February 23rd, 2015
by Daniel J. Wilson - FRBSF Economic Letter, Federal Reserve Bank of San Francisco
State and local governments frequently offer tax incentives to attract businesses to locate in their area. Proponents view these incentives as a valuable tool to encourage economic development. Critics, on the other hand, argue either that incentives have little effect on business location decisions - and hence are wasteful giveaways - or that their benefits come at the expense of reduced economic activity in other areas. A key element in this debate is distinguishing what is best from a local versus a national perspective.
Late last year the electric carmaker Tesla chose Nevada as the site of its future "gigafactory," where it will produce all of the batteries to power its vehicles. The company expects the factory to cost $5 billion to build and to eventually employ around 6,000 people. Tesla cited Nevada's offer of low taxes as an important factor in its decision to locate there rather than in one of the several other states vying to host the factory. In addition to having no individual or corporate income tax, Nevada enacted a package of tax incentives - including refundable credits for job creation and investment, sales tax exemptions, and property tax abatements - specifically for Tesla to enhance the state's attractiveness (Damon 2014).
The Tesla-Nevada deal is just one of the latest high-profile business location decisions in which state and local tax policy purportedly played a critical role. Such deals tend to stir up renewed interest in the long-standing policy debate about the role of state and local taxes and incentives in affecting where jobs and economic activity are located. There are several key questions to consider in this debate: Does local tax policy actually have an important effect, or do businesses base their decisions of where to locate primarily on nontax factors? Do the potential benefits to the local area exceed the cost of the tax revenue that is lost? Are these policies a zero-sum game nationally, such that economic activity simply shifts from high-tax/low-incentive areas to low-tax/high-incentive areas? Finally, would the nation as a whole be better off if such tax competition were banned? This Economic Letter discusses some key concepts and related research to help guide the policy debate surrounding these questions. I start by describing the main tax policy tools used for local economic development.
Local policy tools
State and local governments, hereafter referred to as local governments, have myriad tools for promoting economic activity in their jurisdictions. Regulation, tax policy, direct subsidies, government services, and even government efficiency are all ways local governments create conditions that can attract investment. These tools are interrelated. For instance, more services can only be provided with higher tax revenue or greater efficiency. This relatedness is part of the reason local tax competition is so hotly debated: For a given level of government efficiency, higher subsidies or lower taxes typically require a trade-off of lower public services, at least in the short-run. In the long-run, proponents argue that lower tax rates or higher incentives should boost economic activity and enlarge the tax base so much that the cuts effectively pay for themselves. In this Letter, I focus on tax policy, given its particular importance to businesses. However, it's important to keep in mind that this is only one element of the overall package that a local government can use to attract jobs and investment.
Local governments' policy toolbox includes both taxes and incentives. Common taxes include individual and corporate income taxes, sales and excise taxes, and property taxes. On the incentive side, one prominent policy is income tax credits for such activities as job creation, investment, research and development, and film production. Other common types of incentives include expensing or accelerated depreciation of capital expenditures, sales tax exemptions for equipment purchases, and property tax abatements.
Important and unimportant distinctions
When considering the economic effects of tax policies, there are two important distinctions to keep in mind and one unimportant - at least economically unimportant - distinction that often obfuscates the real issues and can hinder the policy debate.
The first important distinction is between incentives that are nondiscretionary, meaning they are available to all qualifying businesses, and those that are discretionary, meaning they are provided at the discretion of a local economic development agency or government on a company-by-company basis, as in the case of Tesla. Standard economic theory suggests that taxes and incentives should not discriminate between businesses that are economically equivalent. That is, if two companies are expected to provide the same economic benefits to an area, the area's tax policy should treat them equally. Discretionary policies, on the other hand, can be used to tailor local policy so that a company pays lower taxes in proportion to the net benefit that company provides to the area. In practice, however, local governments may not be able to impartially and accurately assess the net benefits that different prospective companies will provide in the future. It is worth noting that discretionary taxes and incentives are generally banned in the European Union because of these concerns as well as concerns over excessive tax competition among jurisdictions within the EU.
This second concern relates to the other important distinction in the debate, the difference between the local effects andnational effects of local tax policy. Policies that are ideal from the view of a local area may be far from optimal from a national standpoint. This is because local policies can have repercussions - what economists call "externalities" - for other areas. For instance, Wilson (2009) found that when a state enacts a research and development tax credit, R&D expenditures in that state increase strongly but the combined R&D investments in other states falls by about the same amount. Thus, the credit has a zero-sum effect nationally. Chirinko and Wilson (2008) found a similar result for state investment tax credits and corporate tax rates.
The unimportant distinction is between what is called a tax and what is called an incentive or subsidy. In economic terms, a tax incentive is just a subsidy provided through the tax code as a negative or zero tax rate. Offering an exemption from local sales taxes on business equipment purchases is no different from setting the sales tax rate on business equipment purchases to zero. A property tax abatement as an incentive is just a reduction in or elimination of the property tax for businesses. A tax credit for a particular business activity is equivalent to a lower effective tax rate for businesses engaging in that activity. For all intents and purposes, any tax incentive can be reframed on the expenditure side of a government's budget as a direct subsidy to the eligible business. Though the economic equivalence of taxes, incentives, and subsidies might seem obvious and trivial, this point is often overlooked in the debate over tax incentives and it has a significant policy implication. Namely, national policy prescriptions relating to local incentives or subsidies - banning them, for example - are not likely to be effective if they do not treat local taxes in the same way.
These key distinctions lay the groundwork to address the questions at the beginning of this Letter. First, does local tax policy actually affect business decisions about the location of economic activity? Research to date has consistently found that it does. In addition to the findings regarding investment and R&D tax credits in Wilson (2009) and Chirinko and Wilson (2008), many studies have also found strong effects for corporate and individual income taxes, property taxes, job creation tax credits, film production tax credits, and other tax policies.
Second, do the benefits of a low-tax/high-incentive local policy exceed the cost from lost tax revenue? This is far more difficult to answer and the evidence is much less clear. Foregone tax revenue is inherently unobservable, and it is very difficult to estimate what economic activity and tax revenue would have been had the local government adopted different tax policies. The benefits are even harder to measure because they require policymakers to impartially estimate the monetary value of the jobs created directly or indirectly, as well as any less-quantifiable benefits to the local area.
The third and related question is, are local tax policies a zero-sum game? Empirical studies tend to focus on the effects of policies within a jurisdiction and not on whether they adversely affect other jurisdictions. However, when Chirinko and Wilson (2008) and Wilson (2009) addressed this question, they found the answer was roughly yes.
The last and perhaps most important question is, would the nation as a whole be better off if competition among local governments using taxes or incentives were banned? In thinking about this question, recall the economic equivalence between taxes and nondiscretionary incentives pointed out earlier. In both theory and practice, banning local incentives would be ineffective without also requiring tax rates and rules to be uniform across all local jurisdictions. Indeed, the European Union has moved in this direction by adopting tax harmonization rules and banning some forms of incentives, although substantial variation in tax policies across individual EU countries still remains.
Standard economic theory suggests that, if there are externalities on other places, tax policies set by local governments will not be optimal from a national perspective because local governments do not factor the negative effects on other areas into their decisionmaking. In other words, local government policymakers are expected to set tax and spending policies to maximize the welfare of their own constituents, without regard necessarily for the ramifications for people or businesses in other places. Thus, according to this theory, it would be better for the central government to set all tax policies. However, an opposing view comes from a classic public finance theory known as the Tiebout model (Tiebout 1956), which posits that people and businesses "vote with their feet" by moving to jurisdictions with the mix of taxes, spending, and regulation that best matches their preferences. This residential mobility provides a competitive pressure on local governments to be as efficient as possible in order to charge the lowest possible tax rate to finance public services. According to this theory, the resulting equilibrium allocation of taxes and services across the nation is optimal both because of this pressure to be efficient and because individuals sort into locations best matching their preferences. Forcing all jurisdictions to have the same tax policies would shut off this channel for allowing individuals to maximize their welfare. In sum, there is a trade-off between the benefits of tax harmonization - eliminating the negative impact one jurisdiction's tax policy has on other areas - and its costs - eliminating the positive effects of competition between different jurisdictions.
State and local tax incentives and their role in affecting business location is a contentious issue. Strong claims are often made for and against incentives. This Letter lays out some key concepts and distinctions that can inform the debate. In particular, it is important to distinguish between what policy appears best from a local perspective as opposed to from a national perspective. Optimal tax and incentive policy from a local standpoint would ideally draw upon unbiased analysis of the costs and benefits locally. Optimal policy from a national standpoint, however, will likely differ. It must weigh the benefits of local choice - individuals and businesses selecting jurisdictions that match their preferences - against the cost of how changes in one area might negatively affect competing jurisdictions.
Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System.
About the Author
Daniel J. Wilson is a research advisor in the Economic Research Department of the Federal Reserve Bank of San Francisco.
Chirinko, Robert S., and Daniel J. Wilson. 2008. "State Investment Tax Incentives: A Zero‐Sum Game?" Journal of Public Economics 92(12, December), pp. 2,362 - 2,384.
Damon, Anjeanette. 2014. "Inside Nevada's $1.25 Billion Tesla Tax Deal." Reno Gazette-Journal, September 16.
Tiebout, Charles M. 1956. "A Pure Theory of Local Expenditures." Journal of Political Economy 64(5, October), pp. 416 - 424.
Wilson, Daniel J. 2009. "Beggar Thy Neighbor? The In‐State, Out‐of‐State, and Aggregate Effects of R&D Tax Credits."Review of Economics and Statistics 91(2, May), pp. 431 - 436.