📈 economics

Screening with Product Mismatch

This paper analyzes how a monopolist strategically uses product mismatch to screen buyers with private horizontal preferences, demonstrating that mismatch can either create or reduce information rents depending on whether willingness to pay is independent of or correlated with the buyers' ideal product characteristics.

Original authors: Teck Yong Tan

Published 2026-07-27

📖 8 min read🧠 Deep dive

CC BY 4.0

Original authors: Teck Yong Tan

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to sell something to a crowd of people, but you don't know exactly what each person wants. This is the world of monopoly screening, a branch of economics that studies how a single seller can figure out what customers are willing to pay without them just telling the truth. Usually, economists have looked at "vertical" differences, where everyone agrees that a Ferrari is "better" than a bicycle, but some people just have more money than others. The seller's job is to offer a high-end Ferrari for the rich and a slightly dented, cheaper version for the less wealthy, figuring out how to make the rich people pay extra without them pretending to be poor.

However, the real world is often messier. Sometimes, what you want isn't about "better" or "worse," but about what fits your specific needs. This is called horizontal differentiation. Think of it like buying a backpack: a hiker needs a huge, heavy pack with many pockets, while a student just needs a small, light one. Neither is objectively "better"; they just fit different lifestyles. The tricky part is that the seller knows the hiker's pack costs more to make, and the hiker probably values it more, but the seller doesn't know who is who. The paper you are about to read dives into this messy middle ground, asking a simple but profound question: When a seller can't just make a "worse" version of a product to trick people, but instead has to give them the wrong version of a product, does that help or hurt the seller?

The Great Product Mix-Up

Imagine a giant, magical vending machine run by a clever shopkeeper. This machine sells a whole line of gadgets, from a tiny, basic "Pea" to a massive, complex "Mountain." The shopkeeper knows that the Mountain costs more to build than the Pea. But the customers? They are a mystery. Each customer has a secret "ideal gadget" in their head. Some want the Mountain, some want the Pea, and most want something in between. The problem is, the shopkeeper can't see inside their heads.

In the old days of economic theory, if the shopkeeper wanted to separate the rich Mountain-lovers from the poor Pea-lovers, they would just make a "cheap Mountain" that was missing a few gears. They would degrade the quality. But in this paper's world, the gadgets are horizontally different. You can't just remove gears from a Mountain and call it a "worse Mountain" that everyone agrees is inferior. If you give a hiker a Pea, it's not just "cheap"; it's the wrong tool entirely. It's like giving a surfer a snowboard. It's a mismatch.

The paper, written by Teck Yong Tan, explores what happens when the shopkeeper's only tool for sorting customers is to hand them the wrong gadget on purpose. It turns out, this strategy has two completely opposite personalities, depending on who the customers are.

Scenario 1: The "Just a Fit" Crowd

First, imagine a group of customers who all have the same amount of money to spend, but they just want different things. One wants a red bike, one wants a blue bike, one wants a green bike. They don't care about the price; they just want the color that matches their soul.

In this world, the shopkeeper wants to save money by giving everyone the cheapest bike (the Pea). But if she gives a green-bike-lover a red bike, that customer gets grumpy. The paper finds that when the shopkeeper tries to trick these customers by giving them a mismatch, she actually has to be gentler than she thought.

Why? Because if she gives a green-lover a red bike, that green-lover might say, "Hey, I'm actually a blue-lover, and the blue bike is even worse than the red one!" This creates a chain reaction of complaints. To stop the customers from lying about what they want, the shopkeeper has to give them a gadget that is closer to their ideal than she would if she were just trying to be efficient. She has to give the green-lover a green bike (or at least a very green-ish one) instead of a red one.

In this case, the "mismatch" creates a cost. The shopkeeper has to give up some profit to keep the peace. The more different the customers are from each other (the more they hate mismatching), the less the shopkeeper can sell, and the less money she makes. It's like a parent trying to get three kids to share one toy; if the kids are too picky, the parent has to give them almost exactly what they want, or the whole game falls apart.

Scenario 2: The "Rich and Picky" Crowd

Now, imagine a different crowd. Here, the people who want the big, expensive Mountains also happen to have the deepest pockets. The people who want the tiny Peas have very little money. This is the real world: the more you need a complex tool, the more you probably value it, and the more you can pay.

Here, the story flips upside down. The shopkeeper realizes she can use the mismatch as a super-powerful shield.

If she gives a rich, complex-need customer a simple, cheap Pea, that customer might be annoyed, but they are so rich they'll still buy it. However, if she gives that same Pea to a poor, simple-need customer, that customer will be thrilled. The problem is, the rich customer might try to pretend to be poor to get the cheap Pea.

But wait! The paper shows that the mismatch actually reduces the rent the shopkeeper has to concede to these rich customers. By giving the rich customer a Pea (which is a terrible fit for their complex needs), the contract becomes less attractive to them relative to the rent they would gain by pretending to be poor. The mismatch acts as a deterrent: the pain of the bad fit outweighs the savings of the lower price, making it unprofitable for the rich customer to mimic the poor one.

In this scenario, the mismatch reduces the cost of keeping secrets. The shopkeeper can actually give the rich customers a worse fit (a Pea) than she would in a perfect world, because the mismatch makes the "poor" contract less appealing to them. This allows her to sell to more people, including some who are very picky, and she might even make more money than before.

It's like a bouncer at a club. If the bouncer makes the VIPs stand in a cold, uncomfortable line (a mismatch), the VIPs find the deal of sneaking in as a regular less attractive because the discomfort of the line eats up the savings. The bouncer can let more people in and still keep the VIPs paying.

The "Crowd-Out" Surprise

There is one more weird thing the paper finds. In the "Rich and Picky" world, the shopkeeper ends up selling her cheapest, most basic product (the Pea) to the rich people who actually need the expensive Mountain. Meanwhile, the poor people who would have been perfectly happy with the Pea get kicked out of the store entirely.

This is called crowding out. The rich people "crowd out" the poor people because the rich people are willing to suffer through a bad fit (a Pea) just to save money, while the poor people can't afford the price of the Pea if the shopkeeper has to charge a premium to cover the rich people's rent. It sounds backwards, but it makes sense: the rich people are so valuable that the shopkeeper is willing to sell them a bad product, but she can't afford to sell that same bad product to the poor people because they aren't worth the hassle.

The Big Takeaway

The paper proves that when you sell a line of products where "better" isn't the same for everyone, the way you sort your customers changes everything.

If customers only differ in what they like (but have the same money), forcing them to take a mismatch is expensive. You have to give them better products to keep them honest.
If customers who like expensive things also have more money, forcing them to take a mismatch is cheap. It actually helps you stop them from lying by making the "poor" contracts less attractive, so you can sell more and make more profit.

The authors didn't just guess this; they built a mathematical model with strict rules and proved it works. They showed that the "mismatch" isn't just a mistake; it's a strategic tool. And the most surprising part? The tool works in opposite directions depending on whether the customers' desires are linked to their wallets. In a world of cloud computing, AI models, or tax software, this means a company might need to deliberately offer a "clunky" version to its biggest clients to keep them from pretending to be small clients, a strategy that would be a disaster if the clients were all equally wealthy.

The paper doesn't say this is easy to do, or that it always works in every single business. It just shows that the logic of "mismatch" is far more complex and powerful than we thought, turning a simple mistake into a sophisticated game of chess between the seller and the buyer.

Technical Summary: Screening with Product Mismatch

Problem and Motivation
This paper develops a theory of monopoly screening for product lines where variants are horizontally differentiated from the buyer's perspective but ordered by production cost from the seller's. Unlike the canonical vertical differentiation models (Mussa and Rosen, 1978; Maskin and Riley, 1984), where products share a common quality ranking, these product lines lack a universal ranking. Buyers possess private information regarding their "ideal" product variant, and their willingness to pay (WTP) may be correlated with this ideal.

The central problem is how a monopolist screens buyers when the only available instrument is product allocation (steering buyers toward specific variants) rather than quality degradation. Because a variant well-suited to one buyer may be a poor fit for another, the seller must induce "product mismatch" to screen types. The paper investigates whether this mismatch creates or reduces information rent, a question that yields divergent results depending on the correlation between horizontal needs and vertical valuation.

Methodology
The author models a monopolist offering a continuum of products indexed by $y \in [0, 1]$ , where production cost $c(y) = cy$ is increasing in the index. Buyers have a private type $\theta \in [0, 1]$ representing their ideal product. A buyer of type $\theta$ consuming product $y$ at price $p$ derives utility $v(\theta) - kM(y, \theta) - p$ , where $M(y, \theta) = m(|y - \theta|)$ represents mismatch disutility and $k$ scales the intensity of horizontal differentiation.

The analysis proceeds via the revelation principle, characterizing optimal direct mechanisms $(\alpha, t)$ subject to incentive compatibility (IC) and individual rationality (IR). A key methodological departure from standard vertical screening is the treatment of the utility schedule $u(\theta)$ . In vertical screening, the envelope theorem implies a monotone utility schedule, allowing the seller to focus on the lowest type's IR constraint. Here, because mismatch disutility is type-specific, the sign of $u'(\theta)$ depends on the direction of the mismatch (upward or downward). Consequently, the utility schedule can be non-monotone, and the "worst type" (whose IR binds) cannot be identified a priori. The author solves this by constructing Lagrangian multipliers on IR constraints for all types, adjusting the virtual surplus to account for the fact that satisfying IR for one type may require reducing mismatch for another.

The paper analyzes two distinct cases:

Pure Horizontal Preferences: $v(\theta)$ is constant. Buyers differ only in their ideal product.
Correlated WTP and Product Fit: $v(\theta)$ is strictly increasing (e.g., $v(\theta) = \theta$ ). Buyers with more demanding needs (higher $\theta$ ) have higher WTP.

Key Contributions and Results

1. The Duality of Mismatch in Rent Creation
The paper's primary theoretical contribution is demonstrating that product mismatch can either create or reduce information rent, depending on the screening environment:

Pure Horizontal Differentiation: When buyers differ only in horizontal fit, mismatch creates rent. Inducing downward mismatch (assigning a product below the ideal) steepens the utility schedule, increasing the rent conceded to lower types. Consequently, the seller induces less mismatch than the social planner (first-best) would. Served buyers receive products closer to their ideals than under the first-best allocation. Stronger horizontal differentiation ( $k$ ) increases the rent cost, reducing market coverage and profit.
Correlated WTP and Fit: When high types have both higher WTP and more demanding needs, mismatch reduces rent. Here, the seller must prevent high types from mimicking low types. Downward mismatch makes the low-type contract less attractive to high types, acting as a deterrent. Consequently, the seller induces more mismatch than the first-best. High types are assigned products further below their first-best allocation.

2. Reversal of Exclusion and "Crowding Out"
In the pure horizontal case, exclusion occurs from the top (high types are excluded because they are costly to serve). In the correlated case, exclusion occurs from the bottom. Crucially, the correlated case generates a "crowding out" of the basic product: the seller sells the lowest-cost product ( $y=0$ ) to high types (whose efficient product is non-basic) while excluding low types who are better matched to the basic product. This occurs because the high types' willingness to pay compensates for the mismatch, whereas low types are excluded to prevent them from mimicking the high types' contracts.

3. Comparative Statics on Horizontal Differentiation ( $k$ )
The impact of stronger horizontal differentiation reverses between the two cases:

In the pure horizontal case, higher $k$ reduces profit and coverage.
In the correlated case, higher $k$ can expand coverage and raise profit. Stronger differentiation makes mismatch a more effective screening tool (a stronger deterrent against high types mimicking low types), allowing the seller to concede less rent.

4. The Role of Individual Rationality (IR)
The paper highlights that in screening with horizontal mismatch, IR constraints shape the optimal allocation alongside IC constraints. Because the utility schedule need not be monotone, the seller must track IR constraints throughout the included set, not just at an extreme type. This leads to "pooling in utility" (where utility is held at zero over an interval) rather than the familiar "pooling in allocation" (ironing) found in vertical screening.

Significance and Claims
The paper claims to provide the first theory of monopoly screening specifically for horizontally differentiated product lines ordered by cost. It challenges the intuition that screening always involves degrading quality; instead, it shows that screening can involve steering buyers toward imperfectly matched products. The analysis demonstrates that the interaction between vertical value and horizontal fit fundamentally alters the screening mechanism, leading to outcomes (such as basic-product crowd-out and non-monotone utility) that are impossible in standard vertical screening models.

The author notes that the model imposes a one-dimensional type structure where ideal product and WTP are perfectly correlated. Future research directions suggested include relaxing this correlation to study how the strength of the correlation determines the dominance of rent-creation versus rent-reduction effects, and extending the analysis to competitive markets. The paper does not propose new applications beyond the motivating examples (cloud computing, AI models, tax preparation) but uses them to illustrate the natural occurrence of the modeled correlation.

The Great Product Mix-Up

Scenario 1: The "Just a Fit" Crowd

Scenario 2: The "Rich and Picky" Crowd

The "Crowd-Out" Surprise

The Big Takeaway

More like this