Just Above the Line: Grant Panels, the Uncertainty Discount, and the Return of Patronage in the Language of Peer Review

This volume extends the Consolidation Trap into research funding by formalizing the Uncertainty Discount, underlying three independently documented empirical findings. The Uncertainty Discount names the tendency of funding panels, required to convert an inherently uncertain judgment into a scalar decision under time pressure, to resolve that uncertainty asymmetrically: not as a symmetric unknown, but as a negative signal resolved by substituting an available proxy for direct assessment. The volume documents three such proxies. The first is past funding success itself. Bol, de Vaan, and van de Rijt (2018), using a regression-discontinuity design on a €2 billion Dutch grant program, find that early-career scientists who received an early grant by a margin indistinguishable from chance accumulated more than twice as much subsequent funding over eight years as observationally identical scientists who narrowly missed the same threshold. The second proxy is demographic prior. Ginther and colleagues (2011), analyzing the complete NIH R01 applicant pool, find Black applicants thirteen percentage points less likely than white applicants to receive funding, a gap narrowing to ten points but not closing after extensive controls, concentrated specifically at the review-scoring stage. The third proxy is intellectual familiarity, more precise than a simple novelty penalty: Boudreau, Guinan, Lakhani, and Riedl (2016), using a randomized experiment generating 2,130 evaluator-proposal pairs, find that evaluators score down proposals both too close to and too distant from their own expertise. The volume formalizes the discount as a decomposition of funding score into a quality signal and a proxy-driven uncertainty penalty, and situates the mechanism historically: the mid-twentieth-century transition from foundation and aristocratic patronage to formalized, panel-reviewed grant allocation, inaugurated by Vannevar Bush's 1945 report and the 1950 founding of the National Science Foundation, reproduces in the vocabulary of statistical proxy substantially the same distributive outcome the reform was designed to end.