Pattern/Economics/No. 0769
Principal-Agent Problem
The principal-agent problem, or agency problem, is an economic conflict between a principal and an agent with different interests. Studied by Stephen Ross, it arises when limited oversight or unequal information allows the agent to act in ways that do not serve the principal.
Also called Agency Problem
- Evidence
- Well established
- Read
- 6 min
- Links
- 11 connections
01You've seen this when…
- in life
You ask a financial adviser where to put your savings. They recommend a fund that pays them a commission, without showing you a cheaper alternative.
- at work
Salespeople earn bonuses when contracts are signed. The implementation team inherits promises the product can’t keep.
- out in the world
A city pays a fixed annual fee for streetlight maintenance. The contractor bills on schedule, but residents keep reporting dark streets that nobody inspects.
02The idea
You hire someone because they can do something you can’t do yourself, or don’t have time to do. Then you face a second problem: how do you know they’re doing it in your interests?
Economists call you the principal and the person acting for you the agent. A homeowner and a real estate agent fit the pattern, but so do shareholders and executives, patients and doctors, or voters and elected officials.
The problem arises when their interests differ from yours and your ability to observe or evaluate what they do is limited. You want the best outcome; they may benefit from finishing quickly, selling a particular product, avoiding difficult work, or delivering something impressive before their contract ends.
This isn’t simply dishonesty. Someone can follow every rule while making choices you wouldn’t make if you had their information. Delegation gives them discretion. The question is what guides that discretion when you aren’t watching.
03Why it happens
- The rewards point in different directions. A seller wants a high price. An agent paid a small percentage of that price may prefer a quick sale. Both benefit from selling, but they value another week of negotiation differently.
- Expertise makes oversight difficult. You hire a specialist partly because they know more than you. That same information gap makes it hard to judge whether their recommendation serves you or them.
- Effort stays out of sight. You can see the finished report, but not whether the consultant tested its assumptions. Hidden effort creates room for moral hazard: someone takes less care because others bear much of the cost.
- Contracts can’t describe every situation. A maintenance agreement can specify inspection schedules, but not every judgment a technician will face. Incomplete contracts leave decisions to whoever does the work.
- Results don’t reveal effort cleanly. A careful fund manager can lose money in a falling market. A careless one can profit in a rising market. Paying only for outcomes can reward luck and penalize good work.
04A worked example
Economists Steven Levitt and Chad Syverson studied home sales in the Chicago area, comparing properties owned by real estate agents with properties they sold for clients. Their 2008 paper reports that agents’ own homes sold for about 3.7% more and remained on the market roughly ten days longer, after adjusting for many observable differences.
What it looks like Homeowner and agent want the same thing: a successful sale at a good price. A percentage commission seems to align their interests.
What’s actually going on They receive very different shares of an extra dollar. For an illustrative agent who personally receives 1.5% of the sale price, negotiating another $20,000 earns $300. The homeowner keeps most of the increase. The homeowner may find more calls and another open house worthwhile despite the risk of losing the buyer, while the agent may find the effort and risk unattractive. Selling their own home changes that calculation.
What would have helped The seller and agent could agree on the seller’s priorities. The seller could then review comparable sales independently and ask for evidence behind advice to accept an offer. A carefully designed bonus for achieving a stronger price might help, but only if it doesn’t encourage unrealistic pricing or excessive delay. The study supports an incentive explanation; its observational design cannot eliminate every unmeasured difference between the homes.
05How to spot it
06What to do about it
- Map the actual rewards. List who gets paid, promoted, blamed, or spared work under each option. Include financial rewards and the effects on reputation and convenience. Ask what the agent would prefer if your interests didn’t count.
- Check incentives at the margin. An overall shared goal can still leave extra effort unrewarded. Does one more hour of effort, one more dollar of quality, or one more avoided failure benefit the agent enough to matter? This is the practical test behind incentive compatibility.
- Verify a sample independently. A second opinion or an outside benchmark can help you check recommendations, while inspecting a few completed jobs lets you assess the work directly. You rarely need to watch everything to make careless work more detectable.
- Choose the agent before choosing the contract. Examine relevant performance records, checking them against references and work samples. Screening helps with hidden differences in competence or priorities that payment rules won’t fix.
- Protect what the reward leaves out. Pair sales rewards with checks on cancellations and customer outcomes. A narrow target can create Goodhart’s law problems while supposedly solving an agency problem.
- Price the oversight too. Audits, reporting, and approvals consume resources. Aim for enough accountability to make delegation worthwhile after accounting for the cost of oversight. Compare that total cost with doing the work yourself.
07When it isn’t disloyalty
An agent who disagrees with you may be using the expertise you hired them for. A doctor declining an unnecessary treatment can still be serving you, even when you wanted it. Different judgments can coexist with shared interests.
A financial conflict can coexist with proper conduct. Professional standards, reputation, personal values, and repeat business can support good work even when payment incentives are imperfect. Disclosure reveals a risk; the recommendation may still be sound.
Perfect alignment can be undesirable or impossible. Making a worker bear every business risk may be unfair and expensive, especially when outcomes depend on events they cannot control. Some remaining agency cost is the price of getting help. Judge delegation by how it compares with your realistic alternatives.
08Roots
In 1932, Adolf Berle and Gardiner Means examined a striking feature of large American corporations: the people who owned them were often not the people running them. Ownership was spread across shareholders, while managers controlled factories, hiring, and spending. A shareholder with a small stake had little reason to shoulder the cost of watching management closely. Their book, The Modern Corporation and Private Property, made this separation of ownership and control central to the study of corporations.
Stephen Ross turned delegation into a formal economic problem in his 1973 paper, The Economic Theory of Agency: The Principal’s Problem. He examined how payment arrangements could induce an employee or representative to pursue the owner’s goal. The problem now included both finding trustworthy people and designing the relationship.
Michael Jensen and William Meckling’s 1976 work brought that reasoning into a theory of the firm. They described agency costs as including spending on monitoring, the agent’s efforts to provide assurances, and the losses that remain despite both. That last category matters: they didn’t promise a contract that removes every conflict. The framework traveled well beyond shareholders and managers because the same tension appears anywhere people entrust decisions to others.
09How solid is this?
A foundational economic framework supported by extensive theoretical and empirical research on delegation and incentives. Particular conflicts require case-specific evidence, and changing payment rules does not reliably solve every agency problem.
10Connections
- Often confused with Goodhart’s Law
- Countered by Incentive Compatibility, Screening, Signaling
- Part of Asymmetric Information, Transaction Costs
- Includes Moral Hazard
- See also Hanlon’s Razor, Peter Principle, Incomplete Contracts, Adverse Selection
+ 1 more in the list
11Origin and sources
Modern economic formulation by Stephen Ross (1973), with agency costs and corporate ownership developed by Michael Jensen and William Meckling (1976). Berle and Means examined the separation of ownership and control in 1932.
- [1]Ross, S. A. (1973). The Economic Theory of Agency: The Principal's Problem. American Economic Review, 63(2), 134–139.
- [2]Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–360.
- [3]Levitt, S. D., & Syverson, C. (2008). Market Distortions When Agents Are Better Informed: The Value of Information in Real Estate Transactions. The Review of Economics and Statistics, 90(4), 599–611.
- [4]Berle, A. A., Jr., & Means, G. C. (1932). The Modern Corporation and Private Property. Macmillan.
Suggest an edit· Updated 2026-10-02