Service Outsourcing and Specialization: A Theory on Endogeneous Task Scope

We develop a model of outsourcing and trade in service inputs where the scope of tasks produced by both manufacturing firms and service providers is endogeneous. Manufacturing firms have to perform a fixed set of tasks in order to produce their final good but can decide to outsource some of these tasks to service providers, which, contrary to manufacturers, have the possibility to sell tasks to different manufacturers and thereby benefit from economies of scale in their task production. The key assumption is that the marginal cost of a firm (manufacturer or service provider) increases in the scope of tasks performed inside the firm: a firm which specializes in a narrow scope of tasks is more productive. Working against this incentive to produce as few tasks as possible "inhouse" is a fixed cost paid by each firm. The model yields several new predictions about trade liberalization and welfare as measured by aggregate productivity. An increase in the size of an economy raises the scale of all firms, facilitates greater specialization and therefore raises each firm's productivity. The model therefore generates gains from trade or larger market size through a "specialization effect" as opposed to the classical "variety effect" usually generated by models building on Dixit Stiglitz utility structures. Welfare increases due to adjustments in task scope allowed by the emergence of specialized service firms. Detailed Swedish data on what tasks (or occupations) are performed by workers is used to test this prediction. Indeed,we find that manufacturing firms in larger cities (controlling for firm size) perform fewer tasks inhouse than firms in smaller cities.