Ask a VC: how does a venture fund make money?
I’ve always said that the best salesman never sells. He makes the buyer buy! This means you have to understand your client’s intentions, his needs, and what he’s after. Here, you’re trying to sell a piece of your equity to an investor.
In this article, I’m going to oversimplify the goals of a VC, and not discuss for example the impact he’s trying to achieve as an impact fund, or as a contribution to a community or region. Let’s just focus on financial returns.
Let’s define first what a VC is :
- VC stands for Venture Capitalist, the person you meet and who is going to give you money. We also call this person a GP = General Partner. There can be several of them. They are usually a shareholder of the Management Company (the ManCo) that manages a fund or several funds. Through this ManCo usually, or directly (for tax reasons sometimes) or via a transparent vehicle, the GPs are going to invest at least 1% (I recently saw an early-stage fund with a 20% GP commitment) in the fund the ManCo manages.
- Some partners in VC firms are not shareholders of the ManCo, and only are part of the GP pool sharing carried interest (a form of capital gain) at the Fund level; the partners of the ManCo are partners usually in all the funds managed.
- This means that 99% of the money in a fund comes from someone else, actually the main shareholders of a fund. We call them LPs = Limited Partners. This can be very wealthy individuals, institutional investors like pension funds or university endowments or banks, corporates, funds of funds…
- Often, after a first fund, a ManCo raises a second fund, and so on. So the ManCo ends up managing different funds, from different vintages (the year the fund made the 1st closing = when it got its first money — actually a signed written commitment for this money), with potentially different strategies, different shareholders, different sizes, different rules of distribution, etc. Since each fund is different, often the management of the fund is done by a shell company called the GP of the fund, which delegates the management in an advisory contract to a ManCo. So you can end up with 5 funds, 5 GPs for each fund, all advised (=managed) by a single ManCo. Yes, it gets messy…
So what you usually call a VC is actually a group of GPs (shell companies) and LPs collectively investing in a fund (that will give you the money) with a specific strategy, delegating the decision-making to the team at the ManCo.
There are variations for every number and way of working that I’m discussing here, but let’s assume a plain vanilla case.
There are 3 main ways a VC makes money.
1st revenue stream: management fees.
Each fund is usually structured to be close-ended (no more money in once there is a final closing; there are variations possible with SPVS and opportunity funds…), and has a duration of 10 years (I’ve seen from 6 to 15 years). This means that all proceeds have to be returned to the shareholders (GPs and LPs) within that timeframe. This means that a company that joins the portfolio in year 3, will need to be sold at the latest by year 10, so only 7 years later. Since you need maybe 1 year to shop the company around and formalize the sale, you have to put it on sale latest in year 6. This is why you have a liquidity clause after year 5 usually in investment contracts.
This 10-year timeframe can be extended by 1 year, up to 2 or 3 times in most funds, after authorization of a committee representing all shareholders. It’s not automatic, and there are consequences for this. Sometimes, if there are a few companies left in the portfolio, a secondary sale at a deep discount to another fund is possible. Or the VC might be stuck with whatever is left in the portfolio.
The team at the ManCo will invest during the “investment period”, usually in the first 3 to 5 years. This is necessary to then let the companies mature 5–7 years before selling them. You do realize here that most seed-stage companies might need 10 years or more before an exit. This means early-stage funds have to invest super early in the investment period (in the first 2 years), hope for a sale before year 10, and assume that the “bad” companies in the portfolio will just go bankrupt and be written off before year 10.
Why does it matter?
Most funds have a percentage of the committed capital as management fees, that is they are paid to source deals, close deals, manage deals, facilitate their sale, do compliance reporting with regulators, and quarterly reporting to the investors.
This is usually 2% of the committed capital per year, which accrues to 20% over the 10 years of the fund. Some new emerging funds are aggressive and go for less (1,5–1,8% per year on average), whereas some established funds might go for slightly more.
The trick here is that there is no standard formula for how these cumulated 20% are split over the 10 years. Usually, as there is more work involved in building the portfolio during the investment period, the ManCo will take more upfront, and as the portfolio shrinks with exits, take less after the investment period. Usually, there are no management fees for the years after the 10-year period, which helps incentivize the ManCo to sell as fast as possible (see a discussion here about selling portfolios in the light of DPI or TVPI)!
The management fee can be calculated on the committed capital (easy to calculate), on called capital, on deployed capital…
This money is really important because it is THE money that goes into the budget of the ManCo and that will allow the ManCo to pay salaries of partners, principals, associates, staff, offices, travel, computers, access to databases, etc. Everything is in there. But, LPs don’t like to overpay management fees to GPs, they don’t want them to become rich with this, so these fees are negotiated tight, and often a business plan for the ManCo is communicated. The regulator will also ask for it, and make sure that after the 1st closing, the ManCo is viable and will be able to deliver on its promise even without extra money. This also means “VCs” (the ManCo really) do not have tons of extra money available, and will not pay for software (I see every now and then startups trying to sell to them — bad idea), not sponsor events, and charge back all costs associated with a deal to a portfolio company (lawyers, due diligence, etc.), or at least a flat fee. There really is not much money at this level and no crazy salaries should be paid here.
In the Sample Fund I below, let’s assume a 100m final close, an investment period of 4 years, with Capital calls of 15% in the first 6 years, and the rest spread out in the remaining years. This is the money that will come into the fund. I’m assuming a GP commitment of 2%, with 3 Partners. so each will have to cough up 2%*100m/3 = 666,666.67€ as a personal commitment for the fund over 10 years. You do have to pay upfront to play as a VC… In my model here, it’s 100.000 euros/year they have to pay each in the investment period and then some! Not everyone can be a VC, as you need prior capital to pay for your GP share!
Capital calls can be done at each investment, or for example once or twice a year, whatever is more convenient. There are costs associated with a capital call, so the less the better; however calling unnecessary money early, means you are holding that money for more time than necessary, therefore deteriorating your IRR (performance of the fund). It’s really important to optimize it as LPs compare different fund managers on this metric and on others. So a fund announced with 100m€ actually will only have 15m€ of money in the bank in year 1, 30m€ (but much less left after usage of management fees and some investments) in year 2, etc.
Let’s now assume a 2,5% management fee on committed capital during the investment period of 4 years, and slowly decreasing thereafter until year 10, nothing after. The average will be 1,8% over 10 years (just to be a bit aggressive).
This is by design the revenue of the management company with just 1 fund. between 2,5m€ in the early years down to 500K€ in the last year.
2nd revenue stream: ManCo dividends and/or capital gains
The idea here is that after the investment period is done, the ManCo is allowed to raise a new fund (it is a common clause in LP investment contracts to not allow fund managers to lose their focus between investing and fundraising). The fees from several funds can start adding up as shown below. I’m assuming funds doubling in size at each vintage, with the same distribution of capital calls, and management fees. I’m however adding 2 partners at each fund, and increasing GP commitment by 1% at each vintage (as LPs like to see “successful” GPs commit more of their own money = ‘skin in the game’).
As you see here, a successful emerging manager can start making significantly more money once he manages several funds. Of course, I’m not taking into account increased costs here (better offices, larger staff, more travel, more software tools), but you start expecting some profit to be left at the end of the year in the ManCo.
That profit can then be shared (or reinvested) by all shareholders of the ManCo, the GPs. As stated, some partners are not ManCo partners, and only by title, or at fund level. Those owner partners then can get dividends paid out yearly, as part of their package.
Of course, if the ManCo gets acquired (and we every now and then see VCs being acquired), then the capital gains will ONLY go to those who had a stake in the ManCo.
Keep in mind that to keep operating, GPs still need to be committed and pay every year their GP commitment. It is very rare to have the fund return money to the original investors before a few years, and only then do the GPs get a share of it. Let’s discuss this below, but let’s assume they won’t get any carried interest (% of the gains) before year 8. In my model below, they will have had to pay up to 993K€ EACH before getting any money back. Not exactly a get-rich-fast model.
There is also an incentive to bring in new partners with each new fund, as the burden of the GP commitment is shared among more partners. I’ve increased the % of the GP commitment in the model, as LPs appreciate successful GPs’ alignment with their own interests.
3rd revenue stream: carried interest
This is supposedly the bulk of how VCs make money. The idea is that GPs have to first return the committed capital to their investors (the LPs), with a certain priority interest representing the opportunity cost of not putting this money to work elsewhere (like almost risk-free government bonds). We call this the hurdle rate and often when compounded over 10 years, it gets pretty significant. Whatever comes on top (committed capital and proceeds calculated with the hurdle rate) is shared 80% for the LPs and 20% for the GPs. Some funds structure it even more aggressively with for example 25% for the GPs above a 3x fund return, and 30% above a 5x return. Some very famous funds ask directly for 30% for GPs, and 70% only for LPs.
But how does it work?
Some early-stage funds get away with 0% hurdle rate, meaning they get carried interest immediately after returning the committed capital.
Other funds negotiate a flat percentage over the length of the fund. We can assume 20% for example (this was true when 10-year T-bonds were at 2%, over 10 years, it was almost 20% (21,90% to be exact)). Nowadays with interest rates back up at 5,33%, that is not acceptable for LPs. It is more common to see hurdle rates between 7–9%.
Einstein famously said “Compound interest is the eighth wonder of the world. He who understands it, earns it … he who doesn’t … pays it.”
If a ManCo negotiates an 8% minimum hurdle rate (a very common request from a very famous fund of funds…), this means that a 100m fund must first return 100m of committed capital, then 115,9m of additional priority return to the LPs before the GPs and the LPs share anything! That’s 215,9m of proceeds going first to the LPs!
To be fair, if you remember the GP commitment between 1 and 5% above, then the GPs actually get between 1–5% (prorata of the A shares they hold) of the 215,9m above as well.
To be precise, the hurdle rate is calculated on each capital call, so the aggregate compound number for the hurdle rate should be lower than the 1.16x above in the end.
It can get a bit more technical. GPs hold shares of Class C (= Carried interest), which gives them the economic right of 20 to 30% described above. All other LPs hold shares of Class A. There can be more classes depending on information rights, number of capital calls, etc. Actually, it is common for tax reasons historically for GPs to have a bit of their commitment in C shares and the rest in A shares. Let’s assume here that they own 100% in C shares for simplicity.
Now, if the fund makes a 3x return on the 100m committed capital, that’s 300m of proceeds, of which
- 100m are returned to the LPs first
- then 115,9m to them as a preferred return (should be less depending on capital calls spread out over time)
- then there is a catch-up for GPs as they are now entitled to the 20% of carried interest corresponding to the 115,9m; that 115,9/0,8–115,9 = 28,975m
- the rest of the money is split 80/20. So GPs get an extra [300–28,975–115,9–100 = 55,125 ]*20% = 11,025m
- Therefore in total: 28,975+11,025 = 40m, which is 20% of the (100m fund x 3 return — 100m reimbursed).
We could into more detail with Carta Insights data or Pitchbook data, per stage, per size, per geography, but as a rule of thumb, I seem to remember that only 4% of funds make more than 3x the committed capital. This means that most GPs never see any carried interest…
And yes, they paid almost 1m to play (and should get back) and potentially get only their salaries back without capital gain: hence the crucial role of negotiating the hurdle rate effectively when setting up the fund.
And food for thought:
If a 100m fund has an 18,5m expense associated with management fees, and probably up to 1% = 1m of setup fees, then only 100–18,5–1 = 80,5m can be put to work to make that 3x return on 100m, which means that the real return of a fund has to be 100 x 3 / 80,5 = 3,73x gross return to achieve this required 3x net performance, and those preferred proceeds of 215,9m before seeing any carried interest. Something all VCs think about when they invest.
We can address this issue to some extent with recycling, which will be maybe covered in a future note.
In summary, VCs shouldn’t be making much money with their salaries, only make some money with dividends if they are invited to be shareholders in the ManCo, have to chip in their own money as part of the GP commitment, wait at least 8 years for carried interest to be paid back to them, if ever…
You should now understand why it’s hard to fundraise at the moment, why mega-rounds have disappeared (although they were led in 2020–22 by non-VC investors such as hedge funds that have left the market since), why valuations are coming down: the numbers just don’t work for VCs.
—
I consult with corporates and startups and help them address issues like this and many more. Don’t hesitate to reach out at www.rodrigosepulveda.com or book a video call with me on intro.co.
