Section 2: What are tax incentives for economic development?
Incentives are “tax breaks, cash grants, loans, or services” that target individual firms or industries with the objective of promoting job growth. Timothy Bartik, economist at W.E. Upjohn Institute and the foremost expert on the topic, classifies incentives into five broad types: property tax abatements, investment tax credits, job creation tax credits, R&D tax credits, and customized job training subsidies.
The primary objective of incentives is to promote local investment and hiring. Following the Great Recession, states have become hard-pressed to create jobs, protect jobs, and reduce unemployment rates. Many have turned to tax incentives to accomplish these goals. These pressures are only amplified by the pandemic-induced economic recession.
The primary target of incentives is the relocation and expansion of large firms. Incentives target large firms for their ability to both create many high-paying jobs and make significant capital investments.
Business incentives offered by state and local governments are substantial at around $50 billion annually for export-based industries (See Table 1). This amounts to about 1.4 percent of all industry value-added—how much that industry contributes to the GDP—and about 30 percent of the average state and local tax liability to businesses (Bartik 2017). These numbers are also comparable to state corporate income tax revenue, which was $45 billion in 2013.
Who gets tax incentives?
Incentives go to firms in virtually every industry both large and small, profitable and unprofitable, domestic and foreign. The largest incentive deals nationally are found in the manufacturing sector, oil and coal conglomerates, technology and entertainment companies, and banks and retail chains. Research suggests that states can get the best “return on investment” by targeting export-based industries. This is because export-based industries have the highest employment multiplier effect.
What is an Employment Multiplier effect?
The employment multiplier is one type of measure used to determine the impact a particular industry will have on the local economy upon its arrival or departure. It estimates the total jobs generated as a result of one job in that industry. According to Upjohn Institute’s estimates, high-tech sectors is believed to have highest multipliers, as high as 2.9. That is, for each new high-tech job in a city, 2.9 additional jobs are created (e.g., lawyers, teachers, nurses, waiters, hairdressers).
When studying tax incentives, economists like to distinguish between export-based industries and non-export industries. An export-based industry refers to a sector in which goods and services produced locally are “exported” (not necessarily to other countries but other areas) and consumed outside the local economy. For example, publishing (which falls under the “Information” classification) is generally considered export-based: almost all the work that goes towards producing Washington Post happens in D.C., but readership is spread across nation. In contrast, real estate is generally considered non-export-based since almost all the work done by realtors is consumed by D.C. residents. (See Table 2).
Research suggests that targeting export-based industries is the most effective incentive-granting strategy. These industries are more likely to create many jobs in the local economy. Giving incentives to non-export industries is likely to displace jobs at existing local non-export industries. Consider that a large retailer, like Walmart, offers to invest in the local economy and hire workers in return for state business incentives. Walmart is in the retail industry, which is a non-export sector with low multiplier effect. If Walmart enters the local market and creates 500 jobs but wipes out 600 jobs in local grocery businesses – the result will be a net decline in 100 jobs, a displacement effect. Given that tax incentives are large, it is likely that the retailer’s long-run tax contribution to the state will be less than the total tax contribution of all mom-and-pop shops, leading to a decline in the total tax revenue for the state.
The next section introduces the database used for the analysis. We use the database to identify which of District’s industries receive property tax abatements as well as the value of the assessment. Next, we compare District’s state and local business incentives with other jurisdictions.
About the data
The Panel Database of Incentives and Taxes (PDIT) estimates marginal business taxes and business incentives for 47 cities in 33 states across 45 industries for a 26 year span, between 1990 and 2015 (See Appendix B for the full list of cities). The 33 states compose more than 90 percent of U.S labor compensation. This constitutes the most comprehensive database on incentives and taxes to date, including all five major types of incentives: property tax abatements, customized job training subsidies, investment tax credits, job creation tax credits, and research and development tax credits.
State and local taxes and incentives are calculated for each year of the assumed 20 years of operation of the new facility, using:
- Balance sheet information
- Information on state and local tax rates
- Information on how incentives are determined based on firm characteristics
The PDIT database computes present the value of taxes and incentives as a percentage of the present value of the new facility’s value-added over the same 20 years using a discount rate of 12 percent. The database does not aim to include all incentives but rather include the incentives that are commonly used by medium to medium-large export-base firms. The database does not include incentives rarely used or those that only apply to a few very large firms.
Figure 1 shows that, for unweighted averages between 1990 and 2015, job creation tax credits and property tax abatements receive the highest amount of incentives as a percent of “value-added”. In 2015, tax incentives for new or expanding businesses in export-industries had a present value that averaged 1.42 percent of business-value added, which is about 30 percent of average state and local business taxes.
State and local business incentives offset a substantial percentage of taxes for businesses, although this amount represents only a small amount of total business expenses. For an average firm that receives incentives, tax incentives offset about 30 percent of the overall state and local business taxes that the firm would be obligated to pay otherwise – a substantial amount. However, in terms of the size of business activity, average business incentives subsidize only 3 percent of the firm’s wages for 20 years or on average about 5 percent of the value of a business’ productive activity (which is referred to as “value-added”). Another way to think about it is in terms of overall state and local budgets. Tax incentives comprise only about 3 to 5 percent of total state and local annual tax revenue of $ 1.1 trillion (for 2019).
 Typically, in the US context, firms employing less than 500 employees are considered small businesses. But often, large firms is used more loosely to refer to firms employing more than 100 employees.
 Industry targeting involves giving benefits to certain sectors that are not given to all sectors.
 The Council for Community and Economic Research (C2ER) maintains a state incentives database with a detailed description of each incentive program but does not provide information on the magnitude of incentives or jobs created/promised. The National Association of State Development Officers (NASDA) annual report also provides incentive summaries. Arguably the best available database on incentives is Good Jobs First’s Subsidy Tracker, which contains profiles of state incentives and incentive history.
However, the major limitation of these databases is the lack of different incentives’ magnitude that vary across state, industry, and year. This is where simulations are most useful. Using a simulation, it is possible to predict what would happen if incentives are modified. The previous attempt to simulate the magnitude of incentives was done by Peters and Fisher (2002), but its data extended only from 1990 to 1998. Cline, Phillips, and Neubig (2011) is probably the latest attempt but again is limited to only a few years and a few industries. Bartik (2017) collects data on the rules of each tax and incentive (by type) offered in a locality/state and predicts incentive magnitude, given estimated activity. The drawback of this rule-based approach is that certain assumptions need to be made about balance sheet of firms. More recently, Slattery (2019) and Slattery and Zidar (2020) proposed expenditure-based and narrative-based measures for state business incentives. These measures provide actual outlays for each program and credit year and allow the researcher to examine individual discretionary incentives. But these alternative measures are available only at the state-level, cannot differentiate by incentive type, and do not observe the contract or rules.
 The 12 percent discount rate is based on research on typical discount rates used by corporate executives (Poterba and Summers 1995). where . And, the measure included in PDIT is .
 The database uses Bureau of Economic Analysis’s (BEA) statistics on value added. The BEA definition is “The gross output of an industry or a sector less its intermediate inputs; the contribution of an industry or sector to gross domestic product (GDP). Value added by industry can also be measured as the sum of compensation of employees, taxes on production and imports less subsidies, and gross operating surplus.” For example, when a baker makes and sells a cake, the baker’s value added is the market price of the cake minus the input costs.
 Refer to Table 3 in Bartik (2017)
Feature photo by Ted Eytan (source).
D.C. Policy Center Fellows are independent writers, and we gladly encourage the expression of a variety of perspectives. The views of our Fellows, published here or elsewhere, do not reflect the views of the D.C. Policy Center.