Summary Background Macrosomia is a risk factor for adverse delivery outcomes. We investigated the prevalence, risk factors, and delivery outcomes of babies with macrosomia in 23 developing countries in Africa, Asia, and Latin America. Methods We analysed data from WHO's Global Survey on Maternal and Perinatal Health, which was a facility-based cross-sectional study that obtained data for women giving birth in 373 health facilities in 24 countries in Africa and Latin America in 2004–05, and in Asia in 2007–08. Facilities were selected by stratified multistage cluster sampling and women were recruited at admission for delivery. We extracted data from the medical records with a standardised questionnaire. We used logistic regression with random effects to assess the risk factors for macrosomia and the risks for caesarean section and adverse maternal and perinatal outcomes (assessed by a composite score) in babies with the disorder. Findings Of 290 610 deliveries, we analysed data for 276 436 singleton livebirths or fresh stillbirths. Higher maternal age (20–34 years), height, parity, body-mass index, and presence of diabetes, post-term pregnancy, and male fetal sex were associated with a significantly increased risk of macrosomia. Macrosomia was associated with an increased risk of caesarean section because of obstructed labour and post-term pregnancy in all regions. Additionally, macrosomia was associated with an increased risk of adverse maternal birth outcomes in all regions, and of adverse perinatal outcomes only in Africa. Interpretation Increasing prevalence of diabetes and obesity in women of reproductive age in developing countries could be associated with a parallel increase in macrosomic births. The effect and feasibility of control of diabetes and preconception weight on macrosomia should be investigated in these settings. Furthermore, increased institutional delivery in countries where rates are low could be crucial to reduce macrosomia-associated morbidity and mortality. Funding None.