Trinity College Dublin

Skip to main content.

Top Level TCD Links

Export mode, Trade Costs, and Productivity Sorting

Ronald B. Davies and Tine Jeppesen

IIIS Discussion Paper No. 408

Abstract

In this paper we directly test the proposed productivity hierarchy of direct, indirect and non-exporters using firm-level data from 105 developing and transition countries. Using both regression analysis and propensity score matching, we find strong evidence to suggest that direct exporters are on average more productive than both indirect and non-exporters. However, only the results obtained using regression analysis support a similar ranking between indirect and non-exporters. Furthermore, we test the underlying relationship between source-specific fixed trade costs and the average productivity differences between the three firm-types. We find a significant and positive relation between such costs and the average productivity premium of direct exporters only. While other studies have shown that exports by trade intermediaries increase with destination-specific fixed costs, our results suggest that this is also true for source-specific costs, as an increase in the average productivity of direct exporters indicate that a larger share of less productive direct exporters choose to make use of a trade intermediary as export costs rise.

non technical summary

In this paper we directly test the productivity hierarchy of direct, indirect and non-exporters using firm-level data from 105 developing and transition countries. We begin with a simple model that suggests that more productive firms export directly, that is, do so themselves. In contrast, intermediate productivity firms export indirectly by hiring an intermediary. This is because, although this costs more per unit, it saves on fixed costs which would otherwise prove prohibitive. Low productivity firms find that even these costs outweigh the benefits. In particular, we show that an increase in exporting costs such as red tape can impact firm sorting.

Using both regression analysis and propensity score matching, we find strong evidence to suggest that direct exporters are on average more productive than both indirect and non-exporters. However, only the regression results support a similar ranking between indirect and non-exporters. Furthermore, we test the underlying relationship between source-specific fixed trade costs and the average productivity differences between the firm types. We find a significant and positive relation between such costs and the average productivity premium of direct exporters only.

While other studies have shown that exports by intermediaries increase with destination-specific fixed costs, our results suggest that this is also true for source-specific costs. Since export costs are under the direct control of developing countries, recognizing their role in firm behaviour is important for developing policy.

 


JEL codes: F1
Keywords: Heterogeneous firms; Export mode; Exporting Costs



Last updated 28 August 2014 by IIIS (Email).