›  Strategy 

Risk factors: taking risk budgeting one step further

An increasing number of pension funds are opting to invest in ‘alternative’ or ‘smart beta’ indices to supplement their passive management activities. Several competing methods currently exist, each with their own objectives. Analysing the risk contribution of each factor by type of approach gives investors a clearer picture of the various competing smart beta methodologies.

Article also available in : English EN | français FR

Many large investors already apply risk budgeting to asset classes in their allocation (although sometimes mistakenly referred to as ‘risk parity’, it is not always a question of equally weighted risks). The main difficulty lies in defining risk budgets that are appropriate for each investor. By taking economic risk factors into account, we can accurately define precise risk budgets to match the specific objectives of each investor profile. This represents a step up from the traditional risk parity approach, which focuses solely on asset classes.

The economic and financial crisis has prompted many pension funds and institutional investors to reconsider their long-term asset allocation strategy. They are calling into question traditional portfolio optimisation as propounded by Harry Markowitz and are increasingly leaning towards the risk budgeting approach and its corollary, risk parity.

The principle underlying risk budgeting-based allocation is simple: the asset allocation is established on the basis of the risk contribution of each portfolio component, rather than the expected return. In the case of a portfolio comprising two assets managed using the equallyweighted risks approach, each asset makes an equal contribution to risk and therefore to performance. To achieve this balance, exposure to the riskier assets is reduced, and vice versa.

Insofar as risk factors can affect more than one asset class at a time, ignoring them and focusing solely on asset classes can lead to a concentration of a limited number of factors.
Thierry Roncalli, Head of Quantitative Research, Lyxor Asset Management

Traditional diversification via weightings leads to markedly different results. For example, in a balanced portfolio composed of 50% equities and 50% bonds, the equity component accounts for almost 90% of the portfolio’s volatility. By way of symmetry, the equity component will also generate the same proportion of performance.

Risk-based diversification thus intuitively seems much more accurate and equitable. Investors can use it to optimise their risk profile beyond mere diversification based on market capitalisation, and thus to obtain a better risk-adjusted return. In addition, it allows for an ex-ante understanding of performance attribution: due to the mirroring effect between risk and performance, investors can anticipate the source of their portfolio’s performance.

Risk factors

A major challenge remains to be addressed, namely defining a risk allocation that is in line with investors’ objectives. Even if a portfolio’s allocation appears to be optimally, or at least neutrally, diversified using the risk parity approach, it may harbour other hidden sources of risk, e.g. financial and economic factors affecting the performance of the asset classes in the portfolio. Insofar as risk factors can affect more than one asset class at a time, ignoring them and focusing solely on asset classes can lead to a concentration of a limited number of factors.

Long-term investors may go a step further in applying this approach by no longer considering only asset classes (equities, bonds, commodities, etc.) but also economic risk factors such as economic activity (GDP, industrial production), inflation (commodity and consumer prices), interest rates (real interest rates, steepening and convexity of the yield curve) and the effective exchange rate.

Using such a framework, the portfolio is constructed on the basis of the risk budget allocated to these various economic criteria. At first glance, this might seem easily achieved by simply linking an asset class with a particular type of risk – e.g. bonds with interest rate risk – and applying the traditional risk budgeting method. However, such an approach would not be effective as an asset class can be exposed to several economic risk factors at a time. While equities are affected by economic growth and industrial production, they are also impacted by interest rates – as shown by the Gordon-Shapiro model – and inflation. Therefore the sensitivity of each asset class to certain risk factors must be identified in order to establish an asset allocation.

It is all the more appropriate to consider economic risk factors rather than financial assets given that many pension funds and asset managers reason in macroeconomic terms. For example, a pension fund expecting a lasting period of growth may try to increase the GDP sensitivity of its allocation. By assigning a budget per risk factor, it can do so with great accuracy by selecting a set of assets sensitive to growth.

Such an approach standardises the concept of risk parity in that several asset allocations can then be compared on the basis of common factors, regardless of the asset classes used. It also reconciles the quantitative approach to strategic asset allocation with the fundamental approach.

Smart beta selection

Furthermore, an increasing number of pension funds are opting to invest in ‘alternative’ or ‘smart beta’ indices to supplement their passive management activities. Several competing methods currently exist, each with their own objectives. These include approaches such as the Equally Weighted Portfolio (EW), Minimum Variance Portfolio (MV), Equal Risk Contribution Portfolio (ERC) and the Most Diversified Portfolio (MDP), among others. Analysing the risk contribution of each factor by type of approach gives investors a clearer picture of the various competing smart beta methodologies.

By way of example, the S&P 100 index is sensitive to economic activity but it is also affected by interest rates as shown in the risk contribution table. This can be explained by the fact that the slope of the yield curve is a leading economic indicator. Interest rate risk is therefore not exclusive to bonds.

As the equities making up an MV portfolio are generally among the least volatile in their universe, they have bond-like characteristics. Result: implementing the MV approach in the equity component introduces bond risk. From a strategic allocation viewpoint, this amounts to transferring part of the equity component allocation to the bond component.
Thierry Roncalli, Head of Quantitative Research, Lyxor Asset Management

Sensitivity to interest rate risk is higher in the case of minimum variance portfolios. As the equities making up an MV portfolio are generally among the least volatile in their universe, they have bond-like characteristics. Result: implementing the MV approach in the equity component introduces bond risk. From a strategic allocation viewpoint, this amounts to transferring part of the equity component allocation to the bond component.

Analysing the risk contribution of each factor thus allows us to assess a portfolio’s economic profile. It also provides a clear idea of sensitivity to economic risk factors and, above all, allows for a more accurate prediction of how “smart beta” indices will behave in different macroeconomic conditions.

Thierry Roncalli , April 2013

Article also available in : English EN | français FR

Share
Send by email Email
Viadeo Viadeo

Focus

Strategy CPR AM has recently launched CPR Invest – Global Disruptive Opportunities | A look back at an accelerating phenomenon: disruption

The recently theorised phenomenon of "disruption" is defined as a process whereby a product, a service or a solution disrupts the rules on an already established market. Technological progress, along with the globalisation of trade and demographic changes are now helping to (...)

© Next Finance 2006 - 2020 - All rights reserved