To Settle or Empanel? An Empirical Analysis of Litigation and Settlement at the WTO

This paper examines the use of the WTO's dispute settlement mechanism by states. In particular, we are interested in understanding what causes disputes to move from the negotiation stage to the panel stage. Standard models of dispute settlement within a domestic context usually focus on the role of informational asymmetries on the probability of settlement. Although informational issues may be relevant at the WTO as well, we test an alternative, though not necessarily mutually exclusive, hypothesis. When states negotiate over a dispute at the WTO, their negotiations are typically focused on the specific source of the dispute. This inevitably constrains their ability to reach an agreement because transfer payments between the parties are more difficult than transfers in the form of, for example, cash. Furthermore, to the extent that the only transfers considered are those that can be made by manipulating the disputed policy variables, certain disputes will be particularly difficult to settle prior to the panel stage. When the subject matter of the dispute has an all-or-nothing character and leaves little room for compromise, as might be true of health and safety regulations, for example, the parties' ability to reach an agreement through the use of transfers is restricted. Settlement through negotiation may be even more difficult when governments cannot easily fashion side-payments to compensate for a major indivisible concession. In contrast, if the subject matter of the dispute permits greater flexibility, such as the setting of a tariff level, the parties can more easily structure appropriate transfer payments by adjusting that variable.