This paper reviews research that uses longitudinal microdata to document productivity movements and examine factors behind productivity growth. The research explores the dispersion of productivity across firms, the persistence of productivity differentials, the consequences of entry and exit, and the contribution of resource reallocation to aggregate productivity growth. It also reveals factors correlated with productivity growth, such as managerial ability, technology use, human capital, and regulation. The more advanced literature addresses the causality between these factors and productivity growth. The review highlights the importance of large, representative microdata sets like the Longitudinal Research Database (LRD) for addressing questions that cannot be answered with aggregate data. The paper discusses the advantages of LMDs, including their large sample sizes, time series data, and ability to link data from other surveys. It also addresses methodological issues in measuring productivity, such as the choice between labor productivity and total factor productivity (TFP) and the challenges of measuring inputs and outputs at the micro level. The paper presents stylized facts on productivity dispersion and evolution, noting the significant heterogeneity in productivity across establishments and firms. It discusses various models of industry dynamics and the statistical descriptions of productivity evolution, including random productivity shocks, vintage capital models, and common evolution of heterogeneous plants. The paper concludes by summarizing the findings and suggesting directions for future research.This paper reviews research that uses longitudinal microdata to document productivity movements and examine factors behind productivity growth. The research explores the dispersion of productivity across firms, the persistence of productivity differentials, the consequences of entry and exit, and the contribution of resource reallocation to aggregate productivity growth. It also reveals factors correlated with productivity growth, such as managerial ability, technology use, human capital, and regulation. The more advanced literature addresses the causality between these factors and productivity growth. The review highlights the importance of large, representative microdata sets like the Longitudinal Research Database (LRD) for addressing questions that cannot be answered with aggregate data. The paper discusses the advantages of LMDs, including their large sample sizes, time series data, and ability to link data from other surveys. It also addresses methodological issues in measuring productivity, such as the choice between labor productivity and total factor productivity (TFP) and the challenges of measuring inputs and outputs at the micro level. The paper presents stylized facts on productivity dispersion and evolution, noting the significant heterogeneity in productivity across establishments and firms. It discusses various models of industry dynamics and the statistical descriptions of productivity evolution, including random productivity shocks, vintage capital models, and common evolution of heterogeneous plants. The paper concludes by summarizing the findings and suggesting directions for future research.