With the progressive dismantling of formal trade barriers as a result of many rounds of global trade negotiations, subsidies have become increasingly important as a way for governments to regulate economic activity within their territories. While subsidies are not necessarily bad policy, it is important to weigh their expected benefits against the possibility of efficiency, equity, or even environmental problems that may result.

Investment incentives are those subsidies designed to affect the location of investments. They are thus distinguished from production subsidies, which are not based on investment, but on normal production.

It is difficult to estimate the total amount of investment incentives given in any country. Even in the European Union, where the requirement to pre-notify subsidies to the European Commission (EC) makes the data on individual subsidies widely available, the classification system does not allow the aggregation of investment incentives per se. While many analysts argue that regional aid programs are largely used to provide investment incentives, some Member States are in their entirety areas that are considered poor by EU standards, and so classify little of their aid as "regional." Yet, much of the non-regional aid in countries like Poland, Hungary, or even Ireland could be considered investment incentives.

In the United States, data are fragmentary at best. In 2000, I estimated that total investment incentives in 1996 came to $26.4 billion, based on data from 8 states comprising 30% of U.S.GDP. The number has surely risen since then, given the introduction of numerous new incentive instruments and their widespread diffusion among the states. Confirming this is a recent estimate of $50 billion by Peter Fisher and Alan Peters of the University of Iowa. In Canada, the provinces and territories report on investment incentives to the Secretariat for Internal Trade, but the reports are not made public. Thus, problems of transparency are widespread throughout the industrialized democracies.

Government fiscal decentralization is causing similar patterns of increasing investment incentives in countries such as Brazil, China, and India. This presents the specter of sub-national governments squandering resources on investment bidding wars in countries that can least afford it. Moreover, investment incentives in the industrialized world can redirect investment to them that might have otherwise gone to developing countries. A number of studies have shown that rich countries can afford to give more incentives than poorer countries, even within areas with relatively small income disparities, such as the U.S.A. or the EU.

In addition, investment incentives are likely to harm the environment. Insofar as they artificially increase production in a given industry, the "extra" production facilities will increase overall pollution. Naturally, the more polluting the industry, the bigger this effect will be.

Regulation of investment incentives is largely non-existent. The most comprehensive control is the EU's Multi-Sectoral Framework on Regional Aid for Large Inward Investment. This approach weighs the size of the incentive, the extent of backwardness in the assisted area, the number of jobs created, and the sector, with demerits for sectors with overcapacity or dominant producers. In fact, Ireland was forced in 2005 to withdraw a proposed 100 million euro incentive to Intel, based on Intel's status as a dominant producer and on the non-assisted nature of the proposed investment location.

Canada has a Code of Conduct on Incentives, but its only full-scale requirement is that provinces not use incentives to induce the relocation of existing facilities from other provinces. While its only legal test was disappointingly inconclusive, there do not appear to be any recent violations of this ban. The Canadian provinces and territories have been involved in long-running negotiations to strengthen the Code, as well as the dispute settlement mechanism in the Agreement on Internal Trade, of which the Code is a part. Voluntary versions of no-raiding agreements have been tried in both Canada and the United States without success, and Australian states have been unable to agree to a no-raiding agreement.

Global regulation of investment incentives seems to be a long way off, given that more countries are moving in the opposite direction. Current WTO rules do not appear to have much effect on most incentives. The Agreement on Trade-Related Investment Measures (TRIMs) only prohibits requiring investors to meet domestic content or trade-balancing requirements. The Agreement on Subsidies and Countervailing Measures (ASCM), in Annex IV, does consider some incentives, especially those of 15% or more of the project cost, as actionable, but so far only one complaint has been made: by the EU against state and local subsidies to Boeing in Washington State. In any event, many common incentives, such as those given to the retail giant, Wal-Mart, to select one store location over another, have no impact on trade and could not conceivably be subject to WTO rules. In these cases, the only recourse is citizen pressure on local, state, or federal governments.

One further reason for the difficulty of incentive regulation is the lack of transparency. As with many other types of subsidies, decisions are often taken behind closed doors, making accountability impossible. (See the October 2006 issue of Subsidy Watch, "Transparency in Farm Subsidies: Is Sunlight the Best Disinfectant?") Thus, in many areas where subsidy reform is being pursued, demands for transparency are rightly the first order of business.

Kenneth P. Thomas is the author of Competing for Capital: Europe and North America in a Global Era(Georgetown University Press, 2000). He is an Associate Professor of political science and a fellow at the Center for International Studies, University of Missouri - St. Louis.