Business Demography Statistics in Europe – Developments and Prospects 

Hartmut Schrör
Eurostat

1. Introduction

Data collections on business demography at European level have become a well established statistical domain among Eurostat’s products and have received attention by policy makers and researchers. Indicators on enterprise births, survivals and deaths form part of the Structural Indicators compiled by Eurostat to enable the European Council to monitor the “Lisbon process”, which aims at transforming the EU into the most competitive and dynamic economy in the world by 2010. Other services within the European Commission use the business demography data for policy recommendations. The page views of the data published on the Eurostat website indicate a significant interest by the public, and researchers contact Eurostat to obtain more detailed information.

In the light of the political visibility of the data collection, this paper outlines its recent developments, ongoing work and future prospects. Business demography data at European level benefit from the harmonization of business registers in terms of concepts, definitions and collected indicators. However, the methodological guidelines for business demography could potentially be applied to business registers in countries outside Europe as well with a view to obtaining comparable data at a global level.

2. Recent developments: Publications on 2003 harmonized data collection

2.1 Output produced from the data collection

The participating Member States submitted their data on enterprise births and survivals of 2001 and on deaths of 2000 to Eurostat in the summer of 2003. There were various delays in the data processing and production of final results, which were due to the Eurostat crisis in 2003. Eurostat finally published results in August and September 2004 in the following formats:

-          A “Statistics in Focus” of 12 pages in English, French and German. This publication is available on paper and in PDF format.

-          A detailed tables publication “Business Demography in Europe – results of 10 Member States and Norway” of 138 pages, which is available in English in PDF format.

-          An updated and revised complete dataset on the Eurostat website, as well as an update of the structural indicators on business demography.

Following the new dissemination policy that Eurostat has implemented this year, the publications in PDF format are available free of charge and can be downloaded from the Eurostat website[1]. The underlying complete dataset is expected to be available free of charge as well on the Eurostat website from October 2004.

2.2 New elements

ICT aggregates

Some new elements were introduced with the harmonized data collection of 2003. Firstly, the breakdown of economic activities according to NACE Rev. 1 was extended with special aggregates on the ICT sector, which are as such not available in the standard hierarchy of the NACE nomenclature. These special aggregates are as follows:

l        ICT: total of ICT manufacturing and ICT services

ü        ICT manufacturing

ü        ICT services

¨         ICT wholesale

¨         ICT consultancy

¨         Telecommunications

These special aggregates were collected because the ICT sector was expected to show a higher dynamism in terms of enterprise birth rates and death rates than the average of the business economy. Indeed, the results confirmed this assumption. Figure 1 shows the percentage points by which the birth and death rates in the ICT sector exceeded those of the business economy average.

Figure 1: Extent to which enterprise birth and death rates in ICT activities exceed those of the business economy average, 2000 (percentage points)

The dynamism observed in ICT services was generally higher than in ICT manufacturing. Within the services, especially telecommunications stood out with high birth rates (roughly between 10 and 30 %) and death rates (roughly between 5 and 20 %).

The following table gives an overview of enterprise birth rates in the ICT sectors, illustrating again the high dynamism particularly in ICT services.

Table 1: Birth rates, breakdown by ICT sector (%) (1)
		EU (2)	BE	DK	ES	IT	LU	NL	PT	FI	SE	UK	NO
ICT total	1998	:	:	18.5	13.5	19.4	18.6	:	11.9	9.6	10.4	:	17.3
	1999	12.6	:	18.6	12.5	12.4	17.9	15.3	9.1	10.0	9.0	:	18.2
	2000	13.8	:	20.1	14.8	12.7	17.1	16.2	9.3	11.4	11.1	:	18.6
	2001	12.4	:	17.0	13.5	11.9	16.4	15.1	9.5	10.6	8.4	:	18.4
ICT manufacturing	1998	:	:	7.5	11.9	11.4	:	:	8.2	6.2	4.8	:	:
	1999	7.4	:	7.9	10.3	6.8	0.0	11.3	9.8	4.3	4.6	:	:
	2000	7.7	:	6.8	11.0	7.2	0.0	10.0	6.7	5.6	5.3	:	:
	2001	:	:	6.8	8.9	5.9	:	7.5	6.4	5.3	3.6	:	4.7
ICT services	1998	:	:	19.2	13.7	20.8	:	:	12.2	9.9	10.7	:	:
	1999	13.2	:	19.2	12.7	13.2	18.1	15.5	9.0	10.5	9.3	:	:
	2000	14.4	:	20.8	15.2	13.5	17.2	16.4	9.4	11.8	11.5	:	:
	2001	:	:	17.5	13.9	12.7	:	15.4	9.7	11.1	8.7	:	18.8

(1) The merging of administrative registers in Denmark in 1999 may have resulted in the over-evaluation of enterprise births in that year; the quality of 1998 Italian data is generally low, reflecting changes made to the business register.
(2) Mean of Denmark, Spain, Italy, Luxembourg, the Netherlands, Finland and Sweden.

Legal form breakdown

A legal form breakdown was added to the data collection for a limited set of indicators, excluding those related to enterprise survival. Data are broken down by the following three groups of legal forms:

¨         Sole proprietors: personally owned and no limit to personal responsibility

¨         Limited liability companies: private or publicly quoted joint stock companies with limited liability for those owning shares

¨         Partnerships and other legal forms: both unlimited and limited liability partnerships owned by more than one natural person as well as cooperatives and associations.

This breakdown was introduced because the legal form that newly born enterprises choose may indicate the availability of capital and the effect of policy measures to support entrepreneurship.

Two thirds of all newly born enterprises in the business economy were sole proprietorships in 2001, one quarter were limited liability companies and slightly more than 10% were partnerships and other legal forms. Table 2 shows the density of the enterprise births relative to the population aged from 20 to 59 years. It clearly illustrates the dominance of sole proprietorships, especially in services and construction. Only in Luxembourg, limited liability companies were most frequently chosen.

Table 2: Density of birth rate: number of enterprise births per 10 000 population aged 20-59 (units)

LL: limited liability companies
PA: partnerships and other legal forms
SP: sole proprietors

(1) Denmark, Spain, Italy, Luxembourg, the Netherlands, Finland and Sweden only.
(2) Sole proprietors not covered.

Employment in enterprise deaths 

An indicator on the number of persons employed in enterprises that died was added to the 2003 harmonized data collection. Figure 2 shows that while services generally account for the highest share of jobs created in newly born enterprises, the same applies to the jobs lost in enterprises that died. Figure 3 shows the average size in terms of persons employed in enterprises that died, broken down by legal form. As one would expect, it indicates that sole proprietorships are the smallest enterprises that die, while in most EU countries limited liability companies are the largest ones.

Figure 2: Breakdown by activity of the number of persons employed in enterprises that died, 2000

Figure 3: Average size (in terms of persons employed) of enterprises that died, business economy, 2000 (units)

3. Ongoing Work: 2004 data collection

Following the success of the first two harmonized data collections, most Member States that had participated announced a contribution to the 2004 data collection as well. During the second harmonized data collection of 2003, eight out of 10 new Member States and Romania carried out the feasibility study that the old Member States had completed two years earlier. The results were promising, and these countries now contribute to the ongoing data collection of 2004, thus almost doubling the geographical coverage of the enlarged EU.

As the business demography data collection is still a statistical development project, there are a few changes in this year’s data request, which are less substantial than the previous ones. Firstly, turnover data are no longer collected, leaving indicators on the number of persons employed as those available for the study of effects of demographic events on the economy. Secondly, the special aggregate on ICT consultancy was dropped, because its use was questionable and because it had led to difficulties in applying rules of disclosure control.

We expect to update the Eurostat website and the structural indicators with a new dataset by the end of 2004. We do not intend to release another detailed tables publication, because Eurostat has reduced their number in its revised dissemination and publication strategy. However, two short “Statistics in Focus” are planned to be published in 2005.

4. Prospects for the future

4.1 Consolidation and extension to more Member States

The methodology of the business demography data collection has become quite stable during the three harmonized data collections carried out so far. The participating Member States have built up their experience and proven that the methodology can be applied successfully to produce meaningful results illustrating the dynamism of economic activities with additional information on size classes and legal forms, and particularly on the effects on employment.

In the near future we intend to enhance the business demography data collection by extending the geographical coverage of the published data. Currently, data from Germany, Greece, France, Ireland, Cyprus, Malta and Austria are missing in the publication and dissemination. Some EU Member States and candidate countries that do not yet participate have announced contributions to future data collections. Grants are available on a limited scale for national statistical institutes to support their activities.

Some outstanding methodological issues are being discussed that do not substantially affect the results, but whose clarification will help to improve further the comparability of the data. For instance, the question has been discussed whether units moving out of the activity scope of the data collection should be considered as active and surviving units. Another issue is whether units in liquidation should be considered as active if the only remaining activity is related to the liquidation process itself, e.g. the sale of production factors.

4.2 Legal basis

While there is a high demand for data on business demography and its effect on employment, this data collection is presently covered only to a limited extent by a legal basis. Thanks to the efforts of the Member States and the existing regulation on business registers, availability of data on business demography is already good today in several Member States. The regulation on structural business statistics and the NACE nomenclature also serve as a conceptual basis for the data collection. Nevertheless, the specifics of the business demography domain in terms of the demographic indicators and the NACE activities call for a clear legal basis. Therefore, the current revision of the SBS regulation by Eurostat and the Member States includes a proposal on a new annex on business demography. The consultation process with the Member States on the amended SBS regulation is ongoing, and the new one is expected to enter into force in 2006.

The legal basis will bring about several changes to the business demography data collection. These changes will not affect the underlying methodology, but rather the scope and presentation of the data.

¨         The definition and coding of the indicators will be revised and restructured to make the dataset clearer and easier to navigate for end users, and to avoid conflicts with other codes and definitions.

¨         The NACE breakdown will be slightly more detailed; however the size class and legal form breakdowns will not be requested at 4-digit level.

¨         Member States will be requested to submit estimated preliminary data on enterprise deaths. Currently, death data are available one year later than data on births and survivals, because enterprises are considered as dead only after two years of inactivity.

¨         As the SBS regulation builds the legal framework for the legislation on statistics on business demography, the scope of the data collection will be limited to market oriented activities. This will raise the issue to what extent NACE sections M to O will be included and how market and non-market units are identified in general. This issue will be addressed in the future at the level of the whole SBS domain and thus also affect business demography.

¨         Member States will be asked to provide quality reports with the data, particularly focussing on the differences observed in the number of enterprises and persons employed between the business demography data and SBS surveys. These differences are not surprising considering the different methodological approaches to the data collections, but they need to be explained to the data users.

¨         Pilot studies are foreseen on the production of data on NACE sections M to O, the demography of local units and on demographic events other than births, survivals and deaths.

The study of additional demographic events such as split-offs, break-ups, mergers and take-overs will depend on the feasibility of identifying these events in a coherent way in the business registers of the Member States. Split-offs, for instance, could be an interesting object of study as an indicator of innovation in the different economic activities. The assumption is that in many cases split-offs occur when highly specialized and innovative activities within an enterprise are outsourced as independent units.

4.3 Project on the factors of business success

As a result of the Eurostat crisis in 2003, the envisaged project on the factors of business success was temporarily put on hold. This year, Eurostat and the Commission’s directorate general for enterprise policy (DG ENTR) have revived it.

While the existing data on business demography provide useful information to policy makers on the dynamism of economic sectors in the Member States and its effect on employment, further insight is required in the factors that determine the success or failure of newly born enterprises under certain conditions. To shed light on these factors of business success, Eurostat in cooperation with DG ENTR intends to award grants to national statistical institutes that will carry out surveys among the population of enterprises having survived for a certain number of years. The project will address

¨         the motivations for starting up own businesses

¨         the factors explaining the dynamics of newly born enterprises, i.e. growth in employment and turnover

¨         risks and obstacles experienced

¨         profiling successful entrepreneurs to identify factors for enterprise survival and growth.

Methodological work related to the sampling, the questionnaire and the pre-processing of results to be submitted to Eurostat is envisaged for the autumn of 2004. A test of the survey tool using a restricted sample is planned for 2005, and final results are expected during 2006.

5. Conclusion

The business demography project at EU level is developing towards a regular annual data collection thanks to the active participation of the national statistical institutes. The publications and datasets released so far have received considerable attention, and the OECD and some EU Member States have used the published data for more detailed analysis and further conclusions. The data collection is being extended to include most new Member States, whose data will be released for the first time towards the end of 2004. A legal basis in the form of an annex to the revised SBS regulation is expected to enter into force in 2006. Some remaining methodological issues are being solved. As a result, and also due to the growing experience of the contributing Member States, the quality and comparability of data will increase in future data collections.

The question arises to what extent data on business demography produced in Europe will be comparable with similar data collections in countries outside Europe. For policy makers it will be interesting to compare these data as indicators of business dynamism in various economies at a global level. With a view to this comparability, it would be desirable to study to what extent methodologies applied to data collections on business demography overlap or diverge, and to what extent a convergence of methodologies is feasible. This statistical domain is becoming well established within Europe, but its common methodology is not confined to the European context. Therefore the methodology could bear potential for a wider use outside the EU.

Notes

[1] http://europa.eu.int/comm/eurostat, then go to “Publications”