Evaluating the macro-representativeness of a firm-level database: an application for the Spanish economy

The availability of a firm-level database that represents the productive sector of an economy at the aggregate level is a necessary condition to undertake both reliable policy analysis and economic research in multiple areas. In this paper, we document the construction of a new representative firm-level dataset for Spain using detailed micro-level information provided by firms to the Spanish Commercial Registry and the Bank of Spain. A comparison with National Accounts figures serves to illustrate that the new micro-dataset is able to replicate the growth rates of output, employment and wage bill of the private sector. Using official statistics from the National Institute of Statistics (INE), we show that the resulting dataset covers more than 80% of firms registered in the census over the years 2000-2013 and, more importantly, the resulting dataset replicates the firm size distribution of the Spanish non-financial market economy. The same representativeness analysis is done for the manufacturing sector indicating that this sector is particularly well-represented in the dataset.