The hold-up problem is a situation in economics where one party in a business relationship underinvests because they fear the other party will exploit them after the investment is made. It happens when someone sinks money, time, or resources into a deal that ties them to a specific partner, and that partner later uses their leverage to renegotiate for a bigger share of the profits. The result: people invest less than they should, and both sides end up worse off.
How the Hold-Up Problem Works
Three ingredients create a hold-up problem. First, one or both parties make a “relationship-specific investment,” something that has value inside this particular deal but loses most of its value outside it. Think of a factory built next to a single customer’s plant, or years of training in a company’s proprietary software. Second, the contract between the parties is incomplete, meaning it can’t anticipate every future scenario or lock in terms for every possible outcome. Third, after the investment is made, the parties bargain over how to split the gains.
The problem kicks in because of timing. Before the investment, both sides might agree to fair terms. But once one party has committed resources that can’t easily be redirected elsewhere, the other party knows it. They can push for better terms, demand price concessions, or threaten to walk away, knowing their partner is now locked in. The invested party, anticipating this squeeze, rationally decides to invest less upfront. Economists call this “underinvestment,” and it’s the core inefficiency the hold-up problem creates.
The canonical model, developed by Oliver Hart and John Moore in 1988, formalized this dynamic. In their framework, both the buyer’s valuation and the seller’s production cost are uncertain when the contract is first written and can be influenced by each side’s investments. Those investments are observable to both parties after the fact but can’t be written into enforceable contract terms beforehand. If bargaining after the investment splits the surplus evenly, neither side captures the full return on their investment, so both invest less than the amount that would maximize total value.
The Classic Example: Fisher Body and General Motors
The most widely cited illustration of the hold-up problem comes from the auto industry. In the 1920s, General Motors relied on Fisher Body to supply automobile bodies. The story, as commonly told in economics textbooks, goes like this: Fisher Body allegedly exploited a cost-plus contract by locating its plants far from GM’s assembly lines and using inefficient production methods, which drove up costs (and Fisher’s profits) at GM’s expense. GM, locked into the relationship, eventually acquired Fisher Body in 1926 to end the hold-up.
The reality is more complicated. What GM actually acquired in 1926 was the 40 percent of Fisher Body shares it didn’t already own. And according to research published in the Journal of Law and Economics, Fisher Body did not in fact locate its plants far from GM’s assembly operations, and there’s no strong evidence that a hold-up actually occurred. The case remains a staple of economics courses, but it’s worth knowing that the real history is messier than the textbook version suggests. The story’s value lies more in illustrating the concept than in being a clean historical example.
Why Contracts Can’t Solve It Alone
If you could write a perfect contract, the hold-up problem wouldn’t exist. You’d specify exactly what each side must invest, what prices apply under every scenario, and what happens if conditions change. But real contracts are always incomplete. People have limited foresight (economists call this “bounded rationality”), the future is genuinely uncertain, and some things are simply too complex or too costly to spell out in advance. Oliver Williamson, whose work on transaction costs shaped this entire field, argued that either party can exploit these gaps to squeeze the other side after investments are made.
This is especially true when the economic or policy environment is unpredictable. Research on Chinese firms found that when economic policy uncertainty rises, companies become more worried about being held up by their suppliers or customers. The incomplete nature of their contracts leaves them exposed, and greater uncertainty makes the potential payoff from opportunistic behavior larger. The response, as the theory predicts, is to look for structural solutions rather than trying to write better contracts.
Vertical Integration as a Fix
The most dramatic solution to the hold-up problem is vertical integration: one company simply buys the other. If your supplier might hold you hostage after you’ve built your business around their components, you eliminate the risk by making them part of your company. There’s no one left to renegotiate with.
Williamson argued in the 1970s that this logic explains a significant share of mergers and acquisitions. When the transaction costs of dealing with an outside partner (including the risk of hold-up) exceed the costs of managing an operation internally, firms integrate vertically. The Chinese firm data supports this pattern: when policy uncertainty increased hold-up risk between upstream and downstream trading partners, firms pursued vertical acquisitions to bring those relationships in-house.
Vertical integration isn’t free, of course. Running a larger organization comes with its own inefficiencies, bureaucratic costs, and management challenges. But when the hold-up risk is severe enough, those costs look like a bargain compared to the alternative.
Repeat Interactions and Reputation
Not every hold-up risk requires a merger. When two parties expect to do business repeatedly over a long period, the threat of future punishment can keep both sides honest. This is the logic behind relational contracts: informal agreements sustained not by courts but by the value of the ongoing relationship itself.
If a supplier squeezes you today, you stop doing business with them tomorrow. That future loss, if large enough, outweighs the short-term gain from exploiting the contract gap. Research from Yale’s Cowles Foundation formalizes this intuition, showing that relational contracts work by treating past agreements as still valid, even if they aren’t legally enforceable. The key mechanism is that both parties know the relationship has a future worth protecting.
But this only works when the parties care about that future. If either side treats each negotiation as a fresh start, ignoring the history of the relationship (what economists call “bygones are bygones”), then hold-up problems fully reemerge. Reputation matters too: in industries where word gets around, a company known for squeezing locked-in partners will find it harder to attract investment or partnerships in the first place.
Hold-Up in the Labor Market
The hold-up problem extends well beyond supply chains. It shows up whenever you invest in skills that are valuable to your current employer but not easily transferable elsewhere. Learning a company’s proprietary systems, building relationships with its specific client base, or mastering internal processes all count as firm-specific human capital. These investments make you more productive at your current job but don’t do much for your resume if you leave.
The problem: once you’ve acquired those skills, your employer has leverage. They know you can’t easily take that expertise to a competitor, so they may not fully reward you for it. Anticipating this, employees rationally underinvest in firm-specific skills, even though both sides would benefit if they invested more.
Employers have found creative workarounds. One is the promotion system. Research shows that companies sometimes set promotion standards lower than pure efficiency would dictate. A lower bar for promotion (meaning more people get promoted) compensates employees for the risk of investing in firm-specific skills. The wage increase that comes with the promotion, if large enough, offsets the investment cost and solves the hold-up problem. This has an interesting side effect: it can lead to people being promoted beyond their level of competence, the phenomenon known as the Peter Principle. The promotion isn’t purely about matching the best person to the job; it’s partly a tool for encouraging investment in skills the company needs.
There’s another layer. When a promotion signals an employee’s ability to outside employers, competitors may bid up wages for promoted workers. This external validation actually helps the employer’s incentive system work, because the promotion carries value the employer doesn’t have to pay for directly.
Patent Hold-Up in the Tech Industry
One of the most active modern battlegrounds for hold-up disputes involves standard-essential patents (SEPs). These are patents that cover features built into widely adopted technology standards like WiFi and mobile broadband. Any company making a standard-compliant product must use these patented technologies, which gives patent holders enormous leverage.
For over two decades, policymakers and courts have worried that SEP owners could use their patent rights to extract royalties far exceeding the actual value of their technological contribution. Once an entire industry has adopted a standard, companies producing standard-compliant devices are locked in. The patent holder can threaten an injunction (essentially forcing the product off the market) to demand outsized licensing fees. This is patent hold-up in its purest form: the investment in standard compliance is relationship-specific, and the patent holder exploits that after the fact.
Courts, antitrust regulators, and standard-setting organizations responded by placing limits on how aggressively SEP holders can enforce their patents. But in recent years, a countervailing concern has emerged: “patent hold-out.” This is the mirror image, where companies that should be licensing these patents deliberately delay, stall negotiations, or use litigation tactics to avoid paying. Research has documented both pre-litigation hold-out (refusing to negotiate before being sued) and in-litigation hold-out (using delay tactics during the court process). The policy debate has shifted from a one-sided concern about patent holders to a recognition that opportunism can flow in both directions.