Production follows the rebate

Where a film shoots stopped being an artistic decision a long time ago. It is a financing decision, made against a spreadsheet of state and national rebates, and the state that built the industry spent a decade discovering it could lose on that spreadsheet.

California responded by making the number much larger. Lawmakers raised the annual cap on the film and television credit from three hundred and thirty million dollars to seven hundred and fifty, which is less a policy adjustment than an admission that the previous figure was not competitive.

The state's own accounting of the first year under the expanded program reports a hundred and seventy projects representing several billion dollars of direct production spending, which is the figure a government publishes when it wants the program renewed.

Independent reporting on production volumes describes American production stabilising after a period of sustained decline, with the caveat that stabilising is not the same as recovering and that the trend has not been tested through a downturn.

New York moved in the same direction and at a similar scale, which is the structural problem with the entire instrument. When every jurisdiction offers a rebate, the rebate stops being an advantage and becomes a cost of remaining in the business.

Economists have been consistently unenthusiastic about these programs, and their objection is not ideological. A subsidy that must be matched by every competitor transfers public money to producers without durably changing where anything is made, because the differential it was meant to create disappears.

The counterargument from the states is narrower and more honest than it is usually given credit for. Film production is a mobile industry with a deep local supply chain, and a crew base that disperses does not reassemble on request, which makes the loss harder to reverse than the gain was to buy.

That is genuinely different from subsidising a factory. The asset being protected is a labour market: electricians, grips, costumers and the businesses that rent to them, none of which can wait several idle years for policy to change.

The credits have also been reshaped around that argument. Newer program designs weight spending on wages, extend eligibility to categories that had been excluded, and add uplifts for shooting outside the primary production hub, all of which target the crew base rather than the balance sheet.

What no state has solved is the exit. A program that succeeds becomes permanent, because withdrawing it hands the work to whichever jurisdiction did not withdraw, and the political cost of visible departures is immediate while the fiscal cost is diffuse.

So the sensible reading of California's result is not that subsidies work. It is that they work while they are the largest, which is a description of an auction rather than an industrial policy, and auctions do not end because one bidder is satisfied.