Exporting and Plant-Level Efficiency Gains

Marginal costs within plant-product categories drop by approximately 15-25 percent during the first three years after export entry.

Trade competition has led to aggregate productivity gains, but some research suggests that those gains come only from selection of the most productive plants into exporting, rather than from efficiency gains within plants. That finding is rather surprising, because exporters can learn from international buyers, and by exporting will have access to larger markets and therefore incentives to innovate or invest in productive technology.

In Exporting and Plant-Level Efficiency Gains: It's in the Measure (NBER Working Paper No. 19033) Alvaro Garcia Marin and Nico Voigtlander use a cost-based measure of productivity and find that within-plant efficiency gains do occur after plants begin exporting. They suggest that other studies failed to find such gains because they used a revenue-based productivity measure, which is affected by changes in prices. Garcia and Voigtlander instead calculate plant-product-level marginal costs for a panel of Chilean establishments and show that those costs drop significantly for new exporters -- that is, a within-plant productivity gain.

The gains are substantial: marginal costs within plant-product categories drop by approximately 15-25 percent during the first three years after export entry. At the same time, new exporters pass on most of the efficiency gains to customers in the form of lower prices (around 20 percent), which are accompanied by a strong increase in export volumes. The fact that plants pass on the gains in physical productivity to buyers in the form of lower prices explains why studies that look at revenue-based productivity measures typically do not find evidence of within-plant efficiency gains.

In the data, export entry goes hand-in-hand with a decline in marginal costs in the entry period, which is not driven by productivity shocks before export entry. And marginal costs drop particularly steeply for plants that are initially less productive. Those two facts suggest that investment complementarity -- the fact that investment opportunities in new technologies become profitable in combination with access to larger markets -- is important. Moreover, marginal costs keep falling in the years after entry, which suggests that learning-by-exporting is also an important driver of the result.

Although the results suggest within-plant productivity improvements, selection into exporting based on revenue productivity is significant. In fact, the exporter revenue-productivity premium is 17 percent in this sample of Chilean firms. The within-plant productivity gains reflect efficiency gains in addition to the typically documented selection effect. Within-plant gains are of roughly the same magnitude as the between-plant differences.

--Claire Brunel