5.3 Incentives to Attract FDI

Uri Dadush
Dadush was a senior associate at the Carnegie Endowment for International Peace. He focuses on trends in the global economy and is currently tracking developments in the eurozone crisis.
More >

The desire to attract the “right” kind of FDI often leads countries to engage in a global competition. Many countries, regions, and states or cities within countries have established investment promotion agencies and enacted policies to incentivize FDI. These normally include incentives of a fiscal or financial nature. The former are designed to reduce a firm’s tax burden and include tax concessions in the form of a reduced corporate income tax rate, tax holidays, accelerated depreciation allowances on capital taxes, exemption from import duties and duty drawbacks on exports. The latter consist of direct contributions to the firm from the government and include grants, subsidized loans, loan guarantees, the participation of publicly funded venture capital in investments involving high commercial risks and government insurance at preferential rates.

It is understandable that authorities compete fiercely to attract FDI that creates jobs and helps revitalize local economies. A successful case is the BMW plant in Greenville, South Carolina, which is located close to the previously under-used Port of Charleston. In the early 1990s, BMW received US$ 130 million in incentives (about US$ 200 million today), including tax incentives, road improvements and job training. In turn, the company invested some US$ 2.2 billion in the region and created more than 5,000 jobs, in addition to the thousands more that were created by the automotive parts suppliers and research facilities that subsequently invested in the area.

But investments do not always succeed, and money spent on incentives can be wasted and even drag down the city or region that courted a firm. For example, the failure of Mamtek’s US$ 65 million investment in 2011 in an artificial sweetener plant in Moberly, Missouri sparked debates on government subsidies for foreign investors. The city of Moberly had issued US$ 39 million worth of special bonds to help the Chinese parent firm finance its US factory, in addition to the promise of US$ 18 million in state aid and tax incentives. Mamtek missed the first bond payment and stalled the construction of the facility. Standard & Poor’s subsequently downgraded the city’s credit rating.

However, dramatic failures are probably the exception. More often, cities or states simply end up paying a higher price than they may need to. For example, in 2004 and 2005, Dell opened a series of call centres in the United States and Canada, which prompted bidding wars among a number of cities. In July 2004, Dell announced that it would locate a call centre in Edmonton, Alberta. Edmonton triumphed over Calgary, Winnipeg and three US cities. It put together an incentive package worth about CAD$ 6 million, but it may have been possible for the city to land this deal without offering such a large and expensive carrot, given its success in attracting Ford, GE Credit, Neiman-Marcus and Convergys without incentives.

Investment incentives reflect a coordination failure among governments and are, like most subsidies, a source of inefficiency. They distort markets as artificial restrictions on investment do, though in different ways. The provision of incentives also has the potential to exacerbate regional disparities, since wealthier and more successful cities/provinces/countries are often able to provide larger incentives. For example, Hyundai received incentives of about US$ 117,000 per job from Alabama in 2002, but only about US$ 75,000 per job from the Czech Republic in 2007. Alabama’s per capita income in 2006 was US$ 29,414, while the Czech Republic’s was US$ 12,680 at current exchange rates and US$ 21,470 at purchasing power parity exchange rates.

Although competition among local authorities can be a good thing, incentives are an unhealthy form of competition. They can also distract attention from the hard decisions needed to improve the business climate and from investing in the skills of the local labour force, for example. In extreme cases, they can encourage investments that are inherently unprofitable and ultimately unsustainable. In the longer run, developing countries that become over-reliant on incentives to attract investment can adversely affect their own development.

One can learn about the challenges confronting attempts to regulate incentives at the global level from the experiences of the EU and federal states such as the United States, Canada and Australia in regulating incentives internally. In general, states have had only partial and modest successes in doing so.

The EU undertook the most comprehensive effort to regulate investment incentives. EU “state aid” rules start from the presumption that subsidies should not be used by Member states unless they contribute to a goal of the EU as a whole, and do so in a way that least distorts trade within the EU. Critical elements of disciplining the use of state aid are transparency and oversight. The European Commission must be notified of all subsidies in advance, and can prohibit or modify them if they are in violation of EU law. State aid disciplines also designate maximum aid intensity levels for every location within the EU. Poorer areas of the EU can provide larger subsidies than richer ones, and the most prosperous regions are prohibited from giving any aid at all. Governments can thus only give support to firms in proportion to the disadvantage of the region.

In the United States, by contrast, disciplines on investment incentives remain very weak. Indeed, the most significant force for reform in the United States has been private nongovernmental organizations. There have been attempts to use the federal tax exemption on certain types of local authority bonds as leverage to limit the incentives given to investors. Industrial revenue bonds are federal tax-exempt bonds issued by local governments to fund a wide variety of projects. Until 1986, there were few restrictions on them; they were very popular with local governments because the entire cost of tax deductibility was borne by the federal government. In the 1986 tax reform, caps were put on the amount of bonds that could be issued, and restrictions were placed on their use.

There have also been efforts to limit incentives competition among individual states, including two unsuccessful regional no-raiding agreements. State governments have entered into two voluntary no-raiding agreements. In the 1980s, the Council of Great Lakes Governors approved an anti-piracy pact, but it collapsed even before it went into effect. A 1991 agreement among New York, New Jersey and Connecticut met the same fate a few days after it started. New York City has been a particular target of nearby jurisdictions and has been subject to local companies threatening to move out of the city in order to receive retention subsidies.

The National Governors Association (NGA) has been active in hosting discussions on how to limit incentives competition. State governors are, of course, aware of the dangers of bidding wars for investment. However, the NGA has consistently argued that there should be no federal intervention to stop incentives, thus preserving state sovereignty, while also arguing that states should refrain from bidding wars because avoiding them is good policy. Clearly, this is a case in which moral suasion only goes so far.

Canada and Australia are two federal states in which competition for investment at the sub-national level has at times been severe. In both instances, the pressures have led to attempts to control incentives.

In Canada, the subsidized relocation of investment was a major problem in the 1990s. In this context, the Code of Conduct on Incentives was agreed to in July 1994 as part of the Agreement on Internal Trade, whose parties include the federal government, all 10 provinces and two of the country’s three territories. The code explicitly prohibited relocation subsidies and provided for an agreement among governments to make their “best efforts” to avoid bidding wars. Canada has enacted other disciplines to limit the use of incentives to attract investment. In eight of Canada’s 10 provinces, local governments are prohibited from giving incentives. Manitoba and Saskatchewan, the two provinces that still allow municipal incentives, have populations of just over two million and few major cities. One often-allowed exception is support for R&D, which is generous in Canada.

In Australia, bidding wars and poaching were also considered to be a problem for the states and territories. Five of the country’s six states (New South Wales, Victoria, Tasmania, South Australia and Western Australia), plus the Northern Territory and the Australian Capital Territory, reached an agreement in 2003 to end bidding wars among them and provide annual reports to each other on their investment attraction. The three-year agreement was renewed for another five years in 2006. A year later, Victoria’s treasurer noted that the signatories had saved “tens of millions of dollars” as a result of the agreement and that “…some jurisdictions [were] noticing a decrease in the number of companies seeking incentives to relocate from one State or Territory to another”. The agreement remains largely informal, however, with no monitoring or enforcement mechanisms.

BITs rarely, if ever, address the disciplining of investment incentives, because their purpose is to protect the investor rather than place limits on benefits that the investor might receive from a host government.

In the United States, investment incentive agreements (IIAs) are entered into by the Overseas Private Investment Corporation (OPIC) with host governments. There are over 150 IIAs between the United States and foreign governments. In general terms, these agreements provide for allowing the operation of OPIC’s programmes in foreign countries, recognition of OPIC’s rights and international arbitration between governments if disputes arise that cannot be settled by negotiation. The “investment incentive” in IIAs is the permission for OPIC to operate in the country, offering benefits to the investor such as political risk insurance, loans or equity investment.

The long-standing stall in the Doha Round trade negotiations has coincided with increased recourse to bilateral and regional FTAs to advance countries’ trade and investment agendas. Typically, these FTAs have imposed stronger disciplines on performance requirements than the WTO Agreement on TRIMS. The 2004 Australia–United States FTA goes beyond the TRIMS agreement by additionally banning export requirements, requirements or preferences for host country purchases, conditioning domestic sales on export performance, technology-transfer requirements and requirements that the investor be the exclusive supplier of its products to any market. However, the investment chapter also explicitly permits governments providing investment incentives to enforce “compliance with a requirement to locate production, supply a service, train or employ workers, construct or expand particular facilities, or carry out research and development, in its territory”.

The WTO has at least three agreements that touch in different ways on foreign investment: the Agreement on SCM, which abolishes export subsidies for most products (agriculture is a notable exception) and regulates the response to them; the GATS, which aims to improve access to markets for services, including by providing the right of establishment; and the TRIMS agreement, which limits the conditions that can be placed on investors, such as setting export targets.

As concerns disciplining incentives, the Agreement on SCM is the most important. It includes the provision that all WTO Members must give notice of subsidies to the WTO Secretariat. In principle, these submissions should cover all subsidies given within a Member country, at all levels of government, and include the amounts spent on such support. If adhered to, this would be a valuable transparency exercise. However, as of April 2013, a number of large WTO Members, including China, Egypt, Indonesia, Pakistan, Philippines, South Africa and Thailand, had not submitted their 2011 “new and full” notifications, and many Members submitted these notifications long after the 30 June 2011 deadline. Furthermore, the next round of new and full notifications is due by 30 June 2013, and as of April 2013, only four had been submitted.

On a different issue, Article 27 of the Agreement on SCM establishes “special and differential” rules for developing countries. An important consequence of this article is that it enables certain poorer countries to maintain incentives that are conditioned on export performance (such incentives are common in the context of export processing zones), which would otherwise run afoul of the prohibition on export subsidies. The list of WTO Members qualifying for this exemption includes all of those designated by the United Nations as “least developed” plus certain other Members with a per capita gross national product (GNP) under US$ 1,000 (calculated, pursuant to a WTO Ministerial Decision, in real terms over three consecutive years). The largest WTO Members on this list that have not graduated based on their GNP per capita are India and Indonesia. In addition, Article 27 provides a mechanism whereby other developing country Members can obtain extensions of the transition period to eliminate their export subsidies. Pursuant to special procedures adopted by the General Council in 2007, a total of 19 developing country Members received the last annual extension, for calendar year 2013, to be followed by a final two-year phase-out period (2014–2015).

This section is part of a broader report by the Global Agenda Council on Global Trade and FDI entitled, "Foreign Direct Investment as a Key Driver for Trade, Growth and Prosperity: The Case for a Multilateral Agreement on Investment."