Modelling Contingent Protection

Anti‐dumping and countervailing duty (AD/CVD) law and policy has become one of the most contentious issues affecting trading relations between members of the World Trade Organisation. A major concern among researchers and policy‐makers is that the decision‐making process of regulatory authorities responsible for the administration of AD/CVD laws is biased in favour of providing protectionist outcomes for applicants. However, to date there is no agreed approach on how to undertake empirical analyses needed to help AD/CVD policy development. What is needed is an agreed general framework for such analyses. In this paper a new approach, the Competing Claims framework is introduced to model empirically the AD/CVD decision‐making process. The Competing Claims approach is more inclusive than approaches used previously and makes it is easier to identify appropriate hypotheses for testing. Using Australian data covering a period characterised by relatively heavy use of AD/CVD processes, the paper compares this new Competing Claims approach to that of previous researchers and concludes that previous approaches, if they had been adopted using the Australian data, would have led to missing variable bias. The implication for future researchers is clear. In order to assess the various potential biases that can inject themselves into the very opaque world of contingent protection decision‐making, as broad an empirical model as possible should be used.