The billions of dollars funneled from New York’s treasury to movie and TV producers had no statistically significant impact on the industry’s employment in the Empire State through 2017, according to a new multi-state study of such tax incentives.

Study author Michael Thom, an associate professor at the University of Southern California’s Price School of Public Policy, focused on the effectiveness of “motion picture incentives,” or MPIs. In addition to state tax credits, the incentives include state spending on location assistance, advertising, preferential regulatory treatment, sales and occupancy tax waivers and building incentives for production infrastructure.

Thom previously issued a 2016 national study of MPIs and their general impact on labor and economic conditions. Using a different method of economic analysis, his latest study drills down into a different approach, focusing on five states that account for the bulk of all MPI expenditures: New York, Louisiana, Georgia, Connecticut sand Massachusetts.

The technical twist in Thom’s new study, published in the September issue of State and Local Government Review, is a reliance on “interrupted time series analysis,” which separates employment and wage data observations into time periods before and after the film credits were in effect. That method makes the findings especially credible—since it recognizes that New York, in particular, had significant movie and TV industry employment even before the 2004 creation of the Empire State Film Production Credit. At $420 million a year, or a total of well over $5 billion since its inception, the credit dwarves other economic development giveaways to private companies, including Governor Andrew Cuomo’s abortive proposal for a jumbo-size package to lure Amazon’s second headquarters to Queens.

Thom’s main finding:

Turning … to the question of how MPI programs impacted employment in the five high-expenditure states, the results show the answer is “not much.”

His employment impact analysis was further divided into three pieces: immediate and permanent impacts, subsequent annual effects adding or subtracting from the immediate and permanent impact and the effect of corporate tax expenditures (such as New York’s film credit) in isolation. He found that the only state in which an MPI program had a significant immediate and permanent impact on employment and wages was Connecticut—but that change was attributed entirely to the non-tax components of that state’s MPI program. Moreover, at about 500 total employees, Connecticut’s total motion picture labor force was so small before the program’s implementation that the proportional impact of adding just 50 workers was bound to be exaggerated, he said.

As for impacts over time, Louisiana was the only state in which Thom measured a statistically significant MPI program effect over time. But that result “should also be interpreted with caution,” he wrote, because of data limitations unique to Louisiana.

New York’s movie and TV production subsidies flow through the Empire State Film Production Credit, which is by far the nation’s largest such tax expenditure —but that, too, was found to have no significant relationship for employment, according to Thom. In fact, he said, his economic model indicates that as employment industry rose in New York, wage gains in the industry actually dropped.

Thom noted that his findings on the ineffectiveness of movie and TV production incentives “further reinforce the existing literature’s general conclusion that, as an economic development strategy, targeted incentive programs that carry large tax expenditures fail to encourage meaningful job creation.”

The penultimate paragraphs of the study bear repeating:

This study’s results should encourage policy makers to exercise caution before pursuing targeted economic development programs, especially those that incent creative industries. When the output is intellectual property, production can occur anywhere, and the jobs created as a result of incentives—if any—are far from long term. In a competitive market, the only hope to retain those jobs is to increase tax and other incentives, the very same “race to the top” observed when state and local governments try to outbid each other for the latest purported engine of economic growth (e.g., Tesla, Foxconn, Amazon, or a professional sports franchise). That inevitably creates a bubble in which policy makers have overinvested in a program relative to the program’s ability to yield a return on investment …

This study … does not provide a direct assessment of MPI cost-effectiveness, yet a separate analysis may not be required. Comparing the tax expenditures reported in [the analysis] against the scarcity of employment gains attributable to that investment suggests MPI programs are anything but a prudent use of taxpayer dollars. This study also does not thoroughly investigate the relationship between MPI programs and industry wages. However, some evidence points toward a trade-off between employment gains and wage gains, particularly in New York and Louisiana. [emphasis added]

Of course, the industry and its friends in the state’s economic development agency disagree. One recent state analysis claims the credit “generated” $8 billion in spending that directly and indirectly “resulted in 85,835 jobs” in 2017 and 2018. However, it assumes (preposterously) that the industry would have created no jobs without the credit, and it effectively counts as two “jobs” a single individual who was temporarily employed by the industry in each year.

The Empire State Film Production Credit was due to expire at the end of 2022 before Cuomo snuck a two-year extension into the final version of the FY 2020 budget last April.