The Sidecar, Part II: Nobody just...walks through the door

I’ll admit when I’m wrong.

So, part one of this essay proposed that independent producers should be able to license dormant studio catalog under a standing structure rather than a bespoke deal. It ran into a veteran executive who took it apart by text message over about two hours, and the exchange was more useful than the draft had been, so this is the version that comes out the other side.

Two of his objections: Access to studio libraries is not constrained; any producer can walk into any library and try. Deals on viable material are not hard to paper.

Two of my counters: he granted that the number of buyers has contracted, and he granted that gatekeepers are not good predictors of what the market will support. I think my earlier framing was one of framing rather than substance: the prior argument I made led with a mechanism before I had fully clarified what the problem the mechanism was correcting. Even with his arguments, and if I were to concede those two points, the question remains: if the door is open, and the paperwork is trivial, and the material is available to anyone who asks, why does almost nobody walk through?

So here it is.

The Sidecar : Part II. Look, there is an economist named Roth that features prominently in this essay, so let’s just say, not only is this the business that we’ve chosen , but the sequel is better than the original. Let’s get into it.

Alvin Roth spent his career on a conviction that most of his profession found slightly embarrassing, which is that economics ought to fix things. Not describe them, not model them, not publish a paper proving certain assumptions…. Alvin Roth wanted to fix problems. Alvin Roth called the discipline, “” and he treated it as engineering: find a market that is visibly broken, work out which part of the mechanism failed, then build the replacement. Roth rebuilt the American medical residency match after it had spent decades collapsing into secret early offers. He rebuilt school assignment in Boston and New York, where the existing system was quietly punishing families who told the truth about their preferences.

And then there was the kidney problem.

In 2004, roughly 60,000 Americans were waiting for a kidney. Thousands of them had a willing living donor who was medically compatible with them. Trading was legal. Donating was legal. There was no statute, no gatekeeper, no price, and no barrier of any kind. Two people who could have solved each other's problem stood in the same country with the same intention, and the transaction did not happen, thousands of times over, for decades. Roth built a matching mechanism, the New England Program for Kidney Exchange became an operating clearinghouse, and paired donation chains started running that had been legally and medically available the entire time and had simply never been organized. Roth won the Nobel Prize for this in 2012.

What insight here is worth stealing? Well…this: Markets fail routinely without anybody forbidding anything. They fail from , when too few participants are in the room at once. They fail from congestion, when there is not enough time to evaluate the offers that exist.

I get that "the film library is like organ donation" is a weird analogy to reach for. Nobody is on a waiting list for a 1987 action picture. But think about the structure and the original claim: an open door is not a market. A market is a mechanism. The reason people win Nobel Prizes for designing them is the observation that markets do not assemble themselves out of available inventory and good intentions, which is the entire hidden premise of "if it were any good, someone would have done it by now."

Ronald Coase came at the same problem from the other side, and almost everyone misremembers his conclusion. The point of the Coase theorem was never that rights allocation is irrelevant. His actual argument was the opposite: the frictionless case is a thought experiment, , and because they are positive, the initial assignment of rights and the cost of moving them determine what actually gets made.

Oliver Williamson asked, if markets are so efficient, why does anybody work at a company? Why isn't every job freelance, every task bid out fresh each morning to whoever quotes the best price? Williamson won the 2009 Nobel for the answer, and the answer is that haggling costs money. Not metaphorically. Actually. Every negotiation consumes time, attention, legal review, and the risk of getting it wrong, and once you are dealing with the same counterparty repeatedly over something specialized, the cost of pricing each instance from scratch exceeds whatever you would have gained by pricing it sharply. So you stop. You build a standing arrangement, you accept slightly worse terms on any individual transaction, and you buy back the enormous cost of negotiating at all.

an open door is not a market. a market is a mechanism.

Williamson analyzed agreements that stood as a middle point between ‘employee arrangements’ and ‘complete freelancers’: franchises, long-term supply agreements, joint ventures, standing frameworks with defined counterparties. His finding was that this middle is where you end up whenever transactions recur and the assets are specific, because the spot market is too expensive to run and full integration is too expensive to build. A dormant library title sits…here. It is , with a recurring counterparty, which implies that a per instance negotiating cost swamps the expected value of trying to properly every individual opportunity.

Which is why the claim, "deals are easy" needs a footnote. Deals are easy conditional on somebody having already paid for the expensive part. The expensive part is easy the way the last mile of a marathon is easy, and anyone describing the sport that way has usually been handed the first twenty-five by an institution.

Oliver Hart and Bengt Holmström shared the 2016 Nobel prize for contract theory, and between them they account for the two failures at the center of this.

Hart's contribution is incomplete contracts. Because no contract can anticipate every contingency, what actually governs behavior is who holds , meaning who decides in the situations the document did not foresee. Whoever holds those rights captures the return on unanticipated value, and whoever does not hold them will rationally underinvest, because they cannot be confident of capturing what they build. A producer developing a dormant title holds no residual control. The studio can change strategy, pursue an internal version, sit on the file, or decline to respond, and all of that is squarely within its rights. So the producer invests nothing, the studio observes that nobody brings it strong packages on dormant titles, and it concludes accurately that there is no demand. Both parties are behaving correctly and the equilibrium is that nothing happens.

Holmström's shows that when performance is measured precisely on one dimension and loosely on another, effort migrates to the measurable one regardless of where the value is. A licensed out hit produces somebody else's press release. A licensed out embarrassment produces a meeting with your name on the document.

There is another version of the argument against the Sidecar: nobody has a monopoly on ideas. Good new material enters this business constantly, from nowhere, without permission, and if a dormant title were genuinely worth reviving somebody would have brought a compelling package and it would have gotten made. There is no hoarding, because there is nothing to hoard. Ideas are not scarce.

I agree with all of that but I do not think it touches the argument. Ideas are not the scarce input and I never claimed they were. The scarce input is a position. A property with a community already attached to it does not give you a better idea. . It lets you walk into a market with a partial answer to the only question anyone asks, which is who shows up, and that partial answer is what allows you to finance something. Every operator holding a portfolio understands this instinctively even when they cannot say it out loud: known quantities subsidize the unknown ones. The property with a built-in audience is not so much a payday as it is a hedge against other (necessary) risks.

So, the claim is not, “there is a monopoly on invention.” Instead, we’re acknowledging the makeup of the balance sheet, and the structural advantages that explain this makeup.

Consolidate the IP holders aggressively enough and you thin the market. Then cut development overhead hard enough on top of it, and you’re left with is enormous quantities of hedge held by entities with no remaining appetite for the swing, which is a portfolio that has stopped compounding, and now the entire industrial ecosystem can feel the weight of diminishing returns without being able to name the mechanism.

Overhead came out for a reason, and this is where my interlocutor ended the argument and where I think he was most right: . The back half of the distribution chain, which is where the profit that funded the front half came from, was dismantled by technology companies whose incentives were never aligned with anyone downstream of them and who were rewarded for that misalignment. When the money at the end of the chain stopped arriving, every cost upstream got squeezed to compensate, and the development class was the softest target in the building. That was not strategy. It was capitulation, and it worked exactly as capitulation usually works.

So the situation is not that studios are hoarding ideas. It is that the people who used to convert dormant positions into working ones were an overhead line item in a business that lost its back end and cut its overhead accordingly. The inventory did not go anywhere but its conversion capacity evaporated.

So…where was I wrong?

I proposed a mandate: an obligation on studios to develop or release a minimum tranche of dormant IP on a clock, dressed up in use-it-or-lose-it language borrowed from patent and land-option law. The argument from my mentor was this: market forces are more compelling than legislation. That is correct, nothing here can be imposed.

But while I’ve accepted his frame, there is a third category of self-regulation that can persist. Somebody won a Nobel Prize for proving it exists. Let’s get into it.

Let’s go into the economist behind what we’re going to discuss next. Elinor Ostrom graduated from UCLA and could not get a job. Not because there were no jobs. Because it was the 1950s and the people doing the hiring looked at a woman with a degree and saw two available futures, teaching and secretary. Elinor took a correspondence course in shorthand. She found work as an export clerk. Eventually a firm made her an assistant personnel manager, the first time that company had put a woman in any role that was not secretarial, and it was that job rather than the degree that made her think she might be capable of a doctorate.

Learning shorthand turned out to matter enormously. Decades later she was using it to take notes during face-to-face interviews with lobstermen and irrigators and forest villagers, which is how the fieldwork that overturned the standing theory of common-pool resources actually got recorded: in a stenographic system she had learned because the labor market of 1954 could not picture her doing anything else. The universe has a plan for us all. You just gotta believe…

The standing theory of the day, before Elinor, held that with exactly two escapes, privatize the resource or have the state regulate it. Ostrom found Swiss alpine grazing commons governed continuously since the thirteenth century, Japanese village forests, Spanish irrigation communities five hundred years old, Philippine water associations, Maine lobster grounds. All of them managing shared resources successfully through private, self-organized institutions that were neither markets nor governments. She then extracted what the durable ones had in common: clear boundaries around who is in and what is covered, rules matched to local conditions rather than imported, the people bound by the rules being the people who write them, monitors accountable to the participants, graduated sanctions rather than one terminal penalty, fast and cheap conflict resolution, and outside recognition of the group's right to organize at all.

In 1973 she and her husband Vincent founded the Workshop in Political Theory and Policy Analysis at Indiana, and she ran the thing as an experimental lab as well as a field operation. Students were handed shares in a simulated common resource and turned loose on it. When they were permitted to sit in a room and talk to one another face to face before acting, their returns more than doubled. No enforcement mechanism. No contract. No regulator. The entire gain came from people being able to say out loud, to each other, what they intended to do.

Read these design principles with a guild in mind.

Sure, citing thirteenth-century Swiss cheese farmers in a film finance essay is a choice. I’ve made peace with it. But the alpine commons and the Producers Guild share the same structure even if split apart by several centuries: this is an ecosystem of people dependent on a resource that none of them individually controls, who worked out that unconstrained competition among themselves degrades the thing they all need, before ultimately writing rules for each other instead of waiting for a king to do it. Ostrom's central finding was that this is not a compromise between capitalism and socialism. It is a distinct institutional form with better empirical outcomes than either in a specific class of situations, and Ostrom had decades of field data to prove it.

The remaining ideas to cite come from Milgrom and Wilson: it’s that markets of this kind do not emerge on their own. They are designed, deliberately. Paul Milgrom and Robert Wilson took the 2020 prize for auction theory and the applied design that came out of it, and the clearest demonstration is the FCC's broadcast incentive auction, which closed in 2017.

The problem was structurally identical to this one. An valuable asset, the television spectrum, sat with incumbent holders who were not putting it to its highest use, and nothing prevented them from selling. It did not trade, because no mechanism existed to discover who would give up what at what price, and because no individual broadcaster could evaluate the decision without knowing what the others would do. Milgrom's team built a two-sided auction, buying spectrum back from broadcasters and a forward auction selling it to carriers, linked so the clearing target moved with the prices. It produced roughly $19.8 billion in gross proceeds, of which about $10.05 billion went to the broadcasters who chose to sell, and it repurposed 84 MHz that had been inert for decades.

Nobody legislated a spectrum quota or seized anything. The incumbents participated voluntarily, got paid, and only then found out what their inventory was worth.

The broadcasters did not know what they were holding. They had owned it for decades, they were sophisticated operators with every incentive to know, and they did not know, because there was no mechanism generating a price. In the absence of a price, everyone substitutes judgment and calls it a valuation.

in the absence of a price, everyone substitutes judgment and calls it a valuation.

So why hasn't a studio built this already? The answer is not incompetence.

Daron Acemoglu, Simon Johnson, and James Robinson took the Nobel Prize for Economics in 2024 studying why institutions generate or suppress prosperity, and their central mechanism is neither corruption nor stupidity. Incumbents rationally block innovations threatening existing rents, including innovations that would expand the total. They call it .

This, then, is the structural reason for why sidecars cannot sufficiently originate on the studio side. Not because studios are the antagonists; every version of this argument that reads as a grievance against them is the wrong argument. A studio holding a library is not being small-minded when it declines to build the instrument, it is protecting an option. The option is real, the executive protecting it is doing the job correctly.

The Producers Guild of America already operates the one piece of machinery this requires. The PGA runs a credit arbitration and determination system: a private panel process, applying published standards, issuing determinations that the studios and the Academy accept. That is not a proposal or a white paper. It is a functioning private adjudication regime, inside this industry, binding parties who never signed a statute, and it has been running for years. It is the most underrated institutional asset in the American film business and almost nobody outside the Guild thinks about it.

The Guild also proved the second half. The Producers Mark, the p.g.a. designation, went from proposal to acceptance by the major studios between 2001 and 2013. Twelve years, and the mark produces no revenue, by design. That is the discouraging number and I would rather put it on the table myself, but it is worth understanding why it produces nothing, because the reason is a statute and the statute is useful. can be cancelled if its owner also produces the goods it certifies. That is Lanham Act section 1064(5).

Which means the law has already answered the question of who is allowed to hold this, and the answer excludes almost everybody. It cannot be a fund, an operator, or any producer with a slate, because anyone who benefits from the standard cannot credibly hold it. It has to be a body that convenes producers without being one. There is exactly one of those, it already runs the determination system.

So, the proposal, stated where it can be argued with.

A Guild-administered registry and standard for dormant-title activation, voluntary on both sides, with four components. Objective eligibility, so titles enter by published test rather than by anyone's selection, which removes the individual executive from the decision and therefore eliminates the Holmström asymmetry outright. Qualification standards for participating producers, funded by participants rather than by the studio, which answers the stewardship objection and with a single rule. A standard instrument with a title schedule, so the transaction cost Williamson identified is paid once at the class level instead of repeatedly at the instance level. And determination through the Guild's existing arbitration machinery, which is the component that already exists and the reason to build this here rather than anywhere else.

No title enters unless the holder puts it in. No studio is obligated to participate and none can be made to. Enforcement is exclusion from the registry, which is what the alpine commons used, what every functioning guild has always used.

And this is where I would revise Part One's framing rather than its instinct.

I kept describing a constrained market as though the constraint were a concession I was asking someone to make. But framing it as a concession is a mistake, and we can look to the NFL for analogies. The NFL runs a salary cap and a salary floor, a reverse order draft, revenue sharing, and pooled broadcast rights that required the Sports Broadcasting Act to survive antitrust, and it is the most competitive league in American sport.

Merit is downstream of an engineered option count. The undrafted player from a school nobody has heard of gets a look because thirty-two clubs times fifty-three roster spots plus a mandatory spending floor guarantees the chances exist. Remove the floor and the draft and you do not get purer merit, you get three superclubs and a thousand players nobody sees. The design is good enough to disappear.

What survived the argument is narrower than what went into it and considerably harder to dismiss. The market failure is real, but it is a mechanism failure rather than a gatekeeping failure. The gate stays, and the stewardship standard gets stricter rather than looser. Nothing is compelled and nothing is redistributed.

Still, an open door with no mechanism behind it is not a market.

The one body in this industry with the standing, the neutrality, and the working determination machinery to hold it is the Producers Guild.

I would rather be argued with than agreed with, because the picture is a bit clearer to see now.