Jia Wang, one of OK Policy’s 2013-14 Research Fellows , is a fourth year PhD student in Economics at the University of Oklahoma. Her research interest lies primarily in public economics, especially government expenditure/taxation policy and economic devel

Oklahoma Policy Institute's picture
Printer-friendly versionPrinter-friendly version

Policymakers and local government officials are constantly under pressure to stimulate economic growth and create jobs. The provision of economic development incentives (used interchangeably as business incentives or subsidies below) is often touted to achieve the above ends through increasing capital. Higher earnings and tax revenues are also expected from new capital formation.

Economic development incentives refer to direct and indirect government subsidies to business that are not inherently part of a generally accepted tax structure. They include tax instruments like property tax abatements, tax increment financing, sales tax exemptions and credits and non-tax incentives such as business grants and loans. In all cases, the firm is the initial recipient of the incentive.

Proponents of incentives have invariably argued that government inducements will lead to more private investment/capital, hence more jobs and higher payroll. Altogether, they are expected to contribute to higher revenue and enhanced public services or decreased tax rates, which may further stimulate private investment/capital. This perspective basically describes a virtuous economic cycle induced by government incentives.

However, those who tout the benefits of incentive programs typically ignore or underestimate their potential cost.  Incentives are likely to cause governments to forgo some tax revenue. This is compensated for by either fewer public services or increased tax rates, which could lead to less private investment/capital either directly or indirectly through less jobs and payroll. In fact, research on the effects of state and local public services on economic development indicates that certain public goods like education, highways, public health and safety are positively correlated with growth. Therefore, it is of paramount importance to investigate if states are giving up  these critical public services while offering incentives.

My study directly addresses the question if business subsidies may crowd out spending that would otherwise go to public goods such as education and infrastructure, a largely unexplored question but one of practical importance to policymakers. More specifically, my study estimates the effect that incentives expenditure has on public expenditure using panel data from the states [the full draft study is available here]

Data on incentives from Good Jobs First’s subsidy tracker has been used as a measure for the activeness or intensity of subsidy use by state governments. My model essentially asks if variation in subsidy expenditure explains variation in public expenditure after other determinants of public expenditure has been held constant. My control variables for government expenditures include grants and state income, which measure resources available to state and local governments; population and population density, which capture potential congestion effects or economies of scale in the provision of public services; as well as percentage of young and elderly, which measure the influence of age composition on demand for public services. In addition, I control for time-invariant, state-specific unobserved factors and common shocks associated with national business cycles.

Preliminary results indicate that business incentives are negatively correlated with total expenditures on public goods, including total education, elementary education, police protection, highway, own hospital and parks and recreation. Incentives expenditure is also associated with decreased expenditures on water utility, sewerage, solid waste management, employee retirement expenditure and benefits. Nevertheless, I did not find any significant negative effect on higher education spending, which may be a result of different funding sources.

The decreases in productive public goods expenditures are worrisome as they have been shown to be critical for local economic growth in the long run. At the same time, I observed that increased business incentives are associated with an increase in cash assistance to categorical welfare programs. This seems to echo some concerns in the media that jobs created by incentive spending are likely to be the low wage jobs, which put more people on welfare programs.

Overall, my findings are in line with studies that conclude that subsidies are not cost-effective. A most recent study by Elizabeth Patrick on the impact of million dollar plants on revenues and expenditures has found little evidence that these million dollar facilities have induced virtuous cycle of economic development. Increases in public expenditure at the county level were financed by increased debt rather than enhanced revenue.  As a work in progress, I’m working on alternative model specification and estimation methods. It is yet to be seen whether my results are robust to those changes.

In times of fiscal stress, it is of paramount importance to understand what states are giving up when offering incentives, especially given the close link between public services and economic growth. Findings of this paper substantiate the long-standing criticism against incentives and should serve as additional warning for policymakers that would like to use incentive policy to stimulate economic growth in the long run.

News Source : Jia Wang, one of OK Policy’s 2013-14 Research Fellows , is a fourth year PhD student in Economics at the University of Oklahoma. Her research interest lies primarily in public economics, especially government expenditure/taxation policy and economic development incentive programs.

Copy this html code to your website/blog to embed this press release.


Post new comment

5 + 5 =

To prevent automated spam submissions leave this field empty.
Page execution time was 555.16 ms.

Memory usage:

Memory used at: devel_init()=2.13 MB, devel_shutdown()=22.44 MB.