Rome – October 27 – 31, 2003
Session No3 “On Statistical and Administrative Units”Paper N°
Peter Schmidt, Bernd Waldmüller; Statistisches Bundesamt, Germany
Concept and Results of a Test Survey on Enterprise Delineation
1. Introduction
As already reported at the last roundtable[1] a concept has been developed by a working group of German statisticians for an enterprise term which may comprise more than one legal unit. So far, in German business statistics the enterprise is generally defined as the smallest legally independent unit with own accounts. The new enterprise term is described by two main characteristics, namely the enterprise as an economic entity and an organisational unit with a certain degree of autonomy.
The main characteristics describe the criteria under which legal units should be combined to form a complex enterprise. If these criteria are applied the large majority of legal units can be considered as single enterprises. But especially separate factors of production and ancillary units (in the following shortly named ancillary units) which are exclusively active for other legal units inside the same group (enterprise group) cannot be regarded as single enterprises. Thus, the correct treatment of ancillary activities respectively ancillary units is one of the basic elements of the enterprise delineation.
For the implementation of a new enterprise term it is important to know the empirical relevance of ancillary units and to get some information about their characteristics. To decide whether a legal unit constitutes an enterprise or is part of a complex enterprise, one needs to have information about ownership relations and possibly exclusive ancillary activities to other legal units. Both kinds of information are not available in German statistics. Especially the question whether a legal unit carries out ancillary activities to another legal unit can only be answered by the unit itself or the controlling legal unit.
Administrative data about legal units do not provide sufficient information to identify ancillary units. In order to find out the frequency of “ancillary units” or “factor units” and their impact on specific industries a test survey was carried out in Germany. The survey aimed at providing information about the empirical relevance of different cases (number of legal units, share of employees and turnover concerned, implications for the economic structure) and about the feasibility and the costs related to the implementation of a new enterprise term.
2. The concept and the survey in practice
The main goal of the survey was to identify legal units which cannot be regarded as independent, i.e. legal units providing ancillary activities or factors of production exclusively to another legal unit or a small number of other legal units. The results of the survey should give indications about the empirical relevance of the concept, possibilities for its introduction into official statistics and consequences of this definition of the enterprise term on the classification of these units.
The business register contained about 3,4 million legal units in 2002. It was the base for the sample being drawn for the survey. Since there was almost no information about ownership relations and the characteristics of ancillary units, criteria had to be defined for the selection. The main criteria were the activity code, the membership in an “Organschaft” and the legal form. Concerning the NACE-Code units were chosen whose activities could be regarded as ancillary activities like sales, bookkeeping etc. or provision of factors of production like real estate or machines.
Finally a sample of 10,000 legal units was divided into four survey subgroups:
- Members of VAT-groups (“Organschaften”) (1,000 legal units)
- Legal units in real estate (1,000 legal units)
- Companies of wholesale and retail trade (2,000 legal units)
- Other companies in selected activity codes (mainly services) (6,000 legal units).
A questionnaire was developed especially for this survey. The aim was to formulate a questionnaire in easily understandable form. The questionnaire passed out altogether eight questions. The first five questions asked for general characteristics. So were asked the legal form, the year of foundation of the legal unit as well as its turnover and its number of employees. The following questions asked for connections with other legal units, activities and customers resp. suppliers relationships (especially in the case of trade units) of the legal units.
3. Results of the survey
The questionnaire was answered by 6,850 legal units. The presentation of the results should give some information about the empirical evidence of ancillary legal units. One important aspect is to show in which industries ancillary units can be found and which special characteristics they have. The results refer to the answers of the survey and cannot completely be regarded as representative for the economy as a whole. Despite of this fact one can draw conclusions on the structure and the empirical relevance of ancillary units.
Ancillary units in this context are legal units
-which are linked with other legal units (inside a group), and
-which carry out their services for other legal units in Germany inside the group, and
-which carry out their services for exactly one other legal unit or a small circle of legal units (up to three).
Frequency of ancillary units and legal units linked with other legal units
The following table gives an overview about the frequency of dependent legal units and ancillary units.
Table 1: Frequency of legal units linked with other legal units
Share of type of legal unit (in %)
Industry /Dependent legal units
/ Dependent legal units whose customers / providers are exclusively German legal units inside the same group / Ancillary unitsLetting of own property / 29,5 / 19,6 / 17,3
Other activities in section real estate, renting and business activities / 30,5 / 7,0 / 4,1
Wholesale and retail trade / 28,0 / 3,1 / 2,3
Other industries / 23,3 / 2,6 / 2,1
Total / 28,8 / 6,8 / 4,7
The table shows that about 29% of all answering legal units indicated being connected to another legal unit. This number also contains connections with units abroad from Germany. However, for the examination on hand it is of special interest, how many legal units exclusively render performances for other members of the enterprise group and how many legal units could be considered as ancillary units. In principle five per cents of these legal units would fulfill this condition.
Furthermore one can see the share of legal units fulfilling the second criterion mentioned above. About 6,8% of all legal units answered that they carry out their services to other legal units in Germany inside a group. Referring to all dependent legal units the share is about 24%. Concerning the degree of linkages one can see some differences regarding the several industries. In the industry “letting of own property” about two thirds of these legal units carry out their services exclusively to other members of their group. In other industries this share is much lower.
The percentage share of ancillary units in the total number of all answers consists of a share of about 2,2% of all legal units which carry out their services to exactly one other legal unit and 2,6% of the legal units carry out their services up to three legal units. Due to the structure of the sample the results can not be regarded as representative for the whole economy. Setting the assumption that the number of ancillary units in other industries (especially in the NACE sections C-F) is fairly low, one can expect a share of legal units being ancillary of approx. 1-2%. Despite the share of ancillary units is remarkable for single industries, especially for the industry letting of own property with a share of 17,3%.
Looking at the shares of ancillary units with regard to the structure of the sample one can consider large differences. The highest share had subgroup a) (member of an Organschaft) with approx. 12%. In this group the activity letting of own property has a share of 32%. In subgroup b) this activity had a share of 10,4%. For legal units of wholesale and retail trade (subgroup c)) and other services (subgroup d)) the share was below the average with 1% respectively 3,3%.
Consideration to several industries
Table 2: Frequency and relevance of ancillary units in survey subgroups and selected industries
Survey subgroup
Industry
/ Share of ancillary units (in %) / Share of persons employed in ancillary units (in %) / Share of turnover in ancillary units (in %)Survey subgroup a): legal units as parts of VAT groups (Organschaften) / 12,1 / 1,7 / 2,2
-Among them in trade (NACE 50-52) / 7,0 / 5,1 / 1,8
-Among them Real estate activities (NACE 70) / 34,0 / 6,4 / 15,6
-Among them Renting of machinery and equipment (71) / 19,3 / 1,7 / 1,8
-Among them other business services (NACE 72-74) / 8,0 / 0,8 / 2,1
Survey subgroup b): Letting of own property / 10,4 / 1,8 / 12,0
-Among them of NACE 70.20.1 / 12,4 / 2,0 / 13,7
-Among them of NACE 70.20.2 / 4,5 / 0,9 / 1,1
Survey subgroup c): Trade units in selected legal forms / 1,0 / 1,8 / 0,9
-Among them wholesale trade (NACE 51) / 1,5 / 2,9 / 1,2
-among them retail trade (NACE 52) / 0,5 / 0,2 / 0,1
Survey subgroup d): legal units in other selected industries in selected legal forms / 3,3 / 0,5 / 0,8
-among them Transport, storage and communication / 2,3 / 1,7 / 0,6
-among them Real estate activities on a fee or contract basis (NACE 70.3) / 7,0 / 1,6 / 3,7
-among them Renting of machinery and equipment (NACE 71) / 5,8 / 0,7 / 0,9
-among them legal, accounting, book-keeping and auditing activities; tax consultancy; market research and public opinion polling; business and management consultancy; holdings (NACE 74.1) / 15,3 / 1,2 / 0,6
-among them Miscellaneous business activities n.e.c. (NACE 74.8) / 2,6 / 0,2 / 2,3
Total
/ 4,7 / 1,2 / 1,8In table 2 the percentage of frequency and relevance of ancillary units in several industries is described in more detail. Since there is only a small number of cases available one cannot derive any sound statements for the whole economy. Therefore the following statements are restricted to single industries. The number of cases in the industry letting of own properties differs significantly from the rest. 34% of the legal units being a member of an Organschaft were ancillary units, an order of magnitude which cannot be observed in any other industry. The share of 1,0% of ancillary units in the industry of wholesale and retail trade is very low, though several ancillary units in this industry have a remarkable high number of employees (sales companies). In other selected industries - mainly in the service sector – the shares of ancillary units vary from 2% to 15%. The industry 74.1 (e.g. consulting companies and associate companies) with a share of 15,3%, the industry 70.3 (Real estate activities on a fee or contract basis) with a share of 6,9% and the industry 71 (Renting of machinery and equipment without operator and of personal and household goods) with a share of 5,8% can be mentioned here.
Consideration of turnover and number of employees (size classes)
Table 3: Frequency of ancillary units with regard to size classes
Size classes / Share of Ancillary units (in %)Size classes turnover
-up to 7 Mill € / 4,9-7 to 40 Mill. € / 3,6
-> 40 Mill. € / 2,3
Number of persons employed
-1-9 / 6,1
-10-99 / 1,7
-100 and more / 1,5
It is obvious that the share of ancillary units is higher for smaller size classes. This corresponds to the share of ancillary units measured by turnover and persons employed in table 2.
Differences considering the genesis of legal units
Table 4 shows the frequency of ancillary units and the genesis of the legal units in selected industries.
Table 4: Percentage share of ancillary units and the genesis of the legal units in selected industries.
Industries / Birth (in %) / Split-off (In %) / Others (in %)Letting of own property / 9,9 / 51,5 / 15,6
Other activities in section real estate, renting and business activities / 3,3 / 13,0 / 3,7
Wholesale and retail trade / 1,9 / 9,9 / 1,3
Other industries / 1,0 / 10,4 / 2,4
Total / 3,3 / 19,0 / 4,1
Altogether shall state that units arisen from split-offs have a higher share of ancillary units than units arisen from births or other ways of genesis. 19% of all units followed from split-offs proved as ancillary units. The share of ancillary units created by splitt-offs in the industry letting of own property is about 50%. This supports the thesis that a lot of companies have outsourced their buildings and properties to so-called property companies.
Industries of main legal units
Table 5 shows the industry of the ancillary units and the industry of the main units for which the ancillary unit is exclusively active.
Table 5: Share of ancillary units acc. to their own industry and the industry of the main unit for which they are exclusively active in relation to all ancillary units (in %)
Industry main unitIndustry ancillary unit / NACE section C-F /
NACE section G, H
/ NACE section K / Others / TotalLetting of own property / 63 / 31 / 8 / 22 / 122
Other activities in section real estate, renting and business activities / 70 / 25 / 34 / 21 / 145
Wholesale and retail trade / 18 / 16 / 5 / 6 / 38
Other industries / 3 / 2 / 0 / 15 / 20
Total / 154 / 74 / 47 / 64 / 325
The ancillary units classified in the industry letting of own property carry out about a half of their services to legal units in the manufacturing industry (NACE sections C-F) and a wider quarter for main legal units in the industry of wholesale and retail trade and for hotels and restaurants (NACE sections G and H). However, it should be taken into account that double countings can be contained in these figures, e.g. if an ancillary unit is active for several other legal units.
1
[1] “The interpretation of the enterprise in business statistics and the business register –theoretical and practical aspects” by Peter Schmidt; paper for the 16th Roundtable