The Cooperative — The Firm, Plate xi

The workshop needs machines and the people who know how to use them. Its investors can own the enterprise, appoint its directors, and hire the workers; its workers can own the enterprise, appoint its directors, and borrow the money for the machines. Both arrangements join labor to capital. They place the authority to decide, and the claim to what remains, in different hands. The second possibility makes the first newly visible. A payment for the use of capital need not carry the right to govern everyone whose work makes it productive. The cooperative asks why those rights travel together in one firm and separate in another, and what the people who take them acquire besides the earnings.

Henry Hansmann's account begins with all the parties who transact with the firm: investors, workers, suppliers, customers. He calls them its patrons. Any one of these groups may face a costly bargain when another group owns the enterprise. A supplier can be stranded beside a specialized buyer; a customer can depend on a monopoly; a worker can acquire skills whose value disappears on departure. Giving the exposed group ownership can remove a conflict across the contractual line. It also gives that group the work of governing. Hansmann's comparison adds the costs of ownership for the owners to the costs of contracting for everyone else. The relevant economy belongs to the whole arrangement; a cheap relationship with investors can coexist with an expensive relationship with labor.¹

1. Henry Hansmann, “Ownership of the Firm”, Journal of Law, Economics, & Organization 4 (1988), pp. 269–281, §§2.1–2.5. His comparison includes market power, lock-in, asymmetric information, monitoring, collective decisions, and risk-bearing; it is a framework for comparing arrangements, not a measured ranking of all firms.

For workers, ownership can make the value of a long commitment partly theirs to defend. They may know which machines fail, which instructions waste time, and which promised improvements never arrive. Their proximity can help them monitor the enterprise, while their dependence on its earnings gives them reason to do so. A cooperative can employ a manager and delegate daily decisions to that manager, just as investors delegate to executives. Its distinguishing question is who appoints, disciplines, and ultimately replaces those executives. Ownership by workers does not require a vote on every delivery. Nor does a share of earnings alone settle the problem of effort: a member who receives one small fraction of a collective return still bears the full personal cost of working harder. Supervision, reciprocal expectations, and a workable division of responsibility remain part of the institution.²

2. Hansmann, §§2.4 and 5, pp. 275–280 and 291–296, especially note 11 on delegated management and note 48 on individual effort within shared ownership. His worker-ownership discussion distinguishes the gains from avoiding labor-contracting problems from the costs of governing the enterprise.

⁂

Put the comparison into a small account. Suppose a workshop creates 150 units of surplus before its organizational costs, under either form. Investor ownership uses 6 units to govern the firm and loses another 16 through difficulties in contracting with workers, leaving 128. Worker ownership uses 12 units in governance and risk-bearing, plus 6 in arranging outside finance, leaving 132. On those assumptions the cooperative preserves four more units. Now suppose disagreements among members add 8 units of real decision costs. Its remainder falls to 124, and investor ownership preserves four more. These figures describe an example, not an estimate. All other costs are held equal, and the entries count resources or surplus lost rather than payments merely passing between parties. Changing the owner has no productive magic in the calculation: its advantage comes from the particular frictions it avoids, weighed against those it introduces.

Agreement is easiest when members can recognize a common interest or an accepted rule for dividing different contributions. Partners who provide comparable services may share costs and settle accounts by a familiar formula. A workforce containing different occupations, ages, and expected lengths of service may disagree about pay, expansion, or an investment whose benefits arrive after some members leave. Even a perfectly conducted vote can reflect that divergence. Suppose a proposed change gives two members net gains of 8 each and a third a loss of 20. A majority supports it, although their combined gain is minus 4. With gains of 4 for each member, the same voting rule supports a combined gain of 12. The problem lies in the pattern of interests and the available terms of agreement. Rules for membership, compensation, and redeeming an ownership share can change that pattern; the cooperative's constitution is part of its productive equipment.³

3. Hansmann, pp. 277–280 and 294–296, including note 50 on tradable or redeemable member claims. The voting and cost-account examples are illustrative. The majority comparison assumes equal votes, no compensating transfer, and net gains measured on the stated common scale; it does not attribute one voting rule to every cooperative.

Finance imposes another discipline. Workers who put their savings into the enterprise may expose both their employment income and their accumulated wealth to the same failure. Outside investors can sometimes spread that exposure more readily. Borrowing moves part of the financing outside, but lenders still need protection against losses and against decisions taken after their money has been committed. These costs can favor investor ownership, particularly where much specialized capital is required. Other transactions produce other candidates for ownership. A professional partnership can pool a name and common services; a mutual insurer gives policyholders an ownership interest in the organization promising to pay their claims. Hansmann's historical account of life-insurance mutuals shows how customer ownership can weaken incentives to exploit a long relationship even when dispersed members supervise managers imperfectly. Every form has a constituency to protect and a cost of protecting it.⁴

4. Hansmann, §§3 and 5, pp. 281–284 and 291–296, on financing, diversification, and partnerships; §7, pp. 298–300, including notes 59–62, on life-insurance mutuals. The insurance discussion is his historical explanation, not a claim about the current relative performance or market share of mutual insurers.

The question of why capital hires labor therefore has a comparative answer. Investors may combine access to finance, diversified exposure, and a relatively common financial objective at lower cost than another owning group. Workers or customers may bring stronger information, a relationship worth protecting, or a less costly way to share decisions. Which arrangement preserves more value depends on those circumstances. Who receives that value is a further question: a partnership can own itself while employing people who never become partners, and an efficient ownership arrangement can leave the division of its gains contested. Ownership makes a group responsible for a set of decisions; it supplies neither a complete account of justice nor an exemption from organizational costs. Whichever group governs the workshop must still decide what to do within it and what to purchase outside. The next plate follows that boundary as it moves.⁵

5. Hansmann, §§2.5, 3.2, and 5.2, pp. 281, 283–284, and 293–296. His ownership-cost framework concerns the relative efficiency of arrangements. The distinction between their aggregate gains, membership, and distribution is the implication developed here.