The objective of the present study was to conduct a meta-analysis that summarizes the effect of reducing dietary rumen degradable protein (RDP) proportions on dry matter intake (DMI) and milk production in dairy cows. The data set was identified using 2 search engines comprising of 41 studies with 109 observations. Means were weighted by the inverse of their variance. Treatments of RDP (% of DM) were organized as very low (<8% RDP), low (8% ≤ RDP <9%), medium (9% ≤ RDP <10%), and high (>11% RDP) and compared with a control (10% ≤ RDP ≤11%). Comprehensive Meta-Analysis v3 (Biostat, Englewood, NJ) was used to evaluate the raw mean difference (RDP treatment – RDP control) in a random-effects model. Variables evaluated were DMI and yields of milk and milk protein. Moderators included were experimental design of the studies (change-over or continuous), concentration of dietary energy [medium (<1.6 Mcal of net energy of lactation per kg of DM] or high (≥1.6 Mcal of net energy of lactation per kg of DM)], and concentration of dietary rumen undegradable protein (RUP; < 6 or ≥6% of DM). Statistical significance was identified at P ≤ 0.05. Heterogeneity was present, and publication bias was absent. High RDP did not affect DMI and production (n = 47). Compared with control RDP, low RDP sustained DMI and milk production (n = 12), and medium RDP sustained yields of milk and milk protein (n = 40). However, medium RDP reduced DMI in diets with medium and high energy content (0.45 kg/d; P < 0.02). Low and high RDP reduced DMI and milk production in continuous design trials (n = 56; P < 0.05); whereas, RDP did not affect the evaluated variables in change-over trials. In trials with RUP <6% (n = 26), medium RDP decreased DMI (0.50 kg/d; P < 0.03), but at RUP ≥6%, lowering RDP did not affect production. In summary, lowering RDP proportions may not always reduce DMI and milk production in dairy cows. However, experimental design of the study, dietary energy, and dietary RUP may influence production responses to lowering RDP.