November 1, 2011
A recent article by Michael Greeley shows that US venture capital (VC) funds are focusing more on short term rather than long term investments. Intuitively, that doesn’t feel right, and it got me thinking about a related phenomenon I’ve been seeing in the VC world recently: secondary sales.
Secondary sales transfer equity from current to new shareholders for (usually) a cash payment. While such transactions are common in the public market, in the past they didn’t really happen in private companies, and practically never happened in pre-revenue startups. And for good reason! Secondary sales are private transactions, unprotected by regulation, and thus open to significant abuse. And yet, recent years have seen an upsurge of secondary sales in the startup world to the point where formal marketplaces now exist to facilitate these sales.
As usual, the introduction of secondary sales was well intentioned. Venture capital requires extremely high return exits (at high risk) to make the math work out for investors. That puts many entrepreneurs in the odd position of having to pass on solid acquisition deals in order to pursue moonshots. Secondary sales during investment rounds has become a somewhat popular way to fix the immediate cash worries of the entrepreneur and keep them motivated for that high risk, high reward, play that VC partners dream about. Depending on the magnitude of the early payout, this doesn’t seem unreasonable. While I didn’t take any money off the table in my ventures, I can certainly attest to the fact that having to live with 6 people for lack of money doesn’t make entrepreneurship any easier.
More recently, we have seen VCs and early investors becoming the dominant benefactors of secondary sales (like in the last Groupon funding round). That’s where the logical argument above falls apart. Frankly, the reasons for taking money off the table just do not apply to initial investors or VCs. Founders might take some cash so that they don’t have to worry about losing their home while working for a tiny salary – It probably makes sense to take that risk off their mind so that they can focus on the business – but VCs don’t have that risk. No VC partner is poor. Nor do they have the same degree of personal risk. Their funds normally have decade-long life spans and they have fairly high confidence that they will receive their ongoing management fee during that period. When founders are encouraged to take money off the table, the point is to provide founders with similar risk profiles to VCs. A VC already has the risk profile of a VC, so why should VCs be able to take money off the table these days?
Let’s take a few steps back to how the VC model works. A venture capital firm is a partnership between limited and managing partners. Limited partners (LP) put the money in, usually from pension funds and other big capital pools. The managing partners are the actual venture capitalists. VCs don’t own the money they invest and thus carry no personal risk (they might make less upside if they perform badly, but nobody takes anything away). They make money in two ways.
- They charge a management fee. Usually for a decent sized fund it’s about a 2% fee, so if you have a $100M fund, you get $2M each year. For a 10 year fund, you clear $20M – just to manage the money.
- They participate in the upside or ‘carried interest.’ The way it works is that if they invest $100M dollars, and get 200M dollars back in exit return, they pay the 100M dollar principle back to their limited partners, and get 20% of the remaining 100M (a $20M upside).
This structure is effectively capped in terms of downside risk: unlike entrepreneurs (who make very little and gain solely based on the upside) and limited partners (who can lose all of their money), VCs have a potential share of the upside, but really no way to make less than their relatively high base income (management fee). In economic game theory this creates what is called a “drunk on the sidewalk” scenario (A drunk randomly walking on a one-dimensional cross-section of a road will end up in the middle of the road, but if you put him on the sidewalk where one side is blocked by a wall, he will inevitably end up lying in the gutter at some point – in other words, capping one direction of risk leads to a statistically guaranteed outcome different than the true uncapped mid-point). Apart from guaranteeing a better payout for VCs than for the other two players in the venture game, the structure also encourages a pyramid scheme setup when VCs are routinely taking money off the table prior to a true exit.
Imagine a VC with both a seed fund and a later stage fund under management. Post-management fee they might have $10M and $100M to invest respectively (let’s call the 10 year management fee $2M and $20M respectively). They invest $10M from the seed fund into a venture and then do a secondary sale of the same shares to the late stage fund for $100M. For simplicity sake, let’s assume that the company dies right after that transaction so that there is no return at all for the second investment. The picture is prettier for the seed fund. After repaying the $12M principal there is $88M left which yields a carried interest of $17.6M for the VC. Net combined gain for the VC partners is $39.6M ($22M in fees, $17.6M in carried interest). The LP on the other hand collectively lost $49.6M ($132M investment vs. $82.4M). The remaining $10M is actual capital for the company so some of it will pay at least a salary for the entrepreneur and staff. The problem is that the VC not only profited from a failed enterprise, but also controlled both ends of the deal. In a pump-and-dump IPO they have to at least temporarily sell the proverbial bridge in Brooklyn to somebody else.
While a simplified example, there is nothing implausible about it once early VC secondary sales become commonplace. Facebook, Groupon, LivingSocial and many of the other “internet darlings of the day” have thriving secondary markets and the rate of VC participation is growing steeply on both ends (buyer and seller). The number of VC players in these deals is small enough that you can easily find circular arrangements (though none with the same VC using two managed funds yet…).
Apart from regulations, the only thing to keep this kind of circular pyramid scheme from happening would be VC expectations that the larger fund can be more profitably employed elsewhere (e.g. investing the $100M in the example into other ventures which would yield more than the ~$40M total VC payout available in the pyramid scheme). That’s of course how things are supposed to work, except that in most markets, including Canada and the US, the VC industry as a whole hasn’t made a dime of net gain in the last decade. In that environment, a guaranteed $40M might sound mighty attractive.
I understand that the example above is pretty simplistic, but I would love to know whether I am missing some mechanic that would prevent these kinds of circular secondary market deals from happening.