Adjustment and the measurement of marginal intra-industry trade

In this paper, we have set out to do two things: first to investigate the properties of the index of MIIT recently proposed by Hamilton and Kniest; second to evaluate how useful the index is as an input to relating IIT and adjustment. In investigating the former, we found that there were a number of problematic properties of the H-K index. Most notably it censors the data in a non-random way by excluding important information; it inflates the measured MIIT by using nominal rather than real data; and it is unsealed. These properties not only bias the measure upward, they make it inappropriate from the standpoint of evaluating adjustment problems. Our discussion concludes with several suggestions as to how we might identify and measure MIIT more effectively from the standpoint of evaluating the adjustment implications of trade expansion.