Background Observational studies have shown that body mass index (BMI) is positively associated with asthma. However, observational data are prone to confounding and reverse causation. In Mendelian randomization, genetic variants are used as unconfounded markers of exposures to examine causal effects. We examined the causal effect of BMI on asthma, hay fever, allergic sensitization, serum total immunoglobulin E (IgE), forced expiratory volume in one‐second (FEV1) and forced vital capacity (FVC). Methods We included 490 497 participants in the observational and 162 124 participants in the genetic analyses. A genetic risk score (GRS) was created using 26 BMI‐associated single nucleotide polymorphisms (SNPs). Results were pooled in meta‐analyses and expressed as odds ratios (ORs) or β‐estimates with 95% confidence interval (CI). Results The GRS was significantly associated with asthma (OR=1.009; 95% CI: 1.004, 1.013), but not with hay fever (OR= 0.998; 95% CI: 0.994, 1.002) or allergic sensitization (OR=0.999; 95% CI: 0.986, 1.012) per BMI‐increasing allele. The GRS was significantly associated with decrease in FEV1: β=−0.0012 (95% CI: −0.0019, −0.0006) and FVC: β=−0.0022 (95% CI: −0.0031, −0.0014) per BMI‐increasing allele. Effect sizes estimated by instrumental variable analyses were OR=1.07 (95% CI: 1.03, 1.10) for asthma, a 9 ml decrease in FEV1 (95% CI: 2.0‐15 mL decrease) and a 16 ml decrease in FVC (95% CI: 7.0‐24 mL decrease) per 1 kg/m2 higher BMI. Conclusions The results support the conclusion that increasing BMI is causally related to higher prevalence of asthma and decreased lung function, but not with hay fever or biomarkers of allergy.