MOST PUBLIC PROCUREMENT SYSTEMS AIM TO ACHIEVE "VALUE FOR (TAX-payer) money." The mechanisms to attain that goal often center on mimicking the working of the market by requiring procuring entities to seek competitive bids for contracts above a threshold. In practice, costs often are not minimized because of a strong "home bias" that leads contracts to be awarded to local companies. Procuring entities may prefer to spend tax revenues at home, and in the process generate political support, or they may want to safeguard national supply capacity in important sectors (e.g., defense equipment) or to achieve social objectives (e.g., to support minorities, small businesses, or disadvantaged communities).
Insofar as procurement policies explicitly favor domestic firms and products, they can act as trade barriers. The market access dimension of discriminatory procurement practices is generally the main rationale for negotiating disciplines on government procurement in international trade agreements. In this respect, the political economy of market access negotiations on procurement are the same as those for trade agreements more generally. Reciprocity is at the core of efforts to liberalize procurement markets, as the loss of a sheltered home market for domestic firms is offset by an increase in contracts won in trading partners. However, procurement liberalization is more complex than tariff reduction because it involves regulatory regimes--the systems that a government puts in place to allocate contracts and seek to ensure that winning bidders are capable of delivering a product or a project, to reduce the scope for collusion or corruption, and to hold firms and procuring entities accountable for performance.
The regulatory features of procurement regimes are critical in ensuring that public objectives are pursued in an efficient manner. As in any area of regulation, different countries may pursue different approaches, and there is no one-size-fits-all optimal procurement mechanism that is appropriate for all situations and all countries. For complex procurement projects involving long-lived infrastructure projects, new technologies, or outsourcing of public services, it may not be clear what the best approach is, and there is much to be learned from international experience. Thus, in addition to addressing spillovers generated by de jure discrimination and differences in procurement regulation that result in de facto discrimination against foreign bidders, the inherent complexity of designing procurement systems to achieve public service objectives efficiently creates incentives for international cooperation.
In this article, I discuss the effectiveness of reciprocal negotiations to open procurement markets to greater foreign competition. The article is motivated by the question of whether (preferential) trade agreements are (can be) a driver for the gradual multilateralization of disciplines on procurement processes. The evidence and experience to date suggests that regulatory cooperation, transparency, and learning may be more effective than negotiated market access liberalization in enhancing the contestability of procurement markets. The implication is that trade agreements should do more to promote regulatory cooperation to enhance learning about good practices.
This article is organized as follows. I briefly summarize the state of play in the WTO and PTAs on procurement and then turn to available evidence on the effectiveness of trade agreements in reducing home bias. I argue that international cooperation should center more on learning from cross-country experience as to what works and why (and why not) in achieving the various objectives that are pursued through procurement.
Trade Agreements and Government Procurement
Government procurement was excluded from the original General Agreement on Tariffs and Trade (GATT) in 1947. It was not until the completion of the Tokyo Round of multilateral trade negotiations in 1979 that a multilateral Agreement on Government Procurement (GPA) was negotiated. This extends the basic GATT nondiscrimination rules to the purchases of goods by selected government entities. The agreement bound only signatories--at the time just twenty-two countries. The GPA has been revised several times since then to expand its coverage. After more than a decade of talks, the third revision of the GPA was adopted in 2012 and entered into force in April 2014. At the time of this writing, there are seventeen parties to the agreement, representing forty-five World Trade Organization (WTO) members. European countries account for 70 percent of the membership. It currently covers some $1.6 in public contracts. (1)
The GPA goes beyond the GATT and the General Agreement on Trade in Services (GATS) by extending national treatment to foreign affiliates. Signatories are required to "conduct covered procurement in a transparent and impartial manner that is consistent with this Agreement, using methods such as open tendering, selective tendering and limited tendering; avoids conflicts of interest and prevents corrupt practices." (2) The preference for competitive procurement methods is implicit in the agreement, reflected in requirements that notices of intended or planned procurement be published (including information on time frame, economic, and technical requirements, and terms of payment) and in disciplines on treatment of tenders and contract awards.
Until relatively recently, the basic presumption in the procurement literature was that the type of arm's-length international competitive bidding procedures called for by the GPA would, as a rule of thumb, generate efficient outcomes by awarding contracts to the lowest cost supplier able to meet the technical project requirements. However, especially for more complex projects, efficiency may require procuring entities to engage in negotiations and to interact with potential suppliers. (3) The advantage of such competitive dialogue is that it permits companies to engage with procuring entities, which allows the latter to consider alternative solutions and technologies and to determine what would be most appropriate in addressing their specific needs.
Price-preference policies, local content requirements, offsets, and similar discriminatory policies are widely used by governments to achieve equity or industrial policy goals. They are in principle prohibited by the GPA, but exclusions are built in to grandfather domestic content requirements for small businesses--for example, US federal procurement preferences for small businesses owned by women or socially disadvantaged businesspeople. Article V of the GPA gives developing countries the right to adopt or retain price-preference policies and offset requirements on a transitional basis and delay the implementation of any and all provisions other than most favored nation (MFN) treatment for up to three years (five years for a least developed country).
Although tax and subsidy instruments are likely to be more efficient in assisting domestic target groups than procurement favoritism, an advantage of using procurement is that it can help the most efficient firms in that group (as they must compete for the contracts) whereas a subsidy to a region or a minority group will be less selective. A weakness of the GPA in this regard is the centrality of market access. The procurement process-related rules in the GPA are primarily aimed at supporting the market-access goal and not designed to help governments achieve their procurement policy objectives.
Few developing countries have joined the GPA in the past thirty years, in part because of a concern regarding the implications for being able to pursue industrial and social policy objectives, and in part because of the fact that they have limited export potential in this area. This has led GPA members to pursue procurement disciplines through the negotiation of preferential trade agreements (PTAs) and in the process extend procurement rules to non-GPA members. One reason a developing country may be willing to sign a PTA that includes procurement liberalization is that the PTA offers a quid pro quo in other areas, something that is not possible in the GPA context because the GPA deals only with procurement. There are over forty PTAs currently in force that include commitments to open access to procurement contracts on a bilateral or regional basis. (4) Many of these are not far-reaching, but many recent vintage PTAs include extensive coverage of procurement.
A feature of the procurement-PTA landscape is that it mostly involves GPA members and that there is limited coverage of developing countries. In some cases, PTAs between GPA members are used to deepen market access on a discriminatory basis--an example is the Comprehensive Economic and Trade Agreement (CETA) between Canada and the European Union (EU). In the case of the Trans-Pacific Partnership (TPP), only three countries are not GPA members or did not have an existing agreement with the United States: Brunei, Malaysia, and Vietnam. The TPP is GPA-minus in terms of coverage of products and entities. Only five TPP signatories included subcentral procurement, but only if there is reciprocity. Non-GPA members Malaysia, Mexico, and Vietnam negotiated exceptions to key provisions disciplining the use of offsets and price preferences and further limited the reach of procurement rules through high thresholds that apply for long transitional periods. (5)
Do Trade Agreements Reduce Home Bias?
To assess the impact of procurement disciplines in trade agreements on home bias, whether the GPA or PTAs, one needs data on procurement awards over time that distinguish between winning bids on the basis of the nationality of ownership of firms. Such data are not reported by most countries. In the absence of detailed statistics on actual awards by entity and type of procurement (goods, services, works), the only sources...