Co-ops may be initiated by the college, the employers, or by the student, with great variation in the wages paid and in the college credit given. A crucial aspect is the selection of students: in general, the colleges have screening criteria related to academic preparation, while employers screen students on the basis of criteria that they apply to their regular employees-sometimes relying heavily on the colleges to make recommendations. In general, employers are more interested in personal competencies-enthusiasm, the ability to work with others, dependability-than they are in either job-specific skills or in conventional academic criteria like grades. The result of such screening is a "high-quality equilibrium," in which colleges provide well-prepared students while employers provide high-quality placements-and if either side violates this informal but well-understood agreement, the quality of the co-op program is likely to fall.
For employers, one principal benefit is the ability to "grow their own" employees-to generate training programs that provide precisely the mix of cognitive, personal, and job-specific skills that they require. Others stressed that co-op placements make excellent screening mechanisms because they allow employers to see the range of capacities-including personal attributes like initiative, the ability to work in groups, and discipline-that are poorly measured except by direct observation on the job. Many employers also cited co-op students as a source of cost-effective labor, a perspective considerably different from the longer-run perspective of employers trying to "grow their own" employees.
For the colleges involved, the principal benefit of co-op education is that it strengthens their institutional links to employers. In Cincinnati, employers are quite familiar with the variety of education providers, and generally supportive of education-in contrast to other communities where their attitudes range from indifference to hostility. Colleges also benefit from having what they consider to be higher-quality education, guided by the participation of employers and complemented by work-based placements, and they also enjoy higher placement rates for their graduates than would be true in the absence of co-op education.
The benefits of co-op are especially powerful given some special characteristics of sub-baccalaureate labor markets in which community colleges operate. This segment of the labor market is quite local-so that local employment related to a student's field of study is crucial to realizing economic benefits. Co-op education strengthens the relationships between colleges and employers and leads directly to higher placement rates, partly addressing the problems that arise in other local labor markets where educational institutions and employers are distant from one another. In addition, hiring in the sub-baccalaureate labor market usually requires experience, which is a good indicator of job-specific skills and certain personal capacities; this makes it difficult for students without experience to break into employment. Since co-op programs provide experience and allow employers to judge the competence of their co-op students directly, they facilitate entry into sub-baccalaureate employment.
However, we stress that the lack of data about co-op programs limits our ability to draw conclusions about their effectiveness. All statements about effectiveness are statements of employers, educators, and students about their experiences, and-powerful though this kind of testimony can be-it is not a substitute for the kind of quantitative information that is lacking.
A second implication for school-to-work programs involves the importance of the "high-quality equilibrium" in Cincinnati, where colleges screen students for academic abilities and personal attributes while employers try to offer high-quality placements. This outcome is clearly important to the maintenance of co-op education over time because employers and students alike would quickly abandon low-quality programs. In Cincinnati, this equilibrium is achieved through clear expectations among employers and educators, established in face-to-face contact and constant discussion-not through skill standards, certificates of mastery, complex written agreements, or other formal regulatory mechanisms. This suggests that such devices, often thought crucial to school-to-work programs, may not be necessary or desirable.
Several factors are important to institutionalizing the work-based programs in Cincinnati, however. One is state financial support through the regular program of financing community colleges. The other is the support of co-op coordinators in community colleges, who are the linchpins of successful programs. The absence of formal regulatory mechanisms and bureaucratic authority in the Cincinnati co-op programs, while surprising, also suggests that a peculiarly American form of work-based learning has developed, embedded in voluntary relations and expectations rather than rules and regulations (as in the German apprenticeship system). The implication is that the development of a culture of expectations around work-based learning-including expectations among employers that co-op education is worthwhile-may be more powerful in the long run to sustaining work-based learning than are more formal mechanisms of institutionalization. Under these conditions, a unique form of school-to-work program has evolved in Cincinnati-and by implication can develop elsewhere.