The effects of the minimum wage (MW) on the labour market have been studied for almost a century, but the results remain inconsistent. To review what is known until now, a global, multilingual database of 588 articles on MW effects is constructed spanning more than a century, from 1900 to 2020. Most of the studies belong to developed countries, and less than a third come from developing countries. The most researched subject during the period under review is the impact of the MW on employment, but the evidence is mixed independently of the country studied. Consequently, it is only possible to perform a meta-regression to suitably evaluate the impact of the MW on employment according to the number of studies collected on this topic. The meta-regression model is useful to explain the reasons for these differences. For this purpose, recently developed meta-analysis methods are implemented distinguishing between developed and developing countries. To the best of our knowledge, this is the first meta-analysis that reviews articles on the effect of MW without limits on the analysis period, the variables relative to the labour market or the countries under consideration. The results show a negative publication bias for developed countries but not for developing countries. The impact of the MW on employment is negative in both groups, adjusting for that bias. This adverse effect is small but robust. Characteristics related to the control variables, the demographic groups concerned, the methods implemented and the inclusion of local fixed effects, trend or time dummies account for the estimated employment effect. Public policy recommendations are proposed based on the studies reviewed.