ObjectiveTo investigate and contrast the effects of non-invasive brain stimulation (NIBS), including repeated transcranial magnetic stimulation (rTMS) and transcranial direct current stimulation (tDCS), on walking and balance ability in patients with Parkinson’s disease (PD).MethodsThe PubMed, Embase, Medline, Cochrane, CNKI, and Chinese WanFang databases were searched up to June 2022. Quality assessment was performed using the Cochrane Collaboration’s risk-of-bias guidelines, and the standardized mean differences (SMD) or mean differences (MD) for each outcome were calculated.ResultsAmong 32 eligible studies, including 1,586 participants were analyzed in this meta-analysis. The results of the meta-analysis showed that NIBS was effective in improving UPDRS-III scores (MD = −2.07; 95% CI, −2.62 to −1.53; P < 0.00001; I2 = 6%) and variables associated with the ability of walk such as step width (SMD = 0.35; 95% CI, 0.16–0.55; P = 0.0005; I2 = 38%), cadence (SMD = 0.3; 95% CI, 0.05 to 0.55; P = 0.02; I2 = 25%), and 6MWT (MD = 62.86; 95% CI, 39.43–86.29; P < 0.00001; I2 = 0%). In subgroup analyses across intervention types, UPDRS-III scores (rTMS: MD = −2.54; 95% CI, −3.16 to −1.92; P < 0.00001; I2 = 0%; tDCS: MD = −1.20; 95% CI, −1.99 to −0.40; P = 0.003; I2 = 0%) and TUGT time (rTMS: MD = −4.11; 95% CI, −4.74 to −3.47; P < 0.00001; I2 = 0%; tDCS: MD = −0.84; 95% CI, −1.48 to −0.21; P = 0.009; I2 = 0%) significantly improved. Moreover, our results also showed that compared to tDCS, rTMS was more significant in improving UPDRS-III scores and TUGT time (p < 0.05).ConclusionNIBS benefits some walking ability variables but not balance ability in 36 patients with PD. The rTMS significantly improved UPDRS-III scores and TUGT time compared to tDCS. Further studies are needed to determine the optimal protocol and to illuminate effects based on the ideal target brain regions, stimulation intensity, timing, and type of intervention.Systematic review registrationhttp://www.crd.york.ac.uk/PROSPERO/, identifier CRD42022350782.