(When) do co-ops work?

Paulie asks a good question: do co-ops work?
The answer's yes - in some cases.
The great merit of worker co-ops is that they can solve the problem of how to motivate workers. In businesses where workers can't be monitored closely, giving them a share of profits can inspire them to work hard, and to encourage their colleagues to do so.
This is why law and accountancy firms and vets and medical practices are typically partnerships. The success of these ventures depends upon employees working well. That effort is best inspired by giving them control of the firm.
Another benefit of co-ops - or at least of profit-sharing - is that they can promote economic growth without inflation. If a conventional firm is to hire another worker, it sometimes has to raises wages for all workers, to attract the additional employee. That naturally causes a trade-off between growth and inflation. However, if the additional worker is paid a share of the profits, this trade-off is softened. This was the theme of Martin Weitzman's idea of the "share economy", which was briefly influential in the 1980s.
Maybe it's no accident that the rise of profit-sharing in the UK since then has coincided with a fall in the NAIRU.
There are (at least) three drawbacks to co-ops.
1. They don't work in physical capital-intensive firms where worker effort can be monitored. Here, the key to success is getting machinery to run well, rather than getting workers to do well. Car plants, for example, are better run by capitalists, not workers.
2. Co-ops have less incentive to expand, because the profits from expansion are spread more thinly. It's probably no coincidence that the Co-op (a consumer, not worker co-op) has lost market share.
In worker co-ops this problem can be mitigated by hiring workers on lower terms than full partnerships. But this moves them away from the co-operative ideal.
3. Co-ops' often lack access to capital with which to expand, even if they want to. This is because they work best in human capital intensive businesses. But such businesses find it hard to post collateral and so get bank loans.
For this reason, co-ops might have to sell equity to outside owners - thus diluting the co-operative principle - if they want to expand.
Goldman Sachs provides a nice example of this. When it was a human capital-intensive business selling advice it was wholly owned by its partners. But when it wanted to expand its proprietary trading activity - which required capital - it had to sell equity.
Mention of Goldmans' raises a paradox here. We think of co-ops as a left-wing ideal. But in fact they have for years been both employers and suppliers to the rich.
Many wealthy Londoners are partners in law firms, save with the Nationwide or through hedge funds, and shop at Waitrose and Peter Jones - and then tell us how wonderful capitalism is.
More: here are two rigorous papers on the economics of co-ops.

Comments

[dearieme]

You've often recommended worker co-ops, and I wonder why anyone need take any action, other than publicising their merits, if he wants to promote co-ops. If they're such a good idea, shouldn't they be set up spontaneously? Are there any obstacles - e.g. in law, in the tax system, in the pension system - that impede the would-be co-op?

[chris]

I know squit about the law and taxes (rational ignorance), but judging from the ease with which hedge funds have been set up, I doubt it.
The traditional obstacle to the emergence of co-ops is access to capital.
However, if you believe that the "new economy" is increasing the importance of human capital relative to physical, we should expect to see even more co-ops emerge over time.
There are already many more co-operative firms than there are publicly quoted ones on the stock exchange. It's just that worker co-ops tend to be small firms.

[james higham]

...the codetermination and industrial democracy movements (in which management participation by labor is required by law)...

Can you give me an example of this working in Britain? How do you feel about this model?

[Alex]

Hmmm, autogestion rides again..

In a sense, all open-source software projects are worker co-ops - which rather bears out Chris's point.

[Laurent GUERBY]

My quest for econometrics is still unsuccesful.

How many co-op worker vs other? Size, wages, productivity, sales, growth vs equivalent non co-op companies in the same sector?

Some co-op are bigs, eg Mondragon in spain (industry and banking), Desjardins in canada (banking, 103 billions CAD in assets), or Credit Agricole in France (banking, went partly to market too).

http://en.wikipedia.org/wiki/Mondrag%C3%B3n_Cooperative_Corporation
http://en.wikipedia.org/wiki/Mouvement_Desjardins

This might be the exceptions, but some do get big access to capital.

[Kevin Carson]

Chris,

Much of the expansion that currently takes place among conventional enterprises involves growing beyond optimal economy of scale. If it weren't for the possibility of externalizing the diseconomies of large scale on the taxpayer, and the effect of regulatory cartelization and other forms of legal privilege in reducing the competitive disadvantage of inefficiency, there would probably be a lot more, and smaller, firms.