Societies in oil and gas-rich states are typically understood as politically quiescent, coopted by state largesse and repressed by a well-funded security apparatus. In the context of the post-Arab Spring ‘counter-revolution’, the Gulf Cooperation Council states have emphasised a repressive response to societal activism, curtailing access to public space, arresting dissidents, and cracking down on public displays of discontent. Yet although the de-mobilising effect of repression has dominated research on the ‘repression effect’ in the Gulf, the wider literature on repression and opposition has been less conclusive. This paper examines the relationship between oil, repression, and political mobilisation in the contemporary Gulf, drawing from interviews with reformers in Qatar, Bahrain, Oman, and Saudi Arabia, as well as tracing of online activities, published manifestos, and opinion pieces published in local newspapers from 2010-2020. The paper widens its consideration of political activity to include micro-mobilisations and those that occur in private, rather than public spaces. In so doing, the paper traces how Gulf reformers move between different types of activism – including formal participation in elected bodies (where these exist), street demonstrations, online social media campaigns, advocacy via the majlis/diwaniyya system, underground networking, and exile activism – in response to the state use of repression. Findings demonstrate the importance of understanding the ‘repression effect’ as dynamic and impacted by intervening variables. Repression did demonstrate the capacity to decrease street demonstrations (forced demobilisation) by greatly increasing the cost of political action. However, reformers did not necessarily cease opposition activity entirely but rather temporarily shifted to other, less costly, forms of political action. Simultaneously, and as has been previously found in Iraq, West Germany, Russia, and other cases, repression also functioned as a mobilising influence, if groups with a collective identity are specifically targeted or if the individual adheres to an ideology that encourages challenges to state authority. In these cases, even when the state’s power appears overwhelming, citizens sensed that the cost of ceasing political activism was higher than the repressive cost of political action – especially for those whose family members have been personally targeted. By moving between different forms of political activity – and by considering repression as simultaneously mobilising as well as demobilising opposition depending on intervening variables – opposition actors may be more adaptable to coercive governance tactics than we think they are, even if widespread street demonstrations cease following a crackdown.