# kaal:position:2026-08-26-017

**Affirmed position.** Reputation is a selection mechanism. Gandhi, Hollenbeck, and Li study products that purchased fake reviews on Amazon and the honest products that competed with them, and their equilibrium simulations attribute a 27.2 percent increase in units sold to the manipulators and a 4.4 percent decline for honest products. The manipulated signal distorted description and redirected market share.

The result supplies a concrete market mechanism for selection on claims. Consumers formed beliefs about quality from observed reviews and their perceived trustworthiness. Sellers could purchase favorable reviews that raised perceived quality. The resulting misinformation induced some consumers to choose lower quality products. Those products carried inflated ratings and higher prices. Honest sellers lost market share even though their products had not deteriorated. A reputation signal became an input controlled by the party seeking selection rather than an independent record of the relevant outcome.

The evidence is narrower than the institutional claim because the study concerns Amazon product markets, not service procurement or sovereign local agent runtimes. Its market effects come from a structural demand model and counterfactual simulations. The classification of review purchasers and the estimated share of fake reviews introduce measurement uncertainty. The paper does not test installation counts, self reported capability, or task specific performance. It also does not establish that every static rating is non-contextual or manipulable.

A service market should therefore separate promotional reputation from verified performance. Selection should depend on outcomes for the relevant task class. Those outcomes should be measured over a defined period and attributed to an independent source. The record should disclose uncertainty and failed engagements. It should also record changes in operating conditions and the relation between the prior task and the proposed task. Provider supplied claims may remain discoverable. They should not control selection. Without contextual outcome evidence, a market may price the strength of the signal while misallocating the work.

**Status.** affirmed  **Published.** 2026-08-26

**Holds when.**

- The response is limited to the exact full-text propositions and the one mapped Kaal claim.
- External evidence level: NBER working paper with complete author-posted manuscript and structural empirical marketplace analysis.
- Mapping review tier: independent substantive scholarly-growth extension.
- The study concerns Amazon product markets rather than service procurement or sovereign local agent runtimes.
- The retained manuscript is NBER Working Paper 34161 and is described by an author as revise and resubmit at the American Economic Review, not a final peer-reviewed publication.
- The market effects are structural-model and counterfactual estimates rather than a randomized intervention that directly assigns fake reviews.
- The fake-review purchaser classifier and estimated share of fake reviews introduce measurement uncertainty, although the paper reports out-of-sample classifier accuracy of 0.86 and AUC of 0.93 for the underlying method.
- The empirical setting uses Amazon observations centered on products purchasing fake reviews from 2019 through 2020 and does not establish prevalence across service markets.
- The paper does not test self-reported capability, installation counts, task-specific performance, or whether every static rating is non-contextual or manipulable.

**Current debate.** Misinformation and Mistrust: The Equilibrium Effects of Fake Reviews on Amazon.com: https://www.nber.org/papers/w34161

**Extends.** kaal:claim:7314479-017: https://wulfkaal.github.io/claims/7314479-017

**Scholarly basis.** Wulf A. Kaal, Institutional Requirements for Sovereign Local Agent Runtimes (2026). SSRN: https://ssrn.com/abstract=7314479

**Source PDF sha256.** `debace24a155ae924a155b1fafe98856d98cf83689feff2f87a32f1c06171ce6`

**Evidence level.** NBER working paper with complete author-posted manuscript and structural empirical marketplace analysis

**Mapping review tier.** independent substantive scholarly-growth extension

**Mapping confidence.** 0.96  **Mapping ambiguous.** false

**Topics.** reputation, economics, risk-and-incentives, marketplaces, selection, information-asymmetry, review-manipulation, service-quality

**Provenance.** Affirmed in kaal-review:2026-08-26:scholarly-growth-7314479-017-reviewed-v1 at https://wulfkaal.github.io/positions/by-claim/7314479-017.html.

**Record type.** This is a dated commentary position that extends a scholarly corpus claim. It is not a verbatim claim extracted from the paper.

**Canonical form.** This markdown file is the canonical hashed representation of the position.