Imported Inputs and Productivity
László Halpern, Miklós Koren and Adam Szeidl
Abstract
How do imported inputs affect firm productivity? We address this question by estimating a structural model of importers using product-level data for all Hungarian manufacturing firms during 1992-2003. We have three main findings. (1) Imported inputs have large productivity effects: increasing the share of imported goods from 0 to 100 percent increases productivity by 11 percent. (2) About 60 percent of this gain is due to imperfect substitution, i.e., the idea that combining different inputs is "more than the sum of the parts." This is consistent with Hirschman's (1958) view about the importance of complementarities along a production chain for economic development. (3) Tariff cuts have a highly non-linear effect on productivity, due to firm entry into import markets for new varieties. This non-linearity can rationalize differences between estimated tariff effects in different studies, and shows how firm level analysis helps understand macro facts. Our structural approach can also be used for counterfactual policy analysis, and to study the different implications of the quality and complementarity mechanisms.
April 2009
Publication status:
Revise and resubmit at the American Economic Review.