A variety of stylized facts about exporters have emerged in the new literature on international trade based on firm-level data. These include low levels of export participation among firms; small shares of export sales in firm revenue; larger firms; and higher levels of productivity, skill, and capital intensity among exporters. In this paper, we seek to examine the extent to which these stylized facts fit the experience of firms in Pakistan, using two cross-sections of firm-level data—the Census of Manufacturing Industries (CMI) 2000/01 for Punjab and the World Bank Enterprise Survey dataset (2006/07) for all Pakistan.

We find similar levels of export market participation but very large shares of export sales in firm revenue for those who do, compared to the US sample studied by Bernard, Jensen, Redding, and Schott (2007). We also find, as do many other studies, that exporters exhibit significantly higher total factor productivity (TFP) and are larger in terms of employment than non-exporters. Exporters’ TFP was 150 percent higher than non-exporters before we controlled for firm size. Considering the eight largest sectors (which comprise more than 80 percent of the CMI Punjab), with a few exceptions, exporters had higher labor productivity and offered higher compensation to workers, but used more capital per worker and more imported inputs.

The government’s recent emphasis on developing the ready-made garments sector is well placed: more than half the apparel producers in the CMI Punjab 2000/01 were exporting—and nearly all of their output (93 percent). The capital - labor ratio and use of imported inputs was modest. Exporters were relatively large employers with 400 workers on average and offered significantly higher compensation than non-exporting firms. A greater understanding of firm dynamics could be gained if the CMI were to resume collecting data on firm-level exports (not collected since 2000/01) and if this data were linked across years so that firm performance could be measured over time.