آموزش کارشناسی ارشد تجارت : کارآموزی و سرمایه انسانی در غنا

This paper explores the institution of apprenticeship in Ghana. A model is presented where apprenticeship training is idiosyncratic, increasing an individual's productivity in the current firm, but not in any other firm. Still, individuals are willing to fund apprenticeships as they can reap the returns to the specific training of apprenticeship if they manage to acquire the capital required to start their own firms, and replicate the technology and business practice of the apprenticeship firm. Predictions of the model for the productivity and remuneration of different workers are developed and tested using both a linked employer–employee survey of manufacturing firms and a national household survey.

The formulation of an appropriate education and training policy for the manufacturing sector in Africa should begin with an understanding of the training currently in place. Given the importance of apprenticeships in Africa,1 where apprentices learn one of the trades used in the manufacturing sector, any analysis should begin by examining this institution. Nevertheless, very few articles exist which use economic analysis to explore apprenticeships in Africa (Velenchik, 1995). This paper is a step towards addressing this dearth and contends that the apprenticeship institution is best understood in the larger context of the specificity of firm training. Apprenticeships are periods of roughly 3 years in length during which an apprentice learns a trade, such as metal-working or carpentry from a master of that trade. At the end of the apprenticeship, the apprentice may end up being hired by the firm where the apprenticeship occurred, begin working at another firm or the apprentice may start a new firm and become self-employed. Apprenticeships occur most often, but not exclusively, in smaller firms, and the master is often the owner of the firm. Becker (1964) showed that, in the human capital model with perfect labor markets, workers (and not firms) always pay for general training, which increases an individual's productivity with all employers, as the value of their outside wage offer increases with general training. However, firms cannot credibly commit, without some formal contracting or other mechanism, to compensate workers for firm-specific training once it has occurred, as it does not increase their outside wage offers. This dichotomy has provided a framework for analyzing the nature of training in a number of studies. Acemoglu and Pischke (1998) treat German apprenticeships as general training and present a model that explains why firms are willing to invest in this training. Acemoglu (1997) and Acemoglu and Pischke (1999) provide distinct models, which explain why firms might invest in general training. Other papers have examined worker investment in firm-specific training. Prendergast (1993) describes the potential for promotion rules to induce worker investments in firm-specific human capital, with the firm's ability to commit to future wage increases through long-term contracts playing a significant role in the model's mechanism. Scoones and Bernhardt (1998) remove the need for long-term contracts to achieve worker investment in firm-specific human capital by introducing asymmetric information. In their model, workers invest in firm-specific human capital, in order to be promoted by the firm, as promotion reveals their higher ability to other firms. In a similar model, Scoones (2000) finds that workers invest in firm-specific capital because efficient turnover transforms former employers into outside options. Other papers (e.g. Felli and Harris, 1996) examine firm-specific human capital as something exogenous to worker decisions, but we are interested in examining investment in human capital that is discretionary from the worker perspective. At first glance, apprenticeships might appear to be general training, that is applicable at least in all of the firms within an industry, if not more generally. As noted, this is how they have been understood in Germany. However, the nature of the training acquired in Ghana is far more specific—specific to the firm providing the training, reflecting the firm's technology and business practice. In this context, the model of apprenticeship to be developed shares the spirit of the model of Jovanovic and Nyarko (1995). In their overlapping generations model, each old agent understands an idiosyncratic (which is unique to the firm owned by the old agent) technology, which is passed on to a young agent who is his apprentice. This interpretation of apprenticeships as training unique to the firm is consistent with the apprenticeships under examination in this paper.2 Apprentices are taught by a master how to work the master's craft, but the way in which that craft is carried out varies highly from firm to firm. In the process of this research investigation, I visited and interviewed dozens of the manufacturing firms in the dataset. A number of the firms interviewed were involved in the manufacture of a single item. When apprentices apprentice at these firms, they may only learn how to manufacture one or a small number of items. While apprentices at other firms did learn how to manufacture a wider variety of items, product homogeneity, using unique technology, is quite common among the practice of firms with apprentices. The training given by the master, both in terms of technology and business practice, is that of the master, and typically idiosyncratic.3 Therefore, the apprenticeship training cannot be applied to work in other firms, which have their own technology and business practice. However, once the apprenticeship is complete, the former apprentice can use this knowledge to start his own firm and pursue self-employment, replicating the apprenticeship firm. In fact, apprentices appear highly motivated to pursue self-employment. Of the Ghanaian manufacturing workers, who had completed apprenticeships, 77% stated that they would prefer to be self-employed rather than working in their current job. Naturally, however, apprenticeship knowledge is not enough—capital is required to start a firm. Those former apprentices who get access to capital can start a new firm, while others cannot. The potential to reap the returns of the apprenticeship, this specific training, by creating a new firm and replicating the technology and business practice of the apprenticeship firm motivates individuals to acquire the human capital of apprenticeship. This apprenticeship human capital is not transferable to any other firm outside the current firm, but it can be used when the apprentice replicates the technology and business practice of the current firm in self-employment. Therefore, while the apprenticeship certainly does not provide general human capital (which is useful in all firms), the human capital is also not ‘firm-specific’ in the pure definition of the term. The human capital can be used outside of the current firm, but only when the apprentice replicates the apprenticeship firm. Perhaps, a term such as ‘technology-specific’ might better describe the nature of the apprenticeship human capital. However, this technology should be understood as unique to the firm providing the apprenticeship training. The exact labelling of the human capital is not as important as understanding its nature. Therefore, it shall be called simply ‘specific human capital’ in this paper. Workers are willing to invest in this type of specific human capital, which is tightly tied to the firm providing apprenticeship training, not because of asymmetric information, or any other labor market imperfection, or the firm's ability to commit to future wage increases, but rather through a worker's potential ability to reap the benefits of the apprenticeship through self-employment. Of course, once the apprenticeship is finished, the apprentice can also continue to work in the master's firm. While the apprenticeship knowledge is certainly applicable in this context, the master, and not the apprentice, reaps the returns to this knowledge. The reason for this is the fact that, while the former apprentice is more productive in the master's firm, he is not more productive in other firms, and so his outside wage option is low. The master, therefore, only needs to pay the former apprentice marginally above his outside wage option in order to retain him in the firm. Only in self-employment will the former apprentice receive his full marginal product (including the apprenticeship-enhanced productivity) as a wage. Fortunately, this model, which will be formalized in Section 3, creates predictions, which can be tested using the data. Although the model explains worker investment in specific training, the mechanics of the model resemble Acemoglu and Pischke (1998). The predictions of the model include the following. Former apprentices who apprenticed in the current firm should be more productive than former apprentices from other firms. However, they do not need to be paid more than former apprentices from other firms. For this reason, former apprentices will seek self-employment and the returns to apprenticeship should be found in self-employment. This model shares similarities and differences with Becker's (1964) original work in this area. In Becker's model, workers and firms share the cost of firm-specific training, although the relative proportions contributed may vary. Here, we will see that either workers or firms or both may cover the cost of the training, depending on the market for apprenticeship. In Becker's model, after the training, firms pay these specifically trained workers less than their marginal product. This fact reflects the firm's ownership of the worker's specific capital and allows the firm to recoup its portion of the training cost. Here, also, specifically trained workers receive less than their marginal products in the training firms as a wage. As in Becker's model, workers with specific training in a different firm will receive their marginal product in the current firm, but this amount is less than what their marginal product would have been in their training firm. While Becker's (1964) discussion of self-employment is implicit rather than explicit, an application of the original Becker framework to include self-employment (without credit constraints, as in the original Becker model) would also see returns to specific training in self-employment, as the worker becomes owner of his or her specific capital. In fact, one purpose of this paper is to explicitly and formally makes this extrapolation—to examine the case of specific capital and self-employment, in the context of apprenticeship in Ghana. An important characteristic of the model is the fact that firms are not allowed to make time-inconsistent wage offers. That is, in any given period, it will only prove optimal for firms to pay workers their outside option. As will be outlined in the paper, this reflects the difficulties in contract enforcement in Africa, through firm reputation or other means. For comparison purposes, the model for the case of completely enforceable long-term wage contracts will also be provided. Firms make less profit in the absence of full contract enforcement and a measure of this loss is calculated in the context of this model. Therefore, this paper also provides a specific example of the costs that African firms incur as a result of the difficulties of contract enforcement there.