In a previous article, Whose performance is it anyway?, we addressed the significance of performance, for both clients and the firm, in goal-based wealth management. Our key observation was that when assets are managed to meet client-specified outcomes (or liabilities) within their own tolerances, performance is mainly a product of the choices of the client rather than the firm. In fact, where firm-wide performance matters it means firms are treating performance as a contest and this will interfere with their proper role of helping clients make the best choices for them.

In this article we focus on how outcomes-driven investing should employ benchmarks to explain the contribution of each party to the returns earned in a period.

Benchmarks in goal-based wealth management

FCA rules require discretionary managers like us not just to measure and report each client’s performance in each reporting period (at least quarterly) but also to compare it with ‘an appropriate benchmark’. We do both at the level of the virtual goal-based portfolio as this is what is being managed to deliver the planned outcomes for the goal. What constitutes an ‘appropriate’ benchmark is not defined by the FCA but inappropriate benchmarks are likely to be misleading and often deliberately so.

The FCA also requires us to show ‘a prominent warning that the figures refer to the past and that past performance is not a reliable indicator of future results’. It is possible to reconcile these two observations if benchmarking is seen as part of the accountability of a manager to its client, whatever the information content.

‘Accountability’ means being able to quantify divergences and put them in an appropriate context. We believe that means we need two forms of benchmark for our portfolios, each aiming to identify different things.

• The effects of selecting our generic liability-driven approach (the client’s decision)
• The tactical effects of our own implementation of that approach (our decisions).

Benchmarking our decisionsPerformance of every individual spending goal is measured against a customised dynamic benchmark. It is dynamic because it has a time profile for ‘derisking’, replacing equity with risk free assets, based on the shortening duration of the liabilities. As a ‘glide path’, it reflects the individual risk approach for the whole plan. In our modelling the risk approach is constant (unless changed as a plan input along the way) though the risk level (balance between risky and risk free) changes with time, as a function of the duration of the remaining cash flows. The derisking profile assumes time is shortening but market conditions are constant, as if always ‘normal’, defined as in line with each asset’s own historic trend. This is shown below for a sample of ‘drawdown’ portfolios with different cash-flow durations and different risk-aversion levels.