BackgroundPublic opinion has a dynamic relationship with policy and real‐world outcomes in liberal settings where reliable information is abundant. In these settings, the public continuously updates its opinions with reliable policy‐relevant information, and the changes in public opinion go on to affect policy and outcomes. It is unknown whether this dynamic exists in illiberal settings where the public's access to reliable information is heavily restricted.ObjectivesThis article advances a theory of public opinion's dynamic relationship with policy and outcomes that applies to illiberal settings.MethodsOur study examines a vital public good in one of the world's most restrictive information environments and estimates a dynamic model of relationships among three variables—public opinion, policy, and outcomes—with a focus on public opinion and outcomes as the key dependent variables. The analysis looks at air pollution remediation in 274 Chinese localities.ResultsWe find that public opinion reacts to objective air pollution outcomes and not to misleading information that downplays air pollution severity, which suggests the public can accurately evaluate the reliability of available information. We also find that local public opinion's impact on local air pollution is substantively meaningful on timescales as short as 1 to 2 years, indicating that the additional policy effort prompted by public opinion change is sufficient to yield tangible real‐world outcomes even in the short term.ConclusionPublic opinion has a dynamic relationship with policy and real‐world outcomes even in highly illiberal settings. We argue that these findings are likely to generalize across issue domains with outcomes that can be directly observed by the public.