The case for film subsidies (and other goodies)
Last week, Michael Kinsley published a jeremiad against film subsidies in the LA Times. Two of my fave Economist-oids, Ryan Avent and Will Wilkinson, follow up, Wilkinson with an endorsement of Kinsley’s piece, Avent with a more nuanced but ultimately very radical comment.
Film subsidies and other state and local programs intended to promote economic and cultural activity are sometimes smart policy and sometimes corrupt boondoggles. I certainly don’t wish to argue that they are always and everywhere good. But Kinsley argues that they are always and everywhere bad, via arguments that are as compelling as they are false. Let’s try to understand the economics a bit.
The most widely quoted, and most plainly wrong, bit of Kinsley’s piece is this:
New Mexico under [former Governor] Richardson was a pioneer in this field. In 2002, it began offering a credit of 15% — later raised to 25% — toward the cost of making a movie in New Mexico… Now, 42 states have followed its lead… In less than a decade, the absurd notion of welfare for movie producers has evolved from the kind of weird thing they do in France to an unshakable American tradition… Richardson says that the film and TV subsidy has brought “nearly $4 billion into our economy over eight years” and has created 10,000 jobs. By “our,” he means New Mexico. He says every state should emulate this success.
But of course every state cannot do that because it essentially is a “beggar thy neighbor” strategy.
Ryan Avent is not ultimately willing to endorse film subsidies. But he is too good an economist to let this go by. He writes
A subsidy allows a business to cut prices and artificially raise demand. Given generous enough subsidies, many more movies would be made, and each state could, potentially, have a thriving film industry.
To put the same point differently, film subsidies reduce the cost of production and thereby increase risk-adjusted expected returns to investors. In a world thick with aspiring directors and clever screenplays, there are always hundreds of potential films getting ranked, accepted, but mostly rejected by investors willing to support film production. At the margin, there are films — perhaps quite a lot of films, it’s an empirical question — that investors would deem almost but not quite worth funding in the absence of subsidies. These films get funded and produced when governments sweeten the pot.
Film subsidies are not entirely or even predominantly a “beggar thy neighbor” strategy. They are certainly not, as Wilkinson asserts, a zero-sum game. In many countries, a large fraction of production depends upon state subsidies, and many films would not have been produced without them. The elasticity of film production to subsidy is far from zero.
Still, this is “welfare for movie producers”, as Kinsley puts it, right? Avent describes the excess demand as “artificial”. To which I say, huh?
Kinsley’s “France” is not nearly communistic enough to discredit this pernicious practice. If we really want to drive home the idea that film subsidies are a booger floating in the soup of red-blooded capitalism, we should associate them with that most Bolshevik of all institutions, the, um, suburban American shopping mall.
The economics of a well-designed film subsidy and the economics of suburban shopping malls are identical. State governments offer film subsidies on the theory that film-making within the state will generate ancillary economic activity that will more than offset the cost of the subsidy. Suburban shopping mall developers offer what are effectively rent subsidies to stores they expect to generate extra traffic and sales for the shopping mall. Many of the “anchor stores” — the big, national-brand department stores — at your local mall pay no rent at all, despite occupying vast territories of prime space for which their specialty store neighbors pay dearly. This phenomenon has been carefully studied. Gould, Pashigian, and Prendergast write
[T]he differential contracts offered to the anchor and nonanchor stores appear to not only offset some of the externalities generated by the anchor, but do so in an efficient fashion, at least on the dimension of total sales and rent in the mall. If this were not the case, the result would likely be a misallocation of space: a failure to internalize the benefits of the anchor stores would imply too little space allocated to anchors, because anchors themselves would not consider the external benefits their presence has on the other stores when deciding how much space to lease.
The arrangement that has evolved among private parties via consensual, contractual negotiation is that shopping mall developers effectively tax non-anchor stores with high rents in order to subsidize anchor stores with mostly free rents. Far from “artificial”, if developers did not do this there would be a deadweight cost. If rents were held homogenous within shopping malls, there would be a lot fewer anchor stores, which would deprive smaller stores of the foot-traffic and sales those anchors generate, which would then deprive shopping malls of a lot of potential rent.
Still, the Macy’s, Sears, and Nieman Marcuses of the world have to live somewhere, right? And it’s got to rankle shopping mall developers — you know it does — that a substantial fraction of their hard-built space is given away for trivial or even zero rent. Suppose that all of America’s shopping mall magnates gathered in a smoke-filled room and decided to ban the practice of subsidizing rent to anchor stores. What would we call that? It turns out we have names: “price fixing”, “cartel”, “conspiracy in restraint of trade”.
If shopping mall developers could pull off such a scheme — or really if they could have pulled off such a scheme years ago — they might narrowly have benefited. There would have been fewer anchor stores and therefore fewer shopping malls, but the loss of scale might have been offset by developers ability to, um, extract rents from anchor chains, leading to increased profitability. But that extra profitability would have been an ordinary monopoly rent, of the sort we typically condemn and even criminalize, wherein higher prices are extracted by virtue of a monopolist’s power to enforce underprovision of goods. We’d have the FTC or the Department of Justice all over their asses if shopping mall developers tried to pull something like that.
Similarly, if it is true that film production generates positive externalities for local and state economies, it still might be true that having local governments band together and refuse to provide film subsidies would lead to greater overall tax receipts. The reduction of taxable economic activity due to cartelized subsidy refusal could be offset by the savings realized from withholding the subsidy. If this is so, then state and local governments (in aggregate) profit only by forcing a reduction of activity below the level economists would ordinarily call “efficient”. By not permitting filmmakers to recover some share of the value of the positive externalities they generate, we force a lot of them to take their ball and go home, leaving us all poorer in aggregate.
I don’t think it’s likely, either with respect to shopping malls or with respect to films, that “local governments” would in fact benefit by forming a cartel. In both industries, I think the externalities are real, and despite some “beggar thy neighbor” competition — between shopping malls as between states — these “governments” come out ahead by rebating some of the external benefit back to those who create it and getting a smaller piece of a bigger pie.
Kinsley, in his column, implicitly recognizes that he is calling for a cartel of state governments. “Government, in order to work, must be a monopoly,” he asserts, without explanation or justification. Governments do maintain certain monopolies within their territories, but must the fifty state governments band together and become an uber-monopoly as well? Isn’t much of the justification for federalism the notion that multiple governments experiment and compete, generating creativity and dynamism that wouldn’t exist in a monopoly public sector? How does the fact that “the landlord” is a government alter the economic logic of efficient contracting within competitive shopping malls, which is precisely the same logic that justifies film subsidy?
Distributional issues arise when subsidizing externalities. The local gag store is more sympathetic than a national chain, but the gag store ends up paying the anchor store’s rent. Local taxpayers are more sympathetic than Hollywood studios, yet local taxpayers end up funding Hollywood studio returns. With respect to private sector shopping malls, there’s little we can do about these distributional concerns. In the local government sector, however, it is perfectly legitimate to discriminate by, for example, offering the subsidies only to local filmmakers, however defined. That choice is full of trade-offs: Restricting subsidies to local filmmakers arguably implies that, from a global perspective, some less valuable films get produced in preference to more valuable films. The restriction also reduces the power of the subsidy to generate activity, both in absolute and bang-for-the-buck terms, as there are fewer films locally than globally. But local films may contribute to local culture in ways that taxpayers value, local subsidies go to people more likely to respend money in state (increasing tax recoveries), and distributional concerns are legitimate and serious. These are tradeoffs for taxpayers and their representatives to make based on particularities.
The most serious case against film subsidies, emphasized both by Kinsley and Wilkinson, is the public choice argument. Wilkinson claims that “the film and TV incentives racket is a hotbed of corruption.” That may or may not be a fair characterization, but the point is well taken. The economic logic behind subsidies is iron-clad, given activities that generate net positive externalities whose value is known to be more than the cost of the subsidy. But the externalities of future projects can only be estimated, and estimates by potential recipients of subsidies are rationally overoptimistic. If politicians, perhaps blinded by “personal friendships” with campaign contributors, fail to form independent and conservative estimates of public benefits, then they may offer excessive subsidies, which destroy value while transferring funds from taxpayers to the subsidized.
This is a very serious issue. But it needn’t be insurmountable. Subsidies can be and sometimes are attached to contractual obligations to generate promised activity. Localities do this routinely, and sometimes sue to recover the subsidies if public benefits fail to appear. There is always a lot of uncertainty surrounding indirect public benefits that may result from various activities. But, as Matt Yglesias reminds us:
Life is full of situations that demand you to make decisions under conditions of uncertainty. In almost all cases, the right thing to do is to try your best, not to simply give up.
It is conceivable that we are simply unable to organize governments capable of resisting corrupt inducements, and therefore our best option in a bad world is simply to forego all subsidies. I’m pretty cynical about government, but I don’t think we’re there yet. If that is your position, however, at least get the economics right. Don’t imagine that a blanket prohibition of subsidies, to film or any other activity with positive externalities, is “efficient”. On the contrary, a blanket prohibition is guaranteed to result in underprovision, and likely to result in lower total tax receipts than an optimal subsidy regime. You can argue that in a third-best world, we’re better off accepting very large deadweight costs than the corruption that attends differential taxation schemes. But there is nothing efficient in either of those choices.
In the real world, all successful governments subsidize activities with putative positive externalities. All unsuccessful governments do so as well. In the history of the world, I doubt there ever was a government which has not differentially taxed or subsidized in order to promote allegedly valuable activity. Under the circumstances, I think we should take Yglesias’ advice and try our best to do subsidy well. If we do it right, both theory and evidence suggest that subsidy can do a lot of good. If we do it poorly, we’ll destroy a lot of value and generate corruption. Not doing it at all, in a practical sense, is not an option. I think the case for positive externalities associated with film production is pretty strong. If so, then the right thing to do is to keep the subsidies but to administer them as wisely and as noncorruptly as we can. Distributional concerns matter a lot to me, so my preference as a taxpayer is to support subsidies that discriminate in favor of local and/or “independent” filmmaking (although that invites its own corruption in the form of “definition arbitrage”). More generally, a world without state subsidy is not a world to strive for. It would be as much a libertarian paradise as a ghost shopping mall.
FD: My wife is a film student, aspiring to become an aspiring filmmaker. Perhaps that colors my view of film subsidies.
- 5-March-2011, 11:15 a.m. EST: Changed “Kinsey” to “Kinsley” throughout. Thanks to Leigh Caldwell for pointing out the error, and my apologies to Michael Kinsley for making it.