Amonis is convinced that traditional financial risk measures are not sufficient to manage pension fund risk. It believes isolated measures are inadequate and often difficult to interpret together. Instead, the fund considers it needs a carefully selected mix of tools which interact closely. Unlike conventional methods, its risk management is not limited to assets: potential risks also include loss of purchasing power, illiquidity of investments and changing legislation. The fund’s risk then becomes a total scheme risk, with different elements needing interactive management.

Identifying separate risk parameters for each type of risk, then analysing their interaction, is very cumbersome, and might produce discrepancies. So Amonis looked for other metrics – maybe less granular, but easier to calculate rapidly, providing consistency and simplicity and much better, albeit approximate, results.

Both macro- and micro-factor modelling is used to analyse and manage total scheme risk: Amonis prefers this to historical time series value at risk (VaR) as a way of looking at embedded risks. But the biggest leap forward was the extension of factor modelling to include liabilities.

Viewing assets as long positions and liabilities as short positions, the difference between the two is the funding surplus or deficit, so Amonis can measure the factor sensitivity of the funding ratio. It can then analyse which factors will drive the pension plan’s funding risk. These “betas” are now used to stress the investment portfolio when deciding on the asset allocation. By using different forms of scenario analysis, a holistic, forward-looking view on the long-term solvency risk emerges.

Using this approach means that calculations can be made repeatedly and very quickly each day. Furthermore, the board is informed accurately, using an easy-to-understand mechanism which translates complex issues into clear risk-integrated options for making investment decisions.