<p>While many countries adopt performance-based research funding systems to stimulate scientific productivity, the design of performance indicators and funding allocation mechanisms may undermine this goal. This study examines three interrelated issues to test how these components shape institutional behavior and research outcomes: the validity of research indicators, the incentive structures generated by allocation formulae, and the long-term performance effects of asymmetric incentives. It focuses on Romania’s performance-based funding instruments as a critical case characterized by rigid policy transfer, persistent resistance to research reform, and hybrid institutional arrangements. Using a mixed-methods approach that combines indicator validation, simulation analysis of allocation formulae, and dynamic differences-in-differences estimation of post-2011 reform institutional performance trajectories, the study identifies several layers of policy misalignment. First, the most prominent and complex indicator shows no meaningful association with external measures of research performance, indicating limited construct validity. Second, the allocation formulae transform raw indicators into funding in ways that generate size-related advantages, partially insulating large institutions from competitive pressure. Third, dynamic analyses reveal an inverted Matthew effect, whereby the most structurally protected universities exhibit relative performance decline compared to more exposed institutions. These findings contribute to the literature by highlighting the central role of funding formulae in shaping research performance and by providing evidence of how structural protection can undermine long-term institutional effectiveness.</p>