This descriptive paper evaluates a specific hypothesis of the U-shaped female labour force function developed by Claudia Goldin (1994). The hypothesis suggests that in middle-income countries like Mexico, female labour force participation rates (FLPRs) tend to be low due to the high percentage of industrial jobs in the country and a social stigma towards married women working in blue-collar jobs. The regression analysis uses microdata from Mexico's ENOE household survey, and it is based on a repeated cross-sectional dataset from the first quarters of 2005, 2010, 2015, and 2019. The empirical strategy relies on probit regressions to estimate woman's likelihood of being economically active depending on the percentage of jobs in agriculture, industry, and services in the municipality where they live. Contrary to the hypothesis, the results show that a higher percentage of industrial jobs at the municipal level is associated with higher female labour participation. Moreover, a higher percentage of jobs in the service sector exhibits a stronger positive relationship. Conversely, women's probability of being economically active decreases as the percentage of agricultural jobs in the municipality increases. The results were obtained after considering both paid and unpaid work in farms and businesses, and they hold after controlling for individual, household, and municipal characteristics. Disaggregated data suggests that the lack of labour demand in rural areas of Mexico could be one of the reasons behind the low participation of women in agricultural activities. Hence, this paper shows that there is an upward trend between FLPRs and different stages of economic development, since agricultural regions have the lowest FLPRs, and service-oriented regions have the highest. This represents new evidence for the literature that have usually find a U-shaped relationship in both cross-country and within-country studies.