A service of the


Volume 52, November/December 2017, Number 6 · pp. 379-380

Letter from America

America's Cities Compete for Amazon

Joseph Parilla

Joseph Parilla, Brookings Institution, Washington, DC, USA.

In October 2017 Amazon, the world's fourth largest company, received 238 proposals from North American cities aiming to become the site of its second corporate headquarters (or HQ2). HQ2 could eventually employ 50 000 workers at an average compensation of $100 000 – a truly transformative investment for any American city. In a uniquely public request for these bids, the company asked cities to highlight several local assets: the education and skills of their workforce, the quality of their transit and built environment, the strength of their schools and universities, and the livability of their communities. Amazon also requested each jurisdiction to list their tax incentive programs so that the company could understand how tax breaks would help defray the initial cost of its proposed $5 billion investment.

Critics looked askance at a $500 billion company demanding that taxpayers subsidize their entry, but Amazon's request should surprise no one. Obtaining employment that places them in the middle class continues to be a top priority for American households, and local and state leaders are under intense pressure to deliver more sustained and inclusive economic growth to their communities, especially in depressed parts of a country increasingly polarized geographically. Since capital has been mobile, U.S. cities and states have competed for it through tax incentives. In 1936 Durant, a small Mississippi town, issued the nation's first industrial revenue bond to incentivize Real Silk Hosiery Mills, and its 4000 knitting-machine operators, to relocate southward from Indianapolis. The Durant strategy soon extended to the rest of the industrializing South as a means to spur new demand for labor, particularly in manufacturing. Northern states eventually responded with their own incentive packages. And as demand for firms went global during the 1970s and 1980s, the economic development incentive regime spread nationwide.

Unlike major megadeals like Amazon, economic development incentives are typically structured as part of existing programs which are quietly administered by cities and states. Incentives may target firms in specific industries (e.g. innovation economy, manufacturing, real estate developers) or geographic locations within cities (e.g. distressed communities, downtowns, enterprise zones). Incentives can target different firm activities (e.g. research and development, infrastructure, job training, equipment investment). Incentives sometimes target firm attraction and other times firm retention. They may be delivered by a city or state government, a development corporation, or a special tax district. The incentive may be a one-time subsidy or may be tracked over many years, with varying degrees of restriction for whether the firm actually qualifies for the subsidy.

This diversity of incentives – by type, activities and provider – has made it difficult to track their total cost. While no consensus exists, recent estimates suggest the figure is substantial. In 2012 the New York Times estimated the country spends roughly $80 billion annually on incentives. The Upjohn Institute for Employment and Research found that total local and state incentives provided to firms in "export-base" industries cost $45 billion in 2015, or about 30% of the total average local and state business tax collections. Incredibly, this represents a tripling of incentive spending since 1990. Significant spending on incentives has invariably led to questions about whether they actually further job creation, income growth and general economic welfare.

Few economists conclude that the city and state incentives competition is a good use of taxpayer money. To avoid this inefficiency, the EU has utilized "state aid control" as a means to prevent its member states from outbidding each other for firms. However, despite agreement that bidding wars between communities are suboptimal, U.S. federal intervention in this manner appears unlikely in the near term. While President Trump has attacked domestic firms for placing investments outside America's borders, he has actually encouraged them to push U.S. states to compete for their jobs and capital. Unlike in the EU, there are no cohesion funds in the U.S. to help struggling regions enhance their economic prospects.

With American cities and states increasingly going it alone, the question then turns to which economic development policies seem to most effectively enhance local welfare. To this question, the unsatisfying answer comes from Sammis White, who writes that "all (economic development) tools work some of the time, none of the tools work all of the time, and a few tools can be said to work only under special circumstances."1

That lack of unanimity noted, the academic literature remains generally skeptical that providing firm-specific tax incentives is good public policy. Skeptical that tax incentives really matter that much to companies, since state, local and property taxes are a relatively small cost compared to labor. Skeptical that tax incentives actually lead to job creation, or if they do, that job creation can be targeted to specific populations or specific communities. And skeptical, even if incentives do deliver all these goals (and sometimes they do), that it can be done in a way that makes fiscal sense for local communities.

Moreover, the incentives governments do provide are often not well targeted to the firms that can offer the greatest local spillover benefits, such as those that pay high wages, conduct research and development, and export their products and services outside the local economy. These activities generate multiplier effects that ripple throughout the rest of the economy. South Carolina, for instance, used incentives to attract BMW's first American auto plant in 1992 and has since made significant investments in skills and applied research to cultivate an advanced manufacturing cluster. For every success like South Carolina, though, there are dozens of failed attempts to seed a cluster by betting big on a single firm.

Going forward, a challenge for any job creation strategy that relies on large firms is that major corporate expansions and relocations declined by 50% between 2000 and 2012. There simply are not enough Amazons to go around. Rather, 72% of net job growth from 2009 to 2014 came from firms that, on average, expanded from 10 to 30 employees. Spurring this kind of endogenous growth requires a long-term commitment to competiveness by making investments in transportation and housing, education, and increasingly applied research and development. Of course, building this platform requires significant up-front spending that many cash-strapped cities cannot afford and high-capacity coordination between public, private and civic actors that many cities cannot muster. By contrast, the cost of bestowing tax incentives on individual companies can be smoothed over time and only shows up in lost tax revenues.

At a certain point this is zero-sum and even self-defeating, as weakened government coffers cannot support the high-quality public goods that firms like Amazon demand. It is likely, therefore, that Amazon ends up choosing an already wealthy, innovation-rich and high-functioning region in North America's eastern half. Bids from places like Boston or Toronto are not leading with tax incentives, but rather with commitments that build on current assets: connecting potential development sites to regional transit, establishing computer science training partnerships with educational institutions and creating research centers in technologies like artificial intelligence. By contrast, New Jersey offered $7 billion in direct tax breaks, one of the largest packages in American history.

The former approach invests in an economic development platform that will benefit not only Amazon but also workers, communities and other local firms for decades. The latter approach provides vast public subsidies directly to an already well-resourced company – with uncertain public benefit. Whichever approach wins out, Amazon's decision will send an important signal as to how U.S. cities and states will compete for jobs and investment going forward.

  • 1 S.B. White: Perspectives on Economic Development Financing, in: S. White, Z. Kotval (eds.): Financing Economic Development in the 21st Century, Second Edition, Armonk, NY 2013, M.E. Sharpe, pp. 361-382.

Comments on this article

There are currently no comments on this article.

Your comment

Comments are moderated and will be posted if they are on-topic and not abusive.

Spam guard* Which letter is missing in the following word?