Contracts Can Be Weird

Particular contract provisions sometimes seem odd. However, it is important to understand the underlying economics before making policy

You are reading Economic Forces, a free weekly newsletter on economics, especially price theory, without the politics. Economic Forces arrives weekly in the inboxes of over 10,000 subscribers. You can support our newsletter by sharing this free post or becoming a paid subscriber:

---

Firms often make investments that increase the marginal product of other inputs in their production process. An interesting question is how these additional benefits are distributed. In a typical supply and demand analysis, if a firm’s investment increases the marginal product of labor, for example, then the demand of labor will increase and the firm will employ more people at wages.

Sometimes the real world doesn’t work like this textbook example. For example, labor markets often have explicit or implicit contracts. An explicit contract might say that workers agree to work at a particular wage for the next three years. The wage might be constant over that time or the wage might increase each year of the contract. An implicit contract might be that a firm hires the worker at a given wage and the worker has something like an annual review. It is understood that if the worker is going to get a raise, it will come at the time of the annual review.

Other issues can complicate the analysis as well. There might be uncertainty about the possible benefits of the firm’s investment. Maybe the investment will make other inputs more productive and maybe it won’t. Even if the investment does make other inputs more productive, there might be uncertainty about the magnitude in which the inputs become more productive. This matters for the firm because most investments are irreversible in the sense that the costs of investment are often sunk (i.e., you can’t get them back). If the input in question is labor, workers will want to share in the benefits of an investment that boosts their marginal product of labor. They are unlikely to want to shoulder some of the costs of an investment that doesn’t make them more productive.

Some investments are specific to the inputs. If the firm trains a particular worker, it might make that worker more productive. If the worker initially agreed to a wage consistent with his or her original marginal product, this could potentially lead to a hold-up problem. After being trained, the marginal revenue product of the worker is higher than his or her wage. The worker could commit to going to work for a competitor at a slighter higher wage if his or her current employer doesn’t increase the wage to the marginal revenue product. At that point, the cost of investment is sunk. A firm that anticipates this problem might just choose not to provide the training in the first place.

Finally, the value of a particular investment might be difficult to identify if there aren’t close substitutes for the input in the market.

To make things more concrete, let’s consider a particular example. Suppose a production company develops a television show. The company needs someone to distribute the show. They shop the show around to television networks and streaming services. A network or streaming service might agree to buy the show for a certain number of episodes, maybe a full season or even multiple seasons. The purchase of the show is an investment in and of itself, but the network will also make an investment in the show by promoting the show with billboards and commercials.

This type of agreement has all of the real-world characteristics I previously mentioned. The investment is specific to the show. The combination of the advertising for the show and the quality of the show itself could lead to more revenue for the network than if the network simply aired the show with no promotion. The advertising doesn’t necessarily help other shows on the network generate more revenue.

The success of this investment in advertising, however, does not guarantee success. The general public might not like the show. In that case, no amount of advertising will convince them to watch.

It is also hard to determine the market value of the show. Other shows, both on that network and other networks, are not necessarily close substitutes for this particular television show.

Given that the investment is specific to the show, this sort of agreement could be subject to a hold up problem. Suppose that the network buys the show for $5 million. Later, if the show is generating $10 million for the network, the production company might refuse to the let the network distribute the show. They might commit to selling the show to another network for $5.1 million if the network doesn’t pay the production company $10 million. The production company might be willing to pay $4.9 million to get out of its current contract. This could cost the network a lot of money in terms of legal fees and potentially anger viewers who just want to watch the television show that they like. Anticipation of potential problems like this might lead to networks not investing in advertising or moving production in-house.

A chalkboard solution to these problems might be to have the network and the production company sign a complete contract that specifies the investment that will be made by the network and the payoffs to the production company that correspond to every possible future state of the world. All they have to do is negotiate the terms in each state.

Of course, in reality, this isn’t possible. It is not possible to know each possible state of the world. Furthermore, contracting over every possible state of the world is going to be prohibitively costly. Thus, one would expect that in the real world, contract characteristics would emerge that would solve these problems.

In a paper that was apparently never published, Armen Alchian, Ben Klein, and Earl Thompson (name a more iconic trio, I’ll wait) proposed one possible explanation. They pointed out that when investments  (a) are specific, (b) have an uncertain payoff, and (c) the market value of the payoff isn’t easily identifiable due to the lack of close substitutes, contracts are likely to include particular options that the party doing the investing can exercise. In particular, contracts are likely to give the investor first rights of refusal and first rights of negotiation.

In the network/production company example, the contract would work as follows. The network would agree to pay the production company a given sum of money to distribute the show and invest in advertising. The network would keep all of the revenue generated in excess of the purchase price. When the contract term was up, the production company would be required to specify a new asking price to the network. The network could agree or refuse (the network has the first right of refusal). In the event that the network refuses to pay that asking price, the production company could then shop the show to other networks and streaming platforms. If the highest outside offer that the production company received is lower than the refused price, the production company would then have to go back to the network and give them the option to match the offer. However, if another network or streaming service offered the production company a higher price, the production company could simply accept.

These provisions solve all of the problems. First, it explicitly deals with the uncertainty of the payoff. The risk of a bad payoff from the investment is borne by the network. If the show is unpopular, the network will lose. If the show is popular, the network will keep the excess revenue generated from the investment in that initial contract. Second, it solves the problem of an unknown market value. At the end of the contract term, the network can either accept or reject the offer made by the production company. In the event that the network rejects, the market value is determined by the best offer the production company receives from other networks. Third, these provisions solve the hold up problem by eliminating the desire for costly renegotiation since the production company has the ability to shop the show around if the network won’t meet their demands at the end of the term of the contract. They’re only locked into that initial value for the original period of the contract when the state of the world (the show’s level of success) is unknown. After the state of the world is revealed, the production company has a chance to capture at least some of the value from the network’s investment.

Contracts often have provisions in the them that seem weird to outside observers. Often times, these provisions seem weird because we are so used to thinking about basic chalkboard examples of how markets work. Nonetheless, it is important to develop an understanding of where these provisions come from because contract provisions can end up in policy debates. We’ve seen this with the recent debate over non-compete clauses. But attacks on contract provisions are actually more prevalent than that. Many contract provisions have been subject to antitrust actions. When these debates arise, it is important to understand the economic arguments that underlie these provisions.