Anxiety is one of the most frequently reported co-occurring conditions for autistic children and adolescents. The relationship between anxiety and social outcomes in autistic youth has been the focus of a range of studies, with mixed results. This meta-analysis aimed to identify the strength of the association between anxiety and a frequently researched social outcome (social competence) in autistic young people and whether that association is influenced by individual or research design factors. A previous preregistered systematic review was updated with a search of the same three databases (CINAHL, ERIC, and PsycINFO) as the original review. Through this, 20 studies with sufficient data on a neurotypically-defined measure of social competence and anxiety were identified. Results were synthesised using a mixed effects model. The meta-analysis on 2,321 participants (from 22 samples) highlighted wide heterogeneity in results. The findings show that anxiety has a significant, small negative impact on social competence (d = − 0.48; 95% CI = − 0.71, − 0.26), meaning that as scores on measures of anxiety increase, scores on measures of social competence decrease. This relationship between anxiety and social competence was moderated by age, becoming weaker as age increased. Whilst this is an important finding for supporting mental health and well-being of autistic young people, the large amount of variance left unexplained suggests that multiple factors, including the use of measures designed for neurotypical people and the potential impact of camouflaging on such measures, need to be considered in future designs.