National Content, Local Consequences: How the Geographic Orientation of Media Shapes Political Accountability
Bühler & Dickens: "National content and local political consequences: Evidence from public and private television" Journal of Public Economics, Volume 259 (2026)
The introduction of television in 1950s Canada provides a unique natural experiment for understanding how media content, rather than just the technology itself, influences political accountability. Using the historical “single-station policy,” which restricted early markets to either a public or private broadcaster, researchers Mathias Bühler (LMU Munich, Project B08) and Andrew Dickens (Brock University) isolate the effects of nationalised versus locally-bundled news. Their analysis reveals that exposure to public television, which prioritised national programming, reduced voter turnout by 2.1 percentage points relative to districts without access. Conversely, in districts served by private broadcasters (which aired local news alongside national content) voter engagement remained indistinguishable from pre-television levels.
The study finds that this shift in the electorate’s behaviour significantly altered the incentives for elected officials. Members of Parliament (MPs) in districts served by public television reduced the local orientation of their parliamentary speeches by approximately 0.29 standard deviations. This retreat from constituency-focused rhetoric was accompanied by a marked increase in party discipline; MPs in public districts were significantly more likely to vote along party lines than those in private districts. These behavioural changes suggest that when media content nationalizes, representatives prioritize party loyalty over local responsiveness due to a perceived reduction in voter monitoring.
Furthermore, the authors document a substantial impact on electoral competitiveness. The value of incumbency rose by roughly 11 percent in districts with public television, whereas private television districts saw no such increase. This divergence suggests that the scarcity of local information on public broadcasts effectively insulated incumbents by depriving voters of the candidate-specific knowledge required to evaluate challengers. To ensure these results are not artifacts of non-random network expansion, the authors utilise a novel measure of expected signal strength to isolate quasi-random variation in television access. These results are also robust to controls for pre-existing radio infrastructure, supporting a causal interpretation.
Ultimately, these findings indicate that the geographic orientation of media content is a primary driver of political behaviour and legislative performance. While the Canadian government’s broadcasting policy aimed to foster a shared national consciousness, it inadvertently weakened the local accountability essential to parliamentary representation. This study emphasises that in the study of media effects, the specific message matters as much as the medium. As local news continues to decline in the digital age, these historical results offer a cautionary template for the future of democratic representation.
Link: National content and local political consequences: Evidence from public and private television
