The foregoing emphasizes the beneftts of eollective organization. They are especially great in circumstances where the labor force acquires (or the man- agement wishes to induce the workers to acquire) firm-specific human capital. But there are additional efficiency benefits, of a simple agency kind, that collective organization can provide in virtually all enterprises.
Unions are the prevailing form of collective organization in capitalist economies. That the union was not introduced carlier or was not received without resistance is presumably because, in addition to the aforementioned benefits, there are also prospective costs. Some of the more obvious are noted here.
1. Monopoly Power
Collective organization can permit workers to improve the bargains they strike with respect to the disposition of the quasi-rent attributable to firm- specific human capital. If, contrary to the argument in subsection 2.2, workers acquire secure property rights over jobs, further improvement in those bargains—at least in the short run—can be realized by expropriating sunk costs in physical plant and organizational infrastructure. Out of recognition of such an expropriation potential, firms and industries in which investments in durable nonhuman capital are greater will be more resistant to union organization, ceteris paribus.
2. Oligarchy
The Iron Law of Oligafchy holds: “It is organization which gives birth to the dominion of the elected over the electors, of the mandatories over the man- ators of the delegates over the delegators. Who says organization says oligarchy” (Michels, 1962, p.365). Efforts to weaken this Law notwith-standing, it has so far resisted repeal. As Seymour Lipset puts it. modem man faced by an unresolvable dilemma: He “cannot have large institutions such as nation state, trade unions, political parties, or churches, without turning over effectwe power to the few, who are at the summit of these institutions” (1962, p.15). The leadership of unions, like the leadership of other large organizations, is thus often in a position to entrench itself and/or pursue its interests.
That is sometimes obvious, as when the union leadership squanders the retirement and hospitalization payments of the membership. It can also be subtle, however, and may be influenced by the institutional rules of the game. Aoki’s recent discussion of the choice of contract renewal intervals is a possible example of the latter.
The issue is, What is the appropriate interval at which to renegotiate contracts? There are great advantages, from an efficiency point of view in concentrating the hard bargaining aspects of the relation at the contract renewal periods and using analytic processes to effect adaptations during contract execution. Long-term contracts offer the apparent advantage of reducing argaining and increasing analytic processes, but that can be misleading. If the basic contract gets out of alignment with the economic realities, one party or the other is apt to press for relief during contract execution. As a consequence the presumption that the parties will cooperate during the period between contract renewals is placed under strain.
Aoki nevertheless notes an interesting empirical regularity that dis- tinguishes contracting practices in Germany and Japan from those in the mted States: Contracts in the first two countries are typically renegotiated at one-year intervals, while in the United States the usual interval is three years (Aoki, 1984, p.148). What explains the difference?
One possibility is that economic disturbances in Germany and Japan are arger and more frequent. That is the uncertainty explanation: Contracts that experience greater uncertainty should presumably come up for renegotiation more frequently than those that experience less. The explanation that Aoki favors, however, traces the difference to a National Labor Relations Board ru ing The contract bar” doctrine prevents a challenge to the incumbent union rom being filed “during the term of an existing contract, with a three year maxunum in the case of contracts that run for more than three years” (1984, p. 148). Aoki contends that the resulting three-year contracts protect incumbents at the expense of efficiency (pp. 148-50). Whether that is correct or not, the possibility that oligarchical outcomes will influence the rules of the game warrants attention.94
3. Heterogeneity
More complex labor governance structures are needed as investments in firm- specific human capital deepen, ceteris paribus. In consideration of that, and given heterogeneity in the typical work force, it may be that a series of bargains rather than a single bargain applicable to the entire labor force should be struck. Among other things a single union operating under a uniform agreement will have difficulty aggregating the preferences of a disparate membership. To negotiate discriminating terms is at variance, however, with the egalitarian purposes of unions. Japanese firms and labor unions have mitigated the problem by spinning off divergent activities through extensive subcontracting and by the creation of subsidiaries (Aoki, 1984, p. 142). Those who would have United States firms and unions imitate their Japanese counterparts more closely but are not prepared simultaneously to accept Japanese subcontracting (and related practices that facilitate discrimination) should acknowledge the strains.
Source: Williamson Oliver E. (1998), The Economic Institutions of Capitalism, Free Press; Illustrated edition.