Regulations mandating consumer consent for data collection and use have led firms to employ "dark patterns"—interface designs that encourage data sharing. We study the causal effects of these designs on consumer consent choices and explore their implications for consumer welfare. We conduct a field experiment where a browser extension randomizes consent interfaces while study participants browse the internet. We find that consumers accept all cookies over half of the time absent dark patterns, with substantial user heterogeneity. Hiding options behind an extra click significantly sways choices toward visible options, while purely visual manipulations have smaller impacts. Users also frequently close banners without active selection, highlighting the importance of default settings. While better-known firms achieve higher baseline consent rates, dark patterns do not amplify this advantage. We use a structural model to show that the consumer surplus-maximizing banner removes dark patterns and defaults to accepting cookies upon consumer inaction. A browser-level global consent choice further improves consumer welfare compared to site-by-site consent interactions, primarily by reducing time costs.

Reviews and other evaluations are used by consumers to decide what goods to buy and by firms to choose whom to trade with, hire, or promote. However, because potential reviewers are not compensated for submitting reviews and may have reasons to omit relevant information in their reviews, reviews may be biased. We use the setting of Airbnb to study the determinants of reviewing behavior, the extent to which reviews are biased, and whether changes in the design of reputation systems can reduce that bias. We find that reviews on Airbnb are generally informative and 97% of guests privately report having positive experiences. Using two field experiments intended to reduce bias, we show that non-reviewers tend to have worse experiences than reviewers and that strategic reviewing behavior occurred on the site, although the aggregate effect of the strategic behavior was relatively small. We use a quantitative exercise to show that the mechanisms for bias that we document decrease the rate of reviews with negative text and a non-recommendation by just .86 percentage points. Lastly, we discuss how online marketplaces can design more informative review systems.