4 ms·
This blog post, and the app it's meant to promote, can be boiled down to one proposition: the marginal product of an employee can be known with precision, and t
by bwestergard 4y ago
This blog post, and the app it's meant to promote, can be boiled down to one proposition: the marginal product of an employee can be known with precision, and the employee should get no more and no less than their marginal product.
I think it is patently absurd to believe that this marginal product can be determined in a manner that all affected parties will agree on, at a cost all parties are willing to agree on and bear (e.g. some A/B testing scheme).
But for the sake of argument, let's say it is possible. As Coase pointed out, in that situation there would be no need for a firm at all: an arbitrarily large group of individuals could arrange an elaborate division of labor, and agree on payment conditions between them, through a series of exhaustive contracts.
Why don't we ever see this in the real world? Why are there any firms at all? Because the premise is flawed, and the transaction costs always end up being prohibitive.
https://en.wikipedia.org/wiki/The_Nature_of_the_Firm https://en.wikipedia.org/wiki/The_Nature_of_the_Firm
- timdel 4y agoInteresting, thanks for the link. In practice, we do see this in certain industries though. For example, sales people earn on a base+commission model, portfolio managers are paid a percentage of their returns etc... what do you think of those cases?
- bwestergard 4y agoWe see piece work or commissions, sure. But who claims that these allocate to the (e.g.) sales person their marginal product as deduced by rigorous exploration of the whole space of possible techniques?