identifying a crucial provision of the GPA. This states that the disciplines of the Agreement apply only to government purchase per se and excludes from its domain trade measures that have been taken under multilateral trade agreements. Foreign suppliers' ability to effectively contest the market for government procurement may thus be limited by restrictive trade measures. Such measures are far more significant in the case of services than in the case of goods, since in the former case innumerable restrictions are still in place on market access and national treatment. The creation of genuine international competition for procurement contracts thus remains crucially dependent on the liberalization of trade.
Section IV describes the relevance of the GPA to the agency problems that arise in procurement. Informational asymmetries lie at the heart of any analysis of procurement policies. The literature on government procurement in the international context, in focusing on the tension between minimization of procurement cost and protection of domestic firms, has paid relatively less attention to the problem of moral hazard on the part of the procurer. A significant benefit of the GPA is in helping to overcome national agency problems in procurement by creating mechanisms for reciprocal international monitoring. It achieves this by shifting the legal scope for monitoring from dispersed tax-payers, who have little interest in monitoring individual procurement decisions, to the bidders for contracts who have a significant stake. Two elements of the GPA are crucial in this context. First, the agency problem is mitigated by creating obligations on the procurer to be transparent. Secondly, foreign suppliers are given the opportunity to challenge the decisions of the procurer before national courts or independent and impartial review bodies.
However, several weaknesses are identified in the GPA's enforcement mechanism. First, the Agreement allows the compensation to a successful challenger to be limited to the cost for tender preparation or protest. Hence, the anticipated gain from private action, and the incentives for it, are not very high. Consequently, the scope for private litigation to exercise a socially desirable corrective or deterrent effect is limited. Secondly, the Agreement imposes no restrictions on the common practice of settlements, i.e. the exchange of a cash payment in return for a promise by the protester to drop its suit. Settlements between the procurement official and a dissatisfied firm circumvent the enforcement role of the protester, while settlements between the successful and unsuccessful firms can lead to collusive outcomes. Finally, the enforcement mechanism does not provide for challenge