Learn it → Apply it → Save as Skill
When one side of a deal knows more than the other, the market doesn't quietly settle on quality. It settles on distrust, discounts and sometimes collapse, unless the informed side signals or the uninformed side screens.
The Dr. Wallace's Favorite Mental Models library is designed to turn useful ways of thinking into reusable tools. Each model gives you a fast visual overview, deeper interactive learning, practical application exercises and an AI-ready Markdown skill you can save for future use.
Instead of merely reading about mental models, you can gradually build your own AI skills library: a collection of portable reasoning skills that can help ChatGPT, Claude, Grok and other AI assistants apply the mental models you value.
In many exchanges, one side knows something the other cannot easily check: the seller knows the car's history, the applicant knows their own health, the founder knows how the numbers really look. When the uninformed side can't tell good from bad, it rationally prices everything as average or below, and the best options quietly leave the market.
Important: the gap is not always about dishonesty. It can be ordinary: one party has simply lived with the facts longer than the other has.
The model is most useful when a deal, hire or deal depends on quality, effort, risk or intent that one party can see and the other can only infer.
Information asymmetry names the gap. Signaling and screening are the two main tools people use to close it.
Name exactly who holds more information, and about what: quality, effort, health, risk, or intent.
Ask what the informed side could do to credibly demonstrate quality, since unverified claims are cheap to make.
Design mechanisms, such as guarantees, probation periods or audits, that sort good cases from bad without needing to trust claims.
Check afterward whether the chosen signal or screen actually separated good from bad, not just whether it felt rigorous.
Open a use case to see what the model changes.
Candidates know their own skill and effort far better than an interviewer can observe in a few hours. Costly, hard-to-fake signals (a work sample, a trial project, a reference who stakes their own reputation) separate candidates better than claims on a page.
A supplier knows its true defect rate and capacity better than you do. Screening tools such as sample orders, penalty clauses and reference checks shift the cost of being wrong onto the party with more information.
A health or driving history is better known to the applicant than to the underwriter. Insurers screen with questionnaires, waiting periods and deductibles rather than relying on self-reported risk alone.
Investors cannot fully verify traction, churn or team dynamics from a pitch deck. Term sheets use vesting, milestones and liquidation preferences as screens, while audited numbers and reference customers act as signals.
A seller of a used item, a property or a secondhand vehicle usually knows more about its history than a buyer can see in a viewing. Inspection reports, warranties and return policies are signals that narrow the gap.
In a merger, acquisition or joint venture, one party has usually lived inside the numbers for years. Due diligence, escrow holdbacks and earn-outs are screens built specifically for this gap.
This is the classic illustrative case that gave the problem its name.
Expand for the distinction between the original 1970 paper and the broader field it founded.
In his 1970 paper "The Market for Lemons: Quality Uncertainty and the Market Mechanism," published in the Quarterly Journal of Economics, Akerlof used the used-car market to show that when buyers cannot verify quality, the market price reflects average quality, not the quality of any individual item. Sellers of above-average goods receive a price that undervalues them and may exit, which can lower average quality further.
Akerlof shared the 2001 Nobel Memorial Prize in Economic Sciences with Michael Spence and Joseph Stiglitz for their analyses of markets with asymmetric information. Spence's 1973 work on job-market signaling examined how the informed side (a job candidate) can use costly, hard-to-fake signals, such as education credentials, to distinguish itself. Stiglitz's work on screening examined how the uninformed side (an insurer or lender) can design contracts that sort applicants by type.
Signaling is initiated by the informed party: it offers a costly, credible action, such as a warranty or a degree, that a lower-quality party would find too expensive to fake. Screening is initiated by the uninformed party: it designs a menu of options, such as a deductible level or a probation period, that different types of the informed party will sort themselves into based on their own private information.
The model applies when one side could verifiably know more about something relevant and acts on it. A bad outcome from shared bad luck is not asymmetry.
The informed party often just has more accurate beliefs, not malicious intent. Treating every gap as deception damages negotiations that honesty could have resolved.
Costly credentials or certifications that exceed what the situation needs waste resources without adding trust beyond a cheaper, sufficient signal.
Asymmetry is often resolved informally, through reputation, referrals or repeated dealings, before any formal signal or screen is needed.
If neither side can know the answer in advance, that is ordinary risk, not information asymmetry. Use probabilistic thinking instead.
A screening mechanism built to sort by risk can end up sorting by a protected characteristic instead if it is not checked carefully.
neither side can know the outcome in advance; the uncertainty is shared, not asymmetric.
the relationship is long-term and repeated, and track record already resolves trust without a formal signal or screen.
verification is actually cheap and available. Build a signaling or screening scheme only when checking directly is not an option.
This browser tool structures your thinking. It does not send your information anywhere or perform AI analysis.
Save as Skill converts the model into portable Markdown instructions that teach an AI when and how to apply information asymmetry, what questions to ask, what evidence to seek, what mistakes to avoid and how to structure its analysis.
Save the skills you find useful and organize them into your own reusable library. The Markdown is intentionally model-neutral, so it can be used wherever an AI system accepts uploaded files, project knowledge, custom instructions or reusable text context. Platform capabilities differ, so this is a portable skill file rather than a claim of one-click installation.
No. It is a normal feature of almost every exchange; perfect shared information is rare. The problem arises specifically when the gap cannot be verified and no signal or screen exists, so the market defaults to distrust. Many functioning markets have simply built effective signals (brand reputation, credentials, warranties) that keep the gap from causing collapse.
Information asymmetry describes the underlying condition: one side knows more than the other. Adverse selection describes the resulting market-level effect when that condition goes unaddressed, namely that higher-risk or lower-quality participants disproportionately remain in the pool. Asymmetry is the cause; adverse selection is one possible consequence.
A single bad purchase is just bad luck. Akerlof's insight was that when every buyer in a market faces the same unverifiable gap, the market price collapses toward average quality for everyone, which then pushes genuinely good sellers out. The problem compounds because it operates on the whole market's pricing, not on one transaction.