Fine-tuning the SAA

  • Written by 
  • 10/10/2016

Our Strategic Asset Allocation (SAA) is the cornerstone of our investment process. We have a well-researched and tested SAA process, which uses robust statistical methods based on how investors react to risk and return. Although our SAA model portfolios are intended to be stable over time, and not change in response to the immediate economic or market environment, we update and fine-tune our approach on a regular basis.

Balancing yin and yang 6 of 7

Residents of the United States, please read this important information before proceeding

Please read this important information before proceeding.

“A good asset allocation should balance risk and reward in a way which suits both the investor’s attitude to taking on risk as well as the extent to which their circumstances allow it”

Asset allocation is the way investments are organised across broad asset types such as stocks and bonds. A good asset allocation should balance risk and reward in a way which suits both the investor’s attitude to taking on risk as well as the extent to which their circumstances allow it.

Strategic and Tactical Asset Allocation – it’s just a matter of time...

The SAA forms the starting point, while the TAA is the set of shorter- and medium-term over- and underweights relative to the SAA, with the aim of improving performance. The TAA usually refers to tilts between asset classes, such as overweight stocks and underweight bonds, or tilts between countries and investment themes within asset classes, for example, overweight US equities and underweight Japanese equities.

Our approach to SAA – deeper and broader

Our approach to SAA is rooted in what we believe our clients value most. We aim for portfolios that will deliver the best possible return for risk, but which won’t suffer too much if adverse scenarios unfold. Our SAAs are built within a robust statistical framework. We emphasise a forward-looking analysis using key drivers of asset class returns, more realistic risk measures and our insights into investor behaviour. How this approach helps to differentiate us is illustrated below.

At the same time, our SAAs are designed to be well-diversified, to avoid the danger of overrelying on any single asset class (‘The most important thing: asset allocation’, Compass Q1 2016). This is achieved through a process which involves selecting an appropriate roster of asset classes, estimating their future returns, simulating many possible future outcomes, and using them to create an allocation for each risk profile that combines strong anticipated performance with robustness and stability.

Our SAA process – a robust statistical framework

Forward-looking: Many SAA approaches are only based on historical returns with the assumption that history will repeat itself. We combine historical information with forward-looking views of fundamental economic drivers, such as growth and inflation, and how these translate into asset class returns.

To reflect uncertainty, we simulate many possible worlds to act as a proving ground for portfolios, whereby the vulnerabilities of overly-concentrated asset allocations are exposed. The process results in prudent diversification.

Volatility: The most common way to measure financial risk is through volatility – how much returns go up and down. However, we prefer measurements that focus on how much returns go down to manage the risk of poor outcomes. We appreciate that risk is one-sided, so we measure it to mirror what investors should wish to protect their portfolio from: the chance of bad outcomes. Using our downside-focused behavioural risk measure in building our asset allocations means we are more wary of assets which have a tendency to drop rapidly in value.

Behavioural finance: We believe that even the most conservative investor would be willing to make a risky investment if the potential return is high enough. Behavioural finance tells us that this relationship between risk and return is not a straight line. Investors typically require much more return to willingly accept additional investment risk. Our focus on this trade-off between risk and return means that we can look at, and access, opportunities across the whole spectrum of risk. As a result, we will take the risks that are relatively well-compensated in terms of expected return and avoid those that are not.

What’s involved in the update?

We revisit our SAAs regularly not because we expect dramatic revisions in the allocations, but because good investment practice requires that we:

  • Review our assumptions to ensure they remain valid;
  • Update our data with the most recent outcomes;
  • Incorporate what we’ve learned, and how the world has changed;
  • Look for ways to improve our quantitative modelling by building in any advances in industry best-thinking and academia;
  • And capture the current thinking of our market experts.

As the world is inherently uncertain, to arrive at the best allocation requires us to think very carefully about what we know about these assets. This knowledge comes from two crucial sources: current market data and historical data on how assets have performed (objective); and expectations of how they will perform (subjective). Effective asset allocation combines both sources of information to arrive at the clearest understanding of the risks and expected returns that investors face today.

We use a methodical and robust process. This includes ensuring expert opinions are captured systematically, and discussed extensively; and how these are combined to arrive at expectations for future returns that are grounded neither too much in the past, nor too much in subjective opinion. By making all the assumptions explicit and testable, and by revisiting them on a regular basis, we get some comfort and confidence that our allocations have been tested to the highest standard, and use the best thinking available.

What are the main changes?

Despite a detailed and thorough review, the resulting changes are relatively minor. The percentage allocations have not changed significantly (Figure 1), and nor have many of our fundamental assumptions. However, we have upgraded our process in a few key areas, as well as reassessed the returns that we anticipate over the business cycle.

Summary of the key changes

We have carried out a comprehensive review of all methodology steps and have made a number of upgrades to several areas.

These include the risk tolerance framework and asset class risk estimates.

  • The spread between the risk profiles has increased slightly with less risk in the lower risk profiles (Risk Profiles 1 and 2) and more in the higher risk profiles (Risk Profiles 4 & 5) while Risk Profile 3 remains relatively unchanged. This result is due to an increase in risk aversion between the different types of clients. This allows for higher risk solutions for clients with both high risk tolerance and high risk capacity, and lower risk solutions for clients with low risk tolerance and low risk capacity. The newly improved and defined methodology therefore creates a more differentiated risk profile offering.
  • Improvements to the way we estimate risk and volatility has led to new forecasts. These have shown that ATS tends to have low but spiking volatility, while bonds experience more stable levels of volatility. This development in risk estimation has a varied effect across different portfolios due to the complex nature of interactions between asset classes. For many of our allocation models this results in the weight to Investment Grade Bonds increasing across most risk profiles (including Risk Profiles 1-5) while the allocations to Alternative Trading Strategies reduces.


Even though our SAA is based on long-term views to support long-term investments and on a well-researched and tested methodology, it is important to constantly look to upgrade and improve the process. This is why we review our SAA process, inputs and methods, at least every 18 months. These reviews may lead to changes in the asset classes we use or the weights of these asset classes but only if we believe that they would truly benefit our clients.