Alexa asked it straight. Fees on fees feel out of control right now, especially with the top ten names, so what are you doing about the feeding frenzy, and do you need access to those hot names.
I expected Dan Fordyce to answer the access question. He answered a different one.
What came back was a set of numbers. Two new names a year in venture. Two in growth equity. Two in buyout. Roughly equal weight across the three. A ballpark check size. A small, defined slice for directs. Then this: “that also forces us to keep the bar exceptionally high.”
I went in thinking the scarce thing in private markets is access. I came out thinking access got cheap, restraint got expensive, and restraint does not survive a hot market unless somebody writes it down as a number first.
Three things I took from this episode that might interest you:
The first is that the cap is the strategy, not a constraint on it. Dan looks at eight or nine hundred things a year and gets to twenty or thirty he would genuinely want. Six get committed. The gap between thirty good options and six slots is not something he apologized for. It is the mechanism. He traced it to 2021 and 2022. “The number one thing to be a good private equity allocator is don’t miss a vintage year.” Be consistent, and do not overallocate into the years that look hot.
That is a pitch count. A coach does not pull a starter in the seventh because the guy looks tired. He pulls him at a number set before anybody threw a warmup pitch, because in the seventh with the crowd standing, nobody’s read on a tired arm is worth much, his own included. Two a year is the same instrument. Not taste. A number that exists to overrule taste at the moment taste stops working.
The second thing cut against what I assumed. I had the direct co-investments filed as the alpha line. Summation writes alongside its GPs in a handful of deals a year, so I read that as picking. Dan took it apart. “We’re not asset pickers.” The directs are there, in his words, “really at good economics to offset our fee load.” If a manager puts money into a company, they sanity check the thesis and write a small check at better economics than the fund itself pays. The purpose of the line item is the fee math. The valuation call stays with the GP. I have heard a lot of allocators describe co-invest. I had not heard one call it a cost instrument.
The third confirmed something instead of teaching it to me, and I would rather say that than dress it up. I asked Dan why the way people treat others, when there is nothing in it for them, matters more now than it used to. I asked because he has said it to me before. His answer was the one I expected: as knowledge and information get democratized and the other barriers fall, what stays durable is relationships and reputation. What moved was where it sits. I had it as the soft layer on top of the model. I think now it is the only input that never gets commoditized, which makes it the last thing you can actually underwrite. Then he described a family office friend who toured the late stage crossover funds and found “something like the same 10 names were in every single portfolio in different weights.” If ten names are the industry’s portfolio, the portfolio is not the differentiator.
Time will tell if this actually works. Summation is about three years in. Dan, who calls himself a “very wise 31 year old,” says this frenzy reminds him of 2021 and 2022, only broader, with more real businesses underneath it. But a pitch count is easy to hold in the fourth inning. The test comes the first time a name they passed on triples in public with an LP on the phone. I am also chewing on the structure. They chose open ended on purpose, to sit as close to an endowment as they could, and Dan was careful that this is not an evergreen or an interval fund. I did not push on where the redemption mechanics land, and I should have, because an endowment has nobody who can ask for the money back and that is doing a lot of work in the comparison.
Shane Goudey, in the Sidley segment, put a price on the access everybody is chasing. Companies have started extending drag, tag and co-sale rights through an SPV down to the underlying LPs inside it. The first time that happened, he said, it “just sort of blew people’s minds.” Now, “if you want the access, these are the kinds of things you may have to agree to.” That is the other half of Dan’s point. The hot name is available. The terms are what nobody priced.
Last week on Carry On, Santosh and I spent an episode on what still counts as venture. This is the LP side of the same argument. Both land in the same place: the industry has gotten very good at manufacturing access and no better at saying no.
Listen here: https://swimmingwithallocators.com/podcast/lp-destroys-fees-on-fees/
With gratitude,
earn