Benchmark selection and performance

Dirk Broeders*, Leo De Haan

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

1 Citation (Web of Science)
26 Downloads (Pure)


Using regulatory data free of self-reporting bias for 2007–16, we decompose investment returns of 455 Dutch pension funds according to their key investment decisions, i.e., asset allocation, market timing and security selection. In extension to existing papers, we also assess the impact of benchmark selection. Over time, asset allocation explains 39% of the variation of returns, whereas benchmark selection, timing and selection explain 11%, 9% and 16%, respectively. Across pension funds, asset allocation explains on average only 19% of the variation in pension fund returns. Benchmark selection dominates this by explaining 33% of cross-sectional returns. We relate the choice for a specific benchmark to investment, risk and style preferences.
Original languageEnglish
Article number1474747219000246
Pages (from-to)511-531
Number of pages21
JournalJournal of Pension Economics & Finance
Issue number4
Publication statusPublished - Oct 2020

JEL classifications

  • g23 - "Pension Funds; Non-bank Financial Institutions; Financial Instruments; Institutional Investors"
  • g11 - "Portfolio Choice; Investment Decisions"


  • Benchmark selection
  • asset allocation
  • investment performance
  • pension funds
  • Asset allocation
  • benchmark selection

Cite this