Bernard Elbaum Nirvikar Singh
University of California, Santa Cruz

Apprenticeship, a pervasive institution within medieval Europe, persists today in various modified forms as an important element in different national systems of education and training. The persistence of apprenticeship has generated a variety of theories regarding its underlying economic rationale. According to one school of thought, apprenticeship results from monopolistic restrictions imposed by trade unions, employers' associations, and/or the government. From another perspective, apprenticeship offers efficiency benefits that either explain why unions, employers, and the government could find it in their joint interest to maintain apprenticeship regulations, or more directly, why apprenticeship managed to persist of its own momentum.

Although the workings of apprenticeship, both in the present and historically, have been much discussed, they have never been the subject of formal economic analysis. The present paper offers a first attempt at a formal economic model of apprenticeship which seeks to highlight the efficiency benefits that may provide at least part of its underlying rationale.

Our model is based on British apprenticeship practices of the late nineteenth and early twentieth century, and draws upon related historical research. The British case is of particular interest because its apprenticeship practices, medieval in origin, persisted into modern times amidst broadly competitive labor markets and laissez faire government. British laws requiring apprenticeship to practice a skilled trade and enjoy the rights of freemen were abolished respectively by repeal of the Statute of Artificers in 1814 and passage of the Municipal Corporation Act of 1835. Subsequently, apprenticeship flourished within Britain into the post-World War II era without benefit of active public promotion.

Because apprenticeship practices in different times and places tend to have at least some features in common, our model may also be of wider applicability. We will, in particular, try to apply it in order to understand why British apprenticeship practices, though transplanted into colonial America, largely failed to persist within the United States.

Apprenticeship, we suggest, was a device which allowed financially constrained youths to exchange indentured labor services for employer financing of training investments. To the extend that it permitted greater investment in training, apprenticeship afforded an efficiency improvement over the alternative of an unstructured labor market. But apprenticeship was only a "second best" arrangement.

The firm that employed apprentices was constrained to pay an initial wage high enough to meet the subsistence needs of the apprentice's family. After investing in training, since the craft skills involved were potentially transferable to other employers, the firm was further constrained to pay a wage high enough to deter quits that would entail the loss of its investment. As a result, apprentices tended to earn quasi-rents. Because enterprise investment responds in general only to the returns that the firm itself can capture, the economic logic of apprenticeship tended to lead, from a societal standpoint, to under-investment in training.

Central to the logic of apprenticeship, according to our model, is the indenture agreement, through which the worker and employer jointly commit to a fixed term of employment as well as to the provision of training. In Britain a variety of mechanisms helped ensure that indenture agreements would be honored, and that apprentices, in particular, would complete their stipulated term of indenture. Indenture agreements, when written, were legally binding, and even when unwritten, as they usually were, possessed the quasi-legal legitimacy of long-standing custom. In addition, employers' associations as well as trade unions pressed firms to require certification of indenture completion for entry into skilled jobs. Finally, and we suggest, most importantly, firms regarded indenture completion as verifying the worthiness of youths for skilled employment--behavior which we describe formally through a model of skill certification.

Our model also suggests that certification requirements were essential in order to forestall an apprentice runaway problem that could lead to the collapse of the apprenticeship system. The decline of apprenticeship in the United States appears to be a case in point. In this rapidly expanding settler society, no certification was needed to obtain well paid work in the skilled trades. Chronic scarcity of skilled labor led employers to the common practice of filling skilled jobs by up-grading less skilled workers. Mobility--both geographic and occupational--was high, especially among youths, and tended to undermine the enforcement of apprenticeship indenture. Although the powers of the law, trade unions, and employers associations were invoked to enforce apprenticeship regulation, they proved unable to stem the tide that erased all but a small trace of British apprenticeship traditions from the American continent.

The paper proceeds by describing British apprenticeship practices in outline, emphasizing the evidence that supports the two key assumptions of our model: (1.) apprenticed workers were subject to a subsistence constraint; and (2.) the possibility of apprentice quits constrained the pay that firms offered in the latter stage of indenture.

We next present our formal model. For simplicity, we assume that an individual's work life consists of n periods, of which the first two may be occupied by apprenticeship. Firms may invest in training only during the initial period of apprentice employment. For apprenticeship to be profitable the apprentice quit rate must be low enough in the second period for firms to recoup their initial training investment.

We consider three models that are ordered in terms of the scope of the factors they seek to explain. We begin by assuming a fixed quit rate that is given exogenously. The propensity to quit may in this context be viewed as being equal to the proportion of employers willing to hire runaway apprentices at their marginal product--a proportion determined by some form of collective regulation on the part of unions, employers' associations, or the government.

We then analyze an alternative model, in which an apprentice's decision to quit or complete his indenture depends on the wage policy of his employers, and in particular, on how the second period wage compares with an exogenous distribution of external wage offers. In this model, the firm optimizes with respect to its choice of a second period wage.

Finally, we present a certification model in which the distribution of external wage offers, and hence of apprentice quits, is determined endogenously. In this model apprenticeship completion may be viewed as as playing much the same kind of a role as a school degree today. By assumption, the firm can only discern the productivity of workers who lack skill certification through a costly and time-consuming process of work supervision. As a result, the firm is only willing to hire such workers at a pay rate which corresponds to its initial probabilistic estimate of their productivity. Apprentices are in turn deterred from running away since if they do they cannot easily be distinguished from workers who lack systematic training, or from individuals who dropped out of apprenticeship because they were unsuited to the work.

After deriving our formal results, we review evidence on British apprenticeship which suggests that collective regulation of indenture agreements was relatively ineffective in engineering and building--by far the largest sectors of apprentice employment--and draw the implication that enforcement of indenture obligations here depended on the preference by individual employers to hire craftsmen of certified skill attainment. We then proceed to examine the decline of apprenticeship in the United States.