Reproducibility package for Productivity Gap Between Women- And Men-Run Private Hotels In Georgia: A Dea Based Meta Frontier Analysis

The paper makes an initial attempt to account for differences in the technologies used by women- and men-run businesses, that is, technological “heterogeneity”, to better understand productivity differences between the two groups. We do so by applying meta frontier analysis to the efficiency of private hotels in Georgia estimated using data envelopment analysis (DEA) methodology. The exercise allows us to distinguish between productivity differences conditional on the available technology to each group (technical efficiency) and due to differences in the available technology (technology gap). We show that gender-based differences in technical efficiency and technology gap are very different in their direction, size, and distribution across low vs. high levels of efficiency. For example, women-run hotels outperform men-run hotels in technical efficiency by 21 percentage points. However, this superior performance is almost fully countered by the inferior technology used by women due to the prevailing socio-cultural and economic environment. We also uncover that the impact of technology gap in widening the productivity gap is much stronger at low levels of efficiency than at higher levels (the “sticky floors” effect). No such evidence is found for technical efficiency or overall efficiency. Thus, the existing literature, which assumes technological “homogeneity”, provides at best an incomplete picture of the true nature of gender-based productivity gaps and at worst, a misleading one. Our main result survives endogeneity checks based on propensity score matching and is robust to several measures of productivity and outlier checks. Policy implications of the findings are discussed.