Profit and the theory of the firm

Sep 27, 2008 12:41

I [originally called] this LJ Thinking Out Aloud because it is a way of exploring and grappling with issues. So, continuing on from this post, profit and loss can be looked at usefully from the basis of the theory of the firm. The Coasian analysis of the boundary of the firm being set by the relative transaction costs of doing things within the ( Read more... )

value, economics, firm, transactions, property

Leave a comment

Comments 4

taavi September 28 2008, 23:46:04 UTC
I'm glad you have clarified this, but I don't think the theory you are advancing matches the facts particularly well. Saying that decision making can consist of hiring a manager is a bit of a dodge - particularly since, in the large joint-stock corporations which make up most of the economy, many shareholders won't hold shares long enough to vote on elections to boards, voting is optional if they do hold them at the time, and on most corporate board elections I've seen the ratio of candidates to positions isn't much above 1. Yet these shareholders, who don't contribute to decision making in any real sense, still receive the residuum. What they have contributed is capital. (Post Gordon Gecko era attempts to align the motives of the actual decision makers (managers) to those of the shareholders through options, etc, have usually resulted in the managers gaming the share price ( ... )

Reply

erudito September 29 2008, 10:19:09 UTC
That there are problems of corporate governance surely emphasises the point about appointing managers.

And the shareholders are still the ultimate guarantors of variability: they essentially "vote" on coordination by buying and selling shares ("exit" rather than "voice"). The problem with the poor taxpayer as owner is that they cannot bail out.

As I pointed out earlier, it is certainly not the return to risk (since all returns to input incorporate some risk element). While it cannot be simply the return to capital, since bondholders are not receivers of residual income.

And no, my view does not say labour is robotic, it says labour does not guarantee variability. That coordination is done at all levels is precisely something that governance structure has to deal with.

The Econ 101 economy has no information problems, no transactions costs and no discovery. (It is so not how business works.) In the Econ 101 economy there is no coordination problem so of course there is no return to it.

Reply

jordan179 October 1 2008, 06:53:02 UTC
Share-based corporations are not perfect decision-making entities. They screw up. OTOH, they screw up less often, and with less serious consequences, than do nationalized corporations -- because they have to worry about pleasing their shareholders, who have a personal financial interest in their performance. The chain of responsibility is much more dilute in nationalized corporations.

Reply

Return to ownership erudito September 29 2008, 10:33:31 UTC
The residuum is the return to ownership, given not all capital is owner-capital. The interesting question is: why is ownership purchased? ("Sweat equity" being a different sort of purchase.)

As the recipient of the residuum, the owner is the guarantor of income variability. Such a guarantee requires capital. The capital put up is the size of the guarantee. So the share of the ownership is the share of the guarantee purchased is the share of the residuum risked/gained.

The owner is also the final authority in the company. Purchasing ownership is purchasing that authority (whether a tiny bit, a more substantial bit or all of it). And matching authority with being guarantor means aligning final say with final return. Hence that authority being the final responsibility for coordination.

Reply


Leave a comment

Up