Database characteristics

The Work Histories Italian Panel (WHIP: see www.laboratoriorevelli.it/whip) is a database of individual work histories set up by the University of Turin in collaboration with LABORatorio Riccardo Revelli and INPS. The reference population includes all the individuals who had a job for which a contribution to INPS was recorded. This population undergoes systematic and random sampling based on four previously determined dates of birth. WHIP is thus representative of the reference population and its high degree of coverage (around one in 90) ensures a sample size that is usually hard to reach in surveys. The dynamic population – that is, the overall number of individuals observed in the period covered by the series – amounts to over 750,000 workers.

The information contained in WHIP includes main personal data about each individual (age, gender, place of birth and place of residence), along with data concerning their work (beginning and end of the work relationship, wage, type of work contract, applied collective labour agreement, occupation, and pay grade) and, for employees only, data on the firm the individual is employed by (firm start-up and closure dates, average number of employees, business sector).

The main value added of the database is represented by the longitudinal identification of labour relationships. For all the individuals included in the sample, the main steps in their career are observed: periods of employment, of self-employment, and of independent contract work, transition towards retirement, as well as periods in which the insured benefited from social protection measures (unemployment benefit, and mobility, sickness and maternity allowances). Moreover, the available information is extremely reliable, which rarely happens in surveys, in particular for variables that interviewees may misperceive or be reluctant to disclose, such as salary.

Despite its richness and precision, WHIP has two limitations. The first is an external limitation, due to the impossibility of observing public employees with open-ended contracts, as well as workers in the agricultural sector, some categories of professionals (since they pay contributions to a different social security fund), and, obviously, irregular work, with the consequence that there might be ‘gaps’ in some work histories, indistinguishable from non-employment. The existence of this limitation makes it impossible to use the WHIP database to observe the entire Italian labour market, although a vast part is covered: all the private non-agricultural sector and a portion of the public sector – the one entailing fixed-term relationships, be they dependent or not.