2025 Theses Doctoral

Essays on Media and Political Persuasion in Autocracies

Rahmani, Bardia

Recent decades have witnessed an explosion in private media in Africa, fueling hopes for political openness in autocratic regimes long dominated by state-controlled media. Yet despite this wave of media liberalization, African autocrats have largely maintained and even strengthened their grip on power. What accounts for the failure of private media to challenge authoritarian rule? The chapters that make up this dissertation shed light on this question. They point to the ways in which autocrats have co-opted private media, turning it from a potential threat into a force for regime stability.

Chapter 1 examines a strategy in which autocrats control private media outlets through proxies in the business world, thereby concealing these outlets' connections to the government and rendering them more credible to regime opponents. Fielding surveys in four African countries, I find that the vast majority of citizens living in autocracies are unaware of the regime's control over proxy-owned outlets. I then present evidence from a series of experiments showing that these knowledge gaps work in the regime's favor: they make opponents more likely to inadvertently consume proxy-owned sources and more likely to be persuaded by their pro-regime news coverage. The results challenge the notion that propaganda only influences an autocrat’s loyal supporters, highlighting conditions under which it reaches and persuades more discerning audiences.

Chapter 2 explores another strategy for controlling private media: suspending critical outlets. I argue that the threat of suspensions generates strong financial incentives for outlets to self-censor. To test this argument, I leverage the 2019 suspension of The Citizen, Tanzania’s most prominent independent newspaper. Applying a difference-in-difference design to a corpus of 30,000 newspaper articles, I find that the suspension caused The Citizen to scale back criticism of the regime and increase positive coverage. I also document indirect chilling effects on other independent outlets. Finally, drawing upon unique data on Tanzanians’ daily media habits, I show that audiences did not respond to the suspension by curtailing their consumption of The Citizen or shifting to other sources. The results imply that the suspension shifted Tanzanians’ media diets in a more pro-regime direction.

In Chapter 3, I explore what happens when citizens in autocracies gain access to independent media. In a field experiment, Tanzanians were randomly assigned one month of free access to 30 newspapers via a mobile news app. The intervention led to a 52 percentage point increase in newspaper readership, with users preferring independent outlets to state-owned sources. Yet rather than foment skepticism, exposure to these independent sources increased respondents' approval of the regime. This effect appears to be driven by participants substituting nominally independent—but self-censoring—newspapers for more critical online sources. The findings highlight why autocrats may tolerate “moderate” independent media: by diverting attention from overtly critical outlets, such media can paradoxically bolster support for the regime.

Taken together, these essays highlight how autocrats use ownership and financial pressure to shape private media coverage in their favor. Crucially, I show that citizens in autocracies continue to perceive private media as independent and credible despite its vulnerability to political capture. As a result, autocrats who co-opt private media may have a greater ability to persuade a skeptical public than those who rely solely on state-run outlets.