Stats › Active, passive and flows › Fund survivorship
Fund survivorship and persistence
When you look at a list of funds and their long term records, you're only seeing the ones that are still here. The ones that did badly enough to be shut down or folded into another strategy aren't on the list at all. Nobody hid them, they just stopped existing, and the average of what's left looks better than what investors actually got.
This page puts numbers on that. Of 2,447 US domestic equity funds alive twenty years ago, 894 made it to the end of 2025. The other 1,553 are gone. Alongside the survival data there's a second question that's just as interesting: do the good ones stay good? The answer is close to no. Across five consecutive years, the share of top quarter funds that stayed in the top quarter was zero in every domestic equity category except small caps.
I put this together because survivorship bias is the single easiest way to fool yourself with data, and it isn't only a fund problem. The same mechanism turns up in stock screens, in strategy backtests, in any list assembled from things that are around today. Seeing it in a dataset this large makes it much harder to forget. Everything here comes from one publisher's scorecards, on one data date, and I've flagged the places where I've done arithmetic on their published numbers rather than quoting a figure they printed. The documents are linked below and they're worth reading directly.
Of 2,447 US domestic equity funds that existed twenty years ago, 36.53% survived to the end of 2025. 1,553 funds are gone. Large cap growth funds fared worst, at 28.96%. Persistence is worse still: across five consecutive years to December 2025, the share of top quartile funds that stayed top quartile was zero in every domestic equity category except small caps, which managed 2.42%. The cleanest test is the top half one, where 4.49% of above median large cap funds stayed above median for four subsequent years against 6.25% expected by chance. Below chance.
Members only
The rest of this page is for members
Below this point there are 9 sections, 1 chart, 1 table and 3 named sources, roughly 1900 words of it. Every figure carries the source it came from and the date the data is from.
You can keep browsing every statistic in the library for free. The intro and the summary are always open.