WHIP is a database of individual work histories, based on Inps administrative archives.
The reference population is made up by all the people – Italian and foreign – who have worked in Italy even only for only a part of their working career. A large representative sample has been extracted from this population: in the standard file the sampling coefficient is about 1: 180, for a dynamic population of about 370,000 people (figures will be doubled in the full edition).
For each of these people the main episodes of their working careers are observed. The complete list of observations includes: private employee working contracts, atypical contracts, self-employment activities as artisan, trader and some activities as freelancer, retirement spells, as well as non-working spells in which the individual received social benefits, like unemployment subsidies or mobility benefits. The workers for whom activity is not observed in WHIP are those who worked in the public sector or as freelancers (lawyers or notaries) – who have an autonomous security fund.
The WHIP section concerning employee contracts is a Linked Employer Employee Database: in addition to the data about the contract, thanks to a linkage with the Inps Firm Observatory, data concerning the firm in which the worker is employed is also available.

WHIP comes from administrative source data. In fact the data on which it is founded is a sample collection extracted from the National Institute of Social Security’s (Inps) administrative archives, and have been provided to LABORatorio Revelli according to an agreement signed between Inps and the University of Torino. The administrative archives from which data have been extracted are:

* unified individual register
* employees' contributions
* artisans’ contributions
* traders' contributions
* contributions of the IV gestione separata (atypical workers)
* retirement benefits
* other social security benefits (unemployment subsidies etc.)
* business register
* monthly contributions of firms

The global supply is rather huge – the quantity is around 6 million records provided, for 1.400 million observations (record per variable) – and is growing with new extractions (see the "future development" section). Inps provides statistical observatories on most part of these archives, the results of which are published online (see the statistical databases pages on the Institute website), where it is possible to consult the methodological notes, which are useful as documentation for WHIP input archives.

The reference population is made up by all the people – Italian and foreign – who have worked in Italy even only for only a part of their working career.
For each of these people the main episodes of their working careers are observed. The complete list of observations includes: private employee working contracts, atypical contracts, self-employment activities as artisan, trader and some activities as freelancer, retirement spells, as well as non-working spells in which the individual received social benefits, like unemployment subsidies or mobility benefits. The workers for whom activity is not observed in WHIP are those who worked in the public sector or as freelancers (lawyers or notaries) – who have an autonomous security fund.

LABORatorio R. Revelli has implemented two policies to make WHIP available, a standard file (WHIP FS) and the full version of the database (WHIP FE).

Diffusion policies of the standard file
The main data diffusion policy was realized through the construction of a standard file of WHIP (WHIP FS), composed by anonymous micro data according to art. 1 of law 675/1996 on the safeguarding of privacy . After the processing of Inps data, the data contained in the WHIP standard file can't be associated with identified or identifiable subjects, and are therefore classified as anonymous data. This allows their diffusion and guarantees their full availability for researchers.
The files are available from the page download . Users need to fill out the request for data access. Upon receipt of the signed statements LABORatorio R. Revelli will give a username and password enabling users to download the data. The request form can be send by mail or fax at the following address: LABORatorio Riccardo Revelli - Via Real Collegio 30, 10024 Moncalieri (To), Italy Fax +39 011 6705061

Diffusion policies for the full version

- Public computer facilities
The Fondazione Collegio Carlo Alberto in Moncalieri has set up public computer facilities to allow students and researchers access to the complete version of WHIP and other databases which belong to the Fondazione. Currently two workstations are available. For additional information or to reserve time at the workstations, please contact the secretariat of the Fondazione (e-mail: databankaccess@collegiocarloalberto.it)

- Distribution for specific projects
Universities, Public and Private Research Institutes, Scientific Societies and individual researchers in the field can ask for access to the complete version of WHIP, based on the presentation of a research plan, in compliance with the ethical norms already cited.
To obtain access to the full edition of WHIP it is necessary to do the following:

1. Fill out the request for data form (download form instantly), which must be signed by the person requesting access or (in the case of a group or company with a legal representative) by the legal representative, where the statistical and scientific objective of data use is clear;
2. Fill out the research protocol (download form) with information about and the signatures of all of the researchers that will be involved in the project;
3. Prepare a detailed research proposal where the specific goals and the benefits of using the full edition of WHIP are highlighted. This proposal should state clearly, in accordance with standard methodology related to discipline, that the treatment used is appropriate for scientific or statistical aims.
4. Send via mail the three documents described above to the following address: LABORatorio Riccardo Revelli - Via Real Collegio 30, 10024 Moncalieri (To), Italy. Please send an e-mail to whip@laboratoriorevelli.it to inform us of the letter posted.
An evaluation group created at the LABORatorio Riccardo Revelli will decide if access will be granted to the full edition of WHIP, based on merit and evaluation of the qualifications of those making the request and of the research proposed.
For the access to the Full Edition of WHIP a contribution of 1800 € is requested; while the contribution requested for updates to users who have already received the Full Edition of WHIP is 240 €.

The files are available from the page download. Users need to fill out the request for data access. Upon receipt of the signed statements LABORatorio R. Revelli will give a username and password enabling users to download the data. The request form can be send by mail or fax at the following address: LABORatorio Riccardo Revelli - Via Real Collegio 30, 10024 Moncalieri (To), Italy Fax +39 011 6705061

To access the full version of WHIP data please follow the instruction reported in the section "Diffusion policies for the full version".

You have to state, in every research report, article or publication, in which the data have been used, the words: 'Work Histories Italian Panel (WHIP)', or the brief indication 'WHIP', and to indicate, in occasion of the first quotation, the consequent complete reference: 'WHIP – Work Histories Italian Panel, is a database of work histories developed by Laboratorio Revelli – Centre for Employment Studies. See http://www.laboratoriorevelli.it/whip'.

Yes, users of WHIP FE and WHIP FS are requested to send a copy of the reports or of the articles produced using the provided data, by mail to Laboratorio R. Revelli, Via Real Collegio 30, 10024 Moncalieri (To), Italy, or through electronic format to whip@laboratoriorevelli.it;

The Laboratorio R. Revelli, with respect to the regulations concerning the copyrights, will give news of the publication through its own website.

Here >>> you can find the WHIP documentation organized into three sections.

- The metadata, which contains a description of all archives and variables which compose WHIP.
- A technical reports section (forthcoming), which will also accept contributions from users who have results related to data issues, or comparisons between WHIP and other databases.
- ... The Frequently Asked Questions where we are now
If you have not found an answer to your question, please fill a "I Have a Question" (IHaQ) form.

To be up to date whit the future developments of WHIP, you can subscribe to the WHIP mailing list by sending an email to whip@laboratoriorevelli.it with subject subscribe whip news or by directly subscribing at our website www.laboratoriorevelli.it/whip.

All information about WHIP, including future developments and updating, will be reported promptly on our website.

If there is something wrong or something you want to ask us please send us an "I Have a Question" (IHaQ) form or write us at whip@laboratoriorevelli.it.

If something it is not clear to you please send us an "I Have a Question" (IHaQ) form or write us at whip@laboratoriorevelli.it

The variable to be used to link the individual spells through all the archives is id_individual.
The variable to be used to link together the job spells through all the archives is id_job.

There is no attrition in this archive, if we exclude updating problems, i.e. delays in the acquisition of information from the firms. It is compulsory to provide records on employees and firms to the social security administration, if the worker and the firm belong to one of the mentioned categories.

The observed period goes from 1985 to 2004; however each archive covers different periods. On the website the observed period as well as the future developments for each archive is reported.

The WHIP database does not provide complete information on spells of non-employment.
The most likely possibility for people of working age who is not in any of the WHIP archives is unemployment without benefits and/or illegal work. There are four other possibilities (i) periods spent out of the labour force; (ii) moves into freelance activities not covered by INPS; (iii) moves into the public sector; (iii) moves into the agricultural sector. Temporary moves into each of these four categories are plausible, but unlikely.

In the FS version of WHIP the information about the birthplace and the job place of the worker are divided into geographic areas (north-west, north-east, centre, etc.)
In the WHIP FE these infomation are available at a regional level. It isn’t possible to get information about the birthdate and and the job place at a municipal or at a provincial level.

The activities of self-employment or dependent self employment can coexist without any problems, so it is normal to find them overlapping.
As regards dependent employment, teorically a full-time employee should not have more jobs simultaneously.
However, sometimes it is possible to observe full-time jobs overlapped, or a full-time job overlapped with part-time jobs. Since Inps has registered contributions paid for these job spells, in the WHIP construction procedure they have been maintained.

Since 1998 on, contribution rates are different according to the fact that the worker is contributing or not to another social security pension fund.
In 1996 and 1997 the contribution rate was 10% for all workers in this archive. In 1998 and 1999 contribution rate was 12% for workers contributing to this social security fund uniquely, while it was 10% for those workers registered also in another social security fund. The variable takes value of 22 for workers contributing during the year under both regimes.

For the collaborators it shows the geographic area of the place where the collaboration activity is carried out, for the self-employee it shows the location of the registered office.

The ’compensation’ variable relative to an ’employer-employee relationship’ is the sum of the current charges and of other charges. The current charges regard the overall amount of the monthly wages due in the solar year, both entire and reduced (wages, cost of living allowance, additional fees, etc.).
The other charges are equal to the overall amount of the non monthly charges (remuneration such as 13th or 14th wages, salary arrears referred to previous years, etc.)

No, it doesn’t. The variable compensation represents the yearly net sum perceived by the employee, without the social contributions paid by the employer, and inclusive of the income tax and social contributions paid by the employee.

The information on the firms are drawn from the Firms’ Observatory published by Inps. At the moment the data are available from 1987 to 1998.

Firms’ legal transformations can originate "spurious" movements of workers; '. From the workers’ viewpoint, job changes will be observed in administrative archives that may have not actually occurred. The WHIP data base detects and corrects legal transformations with two different procedures. The first is based on firms data, the second is based on workers data once individual job spells have been built.

1. Firm longitudinal identification and identification of the firm transformation is fundamental to obtain reliable firm data (and consequently to identify the labour contracts) and is necessary because WHIP has a longitudinal dimension that is superior to the number of years (five) for which INPS build the Firm Observatory records and from which the sample is drawn. The problem is that among different editions of the Observatory the firm identifying codes can change as well as the firm composition because of breakups, mergers and other firm transformations, hence the need of proceeding to the longitudinal identification of the firms among the different Observatory editions.

2. The possibility offered by the archive to observe both individual employees and the firms for which they work enables to deduce the existence of other underlying legal transformations from the contemporaneous flow of workers between two firms. The key is to discriminate between 'normal' movements, deriving from workers' decisions to change jobs, and 'spurious' movements. It is intuitively unlikely that 'many' workers of a company independently and simultaneously decide to move together to another firm, whereas this event will take place if all or part of the activities of the first firm is transferred to the second firm. In order to identify spurious components a threshold for the intensity of such movements has been established. In the WHIP archive the whole population of employed persons is not available as it is a 1:90 sample; however the observation of two workers who move within one month from the same firm (call it A) to a same firm (call it B) would statistically means that on average firm A has handed over about 200 workers to firm B. Thus, if we observe in the same month at least two workers move form firm A to firm B we call it a spurious movement. Once we detect spurious movements in this way, we proceed in keeping connected the job spells of every worker who has made the same movement some months before or after.

Before 1997 the status of Cadre was not distinct, in the Inps forms, to the one of white collar. From 1997, instead, it is pointed out as a separate skill level.

The information about temporary work relationships are in the jobs_yearly archives.Temporary work is identified with the value 4 in the variable cont_rebate.

In the FS version of WHIP the firm’s dimension is aggegated in classes; in the FE version of WHIP for each worker present in the archive in a fixed year it is available the information about the average number of employees in the firm, obtained as the average of the monthly number of employees.

No, it isn’t. The data from which the infomation for WHIP are drawn don’t provide for the moment this information.

At the moment it isn’t possible to have this information in WHIP data. Soon for some workers a series of information about the household will be available, drawn form ISEE (indicator of the equivalent economic situation) statements. These information will concern not only the composition of the household (family number, status...) but also economic information, as incomes, financial assets and real estates.

No, the top coding doesn’t change from year to year.

If the applied contract has the 13th o 14th wages, these amounts are included in the variable remuneration.

Because of problems due to the way of registration of the illness benefit cases, data before 1989 are not to be considered complete.

To build the job_yearly archive it is necessary not only to have information about the contributions and the period to which they refer, but also to have the information about the employer who pays for contributions. At present the employee are updated to 2004, and have converged in the earnings archive.
The firm data are drawn from the Firm Observatory published by Inps, available at the moment only up to 1998.

Part-time workers are identified by the variable ptime.

Yes, it does.

For each year of service, a provision of 1/13.5 of the yearly gross salary is included in the individual T F R account, and is yearly re-valuated according to the following coefficients: 1.5% fixed plus 75% of the CPI inflation of the previous year.
More information about TFR fund in WHIP can be found in Garibaldi Pacelli (2007) .

In the jobs_yearly archive just one record is reported for each employer-employee relationship in the year. If during the year some characteristics in the relationship have modified (for example there has been a change in the skill level, in the contract, in the job location etc.) the information referred to the last registered characteristic is reported. For example, if a worker passes from a qualification of "cadre" to the one of Manager, during the year, in the record referring to that job relationship in that year one will find the last skill level registered, i.e. that of Manager.

The job_spells archive contains the list of all job spells that an individual had in the period covered by whip, with the dates of beginning and end of the spell. The other information on the spell (firm characteristics, wage, contract etc.) are included in thejobs_ yearly archive.
The univocal key to link the data of the job spell and the yearly data is id_rl.

The dep_self_employed archive includes mainly workers with incomes from:
- collaborazione coordinata e continuativa o a progetto
- professional activity
The self employed are Vat holders without a social security fund; the others are defined collaborators.
More details on the categories included in this archive can be found on the INPS website (www.inps.it) in the section of the ’parasubordinati’ and in Berton, Pacelli e Segre (2005) 'Il lavoro parasubordinato in Italia: tra autonomia del lavoratore e precarietà del lavoro', Rivista Italiana degli Economisti, n. 1, 2005.

For the problems linked to the way of registraton of the maternity events, maternity data before 1989 are not to be considered as complete.

Data on the size and sector of activity of the firm in the jobs_yearly archive refer to the firm on the whole and not to the local unit where the worker is employed for which no information is available.
The data on the localization (variable firm_reg or firm_area) show the registered office of the company.

No, it isn’t possible to know what kind of part-time is carried out by the worker. And it is not possible to identify exactly the number of daily worked hours.

WHIP does not contain data on nationality but has information on the place of birth. This variable can be used as proxy for nationality. Details about the identification of foreign workers can be found in the forthcoming technical report on foreign workers or in Villosio (2002) "Foreign workers" in B. Contini (eds) "Labour mobility and wage dynamics in Italy ", Rosenberg & Sellier Publishing.

The apprentice contract is ruled by Law 25/1955 and further amendments. The duration is fixed by CCNL; it can’t however be lower than 18 months and higher than 4 years (a higher duration can be expected only in the handicraft sector).

The training-on-the-job contract has been introduced by Law 863/1984 and modified by further amendments. It is a "causa mista" fixed term contract, with a maximum duration of 24 months, which can’t be renewed.

For the apprentices the variable "skill" takes value of 1.

For workers holding a training-on-the-job contract the variable "cont_rebate" takes value of 2