Classical economic theory presumed agents process all relevant information when forming preferences and making choices. Yet decades of experimental evidence demonstrate that attention itself is a scarce resource, allocated strategically across competing stimuli and frequently misallocated in systematic ways. This recognition has fundamentally restructured how we model consumer behavior, firm competition, and welfare consequences.

The implications extend well beyond academic refinement. When attention operates as currency, markets do not merely aggregate preferences over fully understood options—they aggregate preferences over perceived options, where perception is shaped by salience, framing, and strategic obfuscation. The gap between considered and ignored attributes becomes economically consequential, generating equilibria that diverge sharply from full-information benchmarks.

What follows examines three interconnected dimensions of attention economics. First, the formal modeling apparatus developed by Sims, Bordalo, Gennaioli, and Shleifer that captures how cognitive constraints shape choice. Second, the industrial organization of inattention—how firms exploit, sustain, or sometimes correct attention failures through competitive dynamics. Third, the regulatory architecture required when disclosure regimes confront predictably limited cognitive bandwidth. Throughout, the analytical thread connects behavioral mechanism to institutional consequence, recognizing that choice architecture is inseparable from attention architecture.

Rational inattention theory, formalized by Christopher Sims, models attention as a costly information-processing activity constrained by a Shannon mutual information capacity. Agents optimally choose what to attend to, conditional on the marginal value of reducing uncertainty about a given dimension. The framework preserves optimization while embedding cognitive scarcity directly into the choice problem.

Salience theory, developed by Bordalo, Gennaioli, and Shleifer, takes a different route. Attention is allocated not by deliberate optimization but by perceptual contrast: attributes that stand out relative to a reference distribution receive disproportionate decision weight. Salient thinkers overweight differentiated payoffs, generating predictable departures from expected utility maximization in lotteries, consumer choice, and asset pricing.

These approaches generate distinct empirical signatures. Rational inattention predicts that attention shifts with stakes and prior uncertainty, producing state-dependent stochastic choice. Salience predicts context-dependent reversals driven by the composition of the consideration set itself. Laboratory paradigms—including eye-tracking, mouselab process tracing, and neuroimaging of dorsolateral prefrontal regions—have begun discriminating between mechanisms.

Neuroeconomic evidence suggests both processes operate, with salience-driven attention arising from bottom-up perceptual systems while rational inattention captures top-down allocation under deliberation. Drift-diffusion models incorporating attention weights, as in Krajbich and Rangel's work, demonstrate that gaze duration causally influences valuation, not merely reflecting it.

For applied researchers, the modeling choice matters considerably. Rational inattention generates welfare conclusions sympathetic to revealed preference: ignored attributes carry low value because attending to them was not worth the cost. Salience-based models open space for genuine welfare losses, since the attention allocation itself is mechanical rather than optimized.

Takeaway

Attention is not a passive filter but an active economic variable, and whether you model it as optimized or mechanical fundamentally changes what counts as a market failure.

Gabaix and Laibson's shrouded attributes model demonstrates that firms can profit by hiding add-on prices—printer ink, hotel resort fees, overdraft charges—even in competitive markets with sophisticated consumers present. The unraveling logic that should force disclosure breaks down when educated consumers can simply substitute away from add-ons rather than rejecting the base good.

The equilibrium is sustained by a curse of debiasing: any firm that prominently advertises competitors' shrouded fees inadvertently teaches consumers a substitution strategy that benefits all firms charging the base good honestly, dissipating the educator's competitive advantage. Information provision becomes a public good that competitive markets undersupply.

Empirical work confirms these dynamics across diverse settings. Brown, Hossain, and Morgan documented persistent shipping fee shrouding on online auction platforms. Stango and Zinman showed that consumer overdraft costs rise with attention scarcity. Heidhues, Kőszegi, and Murooka extended the framework to show that even naive consumers can be exploited through complex pricing in equilibrium.

Crucially, competition does not uniformly correct attention failures—it sometimes intensifies them. When marketing technology becomes more sophisticated, firms compete on the ability to direct attention toward favorable attributes and away from unfavorable ones. The competitive frontier shifts from product quality to attentional manipulation.

Yet shrouding is not universal. Markets with sophisticated repeat purchasers, low switching costs, and standardized comparison metrics tend toward transparent pricing. The contrast between these regimes suggests that institutional features—comparison platforms, mandatory disclosure formats, third-party rating systems—are doing substantial economic work in determining whether attention failures translate into welfare losses.

Takeaway

Competition disciplines prices on attributes consumers attend to, but on attributes they ignore, competition can intensify exploitation rather than correct it.

Regulatory frameworks that assume full information processing perform poorly when consumers face binding attention constraints. Disclosure mandates—nutritional labels, financial product summaries, energy efficiency ratings—generate value only when their format aligns with the cognitive processes consumers actually employ. Volume of disclosure can paradoxically reduce comprehension by exceeding processing capacity.

The behavioral evidence supports a hierarchy of intervention. Simplified, salient summary metrics outperform comprehensive disclosure. Bertrand and Morse demonstrated that payday loan disclosures stating dollar costs rather than APRs measurably reduced borrowing. Allcott's work on energy efficiency labels shows that comparative framing alters purchasing decisions where absolute kilowatt-hour figures do not.

Default architecture exploits attention scarcity directly. Because revisiting a default requires costly attention, defaults shape choice asymmetrically across consumers with different attention budgets. Madrian and Shea's retirement contribution research and subsequent organ donation studies established that defaults function not as neutral fallbacks but as outcomes for the inattentive—who are often the population regulators most need to protect.

Sophisticated policy design therefore requires distinguishing inattention that reflects rational economy of cognitive effort from inattention that produces systematic welfare losses. The former argues for preserving choice freedom; the latter argues for protective defaults or even active restrictions. This distinction is empirical, not philosophical, and demands welfare frameworks that incorporate cognitive costs explicitly.

The frontier involves dynamic attention regulation: cooling-off periods, attention-forcing disclosures at the point of consequential decisions, and algorithmic auditing of interfaces designed to manipulate attention. As digital choice environments become increasingly engineered to capture attention, policy must engineer protective architecture with equivalent sophistication.

Takeaway

Designing policy without modeling attention is like designing infrastructure without modeling load—it may look complete, but it fails precisely where pressure is highest.

Attention economics is not a peripheral correction to standard theory—it restructures core questions about welfare, competition, and institutional design. Once we accept that cognitive bandwidth is scarce and strategically targeted, the boundary between preferences and the architectures that elicit them becomes analytically central.

The most productive research integrates formal modeling with experimental measurement and neuroscientific evidence, treating attention as observable rather than residual. This integration yields actionable insight: which disclosure formats actually inform, which defaults actually protect, which competitive dynamics actually discipline firms.

For designers of policy and institutions, the implication is operational. Treat attention as a primary design variable, not an afterthought. The choice environments we build will allocate cognitive resources whether we intend them to or not—the only question is whether that allocation serves the choosers or those who would profit from their distraction.