Manufacturing-Led Growth in Mauritania A Case Study
Description
This dataset accompanies the research paper "Manufacturing-Led Growth in Mauritania: A Case Study (1985–2025)." It investigates the relationship between manufacturing-sector development, economic growth, employment reallocation, and labor productivity in Mauritania over the period 1985–2025. The primary research hypothesis, based on Kaldor's first law of growth, is that manufacturing value-added growth is positively associated with GDP growth. The null hypothesis assumes no statistically significant relationship between these variables, while the alternative hypothesis predicts a positive relationship. A second research objective is to determine whether employment reallocation between manufacturing and non-manufacturing activities contributes to aggregate labor productivity growth. The dataset contains annual macroeconomic and labor-market observations collected from publicly available international databases, including the World Bank's World Development Indicators (WDI), the International Labour Organization's ILOSTAT database, and the United Nations Industrial Development Organization (UNIDO). The main indicators are GDP, manufacturing value added, manufacturing employment, total employment, and manufacturing trade statistics. Annual growth rates are calculated from GDP and manufacturing value added expressed in current US dollars. Sectoral labor productivity is measured as value added per employed worker. Some missing employment observations are estimated using a log-linear time-trend model. The empirical methodology combines ordinary least squares (OLS) regression with a two-sector shift-share decomposition. The regression evaluates the statistical association between manufacturing value-added growth and GDP growth. The shift-share decomposition separates aggregate labor productivity changes into within-sector productivity effects and structural-change effects associated with employment reallocation. The reported regression coefficient for manufacturing growth is positive (0.0835) but statistically insignificant (p = 0.5782), with an R² of 0.0082. Therefore, the results do not provide sufficient statistical evidence supporting manufacturing-led economic growth in Mauritania. The shift-share decomposition attributes approximately 99.70% of the measured productivity change to within-sector improvements and only 0.30% to employment reallocation. These findings indicate that manufacturing expansion has not been demonstrated to be a significant driver of economic growth during the study period, while changes in productivity within existing sectors dominate the measured structural transformation process. The dataset can support replication, descriptive analysis, productivity decomposition, and comparative research on structural transformation in resource-dependent developing economies. However, current-dollar measures may reflect inflation and exchange-rate movements, and estimated employment observations introduce uncertainty.
Files
Steps to reproduce
The dataset was compiled from publicly available secondary data sources to investigate manufacturing-led economic growth and structural transformation in Mauritania over the period 1985–2025. The main sources were the World Bank's World Development Indicators (WDI), the International Labour Organization's ILOSTAT database, and the United Nations Industrial Development Organization (UNIDO). **Data collection:** Annual observations on gross domestic product (GDP) and manufacturing value added were obtained from WDI. Manufacturing employment and total employment data were obtained from ILOSTAT, while manufacturing trade indicators were sourced from UNIDO. The observations were organized chronologically by year and matched to construct a consistent national time-series dataset. No primary surveys or experimental instruments were used. **Data preparation:** GDP and manufacturing value added were expressed in current US dollars. Annual growth rates were calculated as the percentage change between consecutive observations: Growth Rate = [(Xₜ − Xₜ₋₁) / Xₜ₋₁] × 100. Manufacturing labor productivity was calculated by dividing manufacturing value added by manufacturing employment. Non-manufacturing value added and employment were derived by subtracting manufacturing values from their respective national totals. Missing employment observations were estimated using a log-linear time-trend model, in which the natural logarithm of employment was regressed on time using ordinary least squares (OLS). Predicted values were transformed back into employment levels. These estimated observations introduce uncertainty and should be distinguished from directly observed data. **Econometric analysis:** Kaldor's first law was examined using an OLS regression of annual GDP growth on manufacturing value-added growth. A two-sector shift-share decomposition was applied to separate changes in aggregate labor productivity into within-sector productivity effects and structural-change effects associated with employment reallocation between manufacturing and non-manufacturing activities. **Reproducibility:** The analysis can be reproduced by retrieving the corresponding annual series from WDI, ILOSTAT, and UNIDO, aligning observations by year, applying the stated transformations, estimating missing employment observations, and repeating the OLS regression and shift-share decomposition. Researchers should document source revisions, missing values, and estimated observations. The use of current-dollar output measures should also be considered when interpreting the results, since nominal changes may reflect price and exchange-rate movements rather than real production growth.
Institutions
- University of BelgradeCentral Serbia, Belgrade