The data received from Inps are extracted from archives which respond to the managing purposes of the National Institute for Social Security. WHIP building procedure may be seen as a black box which processes the archives provided by Inps and gives as result archives that allow for statistical-economical analysis. If we open the black box we find a block of programs (about a hundred) which carry out four kinds of fundamental operations (and other optional operations) in a completely automatic way:

1. the normalization, cleaning and re-codification of most variables contained in the source archives;
2. firm longitudinal identification;
3. working contracts longitudinal identification;
4. strengthening of the contribution record in the annual data of working contracts.

From the point of view of ease of use, the main passages are the first and the fourth, since they allow the final user to avoid having to untangle the codifications used in the contribution declarations which Inps receives – that are often particularly far from interesting subjects for economists – and which allows one to work on the archives organized per year in a coherent and unambiguous way. From a more substantial point of view, the information produced, the more relevant passages are the second and the third ones, because they produce one of the more important added values of WHIP.

Firm longitudinal identification and firm transformation identification is fundamental to having reliable firm data (and consequently to identifing labour contracts), and is necessary since WHIP has a longitudinal dimension which is superior to the number of years (five) for which Inps builds the record of the Firm Observatory, from which the sample object of supply is extracted. The problem is that between one edition and another the firm identification codes and the composition of the firm itself can change, due to a split up, merger or other corporate transformation, therefore it is necessity to identify longitudinally the firms among the different releases of the Observatory.

Longitudinal identification of labour contracts, for what concerns labour market studies, is probably the greatest added value produced by WHIP. The data, records are organized by contribution declarations produced for a worker, since, simplifying a little, the amount of the contribution represents the most interesting variable for the Institute, which uses it for the computation of retirement benefits and other benefits which are in favour of the insured person. However, a one to one relationship does not exist among the declared contributions events and the labour contracts established by the person (the relationship is many to many). Moreover, the date of beginning of the contract is not explicitly registered, and the date of cessation – even if present – lacks quality, both in terms of missing values and correctness. The steps of the procedure devoted to the longitudinal identification of labour contracts adjust for these problems and produce exactly this: the dates of the beginning and the end of labour contracts established by individuals.