Mixing Up Risk: One Cocktail to Avoid
Companies are increasingly attempting to assess all their business risks on a level playing field. This means they, and corporate stakeholders, are interested in understanding how risk in the pension scheme relates to other risks within the business, both in terms of magnitude and correlation.
Falling interest rates, increasing life expectancy, diminishing equity values and changes to legislation have all increased the focus on the management of past and future pension commitments. The weight of pension issues no longer rests solely on the pensions manager but has drawn in finance directors, credit analysts and fixed income investors. It is now clear that the risks a company is exposed to through its defined benefit pension scheme are no longer ‘off balance sheet’.
Increasingly corporates are using value-at-risk (VaR), a risk measurement methodology developed within financial institutions, to quantify the economic risks they are running in their pension schemes. The VaR95 measure is used to estimate the 95th percentile worst one year move in the pension scheme’s funding level, which equates to the worst 1 in 20 year event. VaR calculations not only allow for movements in the value of the pension scheme’s investment assets but also consider these movements relative to changes in the value of the scheme’s liabilities. Linking the risks of the assets and liabilities in the pension scheme in this way is a particularly attractive element of risk budgeting studies, which is probably why 80 per cent of Watson Wyatt’s largest clients have conducted such exercises. A large proportion of these monitor their VaR annually with a small number now requesting quarterly VaR reports.
Companies using VaR to measure the risk in their pension scheme aim to answer some key questions:
For a company with a large pension scheme, the combination of on and off balance sheet debt and a large pension scheme VaR can be a dangerous cocktail. A large fall in the value of a pension scheme’s investments relative to the scheme’s liabilities will result in a significant increase in the company’s FRS 17 deficit. Credit analysts at both the rating agencies and banks now view this deficit as debt. This increase in the debt burden could precipitate a ratings downgrade, prompt fixed income analysts to issue sell recommendations or cause the company to break its bond or bank loan covenants. A company may then find itself under pressure to repair this increased deficit through extra contributions at a time when there has been a material increase in its cost of capital.
Beyond its use as a tool for measuring the size of the risk, VaR is a useful framework for companies to question their reasons for taking specific economic risks within the pension scheme. Generally, pension scheme risks fall into two areas: rewarded and unrewarded risks. Rewarded risks are investments in risky assets, such as equities or corporate debt taken because the stakeholders in the pension scheme have strong views that the rewards compensate them for the additional risks. Unrewarded risks are those that companies are intrinsically exposed to through their pension scheme but are not compensated for taking. Many companies have been exposed to material unrewarded risks without being aware of their true nature. Examples of unrewarded risks include interest rate and inflation risks that drive the value of a scheme’s liabilities and the unintended currency bets schemes make through unhedged investments in overseas assets.
Analysis of the pension scheme’s risks through a VaR framework also allows a company to ask key questions such as:
Analysis of a pension scheme’s sources of risk, and how they interact, helps companies increase the efficiency of their pension scheme investments. By measuring and understanding a pension scheme’s exposure to different sources of risk, Trustees and their advisors can work together to reduce the level of risk as measured by the VaR to an acceptable level. The efficient removal of unrewarded risks is becoming key for many companies in our tough and unforgiving world. For example in the case of unrewarded interest rate risk it might be that simply increasing the average maturity of the pension scheme’s bond investments will sufficiently reduce the scheme’s VaR. In some cases risk reduction can be achieved without reducing the scheme’s exposure to appropriately rewarded risks. For example introducing more risk diversification by investing globally rather than just in the UK and by utilising asset classes other than equities (such as property or private equity) the VaR can be reduced without sacrificing expected return.
This drive by companies to measure and manage all their business risks means they are increasingly focused on understanding how the risks in their pension scheme relate to other drivers of risk within their business. Analysis of risk correlation means the response to the outcomes of these risk budgeting exercises can be sector specific. Examples of sector specific inherent economic drivers that would influence a company’s assessment of its pension scheme arrangements include:
Another consequence of considering the interaction between business and pension risks would be questioning the rationale for companies whose performance is highly geared to the economic cycle (ie high beta companies) investing in equities.
Many companies now view active measurement and management of their pension scheme risks using VaR techniques as a fundamental discipline to help them understand the strength of their risk cocktail. By then considering how the ingredients of their cocktail might be mixed they can avoid combinations that could have negative impacts on their financial flexibility, credit ratings and profitability.