Public expenditure is often assessed through its contribution to economic growth, even when services pursue broader social and welfare objectives. This study examines public expenditure on recreation and sport (COFOG GF08.1) across 17 European countries over 1995–2023.Using recreation and sport expenditure as a share of GDP, the study evaluates linear, polynomial, threshold, smooth-transition, spline, and generalized additive specifications. These are complemented by dependence-robust inference, heterogeneous-panel estimators, turning-point and marginal-effect analysis, bidirectional predictive-direction tests, and a multiverse of 360 model specifications. The results show no statistically robust association between the expenditure share and GDP growth. Greater functional flexibility does not reveal a robust nonlinear relationship, a statistically defensible interior optimum, or marginal associations that differ from zero across the observed expenditure range. These conclusions remain stable across alternative estimators and analytical choices, and when the expenditure share is lagged or replaced by measures that do not use GDP as the denominator. The association varies with economic development: it is negative in higher-income countries and absent in lower-income ones. Predictive-direction evidence is more consistent with economic growth preceding subsequent increases in recreation and sport expenditure than with expenditure preceding growth, although this temporal ordering should not be interpreted as structural causality. The findings challenge the use of aggregate GDP growth as a sufficient criterion for appraising this expenditure category. Public expenditure performance should instead be evaluated against its intended outcomes, including health, participation, accessibility, social cohesion, distributive equity, and quality of life, rather than inferred from aggregate growth alone.
