Abstract
We study information resonance – the extra weight people attach to information from people similar to them – and examine its implications. In a new lab experiment designed to isolate resonance, we show that people follow the advice of strangers who share characteristics with them at a much higher rate. We find that this is true not only for shared demographics or political identities, but also for a broad range of beliefs, personality traits and even superficial preferences, and we measure these effects in several types of choices. The design sharply limits rational explanations based on information relevance or advisor quality. To examine the consequences of resonance, we adapt a standard information diffusion model, replacing geographic proximity with characteristic similarity. The model explains why role models matter, how social media can erode authority, and why promotion rules that look neutral can still produce homogeneous leadership teams. Finally, field evidence connects micro biases to potential macro outcomes: young workers disproportionately enter occupations already staffed by coethnic incumbents, and are less likely to enter those occupations after coethnic incumbents experience negative labormarket shocks. By weaving together micro behavior, theory tools and macro consequences, the paper uncovers new levers for educators, firms, and policymakers to meaningfully influence their constituencies.