In observational studies, early menopause is associated with lower forced vital capacity (FVC) and a higher risk of spirometric restriction, but not airflow obstruction. It is, however, unclear if this association is causal. We therefore used a Mendelian randomisation (MR) approach, which is not affected by classical confounding, to assess the effect of age at natural menopause on lung function.We included 94 742 naturally post-menopausal women from the UK Biobank and performed MR analyses on the effect of age at menopause on forced expiratory volume in 1 s (FEV₁), FVC, FEV₁/FVC, spirometric restriction (FVC<lower limit of normal (LLN)) and airflow obstruction (FEV₁/FVC<LLN). We used the inverse variance-weighted method, as well as methods that adjust for pleiotropy, and compared MR with observational analyses.The MR analyses showed higher FEV₁/FVC and a 15% lower risk of airflow obstruction for women with early (<45 years) compared to normal (45–55 years) menopause. Despite some evidence of pleiotropy, the results were consistent when using MR methods robust to pleiotropy. Similar results were found among never- and ever-smokers, while the protective effect seemed less strong in women who had ever used menopause hormone treatment and in overweight women. There was no strong evidence of an association with FVC or spirometric restriction. In observational analyses of the same dataset, early menopause was associated with a pronounced reduction in FVC and a 13% higher risk of spirometric restriction.Our MR results suggest that early menopause has a protective effect on airflow obstruction. Further studies are warranted to better understand the inconsistency with observational findings, and to investigate the underlying mechanisms and role of female sex hormones.