03/29/2024

Motion Picture Association Fails to Refute Damaging Film Tax Credit Study

The MPAA is up early this morning with a news release and short (~1,200 word) piece attempting to refute two studies by Professor Michael Thom of the University of Southern California (USC). The news release is full of adjectives and adverbs: “troubling,” “false,” “misleading,” “recklessly,” “otherwise respected,” “tarnishes the reputation,” “academic malpractice,” and “provocative.” Provocative might be the key one: the MPAA is pretty upset that a university in its backyard dared to research the effectiveness of tax subsidies to the movie production industry, let alone come up with a result that suggests they don’t work.

The studies, published by The American Review of Public Administration and American Politics Research, examine why states adopted or terminated film tax incentive programs and measures the effects of film tax credits in 40 states on employment and wages from 1998 to 2013. The authors found that sales tax waivers had no measurable effects; transferable tax credits had a small, sustained effect on employment but no effect on wages; and the most generous form of tax credit, refundable credits, had no employment effect and a temporary wage effect. Spending more on incentives had no lasting impact.
The MPAA obviously needed to respond to this study, which is very damaging to their efforts lobbying states for more film tax credits, and coming as about a dozen states have curtailed, suspended, or ended their programs. The author makes a solid effort to find problems with the Thom articles, but they fall short of a refutation:

The MPAA argues that data used by USC is not precise enough. The MPAA acknowledges that Professor Thom used the most relevant and timely data available, from the Bureau of Economic Analysis (BEA), using the North American Industry Classification System (NAICS) codes for motion picture and sound recording industries. The MPAA says this data is both overinclusive (including cashiers, for instance) and underinclusive (not including freelance workers, for instance). You could say that the data could be more refined and more detailed for any study on anything, but that doesn’t mean that a study using the best available data from the most trusted source for that data is invalid.

View Article