2012 Theses Doctoral

Essays on Firms' Behavior in International Trade with Vertical Specialization

Maitra, Madhura

My dissertation consists of three essays that allow me to investigate two related trade induced economic phenomena -- processing trade and offshoring -- using diverse datasets and theory. In Chapter 1, a joint work with Mi Dai and Miaojie Yu from Peking University, we solve the documented puzzle that exporters in China are less productive than non-exporters in the labor intensive sectors and in the Foreign Invested Enterprises (FIE). We show that this anomalous finding is entirely driven by firms that engage only in export processing -- the activity of assembling tariff exempted imported inputs into final goods for resale in the foreign markets. We find that pure processing exporters are less productive than non-exporters, but other types of exporters -- those doing only non-processing trade and those doing both processing and non-processing trade -- have superior performance relative to non-exporters. Our results show that distinguishing between processing and ordinary exporters is crucial for understanding firm-level exporting behavior in China. In Chapter 2, a joint work with Henrik Bursland Fosse, from Copenhagen Business School, we investigate the effects of offshoring on wages. Offshoring firms are found to pay higher average wages than purely domestic firms. We provide a unifying empirical approach by capturing the different channels through which offshoring may explain this wage difference: (i) due to a change in the composition of workers (skill composition effect) (ii) because all existing workers get higher pay (rent sharing effect). Using Danish worker-firm data we explain how much each channel contributes to higher wages. To estimate the causal effect of offshoring on wages we use China's accession to the WTO in December 2001 and the soon after boom in Chinese exports as positive exogenous shocks to the incentive to offshore to China. Both skill composition and rent sharing effects are found to be important in explaining the resultant gain in wages. We also show that the firm's timing in the offshoring process determines the relative importance of a channel. For firms offshoring to China in 2002 but not in 1999, only rent sharing explains the gain in wages. However, for firms offshoring to China both before and after China's WTO accession the wage increase is explained more by the skill composition effect. Moreover, these patterns are not discernable from the measures of skill composition and rent sharing available in typical firm level datasets-- such as ratio of educated to uneducated workers and sales per employee. In Chapter 3, I extend the Sethupathy (2008) model to investigate the wage effects of offshoring in the presence of heterogeneous firms, heterogeneous workers, and imperfections in the labor market with rent sharing. The salient features of the model are: first, there are heterogeneous firms who differ in terms of productivity; second, presence of heterogeneous workers who vary at the skill level; third, imperfect labor market with presence of search costs, wage bargaining leading to rent sharing between firms and workers; fourth, performance of high-skilled and low-skilled tasks are required for production of the good; fifth, there is opportunity for offshoring each type of task, requiring a marginal cost that varies with the degree of non-routineness of the task and a fixed cost. In this framework I show that a fall in the cost of offshoring increases average wage in the offshoring firm due to a rent sharing effect. This effect can be further reinforced or weakened by an accompanying skill composition effect. Average wages in the non-offshoring firms decline due to a rent sharing effect only; there is no skill composition effect for these firms in the model.