Every new generation of venture investors since time immemorial has asked themselves “why don’t we forgo the all or nothing model in favor of consistent returns?” /1
That is, why not just invest in companies that will return 3x and not shoot for a mix of 10x and zero? /2
But asking this question is the same as asking why you can’t have your cake and eat it too. “Why can’t I have my return and not take risk too?” It’s a great goal, almost never achieved. Why? /3
I’ve talked about this before in terms of power law outcomes, but let’s address it just using plain vanilla finance theory. This is what I think is going on /4
Getting a 3x over 7 years, say, is a 17% per year IRR. This is about twice the long-term return from the stock market, so should be twice as risky, according to finance theory. This doesn’t seem that bad, and is the realm PE funds play in. Why not venture? /5
The problem is that risk measures the degree of volatility in value. The riskier the company, the higher the chance that the value of the company rises or falls below some level /6
This is good on the upside, but on the downside there is a chance that the value of the company falls below some level that causes the owner-operators to give up /7
For instance, if their stake in the business becomes worth less than they could earn at some other opportunity it would be rational for them to walk away (a preference overhang exacerbates this) /8
In large companies, this threshold is a small percentage of the current value; in small companies, it is a relatively large percentage. Plus, in addition to the higher threshold, small companies have a higher risk/volatility, making it more likely they sink below any threshold /9
So small companies will fail far more frequently than large companies. If you want to have a similar failure rate at small companies as large, you would need to aim for a *lower* return than the stock market /10
This only makes sense if you are *not an outside investor*: if you are a lawyer starting a law firm, you are not seeking to increase the valuation of the firm...you probably don’t care because you never intend to sell it /11
This, of course, is all much worse with power law distributions because volatility rises much faster with risk. And there’s good evidence that PLDs are not just a feature of VC investments, but of the set of all companies /12
Consistent returns are obviously a goal. That it is so rarely achieved in venture is not because no one else has tried, it’s because odds-on they failed /13
