Skip to content
Preprint

Opinions Before Evidence: Dynamic Information Quality and Source Familiarity

Sep 2026 · 1 citation · 11 references
Economics

Abstract

New information services begin without a record that teaches customers how to interpret them. We study a two-period monopoly selling standardized forecasts whose committed scoring technology combines noisy evidence with a persistent, initially unknown calibration. More evidence improves a forecast's current decision value but makes the source's calibration harder to learn. The first forecast is released publicly only after its immediate use expires, so it builds a common decoder for the next forecast. For an arbitrary stakes distribution, monopoly pricing uses the same static cutoff in both periods. Provider patience lowers initial evidence intensity; with identical technologies and a binary format menu, transparent conditions generate an endogenous low-evidence first forecast followed by a high-evidence second forecast. A constrained planner switches later because monopoly undercaptures the return to costly evidence. A calibration audit eliminates the accuracy--interpretability trade-off, but the provider may not adopt it even when adoption is socially valuable. The mechanism requires neither distorted reporting nor confirmation preferences: the strategic choice is the auditable evidence input to a mechanically generated score.

View source

Similar papers

Preprint Sep 2026

The Price of a Familiar Perspective

Access to past reports changes how consumers learn about competing information providers. A subscriber who has learned to interpret a familiar source may find an equally competent rival less useful. We study how opening the rival's archive changes renewal prices and the effect of outcome feedback. Better feedback stren...

G. Lukyanov, Ilya Pototskiy · 1 citation

Department of Economics

Yan-Lin Chen, Jun Zhang · 0 citations
Open access Oct 2026

Learning Whom to Trust: Machine-Learning Aggregation of Analyst Earnings Forecasts

Analyst consensus forecasts are widely used in investment decisions and as proxies for market expectations, yet they typically assign equal weight to component forecasts despite predictable differences in quality. We develop a constrained machine-learning framework that predicts each forecast’s relative quality from 29...

Levent Bulut, Sami Keskek, Abdullah Kumas · 0 citations
Preprint Oct 2026

Rewarding Noise: Incentives in Decentralized A/B Testing

Many firms that run A/B tests struggle with low statistical power, increasing the risk that launch decisions are driven by noise rather than true impact. We argue that in decentralized organizations, low power can reflect misaligned incentives: product teams choose what and how to test on a customer-facing experience t...

Ryan E. Kessler · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.