Trade Policy and Industrial Sector Responses in the Developing World: Interpreting the Evidence

Most students of economic development believe that liberal trade regimes are a good thing and costs of protection can be substantial. In significant part this belief is based on the notion that foreign competition disciplines domestic firms, forcing the firms to eliminate waste, accelerate innovation rates, or shut down. In turn this import discipline notion traces to numerous empirical studies of firm- and plant-level liberalization episodes.¹ These studies conclude that developing countries' manufacturing sectors have become more efficient after trade liberalization episodes. This has been accomplished partly through producer turnover, with heightened competitive pressure from imports as the motivating force.

Although the empirical literature that supports the import discipline hypothesis offers some robust findings, many basic issues remain unresolved. One source of ambiguity is that the literature is based on flawed measures of firm performance. But more fundamentally, this literature fails to identify empirically the mechanisms that link import competition to efficiency, it only describes the short-run effects of trade liberalization, and does not translate firms' performances into welfare measures. As such, it is of limited use for policy analysis. Hence we develop a coherent dynamic model that is consistent with the findings of firm- and plant-level empirical studies, but more informative concerning managerial behavior, transition paths, and welfare effects.

Our analysis is based on a modified version of the model of industrial evolution of Ericson and Pakes as well as Pakes and McGuire, hereafter referred to as the PEM model.² It describes an industry populated by a changing set of firms, each producing its own differentiated product. New firms enter the industry when the expected present value of their future earning stream exceeds their entry costs. Incumbent firms exit when the expected value of their future earnings stream falls below the scrap value of their assets. While active, firms can invest an amount of their choosing to increase the likelihood of a quality-improving product innovation. All economic agents make optimal choices, given their current information sets and the idiosyncratic shocks they experience. (Among other things, these choices reflect accurate perceptions concerning the stochastic processes against which they optimize and the behavior of their competitors.) We modify the PEM framework by introducing an imported product variety that competes with the domestically produced varieties and increases in quality at an exogenous rate.

Simulations of our version of the PEM model reproduce the well-known features of short-run adjustment to trade liberalization by import-competing sectors: price-cost margins fall and efficiency improves, largely because of the elimination of weak product lines and closings of inefficient plants. But our results also demonstrate that the intraindustry efficiency effects of foreign competition are probably more nuanced than commonly believed.

Specifically, we find that productivity gains due to the purging of weak firms are transitory and likely to dissipate within ten to fifteen years of trade liberalization. As they fade, the cumulative effects of reform-induced changes in the incentive to innovate become more important. These are often negative, so foreign competition can create a longer-term tendency for the quality of domestic goods to deteriorate relative to imports. Depending on the nature of the trade reforms, this tendency may or may not be offset by quality or efficiency gains due to embodied technological progress in imported capital. In any case, heightened import competition is likely to be accompanied by permanently higher plant or product line turnover, plus more rapid job creation and job destruction. Finally, there is a strong possibility of welfare losses on the part of domestic producers, but welfare gains among consumers due to lower prices are likely to be larger.

Import Discipline Hypothesis

The conjecture that import competition might force domestic firms to become more productive has been popular for at least three decades. When it first gained popularity there was little in the way of formal theory to explain it. Similarly, the supporting empirical evidence was limited to a handful of case studies and rudimentary productivity analyses. But over the intervening years the case for the import discipline...