Live data from Hacker News

Counterfactual Theory of Value

perell.com

21–30 of 59 posts

Re: Counterfactual Theory of Value

#21

> In startups, why do the founders receive such a disproportionate percentage of equity? Because on day zero, 100% of the company must be owned, and the founders are the only ones there. Equity to founders isn't handed out based on an arm's-length negotiation, or on the basis of work done.

But all other stock allocations are an arms length transaction; so all you've done is reframe the question to "why do the founders retain so much equity?" Why dont seed fund demand 16% instead of 8%? Why doesn't the 6th employee demand 5% instead of 3%?

>reframe the question to "why do the founders retain so much equity?"

The founders can only retain as much equity as the marketplace allows. Each side has a threshold to do a business transaction.

>Why dont seed fund demand 16% instead of 8%? Why doesn't the 6th employee demand 5% instead of 3%?

The parties can demand any percentage they want but the ultimate resolution is will the other side agree to it? In other words, the question is, "Why do angels _agree_ to 8% instead of 16%"? Because an offer requiring 16% would be rejected by the founder and lose to other angels only requesting 8%.

E.g., back in 1999 during the dot-com craze, VC Sequoia Capital offered MP3.com (founder Michael Robertson) $10 million for 45% of the company. He said no deal. They later negotiated it down to 20% ... which is in the more reasonable ~15% to ~20% range of other VC deals.

Lesson: Demanding a high 45% so that the founder only retains 55% instead of 80% doesn't automatically mean the founder will say "yes" to the reduced equity. People can still voluntarily choose not to do business with you at all. All "demands" are competing in the marketplace. Can an employee demand a higher 10% equity? Sure. Whether the founders _agree_ to it depends on the marketplace and the employee's particular leverage (e.g. a very rare skill).

Re: Counterfactual Theory of Value

#22

Earlier quoted context omitted.

You can't prove it, but Musk is not a one-hit wonder. He's transformed the space industry by sheer force of will (as a part time founder & CEO), as Tesla has done for EVs, so it's quite reasonable to assign the success of Tesla to him.

In this case it makes sense but in general you have to also ask about the environment. Was money cheap in that era? Was there a favourable media environment for that type of business? Was politics favourable? Etc.

To be fair, Elon took over as CEO in 2008. Hardly a favourable environment for risky endeavours or capitalists in general.

Re: Counterfactual Theory of Value

#23
The problem I have with this is that it presupposes that you can change 1 thing and view the difference. What if the only reason a company is successful is because of a unique idea that 3 people came up with together? Are each of them worth 100% of the company- take any of them away and it’d be worth nothing. Any company or endeavour is made up of complex interactions of many different people it’s impossible to differentiate with respect to just one variable.

Re: Counterfactual Theory of Value

#24
Labour theory isn't wrong for the reason he states. Causing people to work harder isn't actually a problem with the theory. It's wrong for other reasons.

The moneyball theory of value is not terrible, is actually pretty sound. But this kind of analysis works well in a context like sports because there's only so many players in a team and the team is basically just struggling in a win or lose world against other similar teams. Your wins come from someone else's losses.

The sports metaphor is also applicable because sports are reproducible. It's the same little world over and over again, with the same people. And there's just an enormous amount of stats to back up or reject some hypothesis. Goalkeeper is good or bad? Let's check how many shots he faced and what quality they were.

In the general market this is not the case. The firm only launches its first product once, in a world where you don't get to repeat the conditions. We also do have a history of each contributor's actions. Basically it's very hard to make comparisons.

Re: Counterfactual Theory of Value

#25
post #23

The problem I have with this is that it presupposes that you can change 1 thing and view the difference. What if the only reason a company is successful is because of a unique idea that 3 people came up with together? Are each of them worth 100% of the company- take any of them away and it’d be worth nothing. Any company or endeavour is made up of complex interactions of many different people it’s impossible to diffe…

> The problem I have with this is that it presupposes that you can change 1 thing and view the difference.

You should have other problems with it. Value that would be lost in the absence of a thing is unrelated to the value gained by its presence, because you have more choices than "pay for the thing" or "do without".

The theory of value espoused in this post would tell you that you should pay more for the oxygen you consume than you do for your computer.

(There is some self-contradiction in the post - it defines counterfactual value thus:

> what if pay is determined by asking: “How much would this company be worth without this individual?”

But then it provides an example of something very different:

> Baseball managers use the Counterfactual Theory of Value all the time, using a statistic called “Wins Above Replacement.” It predicts how many more wins a player gives their team, compared to whoever would replace them.)

Re: Counterfactual Theory of Value

#26
post #15

This thinking is almost exactly how most pricing works in derivative markets and is the real breakthrough in the original Black/Scholes/Merton model[1] that has been reproduced many times since. They showed how you could replicate the payoff from a European-style option by hedging using the underlying and a risk-free instrument (under certain unrealistic assumptions, notably constant vol and the ability to trade in c…

> Since then, pricing using a replication portfolio in this way has been a cornerstone of financial maths - the price of a thing and a perfect hedge/replacement for the thing must be the same. And the reason for this is arbitrage. If you can perfectly hedge your position in Asset A using Asset B (and vice versa), then you can make a profit by simultaneously buying the lower-priced asset while selling the higher-price…

Is the implication here that if we had a Black/Scholes/Merton method for salaries we would see a readjustment of salaries across the board as people could identify their value as wins above replacement - probably a huge shock to CEOs.

Is it possible that automation provides that arbitrage opportunity?

Re: Counterfactual Theory of Value

#27
One thing that is important here is whether the counterfactual is the marginal employee, or nobody.

In the sports roster case, a soccer/baseball team would be in trouble if they lost one of their average-talent players if they didn't have anyone on the pitch but there might be a plentiful supply of low-cost players willing to work for a similar amount.

This means that the counterfactual of 'lose average player and replace with noone' suggests that the player has a very high counterfactual theory of value, but 'lose average player and replace with similar average player' suggests that the counterfactual value is not high.

There might be a star player such that they are not replacable by anyone, or by anyone also able to demand a high salary from an alternative team, and they would have a very high value according to both counterfactuals.

It seems to me that many sports teams operate with the 'similar average player' counterfactual.

I think you could draw analogies to businesses and employee pay.

Re: Counterfactual Theory of Value

#28
post #23

The problem I have with this is that it presupposes that you can change 1 thing and view the difference. What if the only reason a company is successful is because of a unique idea that 3 people came up with together? Are each of them worth 100% of the company- take any of them away and it’d be worth nothing. Any company or endeavour is made up of complex interactions of many different people it’s impossible to diffe…

> The problem I have with this is that it presupposes that you can change 1 thing and view the difference. You should have other problems with it. Value that would be lost in the absence of a thing is unrelated to the value gained by its presence, because you have more choices than "pay for the thing" or "do without". The theory of value espoused in this post would tell you that you should pay more for the oxygen you…

You'd be willing to pay more for the oxygen than the computer (its value is very high) but the price is low (it's free).

Usually when you buy something, the value (to you) is greater than the price, which is greater than the cost (to the seller).

Re: Counterfactual Theory of Value

#29
post #5

> In startups, why do the founders receive such a disproportionate percentage of equity? There is a really straightforward answer to this - equity isn't distributed according to a theory of value and nobody ever claimed it was. I own some equity in the Australian company CSL. It is unfair to claim I've ever added any value whatsoever to the company at any point, so obviously I don't get that equity because of a value…

It's based also on risk. Like, primarily. Thus the whole of CAPM finance underlying trillions of dollars in wealth worldwide.

Re: Counterfactual Theory of Value

#30
This doesn’t account for frequentist past looking vs Bayesian future looking value. At some point, the value created by an individual is no longer what it used to be, but ownership compensation persists, while cash compensation tapers off.
Post reply on HN