be as limited as possible. Now Elisa hears that ETFs can be used to easily benefit from profits of public companies all around the world. An index called MSCI ACWI and ETFs that try to replicate this seem suitable to Elisa at first glance. However, Elisa notes that the MSCI ACWI includes more than 50% shares in US companies. Represent the US really half the global economy? Following the criterion of market capitalization used here, definitively. There are other criteria for tracking the global economy, see, e.g., Justetf. A popular alternative to 100% ACWI includes the MSCI World Index at 70% and the MSCI Emerging Markets Index at 30%.

Geographic rebalancing

Suppose Elisa chose the 70% MSCI World and 30% Emerging Markets option. Furthermore, after some time, the portfolio has shifted to 75%-25%. Now Elisa is wondering if and when she should re-balance to the original split of 70%-30%. Various channels mentioned above such as Justetf or Finanzfluss postulate that the rebalancing of broadly diversified equity ETFs is beneficial to returns. If that’s true, a poorer return over time increases the likelihood of better returns later. Finanzfluss simply recommends annually reviewing the deviation from the original allocation strategy and acting accordingly. Justetf even goes one step further and suggests constantly monitoring the allocation, which is much more time-consuming. For more insights, Elisa can use this calculator*Implemented in Rust and compiled to WebAssembly. There are the following options:

• Simulation of price developments as pure random walks*i.e. without the increased probability of better returns if things went worse for a while
• Backtesting some historical indices like the MSCI World and the MSCI Emerging Markets*I found the data on curvo.eu/backtest. On curvo.eu are significantly more precise options for historical backtesting.

The following table shows the values for the 70/30 split. Please note that any costs such as taxes, transaction fees, or Elisa’s workload are not taken into account.

Indeed, in the historical data re-balancing is actually profitable*or the calculator still has one or the other bug. Just that 6 years as re-balancing interval had made so much additional return in the past is surprising to me. However, choosing the right re-balancing strategy does not seem trivial to me. Now I suspect that Elisa doesn’t need to geographically re-balance once a year and should not react immediately to every minor change. Consider that changes cost time and possibly Euros. As an alternative to strictly temporal rebalancing, one could act from based on the level of deviation as justetf suggested, too. The following corresponding table was re-created using said computer.

The deviation stragies would have performed worse than the 6-year strategy. Our calculator determines retrospectively that re-balancing every 81 months with a simultaneous deviation of 9% or more is the best*We optimize over integer deviations in percent and months. of all strategies leading to 10.28% annual return. Deriving recommendations for the future is not obvious to me.