Data description for 
Earnings Dynamics, Inequality, and Firm Heterogeneity 
by Paul Bingley and Lorenzo Cappellari

The data are drawn from Italian administrative registers held by the Social Security Institute (INPS) and are confidential. Physically these administrative micro data are located on servers at INPS and may not be transferred to computers outside INPS due to data security considerations. Our access is not exclusive and other researchers are allowed to use these data if their research project is approved by INPS, following a procedure described at https://www.inps.it/dati-ricerche-e-bilanci/attivita-di-ricerca/programma-visitinps-scholars. Researchers need to submit a written application to INPS. The application should include a detailed research proposal describing the goals and methods of the project. Once received, applications must be approved by the INPS Research Directorate in order to ensure that data are processed in a manner that protects the confidentiality of registered individuals. To access data for replication purposes researchers should contact the INPS DC Research (dcstudiericerche@inps.it).

We use four archives, which can all be linked thanks to unique identifiers of workers and firms:
1. Workers' registry ("Anagrafica")
2. Firms' registry ("Anagrafica aziende")
3. Employment spells ("Rapporti di lavoro")
4. Firm charactersitics ("Visitinps_aziende")

Data are organized by job spell. In case of multiple spells per year we retain the longest. The data used in the paper refer to male employees born between 1939 and 1982 and aged 25 to 55. The working sample is an unbalanced panel of individual real daily wages between 1985 and 2016, corresponding to 12,339,989 men and 3,067,752 firms, resulting in a total of 152,470,973 observations.

The variables used in the analysis are:
id_soggetto			worker identifier
id_impresa			firm identifier
qualifica			occupation
anno_nascita			year of birth
anno_riferimento		year of observation
anno_inizio_lavoro		year of first job
mese_inizio_lavoro		month of first job
data_assunzione			date of hiring 
data_cessazione			date of termination
imponible			yearly gross earnings 
settimane_lavorate		yearly weeks of work
settimane_utili			yearly weeks actually worked
giorni				yearly days of work
percentuale_parttime		share of part time work
gen-dic				12 binary indicators for job spell active in any given month
id_azienda			plant identifier
ateco_91 			SIC code
provincia_contr			province of work
data_costituzione		date of firm establishment
