Economists call the gap in knowledge between market participants “information asymmetry.” When sellers know product quality while buyers do not, markets tend to let “bad money drive out good,” leaving mostly poor-quality goods. Akerlof used the used-car market to formalize how a “market for lemons” emerges. Spence showed that costly actions—education, heavy advertising—can work as “signals” that allow high-productivity workers or firms to distinguish themselves. Stiglitz analyzed how insurers or lenders, who lack information, can design a menu of contracts that induces clients to reveal their private risk levels; this is “screening.” These ideas underpin modern game theory models with incomplete information, shape financial regulation, and influence labor-market policies.

Akerlof’s 1970 paper models a used-car market in which quality q is unobservable. Under a single price based on expected quality, low-quality cars crowd out high-quality ones, producing adverse selection. Spence’s 1973 signaling model treats worker productivity θ and education effort e as strategic variables and derives separating-equilibrium conditions such as e(θ_h)−e(θ_l)<w(θ_h)−w(θ_l). Rothschild & Stiglitz (1976) build an insurance-market screening model: incentive-compatibility and self-selection constraints yield a two-contract separating equilibrium. Stiglitz & Weiss (1981) explain credit rationing, showing that non-monotonic expected returns make quantity rationing preferable to raising interest rates. These insights permeate mechanism design, auction theory, and corporate finance. On the policy side, they justify disclosure rules, risk-based insurance premia, and other tools that mitigate information asymmetry.

The prize consolidates the theoretical architecture of market equilibria under asymmetric information. Akerlof’s model shows that when quality is unobservable, expected quality E[q|p] becomes downward-biased; if the uniform price falls below marginal cost, trade collapses—an adverse-selection equilibrium. Spence extends a complete-information repeated game to a Bayesian game of incomplete information, classifying pooling vs. separating equilibria using the monotonicity of the education cost function c(e,θ). The Rothschild–Stiglitz insurance model imposes a no-cross-subsidy condition to exclude self-fulfilling expectations and highlights self-selection equilibria, often contrasted with Wilson equilibrium. Stiglitz & Weiss endogenize the link between project return distribution F(r|θ) and interest rate i, proving that raising loan rates attracts only high-risk borrowers, reducing expected bank profits and leading to credit rationing. These contributions underpin subsequent results such as the Myerson–Satterthwaite impossibility theorem and dynamic moral-hazard models à la Holmström–Milgrom.