Principal-Agent Problem: Definition, Causes, Solutions, and Examples

By Markets & Investing DeskUpdated Nov 10, 2026

Definition

The principal-agent problem is a conflict of interest that arises when a principal hands decisions to an agent whose goals differ and whose actions the principal cannot fully observe.

Key takeaways

It requires two ingredients: goals that differ and information the principal cannot fully observe.
The losses and expenses created by the conflict are called agency costs, a concept formalized by Michael Jensen and William Meckling in 1976.
Remedies work by aligning incentives, improving information, or both, and each remedy carries its own cost and side effects.

The principal-agent problem is a conflict of interest that arises when one party (the principal) delegates decisions or control over an asset to another party (the agent), and the agent has the ability and incentive to act against the principal’s interests. The principal keeps ownership and bears the consequences; the agent makes the day-to-day choices. Because the principal cannot verify everything the agent does, the gap between their goals can cost the principal money, effort or outcomes.

How the Principal-Agent Problem Works

Delegation splits ownership from control. The principal grants the agent decision rights but keeps the assets and the liability for losses. A shareholder owns part of a company but does not run it; the CEO runs it but owns, at most, a small part.

That split only creates a problem when two further conditions hold.

Divergent interests. The agent has its own goals: a higher salary, less effort, job security, prestige, a commission. The principal wants a different result, such as a higher share price, a lower legal bill or the best sale price.
Asymmetric information. The agent knows more about its own effort, skill and choices than the principal does. If the principal could judge every decision costlessly, poor behavior would be caught and corrected, and divergent interests would matter little.

Economists describe two forms the information gap takes:

Moral hazard (hidden action): after the agreement, the agent’s effort or choices cannot be fully observed. A manager may shirk, take excessive risk with other people’s capital, or spend on perks.
Adverse selection (hidden information): before the agreement, the principal cannot fully judge the agent’s ability or intentions, so it may hire the wrong agent.

The idea predates its modern name. Adam Smith warned in The Wealth of Nations (1776) that company directors managing “other people’s money” would not watch over it as carefully as their own, and Adolf Berle and Gardiner Means described the separation of ownership and control in large U.S. corporations in 1932. Jensen and Meckling formalized it in 1976, and the broader framework is called agency theory.

Agency Costs: What the Problem Costs

Because the principal cannot constantly monitor the agent, it bears the risk that the agent will shirk, decide poorly or act against its interest, plus the cost of guarding against that risk. Together these are agency costs, which Jensen and Meckling split into three parts:

Monitoring costs: what the principal spends to observe and control the agent, such as audits, board oversight, reporting requirements and performance reviews.
Bonding costs: what the agent spends to commit to good behavior and reassure the principal, such as providing financial statements, accepting contractual limits or posting a bond.
Residual loss: the value still lost because the agent’s decisions differ from what the principal would have chosen, even after monitoring and bonding.

Agency costs are a type of transaction cost. Total elimination is rarely worth paying for: a principal adds safeguards only while they cost less than the misbehavior they prevent, aiming for the cheapest combination of controls and residual loss.

Worked Example: A Commission That Points the Wrong Way

Hypothetical example for illustration; the figures are not any firm’s terms.

A homeowner hires a listing agent who earns 1.5% of the final sale price. An offer of $400,000 is on the table. The agent believes that two more weeks of showings would likely bring an offer of $410,000.

	Sell now at $400,000	Wait for $410,000
Agent’s commission (1.5%)	$6,000	$6,150
Seller’s proceeds after commission	$394,000	$403,850
Gain from waiting	Agent: $150	Seller: $9,850

Assuming no other costs, the seller gains $9,850 by waiting, while the agent gains only $150 for two more weeks that carry an opportunity cost in time taken from other listings. The agent therefore has a reason to recommend the first offer, and the seller, who knows less about local demand, may not be able to tell whether that advice is honest. Nothing illegal has to happen; the misalignment is built into the payment structure.

Solutions to the Principal-Agent Problem

The burden usually falls on the principal, who is best placed to fix the arrangement. Remedies fall into three groups.

Contract Design

A contract can align interests before the relationship starts and give the principal tools afterward. Typical terms include:

Defined goals and clear expectations from the outset
Regular reporting of results to the principal
Rights to hire outside monitors or auditors
Limits on the agent’s discretion over key decisions
The right to replace the agent if results fall short

Performance-Linked Compensation

Tying pay to outcomes the principal cares about is the most direct way to align interests, because compensation ranks high among an agent’s own priorities. Common tools include stock options and restricted stock for executives, deferred compensation, profit sharing, bonuses tied to performance evaluations, and clawback provisions that recover pay if results are later restated.

Monitoring and Governance

In companies, independent boards of directors, external auditors, shareholder voting and disclosure rules all reduce the information gap. In markets with competition, the threat of being replaced also disciplines agents: a client can switch lawyers, a seller can change agents, voters can elect someone else.

Examples of the Principal-Agent Problem

Shareholders and a CEO. Investors want the stock to rise, dividends, or both. A CEO might instead pour profits into empire-building expansion or large management bonuses.
Client and lawyer. A client billed by the hour may worry the lawyer is logging more hours than the case needs.
Homebuyer and real estate agent. A buyer may suspect the agent cares more about closing and collecting a commission than about the buyer’s concerns.
Investor and financial advisor. A conservative investor could discover that family savings were placed in a speculative asset, or in products that pay the advisor higher fees.
Homeowner and contractor. A homeowner rarely can judge whether a recommended repair, such as replacing rather than fixing an air conditioner, is truly needed.
Landlord and tenant. When a landlord buys appliances and the tenant pays the energy bills, the landlord has little reason to pay more for efficient equipment.

The Principal-Agent Problem in Politics

Voters are principals and elected officials are their agents. Unlike in a business contract, the agent answers to many principals with conflicting preferences, whose main enforcement tool is a vote cast at fixed intervals. Lobbying and the difficulty of observing what officials do between elections widen the information gap. Similar chains link legislators and the agencies that carry out their laws.

Limitations of the Remedies

Each fix can create a new version of the problem:

Stock options reward rising share prices but impose no loss when the price falls, which can encourage excessive risk-taking.
Short-term targets such as quarterly earnings can push managers to cut research or maintenance to hit a number.
Measurable goals pull effort away from what is hard to measure, such as quality or long-term reputation.
Monitoring costs money and is never complete; an auditor is itself an agent with its own incentives.
Pay-for-performance shifts risk to the agent, and a risk-averse agent will demand higher total pay to accept it.

Agency theory also assumes self-interested, rational actors. Real agents are often guided by professional norms, loyalty or reputation, and failures can stem from plain incompetence or poor communication rather than self-interest.

Agency theory vs. the principal-agent problem: agency theory is the overall framework for studying delegated relationships; the principal-agent problem is the conflict that framework analyzes.
Principal-principal problem: a conflict between owners themselves, typically controlling shareholders who extract benefits at the expense of minority shareholders. It is common in companies with concentrated ownership, while the classic principal-agent problem is associated with widely dispersed ownership, a feature of large firms under capitalism.
Principal vs. agent in accounting: under revenue-recognition rules, the “principal versus agent” test decides whether a company reports gross revenue or only its fee. It shares the vocabulary but is a reporting question, not an incentive conflict.
“Principle” agent problem: a common misspelling; the term uses principal, meaning the party on whose behalf the agent acts.

The Bottom Line

The problem touches shareholders, clients of lawyers and advisors, home sellers, voters and anyone who delegates decisions over money or assets. Principals can reduce agency costs through contract terms, pay tied to results and monitoring, but each remedy has its own cost and can create new distortions. The practical goal is the cheapest mix of safeguards and remaining loss, since eliminating the problem entirely is rarely worth the price.