Discussion Paper No. 565
February 16, 2026
Designed Uncertainty in Mystery Products
Author:
Abstract:
Mystery products deliberately hide key attributes until after purchase and have become a common strategy in retail and services, yet systematic evidence on how to design them effectively remains limited. This research studies two managerial levers---outcome-set composition and uncertainty framing (risk vs. ambiguity)---in two incentive-aligned choice experiments: an induced-value lab study with vertically differentiated outcomes and a large-scale choice-based conjoint on apparel with horizontally differentiated brands. Willingness-to-pay is shaped primarily by the structure of the outcome set: when a dominant outcome is included, consumers discount the mystery product; when outcomes are similar in value, a premium can emerge. Ambiguity reduces valuation primarily when outcome differentiation is high, and it shifts attention away from brand and ``mystery'' cues toward tangible attributes such as fit and color. In market simulations, mystery products are more price-elastic than fully specified alternatives and shift profits toward participating brands, especially weaker ones, while non-participants lose. Overall, the results inform when and how designed uncertainty can be used as a marketing instrument.
Keywords:
mystery products; choice-based conjoint; hierarchical bayes; market simulation; price competition;
JEL-Classification:
D81; D12; C25;
Download:
Discussion Paper No. 564
Benefits and Challenges of Ambiguous Product Information
Author:
Abstract:
We study the welfare effects of ambiguous product information for a buyer with α-max-min preferences and a price-setting seller. The buyer privately receives information about her valuation. We show that the seller or the buyer can benefit when this information is ambiguous, and we characterize all possible combinations of producer and consumer surplus, as evaluated under ambiguity-sensitive preferences. Ambiguity concerning the valuation perceived by the buyer when making the purchase decision can induce the seller to change the price. Before receiving information, ambiguity concerning the purchase decision can make the buyer optimistic about buying only for high valuations, which relaxes the participation constraint.
Keywords:
Ambiguity; uncertainty; information design; bayesian persuasion; strategic learning; pricing; bargaining;
JEL-Classification:
D42; D81; D82; D83; L12;
Download:
Discussion Paper No. 563
The Price of Productivity
Author:
Abstract:
We construct a new micro-geographic commercial rent index for Germany to study the capitalization of agglomeration economies into floor space prices. In large local labor markets, commercial rents decline by -17% per kilometer from the central business district, compared to 13% for residential rents, reflecting stronger agglomeration benefits at the center. Commercial rents in central business districts increase with local labor market size at an elasticity of 15%, implying that wage responses capture only about half of the agglomeration effect on total factor productivity.
Keywords:
floor space; rents; spatial equilibrium; total factor productivity;
JEL-Classification:
L2; R3;
Download:
Discussion Paper No. 562
January 26, 2026
A Horserace of Methods for Eliciting Induced Beliefs Online
Author:
Abstract:
This study evaluates the effectiveness of three widely used belief elicitation methods in an online setting: the binarized scoring rule (BSR), the stochastic Becker-DeGroot-Marschak mechanism (BDM), and unincentivized introspection. Despite the theoretical advantages of incentive-compatible methods (BSR and BDM), we find that they impose significantly higher cognitive costs on participants, requiring more time and effort to implement, without delivering clear improvements in belief accuracy. In fact, BSR systematically leads to greater errors in reported beliefs compared to introspection, while BDM also reduces accuracy, though to a lesser extent. Surprisingly, individual differences in probabilistic reasoning skills do not mitigate these errors for BSR but do help improve accuracy under BDM. Our findings suggest that simpler, unincentivized approaches may offer comparable or even superior accuracy at a lower cognitive cost. These results have broad implications for the design of experiments and the interpretation of belief data in behavioral and experimental economics.
Keywords:
belief elicitation; induced beliefs; incentives; online experiment;
JEL-Classification:
C81; C89; D83; D91;
Download:
Discussion Paper No. 561
The Economics of Architecture
Author:
Abstract:
We illustrate the coordination problem in the provision of distinctive architectural design that arises from design externalities within a quantitative model. To quantify the model, we conduct a quantitative review of a growing literature concerned with the costs and benefits of distinctive design as well as a survey of architectural design preferences. We find that distinctive buildings sell at a 15% premium, on average. Positive design spillovers from distinctive nearby buildings result in a 9% premium. Distinctive buildings, however, are about 25% more expensive to build. The distribution of design ratings within buildings is well described by a Fr´echet distribution with a shape parameter of about 4. Parametrising the model to match these moments, we show in counterfactual simulations that the optimal subsidy of distinctive buildings amounts to 10% of construction costs.
Keywords:
architecture; design; economics; regulation; welfare;
JEL-Classification:
R3; N9;
Download:
Discussion Paper No. 560
Quantile Selection in the Gender Pay Gap
Author:
Abstract:
We propose a new approach to estimate selection-corrected quantiles of the gender wage gap. Our method employs instrumental variables that explain variation in the latent variable but, conditional on the latent process, do not directly affect selection. We provide semiparametric identification of the quantile parameters without imposing parametric restrictions on the selection probability, derive the asymptotic distribution of the proposed estimator based on constrained selection probability weighting, and demonstrate how the approach applies to the Roy model of labor supply. Using German administrative data, we analyze the distribution of the gender gap in full-time earnings. We find pronounced positive selection among women at the lower end, especially those with less education, which widens the gender gap in this segment, and strong positive selection among highly educated men at the top, which narrows the gender wage gap at upper quantiles.
Keywords:
quantile regression; sample selection; roy model; rank invariance; semiparametric inference; gender wage gap; wage inequality;
JEL-Classification:
C14; C31; C36; J16; J21; J31;
Download:
Discussion Paper No. 559
January 8, 2026
The Impact of Behavioral Design and Users’ Choice on Smartphone App Usage and Willingness to Pay: A Framed Field Experiment
Author:
Abstract:
Behavioral design in smartphone apps aims at inducing certain, monetizable behavior, mainly increased engagement, measurable by usage time. Such design is rarely transparent and often restricts users’ ability to make alternative choices. In a framed field experiment, we document that behavioral design doubles app usage time compared to a version without behavioral elements. Providing users with choices—simply explained and conveniently adjustable design features—reduces usage time and increases their willingness to pay for the app. These findings suggest that offering choice could pave the way for new business models based on more responsible app design.
Keywords:
smartphone app; behavioral control; filtering algorithm; transparency and choice; self-determination; corporate social responsibility; field experiment;
JEL-Classification:
C93; O33; D83; L86; M14;
Download:
Discussion Paper No. 558
December 22, 2025
Delegating in the Age of AI: Preferences for Decision Autonomy
Author:
Abstract:
Despite the documented benefits of algorithmic decision-making, individuals often prefer to retain control rather than delegate decisions to AI agents. To what extent are the aversion to and distrust of algorithms rooted in a fundamental discomfort with giving up decision authority? Using two incentivized laboratory experiments across distinct decision domains, hiring (social decision-making) and forecasting (analytical decision-making), and decision architecture (nature and number of decisions), we elicit participants’ willingness to delegate decisions separately to an AI agent and a human agent. This within-subject design enables a direct comparison of delegation preferences across different agent types. We find that participants consistently underutilize both agents, even when informed of the agents’ superior performance. However, participants are more willing to delegate to the AI agent than to the human agent. Our results suggest that algorithm aversion may be driven less by distrust in AI and more by a general preference for decision autonomy. This implies that efforts to increase algorithm adoption should address broader concerns about control, rather than focusing solely on trust-building interventions.
Keywords:
algorithm; delegation; artificial intelligence; trust in ai; experiment; preferences;
JEL-Classification:
C72; C91; D44; D83;
Download:
Discussion Paper No. 557
AI Tutoring Enhances Student Learning Without Crowding Out Reading Effort
Author:
Abstract:
We study how AI tutoring affects learning in higher education through a randomized experiment with 334 university students preparing for an incentivized exam. Students either received only textbook material, restricted access to an AI tutor requiring initial independent reading, or unrestricted access throughout the study period. AI tutor access raises test performance by 0.23 standard deviations relative to control. Surprisingly, unrestricted access significantly outperforms restricted access by 0.21 standard deviations, contradicting concerns about premature AI reliance. Behavioral analysis reveals that unrestricted access fosters gradual integration of AI support, while restricted access induces intensive bursts of prompting that disrupt learning flow. Benefits are heterogeneous: AI tutors prove most effective for students with lower baseline knowledge and stronger self-regulation skills, suggesting that seamless AI integration enhances learning when students can strategically combine independent study with targeted support.
Keywords:
ai tutors; large language models; self-regulated learning; higher education;
JEL-Classification:
C91; I21; D83;
Download:
Discussion Paper No. 556
Visual and Social Anchoring in a Framed Online Rating Experiment
Author:
Abstract:
We conduct an online experiment to assess the effect of the anchoring bias on consumer ratings. We depart from the canonical anchoring literature by implementing non-numerical (visual) anchors in a framed rating task. We compare three anchoring conditions, with either high, low, or socially derived anchors present, against two control conditions – one without anchors and one without framing. Our framing replicates the common observation of overrating. We unveil asymmetric non-numerical anchoring effects that contribute to the explanation of overrating. Both high anchors and socially derived anchors lead to significant overrating compared to the control condition without anchors. The latter finding is driven by instances of high social anchors. The upward rating bias is exacerbated in a social context, where participants exhibit more trust in anchors. In contrast, low anchors and instances of low social anchors have no effect compared to the control condition without anchors. Beyond consumer ratings, our results may have broader implications for online judgment environments, such as surveys, crowdfunding platforms, and other user interfaces that employ visual indicators such as stars, bars, or progress displays.
Keywords:
anchoring bias; consumer judgment; economic experiment; online feedback systems; user interface design;
JEL-Classification:
C91; D80; D91;