The long-term investor has to make many decisions in response to certain problems within the investment process. One of these pertains to the tactical asset allocation (TAA) decision, which is the portfolio’s current deviation from the strategic asset allocation (SAA). This article considers how decisions on deviations from SAA should be made – actively or passively.
The most common approach is to leave it to an external asset manager who continuously decides the relative size of the asset classes. The opposite is passive TAA. There are different ways of passively managing the asset allocation, one of which is to restore the SAA whenever the limits set on the maximum deviation from it have been exceeded. This method is commonly called a passive rebalancing strategy and the findings of this research indicate excess return (ER) could be as high as 0.5% annually.
The SAA is usually expressed as a fixed distribution between domestic and international stocks and bonds, as well as property and alternative investments. Considering the research effort spent in this process, it is not desirable that the portfolio’s exposure dramatically deviates from SAA. This could produce higher risk levels than those outlined in the SAA process. Thus there are good arguments for creating rules regarding the portfolio’s maximum deviation from SAA, or in other words, limiting TAA.
In addition to these limits, the investor must decide on how to run TAA. There are two answers to this question: an active decision, based on current analysis of the asset classes’ relative attractiveness, active tactical asset allocation (ATAA), or passive tactical asset allocation (PTAA). PTAA involves no analysis or judgement, the investor (or the investment manager) merely rebalances the asset weights to SAA whenever the limits for TAA are exceeded.
The main purpose of this article is to shed some light on how the institutional investor should run TAA (actively or passively) and on who should run TAA – the investor or the manager. In particular, the aim is to illustrate the consequences associated with a passive rebalancing strategy. The questions outlined will be analysed from a return as well as a risk perspective. The costs and the construction of the benchmark will also be discussed.
The methodology involves a comparison between the return of a portfolio using PTAA as well as passive management within each asset class and the return of the benchmark. The benchmark consists of 50% Swedish equities and 50% Swedish T-bonds (monthly weighting of the return of each index) throughout the research. This methodology produces a return difference between the portfolio and the benchmark that is attributable only to the PTAA effect. The initial maximum deviation from SAA is ±10 percentage points for equities and bonds. Rebalancing to SAA takes place at month ends when any of the limits are exceeded.
The analysis shows that the return of the PTAA portfolio is 19.14% a year for the 20-year period. The benchmark return is 18.8% a year. Consequently, the PTAA produces a positive excess return of 0.34 percentage points per year.
Figure 1 illustrates the real asset exposure of the portfolio during the 1980s and the 1990s. Rebalancing has occurred seven times during the period (indicated by arrows) – on average at an interval of less than three years. The limits have been exceeded because of over-weighting of T-bonds on only one of these occasions (30 November 1990). The average exposure to equities was 52.41% and 47.59% to T-bonds.
How is it that the PTAA produces positive excess return? The intuitive reason actually appears in Figure 1. During the seven sub-periods delimited by the starting point and the rebalancing occasions, the investor has gradually had a larger exposure to an asset class with higher returns compared to the fixed weighted benchmark. Simultaneously, the investor has gradually had less exposure to an asset class with less return compared to the fixed weighted benchmark.
In terms of the returns for PTAA portfolios at varying limits for maximum deviation from SAA, the average ER is +0.43 percentage points a year for maximum deviations between ±6 and ±13 percentage points. In these instances, ER varies between +0.12 and +0.79 percentage points. The number of rebalancing occasions varies between 18 and six times during the 20-year period. Figure 2 illustrates ER depending on the size of the maximum allowed deviation from SAA.
Figures 1 and 2 tell us that the ER associated with PTAA is dependent on: the size of the maximum deviation from SAA and the relative volatility of the asset classes. Another way of putting this is to say that the investor wants to put the rebalancing strategy into effect as many times as possible (which suggests small deviations from SAA) and, each time, the investor wants the effects to be as big as possible (which suggests large deviations from SAA).
Eventually, the size of the maximum deviation from SAA should be governed by the short-term risk aversion of the investor and it should be decided upon in the ALM process.
It is not meaningful to talk about return without talking about risk – or rather, the specific risk associated with the specific return. The average excess risk (standard deviation) is +0.29 percentage point a year for maximum deviations from SAA between ±6 and ±13 percentage points. In these instances, excess risk varies between +0.11 and +0.53 percentage points. The average active risk (tracking error) is 0.97%, and varies between 0.57 and 1.39%.
The findings suggest excess return also produces excess risk (measured traditionally). This causes a discussion on the alternative at hand – active tactical asset allocation (ATAA) – and the risk associated with it. The asset manager (or the investor) has to deviate from SAA to be able to produce ER at the TAA level. The risky part of this is that a wrong decision is being made, resulting in negative ER (ie, the benchmark return exceeds the portfolio return).
We have just seen that PTAA does not produce negative ER. Further, it can be argued that PTAA is not associated with risk at all (who would feel that positive ER is risky?), though the investor should make sure that the one’s responsible for the PTAA process have sufficient competence and resources to fulfil the implementation.
The conclusion is that the investor can render positive ER possible through the implementation of a passive strategy at the tactical asset allocation level. The size of the ER will ultimately depend on the relative volatility of the asset classes outlined in the strategic asset allocation process and the size of the maximum allowed deviation from SAA. The findings of this research indicate that ER could be as high as 0.4–0.5 percentage points annually.
The investor can choose to be responsible for the implementation himself, or hire some external party to perform the tasks associated with PTAA.
It is important though, that the investor makes sure that the one responsible for this management level is duly equipped in terms of competence, resources and access to information necessary to implement PTAA.
The risk of making wrong decisions at the TAA level is minimised through PTAA compared to ATAA. PTAA automatically answers the (difficult) question on which asset class to over- or underweight.
There is reason to believe that the costs associated with PTAA are below those associated with ATAA. The grounds for this conclusion are that cashflows between asset classes will with PTAA only take place when they must according to investment policy guideliens. With ATAA they will take place whenever the asset manager feels the relative attractiveness of the asset classes has changed. Eventually, the result of this research paper suggests that the investor should implement specialist management at the asset class level, and PTAA at the total portfolio level. Another way of putting this is to say that the ATAA manager, in order to claim outperformance of the benchmark, has to achieve a return that is higher than the combined benchmark return and PTAA excess return.
Nicklas Fahlstrom is a senior consultant at Wassum Investment Consulting in Stockhol

Topics