France 1979 - Documentation

Table of Contents

 

A. GENERAL INFORMATION back

Official name of the survey:
Official name of the survey/data source: Revenus Fiscaux 1979
(Survey of Individual Income Tax Returns)

Administrative unit responsible for survey:

Division Revenus
Institut National de la Statistique et des Etudes Economique (INSEE-DG)
18 bld Adolphe Pinard
75014 Paris
FRANCE

The organization responsible for this survey is the Institut National de la Statistique et des Etudes Economique (the National Statistical Office). Funding for the survey is also provided for by INSEE.

The data are stored at:

Division Revenus
INSEE-DG
18 bld Adolphe Pinard
75014 Paris
FRANCE

A copy of the original codebook and other documentation can be obtained from the Division Revenus at the above address. The primary research group which has analyzed the data is the Division Revenus. Other principal users of the data are statistical administration and research institutions.

The Revenus Fiscaux is conducted every 4, 5, or 6 years. The first year in which the Revenus Fiscaux was conducted was 1956, and was first made available as microdata in 1970. The main purpose of the survey is to measure household income distribution.

The data are not collected from interviews, but are taken from the income tax files of the French income tax administration.

There are restrictions on the use of this data by the public. The Revenus Fiscaux microdata have been made available to the Luxembourg Income Study Project under the condition that the data not be given or sold to any other body by LIS. They are available for use by LIS researchers only.

A bibliography of the most important publications based on the Revenus Fiscaux data is provided in section N. Also included in the bibliography in section N is a list of the user documentation for the Revenus Fiscaux data.

 

 B POPULATION AND SAMPLE SIZE, SAMPLING METHODS back

Although the Revenus Fiscaux data have been extracted from income tax files, the sampling frame for the survey consists of housing units recorded in the 1975 population census, updated to 1979 from a list of newly constructed housing units. A three stage stratified sampling design was used to draw a sample of addresses, which was then given to the income tax administration, which matched the taxation records with these addresses.

Household heads who did not complete tax returns were not included in the survey. In addition, people living in institutions or mobile homes were not included in the survey.

Some military personnel were included in the sampling frame, however this depended on where they registered their tax documents.

1. The Sample Design

The Fiscaux Revenus survey utilized a three stage stratified sample design, which was constructed in the following way:

  1. For the sampling frame the 1975 census (Reconsement Population 1975) was used to compile a list of addresses for the Revenus Fiscaux survey. The sampling frame was expanded to include new houses which were built between March 1975 and December 1979.
  2. The sample design for the Fiscaux Revenus was then stratified by three factors: geographic location (21 regions), socio-professional status (23 categories), and size of household (3 categories).
  3. A sample of addresses was then selected which was given to the taxation administration. These were matched with taxation records.

There were different selection probabilities for each group. High-income recipients had a higher probability of selection than other income categories. For example, industrialists had a 1 in 40 selection probability, whereas pensioners and workers had a 1 in 800 selection probability.

Those households or individuals which fell into the sample but which could not be located or for which income data was unavailable can be broken down into the following two categories:

  1. Unoccupied, or destroyed household units (11.3%)
  2. Unknown by the fiscal administration (4.5%).

A total number of 34,062 respondents were selected. There were 6,389 sample units from which data was not obtained because they could not be identified, because they were a second residence, or because they were unoccupied. The sample that was made available for the LIS database was comprised of 11,044 cases. LIS was able to make available to its users all 11,044 cases.

The geographic regions of Reunion, Martinique, Guadeloupe and Guyane were excluded from the Revenus Fiscaux survey.

Estimates of sampling variability (standard errors) have not been computed for the Revenus Fiscaux.

  

C. MEASURES OF DATA QUALITY  back

1. Reporting and Under-Reporting and Income Data Quality

An evaluation of the quality of the income data has been made by comparing the Revenus Fiscaux sample with the "Direction générale des impôts" DGI (tax administration). The comparison between the Revenus Fiscaux and Statistics D.G.I. is shown in Table FR79.1. This evaluation suggests that taxation values in the Revenus Fiscaux 1979 have been accurately estimated.

However, the Revenus Fiscaux have underestimated by 2.5% the number of households which were taxed. Also, average net tax has been overestimated in the survey. Detailed information can be found in Geneviève Canceill; 1987, Les Revenus Fiscaux des ménages en 1979, no 127, pp. 25-33. An evaluation of the quality of tax, demographic, wealth, expenditure and other data has also been made.

Table FR79.1
General Taxation Registration Statistics D.G.I. Revenus Fiscaux

Overall registration:
Number of Persons 15,000,673 15,375,452
Taxable Income 755,400,000,000 F 768,700,000,000 F
Net Tax 113,500,000,000 F 115,900,000,000 F
Tax Rate 15,0% 15,1%
Amounts per Person Taxable Income 50,357 F 40,999 F
Net Tax 7,566 F 7,736 F

2. Data Corrections and Adjustments

For some households the income structure is unknown. Only the level of taxable income is known. Incomes are estimated using information from households of the same social category and income level.

3. General Assessment of Income Data Quality

Users of this data should be aware of the following problems regarding the quality of the income data collected in the Revenus Fiscaux:

  1. Self employed incomes are probably underestimated.
  2. Farm incomes, which are located at the bottom of the income distribution, are probably imprecise. Consequently, estimates of inequality should only be measured for wage earners and pensioners.
  3. Property incomes are allocated either to previously deducted taxes or they are not taxed. Neither are property incomes declared to the fiscal administration. These represent approximately 45% of total cash income from property.

 

D. DATA COLLECTION AND ADMINISTRATION back

1. Method of Data Collection

The data was collected by tax administration employees, who completed the questionnaire by consulting the tax file documents corresponding to the household addresses which they were provided with. No interviews were conducted.

2. Structure of the Survey Instrument

For each household one questionnaire was completed which provided demographic information for each household. A second questionnaire was completed which provided information on income and taxation.

3. Survey Respondent Rules

Because the data collection was done by tax administration employees there were no survey respondent rules.

 

E. WEIGHTING PROCEDURE back

In the Revenus Fiscaux weights have been assigned to each sample case. These weights were assigned in order to adjust for the differing sampling rates of the various strata. Because of the existence of unoccupied, destroyed and secondary housing, the sum of the survey weights is not equal to the total number of units in the sampling frame. However, estimates derived from the sample are representative of the total population as defined by the sampling frame.

 

F. DETERMINATION OF SURVEY UNIT MEMBERSHIP back

All individuals who can be identified as living in a particular housing unit are considered members of the same household.

The basic unit of aggregation which was used in this survey was the household unit, which is an aggregation of fiscal units ("foyer fiscal").

The sampling unit can be divided into subgroups in those situations where several families share the same housing unit. If there is one family, the head is the male adult. If there is more than one family in a household, the head is the oldest working person, or the oldest person when no household member is working.

Other than the sampling unit head the following individuals can be identified in the Revenus Fiscaux:

  1. head of household
  2. spouse (legal or not)
  3. children
  4. grandchildren
  5. elderly persons of the family (father or uncle of the head)
  6. siblings
  7. friends, pensioners, sub-tenants, foster children
  8. (domestic) servants, employed persons
  9. undefined

It is possible to identify the relationship of these individuals to the sampling unit head.

 

G. CHILDREN AND SPOUSES back

In this survey children are defined as all supported persons in a family who have never been married and were under the age of 18 years before January 1, 1979, or were born after 1979. If they are disabled the age is not taken into consideration for this definition. Individuals who were supported by their parents and are between the ages of 18 and 21 years, or who were in military service, or were studying and are between the ages of 18 and 25 years were also defined as children. In addition to this children who are married but are supported by their family and are under 21 years of age (or under 25 years if studying) are considered children.

In this survey spouses are defined as the legally married partner of the household head. Cohabiting individuals are also considered spouses if this information was available from the census form. However, in this case they are considered to be two separate fiscal units.

Variables exist which allow one to link the household head and the fiscal unit head.

 

H. AVAILABILITY OF BASIC SOCIAL AND DEMOGRAPHIC INFORMATION back

In Table FR79.2 is summarized the basic social and demographic information which is available in the Revenus Fiscaux.

Table FR79.2
Basic Social and Demographic Information Category Available Persons for whom information available

Category Available Comments
Sex Yes All persons
Age Yes All persons
Date of Birth Yes All persons
Relationship to unit head Yes All persons
Ethnicity/nationality No -
Race No -
Legal marital status or cohabitation Yes Head and spouse/partner
Highest level of education No -
Disability status Yes Head and spouse

 

I. AVAILABILITY OF LABOUR MARKET INFORMATION back

Labor force status information is available for the Revenus Fiscaux. The quality of the information for persons who had been unemployed for more than 6 months is uncertain. Labor market information which is available in the Revenus Fiscaux is summarized in Table FR79.3.

Table FR79.3
Labour Market Information

Category Available Comments
Labour force status Yes -
Hours worked No -
Full/part-time No -
Type of occupational training/apprenticeship No  
Occupational group No  
Industry group No  
Worker or professional status Yes Head and spouse
Weeks employed last year No -
Duration of unemployment last year No -
Wage/salary income Yes Head and spouse
Self employment income Yes Head and spouse

 

J. AVAILABILITY OF GEOGRAPHIC INFORMATION back

In the Revenus Fiscaux it is possible to identify the geographic categories and size of location. (See Variable D20.1)

D20 has the following categories :

0. small area
1. urban area 5,000 or more inhabitants
2. urban area 5,000 to 10,000 inhabitants
3. urban area 10,000 to 20,000 inhabitants
4. urban area 20,000 to 50,000 inhabitants
5. urban area 50,000 to 100,000 inhabitants
6. urban area 100,000 to 200,000 inhabitants
7. urban area 200,000 to 2,000,000 inhabitants
8. agglomeration of Paris

 

K. SOURCES AND AMOUNTS OF CASH INCOME back

In the Revenus Fiscaux sources and amounts of income are recorded for an annual period. The amounts recorded were the actual amounts received. Income sources and amounts are not recorded for each person. Sources and amounts of income were based on fiscal declaration and imputed allowances.

 

L. TAXES back

Because the data were taken from taxation records there was no need to estimate direct taxation amounts.

 

M. BIBLIOGRAPHY OF IMPORTANT PUBLICATIONS back