In this paper I analyze the productivity gains from trade liberalization in the Belgian textile industry.So far, empirical research has established a strong relationship between opening up to trade and productivity, relying almost entirely on deflated sales to proxy for output in the production function.The latter implies that the resulting productivity estimates still capture price and demand shocks which are most likely to be correlated with the change in the operating environment, which invalidate the evaluation of the welfare implications.In order to get at the true productivity gains I propose a simple methodology to estimate a production function controlling for unobserved prices by introducing an explicit demand system.I combine a unique data set containing matched plant-level and product-level information with detailed product-level quota protection information to recover estimates for productivity as well as parameters of the demand side (markups).I find that when correcting for unobserved prices and demand shocks, the estimated productivity gains from relaxing protection are only half (from 8 to only 4 percent) of those obtained with standard techniques.
Stored only when both papers are in the corpus — 389 citations exist upstream.
Imported from OpenAlex (CC0) — nothing here is generated.