Benchmark selection for trustees: when the index does (and doesn’t) tell the right story

Insights

Why the wrong benchmark is worse than no benchmark, and how mandate-relevant comparisons should be specified in the IPS, not chosen by the manager

4 mins

Topics

Industry & Peer Groups Monitor Connect

insight-article-benchmark-selection-thumbnail

Author

Tom Wiseman

CEO

Tom is the Chief Executive Officer of Enhance Group overseeing our multi-jurisdictional Monitor and Select solutions from our Jersey headquarters.

View all insights

Every investment manager produces a benchmark. Most trustees accept the one the manager chooses. The two facts are not coincidental: a benchmark is the single most powerful framing device in an investment report, and the manager has every commercial incentive to choose the one against which they look best. This article explains why benchmark selection is a governance decision rather than an analytical one, the three principles trustees should apply when specifying benchmarks in the IPS, and how Enhance’s framework combines mandate-relevant benchmarks with peer-group context to produce comparisons that survive scrutiny.

Why a benchmark is a governance decision, not an analytical one

A benchmark is not a number. It is a frame of reference, the standard against which performance, risk and decision-making are tested. Choose the benchmark, and the conclusion of the analysis is half-written before the analysis begins. Compare a balanced portfolio to a 60/40 equity-bond composite and it looks broadly tracking; compare the same portfolio to a global equity index and it looks materially behind. Both numbers are true. Only one of them is appropriate for governing a balanced trustee mandate.

Benchmarks chosen by managers tend to share two features: they look most favourable to the manager’s recent positioning, and they migrate when the positioning shifts. A manager who outperformed by overweighting US technology will present against a global benchmark; the same manager, having lost ground when technology corrected, may present against a technology-tilted reference instead. The migration is rarely flagged. Trustees who accept the moving benchmark accept the manager’s framing of their own performance, the exact failure mode discussed in Reviews vs reports.

A benchmark must reflect the mandate the portfolio is actually running. A global balanced mandate should be benchmarked against a global balanced composite, not against a domestic equity index that happens to be familiar. A multi-currency portfolio should be benchmarked against a multi-currency composite, not against a single-currency index that ignores the structural exposure. A growth mandate should not be benchmarked against a balanced index just because that is what the previous IPS used. Relevance is the trustee’s first test of a proposed benchmark, and it is the test against which every subsequent performance discussion will be filtered. Get this one wrong and every quarterly review report will produce a comparison that points to the wrong conclusion.

The benchmark must be specified in the IPS by the trustee, not by the manager whose performance is being tested against it. The trustee’s benchmark is a fixed reference point against which performance is judged across managers, across years and across mandates. The manager’s preferred benchmark is whatever produces the most flattering comparison this quarter. The two are different artefacts. A manager-chosen benchmark can quietly migrate over time, with one benchmark in the marketing pitch, a different one in the appointment letter, and a third in the most recent quarterly statement; the trustee notices only when the comparison stops looking good. Enhance tests every portfolio against the IPS-specified benchmark, which is what matters for governance, and reports the manager’s stated benchmark alongside it for context. Where the two diverge over time, the divergence is itself an action point.

A benchmark must be stable across reporting periods. A trustee comparing a portfolio against one benchmark in Q1 and a different benchmark in Q2 cannot draw conclusions about manager performance from the comparison. Where a benchmark change is genuinely warranted, because the mandate has changed, because the original benchmark has been discontinued, or because the underlying market structure has shifted, the change should be documented in the IPS, dated, justified and applied prospectively. The pre-change history should remain reported against the original benchmark. Stability is what makes long-run performance evaluation possible.

How Enhance combines IPS benchmarks with peer-group context

Every Enhance review tests portfolio performance against two reference points: the IPS-specified benchmark (or composite) for the mandate, and a relevant fund peer group. Enhance Peer Groups, calculated from the real portfolios Enhance independently monitors, are published separately to Monitor reviews and provide an additional peer group perspective to trustees (see Enhance Peer Groups). This gives the trustee a complete picture: how is the portfolio performing against the standard the trustee set, and how is it performing against the broad sample of comparable trustee-managed portfolios?

Benchmarks and peer groups answer different questions. The benchmark answers: is the manager beating an unmanaged index for the mandate? The peer group answers: is the manager keeping pace with what other funds or managers running similar mandates are actually delivering for trustees? A portfolio that beat its benchmark but lagged the peer group is a different conversation from a portfolio that lagged both. The combination produces comparisons that survive scrutiny by Investment Committees, Boards and beneficiaries (see Investment Committees).

What good IPS benchmark clauses look like

The IPS benchmark clause should specify the benchmark or composite, the rebalancing rules, the reporting frequency and the conditions under which the benchmark may be changed. The example below is illustrative for a global balanced GBP mandate.

Clause	Illustrative wording
Benchmark composite	60% MSCI All Country World Index (Net, GBP-hedged where applicable), 40% Bloomberg Global Aggregate Bond Index (GBP-hedged).
Rebalancing	Quarterly rebalancing of benchmark weights.
Reporting frequency	Quarterly portfolio performance vs benchmark, reported in GBP.
Change conditions	Benchmark composition may be changed only by trustee resolution recorded in the trust minute book, with reasoning documented. Pre-change history retained against original benchmark.
Peer-group context	Quarterly review to include relevant peer group comparison.

The bottom line

Benchmark selection is a governance act, not an analytical preference. The wrong benchmark is worse than no benchmark because it produces confident-looking comparisons that point to the wrong conclusion. If you specify mandate-relevant, independently chosen, stable benchmarks in your IPS, you protect yourself against benchmark migration and give every subsequent review report something defensible to test against. Combined with peer group context, the framework produces performance conversations your Investment Committee, Board and beneficiaries can act on.

Featured resource

Featured resource