The Zero-Sum Game: States Cannot Stimulate Their Economies by Cutting Taxes
March 2, 2010
Policymakers in a number of states are considering proposals to cut taxes and institute job-creation tax credits. While state policymakers are understandably eager to do something to improve their economies, such measures generally will not increase economic growth.
State balanced-budget requirements prevent states from stimulating their economies by cutting taxes. If a state cuts a tax, it generally has to make an offsetting cut to expenditures for a program or service in order to maintain balance. This spending cut is likely to reduce demand in the state just as much as the reduction in taxes may stimulate demand. It is at best a zero-sum game, where the gains in one area are offset by the losses in another.
Moreover, a tax cut designed to induce the hiring of additional private-sector workers may also cause the layoff of other workers in the public or private sector because of the loss in state or local revenue. When states cut spending, they lay off public employees, cancel contracts with private-sector vendors, and eliminate or lower payments to nonprofit organizations that provide direct services. Such steps lead to job losses in the private and nonprofit sectors, as well as the public sector. Thus, state-level tax cuts may shift employment from one sector (or business) to another, but the net effect is unlikely to be positive.
Because of this dynamic that occurs under a balanced budget requirement, a state cannot stimulate its economy during a fiscal crisis by cutting taxes -- either through a general tax cut or one targeted to specific sectors of the economy.
Cuts to Broad-Based Taxes
Though Arizona is facing a large deficit for fiscal year 2011 (equivalent to 25 percent of the budget) and is considering huge budget cuts, in January the state House approved a proposal to cut both the corporate and individual income tax rates as well as the business property tax. Similarly, governors in Delaware, Florida, South Carolina, and Rhode Island have proposed corporate income tax cuts as stimulus.
Such proposals are highly unlikely to work. When a state cuts a general tax such as the corporate or individual income tax, the impact on the state economy depends on what the business or the individual does with the money freed up by the tax cut.
- If a tax cut to a corporation increases its profits, it may distribute those profits as dividends to shareholders who live throughout the country; those funds will not necessarily create additional in-state demand.
- A corporation will not necessarily use the funds provided by the tax cut to make additional investments in the state in the short term. If there isn't additional demand for a business's good or service, the firm might keep the funds in reserve until demand picks up at a later time -- by which time it would not need any government inducement to expand.
- If a tax cut goes to a higher-income person, that person might save most of those extra dollars -- invest them in the stock market, for example -- so the tax cut would create little or no additional demand within the state.
Moreover, broad-based tax cuts in these circumstances can inflict damage on public investments seen by many economists as key avenues for both short- and long-term economic development -- including education, infrastructure, and other public investments. For example, Timothy Bartik, a widely respected economist at the W.E.Upjohn Institute for Employment Research, has proposed eight policies that would be especially cost-effective in promoting Michigan's economic development. Six of them would institute or expand public spending for job training, apprenticeships, or general education. Bartik specifically cites Michigan's budget gap as a threat to the state's economic progress. Rather than advocating tax cuts, he recommends broadening of the state's major tax bases and increasing the progressivity of its income tax, so that the government will be able to fund these programs.
Targeted Tax Cuts
Some proposals seek to overcome the uncertain effect of broad-based tax cuts by requiring the recipient to do something in order to receive the tax cut. For example, several states (including Alabama, Connecticut, Hawaii, Maryland, Massachusetts, Rhode Island, and Virginia) are considering tax credits that businesses can receive only if they hire additional workers. The credits typically would range from $1,500 to $3,000 for each new hire or new job created.
Job-creation tax credits raise a number of issues of cost and effectiveness. While 22 states have broad, statewide credits similar to those being proposed, and about another 12 states have narrower credits targeting specific industries or areas of the state, there is no evidence that these states' economies are doing better than other states' economies. The unemployment rate in states with and without credits is essentially the same.
- Research has shown that many companies receive job-creation tax credits for hiring employees that they would have hired anyway. Thus, the credit has little or no effect on these companies' hiring decisions. Many employers will be expanding employment or payroll anyway over the next year or two to meet increased demand for their goods or services as the economy strengthens. And even in a recession, many companies do expand and hire new workers.
- Particularly at the state level, it can be difficult to publicize a credit to a degree that changes employer behavior. Some employers, especially smaller ones, may learn about the credit only when they meet with their tax preparer at the end of the year -- and are told that they will get an unexpected tax credit for the person they had hired during the year. A Census survey regarding a federal jobs tax credit enacted in 1977 found that even among firms that knew about the credit, a quarter did not know whether they qualified.  In these cases the credit can have no affect on the firm's hiring behavior.
- Because some jobs would be created even without a credit, the revenue loss for each job that the credit would actually create is much larger than the amount of the credit. There is considerable dispute in the literature about the magnitude of this gap. If employers respond to jobs tax credits as if they were reductions in wages, as some researchers suggest, then some additional new jobs will be created. But for every additional job induced by the credit at least four others would have been created even if there had been no credit. In other words,at least 80 percent of the cost of the credit would finance hires that would have taken place anyway. 
For example, if a state offers a credit of $3,000 for each additional person hired, the cost for each additional job that the credit induces would be at least $15,000 ($3,000 plus four non-induced jobs at $3,000 apiece). Moreover, for reasons delineated below, a jobs tax credit packs nowhere near the "bang for the buck" that these relatively optimistic estimates suggest.
Federal Versus State Job-Creation Tax Incentives
The consequences of a state jobs credit would be quite different than for the federal jobs credits Congress is considering.
One difference is that the federal government does not necessarily have to make offsetting budget cuts or tax increases to pay for the credit. The federal government is running a large deficit and could borrow the funds to support the credit.
In addition, some of the ways a credit can benefit the national economy do not apply at the state level. In a recent study, the Congressional Budget Office explained four ways in which a reduction in employers' payroll taxes (the major option Congress is considering) could improve the economy -- although it also suggested that the overall effect would be relatively small. Some firms would reduce the prices of the goods and services they sell, some (a small number) would pass the tax savings on to employees in the form of higher wages, some would retain the tax savings as profits (which would in turn raise stock prices), and some "would use slightly more labor during a period when it was temporarily less expensive."
Unlike the federal government, which can consider these nationwide economic benefits, a state is unlikely to benefit from lower consumer prices or higher stock prices when multi-state companies take advantage of a state tax credit. Only a small portion of the benefit would stay in the state. In addition, a state generally would have to offset the cost of the credit with budget cuts in some other sector of its economy.
Thus while a federal credit might do some good for the national economy, a state credit is unlikely to benefit a state economy.
- A poorly designed credit can result in even greater revenue loss per induced job. If the credit turns out to be too small to induce incremental hiring, almost all of the revenue loss may pay for employees who would have been hired anyway. Similarly, if the credit's design is not appealing to employers, or if it is poorly publicized, it may induce little hiring
- Job-creation credits are difficult to design in a way that ensures that they create incremental jobs. The proposed state jobs credits provide a benefit for each new employee hired. That raises a number of issues. For example, employers could reduce the hours of their existing workers and then claim a credit for hiring new workers. They could even replace certain full-time workers with a larger number of part-time workers and thereby multiply their tax credits.
In addition, if the credit measures employment over a base period, it is difficult to define a base period that is fair to all regions and industries. And if the credit is available only to start-up businesses, a business could close one subsidiary and open another (or otherwise reconstitute itself as a "new company") in order to receive credits for its workers. There are many potential variations on this theme. All of these cost state revenue but yield no net gain in employment.
A credit's design has to be relatively complex to guard against these types of abuses, but employers are less likely to understand and respond to the incentive of a credit if it is highly complex.
- Many businesses wouldn't benefit from the credit because they have no taxable profits due to the recession. Businesses that have experienced losses over the past couple of years in the recession are unlikely to have any state tax liability next year or for the next few years, because those losses can be carried forward to eliminate or reduce taxes once the company again becomes profitable. A company in that situation would not get any immediate benefit from a job-creation tax credit (unless it is refundable). While the company may be able to carry the credit forward to a future year, an unusable credit would not be much of an incentive to hire new employees now, when a stimulus is needed.
- Even for profitable firms, the value of the credit is less than it appears because state taxes are deductible for federal tax purposes. Thus, a company that pays less state tax will pay more federal tax. The state loses the total amount of the revenue, but the value of the credit is shared between the firm and the federal government. This makes a modest-sized credit even less likely to call forth additional job creation. It also increases the likelihood that most of the state revenue loss would subsidize jobs that would have been created anyway.
Given states' balanced budget requirements, neither a broad-based tax cut nor a jobs credit can do much to increase overall economic activity in the state. If the state pays for a tax cut by reducing other spending, then the private-sector job induced by the credit replaces a job in the public sector, in a nonprofit organization that contracts with the state, or in another private-sector company that provides goods and services to the state. There is little or no net gain to the economy. Indeed, a state-level effort to stimulate the economy in this way can inadvertently create a fiscal drag on the state and national economy.
 Some states may contemplate paying for a stimulus tax cut with rainy day funds or other reserves, in which case it would be important to evaluate the relative value of using the funds to avoid demand-reducing expenditure cuts or to pay for a tax cut that may or may not create much employment growth. Another possibility would be to pay for the stimulus tax cut with an increase in a different tax, but that too could be a zero-sum game.
 Timothy J. Bartik, "What Should Michigan Be Doing to Promote Its Long Run Economic Development," Upjohn Institute Working Paper 09-160, November 2009. Note that Bartik is a proponent of targeted tax credits for employers who create jobs, but only at the federal level (See footnote 9).
 The most thorough study of a jobs tax credit program, involving extensive interviews of actual employers, was performed by Robert Tannenwald in his doctoral thesis. This research is summarized in Robert Tannenwald, "Are Wage and Training Subsidies Cost-Effective? -- Some Evidence from the New Jobs Tax Credit," New England Economic Review, September/October 1982, pp. 25-34.
 Federal Reserve Bank of San Francisco, FSRB Economic Letter, February 20, 2009.
 According to the latest unemployment statistics (December 2009), the average unemployment rate for states without jobs tax credits is 8.9 percent, nearly identical to the 9.0 percent rate in states with jobs tax credits. (Authors' calculations from www.bls.gov.) While such a comparison does not "prove" that the tax credits are ineffective, it does raise questions.
 See, for example Scott Schuh, Michael W. Klein, and Robert K. Triest, Job Creation, Job Destruction, and International Competition, The Upjohn Institute, Kalamazoo, MI, 2003.
 See Tannenwald, pp. 25-34.
 Congressional Budget Office, Policies for Increasing Economic Growth and Employment in 2010 and 2011, January 2010.
 See, for example, estimates in Timothy J. Bartik and John H. Bishop, Complementing Recovery Policies with a Jobs Creation Tax Credit, Economic Policy Institute, October 20, 2009. This report advocates for a federal jobs tax credit, but also points out that the strategy does not work at the state level.
 This is distinct from an intentional job-sharing policy, which some people have advocated as a way to reduce unemployment.