Alternate business structures

In Paul Graham's latest essay, You Weren't Meant To Have A Boss, he applies evolutionary psychology to organizations and discusses how people are naturally ill-suited for working in large corporations. There is a similar libertarian argument about government, which views The State as The Tribe writ large, with people's intuitive belief that the state can work well deriving from their hard-wired expectation of a tribal-sized group. Anti-market biases such as those Bryan Caplan has documented, or Arnold Kling's "Folk Economics" have a similar origin.

As an employee, i've spent the past few years working at, and thus observing, a large corporation. And as an anarchist, a mechanism designer, and a dreamer, I can't encounter a large heirarchical system without wondering whether it could be done better. So, like Paul, I've pondered alternate organization schemes:

A large organization could only avoid slowing down if they avoided tree structure. And since human nature limits the size of group that can work together, the only way I can imagine for larger groups to avoid tree structure would be to have no structure: to have each group actually be independent, and to work together the way components of a market economy do.

That might be worth exploring. I suspect there are already some highly partitionable businesses that lean this way. But I don't know any technology companies that have done it.

It seems to me that there are some clear advantages, if the business is seperable, in seperating it, in resisting the empire-building temptation. Especially if there are alternate suppliers for what some of the units create - that way you stay open to using a different supplier if they are better, and keep competitive pressure on the unit. In a lot of cases, this is just not possible - businesses can be highly interdependent, with some functions that are very speculative or don't have clear bottom-line contributions. But I think this could be done a lot more than it is.

Let's take, for example, a hypothetical technology company. This company has a core business which generates a lot of money, and results in it creating a lot of generally useful infrastructure. As a result of this, it has a variety of small, speculative projects which make use of this general infrastructure. These projects are organized in the standard ways: employees are employees of the main company. Their career path and rewards depend somewhat on the success of the project, but not like a startup.

This seems, to me, like a pretty bad setup. The project's employees are far less invested than if they were at a startup, yet they are basically doing a startup-like job. Why not structure it that way? The big company could invest and provide the general infrastructure along with some cash, the employees could get equity and control, and you get a more "natural" structure.

To be fair, problems of scale are good problems to have - you only achieve them by succeeding wildly. But just as it's worth thinking about whether there are radical alternatives to democracy which might better meet our needs as citizens, it's also worth thinking about whether there are radical alternatives to large corporations which might better meet our needs as employees, consumers, and investors.

Share this

Hierarchy has some benefits too

There have been and continue to be attempts to create "internal markets" for goods within a corporation. They have varying degrees of success. Whenever a spinoff occurs, like GM's spinoff of Delphi or Ford's of Visteon, it means that to some extent hierarchy has failed to justify itself.

That said, there are benefits to hierarchy. If you and I are to work on a book, we could negotiate how much we should each get paid for each sentence, weighted by quality which we would somehow have to measure. Or, we could just trust each other to each do a good job and split the profits at the end. Let's say we work in a bakery: you make the dough that I bake. We would similarly waste a lot of time trying to negotiate what price I should pay for your dough. We're better off working on trust that we will put in the our best efforts and splitting the profits.

Thus, there is a trade-off between the evils of hierarchy and the evils of "transaction costs". Transaction costs occur because negotiating prices, verifying quality, etc. take time.

(All of this is me channeling economists. See Oliver Williamson and his predecessor Ronald Coase for more.)


Y'all are talkin' sideways about agorism. Fun idea, but it does bring up a comparison with structures in which scale is exploited, frequently to the benefit of ... well, lots of us. If you are willing to limit your scope of regard, that is. Ultimately, I think we'd be better off with more "organic" scale structures, such as partnerships, but we are currently living with massively scaled structures in which the corporate fiction allows for both good (widely accepted standards, lower transaction costs, and the like) and bad (everything else, including problems of agency, hierarchy, hive mentality, and the like). That bad is not only bad in and of itself, but it serves as pretext for more state-sponsored shenanigans while tricking vulgar libertarians (Kevin Caron's meaning) into defending the indefensible.

The Koch "market based management" seems to be a common example of a successful hybrid. I haven't read Koch's book (it seems unlikely, p<=0.05, to be unbiased), but I have read Cowen & Parker's Markets in the Firm, and I have yet to read anything that convinces me that you can achieve an agoric success in the context of state capitalism (i.e., modern US or Europe). Best thing I've read so far on the subject is Kirkpatrick Sale's Human Scale.

Oh, and see also this, this, this, this, this, and this.


I guess I could have linked this and saved myself some typing:

Commons-based Peer Production

This may be an argument for moving (some) production out of businesses altogether.