Demand-Mediated Rather Than Productivity-Led Wages: Evidence from Albania

This replication package supports the study Demand-Mediated Rather Than Productivity-Led Wages: Evidence from Albania. It contains the final manuscript, original and harmonised datasets, intermediate data transformations, and exported econometric results. The principal annual dataset covers Albania over 2000–2024, although the effective estimation period varies according to data availability and the lag structure. The variables include private- and public-sector wages, the public–private wage differential, gross domestic product, labour productivity, consumer prices, unemployment, investment, final consumption, remittances, migration-related indicators, human development, capital, labour, education and technology measures. The original data originate from the Institute of Statistics of Albania (INSTAT), the World Bank’s World Development Indicators, Eurostat and the author’s dissertation databases. The harmonised files document the construction of consistent annual series, logarithmic and growth-rate transformations, indexed variables and supplementary macroeconomic indicators. The results files contain ARDL and UECM estimates, lag-selection exercises, bounds tests, unit-root and residual diagnostics, correlation and variance inflation factor analyses, principal-component results, fixed-sample robustness checks, European wage-benchmark comparisons and supplementary historical tests. The package is organised into manuscript, raw data, clean/intermediate data and model-results sections. A SHA-256 inventory is included to verify the integrity and provenance of every file.本复现数据包用于支撑《需求介导而非生产率主导的工资：来自阿尔巴尼亚的证据》这项研究。本数据包包含研究的最终手稿、原始与标准化数据集、中间数据转换脚本，以及导出的计量经济学结果。核心年度数据集覆盖2000年至2024年的阿尔巴尼亚数据，但实际估计时段会根据数据可得性与滞后结构进行调整。所涉变量涵盖私营与公共部门工资、公私部门工资差距、国内生产总值、劳动生产率、消费者价格、失业率、投资额、最终消费、侨汇、移民相关指标、人类发展水平、资本、劳动、教育与技术测度指标。 原始数据源自阿尔巴尼亚国家统计研究所（Institute of Statistics of Albania, INSTAT）、世界银行世界发展指标、欧盟统计局（Eurostat）以及作者学位论文数据库。标准化文件记录了一致化年度序列的构建流程、对数与增长率变换、索引化变量以及补充宏观经济指标的生成方法。结果文件包含自回归分布滞后模型（ARDL）与无约束误差修正模型（UECM）的估计结果、滞后选择检验、边限检验、单位根检验与残差诊断、相关性与方差膨胀因子分析、主成分分析结果、固定样本稳健性检验、欧洲工资基准对比以及补充历史检验。本数据包分为手稿、原始数据、清洗/中间数据以及模型结果四个模块。数据包附带SHA-256校验清单，用于验证所有文件的完整性与来源。