Explain why hedge fund index returns as reported can be biased
The idea
Hedge fund indexes are built from voluntary reporting, and each of the main biases this creates pushes returns upward. Survivorship bias removes funds that closed, which are disproportionately the poor performers. Backfill bias occurs when a fund joins a database and its historical record is added retrospectively — a manager does not volunteer a bad history. Selection bias means managers choose whether to report at all. Together these can overstate index returns by several percentage points a year, so a reported index is an upper bound on the experience of an actual investor.
A practice question
What is survivorship bias in a hedge fund index?
- Closed funds are left out of the index — correct
- Funds report returns before feesWhy not: A basis-of-reporting question, and most report net.
- Only the largest funds are includedWhy not: Size cut-offs exist in some indexes and are a separate matter from funds disappearing.
They were disproportionately the poor performers. The index describes the funds still standing, and the record of those that failed leaves the sample with them.
A harder one
How does backfill bias arise?
- Managers restate returns after an auditWhy not: Restatements are occasional corrections rather than a systematic upward pull on the index.
- Indexes are rebalanced annuallyWhy not: Rebalancing changes weights, not whose history is included.
- A joining fund's prior history is added — correct
Managers with poor histories do not join, so the record that comes with them is a favourable one. The decision to start reporting is itself informative, and adding the retrospective track record of self-selected entrants raises the index's history without any investor having earned it.
The hardest tier
What is the practical implication of these biases for an investor assessing hedge funds?
- Indexes should be ignored entirely in favour of manager marketing materialWhy not: Marketing material is the most selected data of all, which makes it worse rather than better.
- A reported index return is an upper bound — correct
- The biases offset each other and can be disregardedWhy not: Both push in the same direction, so they compound rather than cancel.
It describes an experience better than the achievable one, so it should not be used as a neutral benchmark. All three biases push the same direction, so there is no offsetting. Benchmarking a manager against an inflated index also makes underperformance look worse than it is, which cuts both ways.
In the app: 9 practice questions on this outcome, a narrated walkthrough with diagrams, spaced review that brings it back before you forget it.
Take the free diagnostic →