Panel Data on TFP and Labor Market Outcomes across Development Levels (2000-2023)

This dataset provides panel data for 115 countries spanning 2000-2023, constructed to analyze nonlinear relationships between total factor productivity (TFP) and labor market outcomes across different development levels. The dataset includes employment variables (vulnerable employment, unemployment, employees, agricultural employment, industrial employment, and services employment), productivity measures (TFP at PPP relative to the U.S. frontier), and control variables (human capital, capital stock, GDP growth, inflation, working-age population, labor regulation, dependency ratio, patents per million, and industrial intensity). Countries are classified into developing and developed economies based on World Bank income classifications. The data supports Panel Smooth Transition Regression (PSTR) and GMM-IV estimation methodologies. Outliers in employment variables have been winsorized at the 1st and 99th percentiles by country to preserve cross-country heterogeneity while mitigating measurement distortions.

本数据集涵盖2000年至2023年115个国家的面板数据，旨在分析不同发展水平下全要素生产率（Total Factor Productivity, TFP）与劳动力市场结果之间的非线性关系。数据集包含就业类变量（脆弱就业、失业、雇员就业、农业就业、工业就业及服务业就业）、生产率测度指标（以购买力平价（Purchasing Power Parity, PPP）计算的相对美国前沿水平的TFP），以及控制变量（人力资本、资本存量、GDP增长率、通货膨胀率、劳动年龄人口、劳动规制、抚养比、每百万人口专利数与产业强度）。本数据集依据世界银行收入分组标准，将样本国家划分为发展中经济体与发达经济体两类。数据支持面板平滑转换回归（Panel Smooth Transition Regression, PSTR）与广义矩估计-工具变量（GMM-IV）两种估计方法。为在保留跨国异质性的同时减轻测度扭曲，就业变量中的异常值已按国家分别在1%和99%分位点处完成缩尾处理。