Behavioral Meets IO: Inattention, Shrouding, and Prices

The Informed Consumer Who Isn’t

Industrial organization — the branch of economics that studies how firms compete, set prices, and structure their offerings — has traditionally rested on a foundational assumption: consumers know what they are buying and what it costs. They may face search costs, they may lack information about quality, and they may be uncertain about the future. But the prices they see are the prices they use to make decisions. A rational consumer comparing two printers examines the price of the printer. A rational consumer choosing a bank account examines the fees.

The assumption was always a simplification, but it was thought to be a useful one — close enough to reality that the models built on it would generate reasonable predictions. Over the past two decades, a body of research has shown that this assumption is not merely imprecise but actively misleading in important markets. Consumers frequently do not know what they are paying. They do not process all the price information available to them. They are systematically inattentive to certain cost components. And firms — far from being passive participants in consumer ignorance — actively exploit it.

This collision between behavioral economics and industrial organization has produced some of the most practically important research in modern economics. It has changed how we think about consumer protection, pricing regulation, and competition policy. And it has raised a troubling question: in markets where firms profit from consumer confusion, does more competition make things better or worse?

Gabaix and Laibson: The Shrouded Attributes Model

In 2006, Xavier Gabaix and David Laibson published a paper that formalized something every consumer already suspected: firms deliberately hide the true cost of their products. The paper, “Shrouding: The Equilibrium of Consumer Protection,” introduced the concept of shrouded attributes — product characteristics, usually add-on costs, that firms intentionally make less visible to consumers.

The examples are ubiquitous. A printer is advertised at an attractive price, but the replacement ink cartridges cost more per ounce than fine champagne. A hotel room is listed at $149 per night, but the “resort fee” of $45, the parking charge of $30, and the Wi-Fi charge of $15 are disclosed only at checkout. A bank account is “free,” but the overdraft fees, ATM surcharges, and paper statement charges add up to hundreds of dollars per year. A rental car is quoted at $25 per day, but the insurance, fuel charges, and taxes double the actual cost.

In each case, the headline price — the salient, advertised price — is low. The hidden costs are high. The total cost to the consumer is substantially more than the price they used to make their decision. And firms have strong incentives to maintain this structure.

Gabaix and Laibson modeled a market with two types of consumers: myopic consumers who see only the base price and ignore add-on costs, and sophisticated consumers who see everything. Firms choose both a base price and an add-on price. The key result is that in equilibrium, firms shroud the add-on prices and compete on the base price. Sophisticated consumers avoid the add-ons (they bring their own ink, skip the resort fee, avoid overdrafts), but myopic consumers pay them. The revenue from myopic consumers’ add-on purchases allows firms to lower the base price, which benefits sophisticated consumers. Sophisticated consumers are effectively subsidized by myopic ones.

The truly disturbing result is about unshrouding. Suppose a new firm enters the market and tries to compete by disclosing all costs transparently — no hidden fees, just a single honest price. You might expect this firm to attract consumers. But Gabaix and Laibson showed that the transparent firm can be competed away. Its honest price must be higher than the shrouded base price (because it cannot recoup costs through hidden add-ons), so myopic consumers — who compare only base prices — avoid it. Sophisticated consumers, meanwhile, already avoid add-ons at the shrouding firms and benefit from the low base prices there, so they have no reason to switch either. The transparent firm has no customers.

This is what makes the model devastating: competition does not eliminate shrouding. In fact, competitive pressure can intensify shrouding, because firms that shroud have lower visible prices and attract more customers. The market equilibrium is one in which all firms shroud, and any firm that deviates toward transparency is punished by the market.

Rational Inattention: Information as Scarce Resource

Gabaix and Laibson’s model takes consumer inattention as given — some people notice add-on prices and some do not. But why don’t people notice? The theory of rational inattention, developed by Christopher Sims beginning in the early 2000s, provides a microfoundation: processing information is costly, and rationally optimizing agents will choose not to process all available information.

Sims, who had already won the Nobel Prize (2011) for his work on macroeconomic causality, proposed that human beings face a finite information-processing capacity, analogous to a bandwidth constraint in communication theory. He borrowed the concept of entropy from Claude Shannon’s information theory and used it to measure the amount of information a person can process per unit of time. A person with limited bandwidth must choose what to attend to and what to ignore.

In the context of consumer choice, this means that a person shopping for a bank account cannot costlessly process every element of the fee schedule. Reading the fine print takes time. Understanding the interaction between monthly fees, minimum balance requirements, overdraft policies, and ATM networks requires cognitive effort. A rational agent with limited bandwidth will focus on the most salient features — the ones that are largest, most visible, or most frequently encountered — and ignore the rest. This is not a bias; it is an optimal allocation of a scarce resource (attention) in an environment with too much information.

The implication for markets is that firms face consumers who are rationally inattentive, and they can exploit this by making unfavorable terms less salient. The information is technically available — it is in the contract, on the website, in the fine print — but the cost of processing it exceeds its expected value for many consumers. Firms do not need to hide information; they need only make it costly to find and process.

Rational inattention theory has been applied to a wide range of economic phenomena beyond consumer markets. In macroeconomics, it helps explain why prices are “sticky” — firms rationally choose not to continuously adjust prices because processing the information needed for optimal pricing is costly. In financial markets, it helps explain why investors hold undiversified portfolios — monitoring a large portfolio requires attention that has alternative uses. In labor markets, it helps explain why workers accept jobs without fully understanding their compensation packages.

Chetty, Looney, and Kroft: Tax Salience

One of the cleanest demonstrations of inattention’s economic consequences came from a 2009 study by Raj Chetty, Adam Looney, and Kory Kroft on tax salience. The question was simple: does it matter whether a sales tax is included in the posted price or added at the register?

Standard economic theory says no. A rational consumer buying a $1.00 item in a jurisdiction with a 7.375% sales tax knows the total cost is $1.07 (rounded). Whether the tag says $1.00 or $1.07 should not affect behavior — the consumer integrates the tax into their calculation regardless.

Chetty, Looney, and Kroft tested this in two ways. First, they ran a field experiment in a Northern California grocery store. For a selection of products, they posted tags showing the tax-inclusive price (e.g., “$1.07 incl. tax”) alongside the standard pre-tax price. Sales of the experimental products fell by about 8% relative to control products whose tags were unchanged. This was not a price change — the actual cost was the same before and after the tag change. It was a salience change. Making the tax visible reduced demand, implying that many consumers had not been fully incorporating the tax into their purchasing decisions.

Second, they analyzed alcohol and cigarette taxes across U.S. states, comparing excise taxes (which are included in the posted price, because the retailer pays them and passes them through) with sales taxes (which are added at the register). They found that excise taxes reduced consumption much more than equivalent sales taxes. A one-cent increase in the excise tax on beer, which changes the shelf price, reduced consumption by about three times as much as a one-cent increase in the sales tax, which does not change the shelf price. The taxes were economically identical but perceptually different.

The implications are profound. If consumers do not fully attend to taxes, then the behavioral response to taxation depends not just on the tax rate but on how it is presented. This means that policymakers have a design variable they may not have realized: the format of a tax affects its impact. It also means that the standard tax incidence analysis — which assumes consumers react to the full price including tax — may overstate the demand response to taxes that are not salient.

More broadly, the tax salience findings illustrate a general principle: the economic impact of a price or cost depends on its psychological impact, and psychological impact depends on salience. Identical costs that differ in visibility produce different behavior. This is a fundamental challenge to the standard model of the informed consumer.

Add-On Pricing in Practice: Banks, Hotels, Airlines

The theoretical insights of Gabaix-Laibson and the empirical findings on tax salience have been confirmed and extended by a large body of research on specific industries.

Banking. Overdraft fees are the canonical shrouded attribute. Banks advertise “free” checking accounts and earn billions from overdraft charges — $15.5 billion in the U.S. in 2019 alone. The fees are technically disclosed, but they are triggered by situations (small negative balances, often caused by the timing of deposits and debits) that consumers do not anticipate. Stango and Zinman (2009, 2014) showed that consumers systematically underestimate how much they pay in overdraft fees, and that the degree of underestimation is correlated with measured inattention and financial illiteracy. Regulatory efforts to increase disclosure — such as requiring opt-in for overdraft programs under the Federal Reserve’s 2010 rule — reduced fee revenue by making the cost more salient, exactly as the shrouded attributes model predicts.

Hotels. The rise of online travel agencies (OTAs) intensified price competition on the base room rate, which is the price that appears in search results. Hotels responded by unbundling: resort fees, cleaning fees, and surcharges that were once included in the room rate became separate line items, disclosed only later in the booking process. Economists have documented that this “drip pricing” strategy increases revenue because consumers anchor on the initially displayed price and do not fully adjust for the add-ons, even when the total is displayed before the final purchase.

Airlines. The unbundling of airfares — separate charges for checked bags, seat selection, carry-on luggage, boarding priority, and food — is a textbook application of the shrouded attributes model. The base fare, which is the search criterion, is kept low; the ancillary fees generate revenue from consumers who do not incorporate them into their initial comparison. Spirit Airlines and Ryanair pioneered this model, and legacy carriers adopted it. By 2023, ancillary revenue accounted for over $100 billion globally for the airline industry.

Credit cards. Late payment fees, foreign transaction fees, annual fees waived for the first year but charged thereafter, and balance transfer fees that are prominently disclosed in the offer letter but buried in the fine print of ongoing terms — all of these are shrouded attributes. The Credit Card Accountability Responsibility and Disclosure (CARD) Act of 2009 attempted to increase salience by requiring clearer disclosure of fees, minimum payment consequences, and interest rates. Research by Agarwal et al. (2015) found that the CARD Act reduced some fees and changed consumer behavior, but firms partially offset the regulation by introducing new fees in different locations.

The Curse of Debiasing

The natural policy response to consumer inattention is to provide more and better information: require clear disclosure, mandate total-price posting, educate consumers. The assumption is that if consumers knew what they were paying, they would make better choices, and competitive pressure would drive out exploitative pricing.

The evidence on this assumption is mixed at best. Disclosure requirements often fail, for several reasons.

Information overload. Requiring more disclosure can actually reduce comprehension if it increases the total volume of information consumers must process. The Truth in Lending Act requires lenders to disclose APR, total interest cost, and payment schedules — but the disclosure forms are so long and complex that most borrowers do not read them. Ben-Shahar and Schneider, in More Than You Wanted to Know (2014), documented that mandated disclosure has become “a policy tool of first resort” despite persistent evidence that it does not work as intended.

Strategic response by firms. When regulators mandate disclosure of one attribute, firms can increase the complexity of other attributes. This is the “whack-a-mole” problem: making one fee salient may cause firms to shift revenue to a different, less salient fee. The CARD Act reduced certain credit card fees but saw offsetting increases in others. Airlines responded to baggage fee disclosure requirements by introducing new fees for different services.

Consumer heterogeneity. Debiasing interventions benefit consumers who are capable of processing the newly disclosed information — typically those who are already relatively sophisticated. Less sophisticated consumers may not benefit, and may even be harmed if firms respond to the loss of revenue from sophisticated consumers by increasing charges on attributes that less sophisticated consumers still ignore.

The paradox of transparency in competitive markets. If all firms are forced to disclose all costs transparently, the base prices will rise (to compensate for the loss of add-on revenue), and competition will occur on the total price rather than the base price. This could benefit consumers overall — but it will raise the visible price, which may have political consequences. Consumers who previously saw a $149 hotel room and ignored the $90 in fees may be unhappy to see a $239 room, even though they were always paying $239. The salience of the higher price may reduce demand, harming both consumers and firms.

This is the “curse of debiasing”: informing consumers does not always improve outcomes, because the market equilibrium adjusts in response to changed consumer behavior, and the adjustment can partially or fully offset the direct effect of the information. This does not mean disclosure is never worthwhile — it sometimes is — but it means that the case for disclosure as a policy tool requires more than the intuition that “information is good.”

Implications for Competition Policy

The behavioral IO literature has important implications for competition policy — the body of law and regulation designed to prevent monopoly, promote competition, and protect consumers.

Traditional competition policy focuses on market structure: the number of firms, barriers to entry, market concentration. The assumption is that more competition produces better outcomes for consumers — lower prices, higher quality, more innovation. This is generally true, but the behavioral IO literature identifies conditions under which more competition can make things worse.

In a market with shrouded attributes, competition drives down the visible base price, which attracts consumers, but the revenue must come from somewhere — and it comes from add-on fees that consumers do not see. Intensifying competition on the visible price dimension can intensify exploitation on the invisible price dimension. A market with five banks competing aggressively on the advertised interest rate may generate more overdraft fee revenue than a market with two banks competing gently on the all-in cost.

This creates a tension for regulators. Promoting entry and competition — the standard prescription — may not help if entrants compete on the same shrouded-attribute model. What may be needed instead is regulation of the format of competition: rules about how prices must be disclosed, restrictions on drip pricing, requirements for total-cost advertising. These regulations constrain the dimension on which firms compete, forcing competition onto the total price rather than the base price.

The European Union has moved in this direction with regulations requiring airlines to display total prices including all mandatory charges, and with the Payment Services Directive requiring transparent fee disclosure for bank accounts. The United States has been slower, though the Consumer Financial Protection Bureau (CFPB) and the Federal Trade Commission (FTC) have pursued enforcement actions against drip pricing and hidden fees, particularly in the hotel and ticketing industries.

A New Kind of Market Failure

The behavioral IO literature describes a market failure that does not fit neatly into the standard taxonomy. It is not monopoly — the markets in question are often competitive. It is not an externality — the costs fall on the parties to the transaction. It is not asymmetric information in the traditional sense — the information is available, just not attended to. It is a failure of salience: the price mechanism, which is supposed to guide consumer choice, is distorted because consumers do not perceive the full price.

This failure is pervasive, affects billions of dollars of consumer spending, and is resistant to the standard remedies of more competition and more disclosure. Addressing it requires a behavioral approach to market design — one that takes seriously the cognitive limitations of real consumers and designs institutions (disclosure formats, pricing rules, default options) that help the price mechanism do its job.

The research program initiated by Gabaix, Laibson, Sims, Chetty, and their colleagues has moved behavioral economics from the psychology lab into the heart of industrial organization. It has shown that the question “do consumers know what they are paying?” is not a minor empirical detail but a first-order determinant of how markets function. And it has demonstrated that the answer, in many important markets, is no — and that this ignorance is not accidental but systematically produced and profitably maintained.