Creating and maintaining film clusters: Synthetic control method analysis of the enactment and repeal of US state film incentives

Abstract

The proliferation of US state incentives for film production led to numerous studies of whether the incentives affected production location, including case studies of key early incentive‐adopting states. The overall evidence on the efficacy of incentives is mixed. We carry out numerous additional case studies using the synthetic control method (SCM). A unique contribution of the paper is an examination of both the adoption and repeal of incentives. We also assess whether incentives have spillover effects between production and other activities within the film industry, whether spatial spillovers affect the SCM estimates, and whether the incentive effects are long lasting.

Estimating the economic effects of US state and local fiscal policy: A synthetic control method matching‐regression approach

“…for all j not equal to i and where the w ij are the SCM weights of the other states. We briefly summarize how the SCM method produces the w ij while relying on the presentations of Abadie et al (2010) and Rickman and Wang (2023). Letting Y it denote the economic outcome in state i at time t, Y can be decomposed into what the outcome would be in the absence of changes in fiscal variables (Y N it ) plus the effect on the outcome from fiscal variables (α it D it ).…”

“…Rickman & Wang, 2023).5 We do not adjust income for prices because state implicit price deflators are only available from 2008 and the use of the US consumer price index would not affect variable coefficients in a cross-sectional growth regression.6 https://www.census.gov/data/tables/time-series/demo/income-poverty/historical-poverty-people.html 7 State expenditures and taxes are from the Annual Survey of Government Finances: Urban Institute-http:// slfdqs.taxpolicycenter.org/pages.cfm.8 In 2017, across the nation, total own-source revenues equaled 19.2 percent of personal income, while current direct expenditures equaled 19.6 percent.9 We use the program package Synth in Stata to perform the SCM analysis http://web.stanford.edu/~jhain/ synthpage.html.10 The states are Arizona, Connecticut, Delaware, Georgia, Kansas, Nevada, New Jersey, North Carolina, Ohio, Oregon, Pennsylvania, Rhode Island, Washington, and Wisconsin.11 11 of the 14 states are reported byPartridge and Rickman (2006) in structural vector autoregression analysis as having population growth dominated by labor supply over the period 1970-1998; the three exceptions are Connecticut, Oregon, and Wisconsin.…”

“…Neighbors may not be sufficiently similar and outcome variables may violate the parallel trends assumption required for difference‐in‐differences analysis. Comparisons to neighbor states also may understate policy effects because the neighbors may be similarly influenced by the policies, leading to an understatement of the economic effects of taxes and expenditures (Di Stefano & Mellace, 2020; Rickman & Wang, 2023).…”