A fund's vintage year is the year of its first capital call (or first close). Two facts compound to make a young fund's own reported numbers systematically look weaker than its ultimate quality, purely from where it sits on the J-curve (see capital calls, distributions, and the J-curve): the denominator effect: PIC is large early (most of the commitment has been called to fund investments) while the numerator (distributions plus a conservatively-marked NAV) is still small, so TVPI/DPI print low almost by construction in years 1–4 regardless of whether the underlying picks are excellent, and the marking convention. GPs conventionally hold new investments at cost or a modest step-up for the first year or two (a "conservative marking" norm most LPAs and IPEV guidance both encourage), so even a company that is executing well may not show it in NAV yet. Worked Example 3 makes this numeric: Konkan's own interim snapshot at year 7 of its life shows a materially lower TVPI than its eventual, fully realized figure, the identical fund, the identical eventual outcome, simply observed at an earlier point on its own curve.
Advertisement
Why are vintage-year benchmarks biased upward?
The corollary is the selection effect in vintage-year benchmarks: databases such as Cambridge Associates, Burgiss, and Preqin report median/quartile IRR and TVPI by vintage year, and those benchmarks are built from funds that voluntarily (or via LP-agreement) report their performance, funds that wind down badly, or GPs that stop reporting after a poor run, are underrepresented, biasing the reported benchmark upward (a survivorship/self-reporting bias, the same species of measurement problem M2's forensic modules teach you to distrust in company-reported metrics). The practical discipline: never compare a fund's TVPI to a different vintage's TVPI without adjusting for how many years each has had to mature, and treat any single-vintage benchmark figure as directionally useful, never as a precise population statistic, pull the provider's own methodology note (named primary source) before quoting one. This is also where M6.04's lens earns its keep directly: an LP committee that downgrades a manager, or an analyst who writes off a fund, purely because its year-3 TVPI sits below a stale benchmark is pattern-matching on noise the way M6.04's recency and narrative-fallacy entries describe, the discipline that catches it is the same "know the base rate, then look at the specific evidence" habit taught there.