Services trade by enterprise characteristics - STEC
Data extracted in 2017
Planned article update: October 2019
This article describes how enterprises trading in services operate, by analysing enterprise-level data and examining in detail the statistics of the characteristics of such enterprises: size (the number of employees), ownership (foreign-controlled or domestic) and economic activity of the enterprise trading in services. A study looking at services trade by enterprise characteristics (STEC) was carried out in 15 European countries, the experimental statistical results of which are set out in this analysis.
A study looking at services trade by enterprise characteristics (STEC) was carried out in 15 European countries between 2011 and 2014. In this analysis, we set out the results of the study, which are mostly based on 2014 data. For all the statistics used, the trading partner is the ‘rest of the world’ (i.e. any other country). Due to the limited scope of the experimental study and the fact that it is a pilot, the conclusions are descriptive and cannot be generalised beyond the participating countries.
Increasing international trade in services is an important component and key driver of economic globalisation. Many services have become tradable due to digitalisation, such as health and educational services. Also, many new services have been launched on international markets by means of information and communication technologies (ICT) and digital tools — and some have led to the creation of new markets.
There are some clear differences between the countries covered by the study in terms of how the enterprises have responded to globalisation. As regards international trade in services, Luxembourg and Ireland are more open to international markets than the other countries: typically, foreign owned affiliates enjoy a strong position in their markets and account for more than 85 % of the services they export. In contrast, domestic service providers have a strong position in Denmark, Iceland and Lithuania, accounting for around three quarters of the services exports.
The STEC project enables us to look at the size of the enterprises involved in services trade. For example, small enterprises (up to 49 employees) are responsible for over half the services exports from Luxembourg, Estonia and Denmark. These enterprises tend to be domestically controlled, while large enterprises are more likely to be foreign controlled. The exception is Luxembourg, where services exported by small enterprises are more often the foreign controlled enterprises.
In many countries, export intensity is highest in transport and storage, followed by information and communication enterprises.
Together, STEC and trade in goods by enterprise characteristics (TEC) statistics provide a picture of the traders active in international goods and services markets. We illustrate this with an example later in this article.
Trade in services by enterprise size
Data on enterprise size class help us determine whether trade in services is dominated by large enterprises (250+ employees) or enterprises with less than 250 employees. The role of the smaller enterprises is dominant in small countries (Luxembourg, Estonia). In bigger countries, large enterprises (which tend to be foreign controlled) have a dominant role, e.g. they accounted for over half of the services exported from Ireland, Finland, the Netherlands and the Czech Republic.
Smaller enterprises (less than 250 employees) were responsible for the biggest proportion of services exports from smaller economies such as Luxembourg (87 %), Estonia (81 %) and Hungary (78 %). They made a much smaller contribution in larger economies such as Ireland (23 %), the Czech Republic and Finland (both 40 %). By size of the enterprise, there are also clear differences across Nordic countries: large (and often multinational) enterprises have a much larger market share in Finland (59%) and Sweden (50%) than in Denmark (30%) and Norway (27%) (Figure 1).
The pattern for imports of services is similar to that for exports. In Ireland, the Czech Republic and Finland, large enterprises are responsible for a very high percentage (85 %, 71 % and 66 % respectively) (Figure 2). Smaller enterprises were responsible for the biggest proportion of imports in Hungary (82 %), Luxembourg (81 %), Lithuania (77 %) and Estonia (76 %).
We can also analyse the relationship between enterprise size and the nature of the services that they export (crossing the ‘number of employees’ and ‘type of service’ variables). Figure 3 takes the case of the Netherlands as an illustration. There is some variation across the types of service, but on average small enterprises account for a quarter of Dutch services exports. Research and development services are exported mainly (80 %) by large enterprises, while exports of financial services are dominated by smaller enterprises.
Trade in services by enterprise ownership
Domestically controlled enterprises account for around three quarters of services exports from Denmark, Iceland, Lithuania and Norway. Foreign controlled enterprises are responsible for most services exports from Luxembourg and Ireland.
The type of ownership tells us whether a enterprise is domestically or foreign controlled. A high proportion of foreign controlled services traders means that the economy in question is well integrated in international markets and global value chains. Typically, this is reflected by foreign affiliates enjoying a strong position in trade in services.
Foreign-controlled enterprises clearly dominate services exports from Luxembourg (90 %), Ireland (85 %), Hungary (77 %) and Belgium (73 %). The proportion of services exports by domestically controlled enterprises is highest (around three quarters) in Denmark, Iceland, Lithuania and Norway. The situation is more balanced in Sweden, Estonia and Austria, where domestic enterprises have a slightly larger share of the export market (Figure 4).
The relative contributions of domestically and foreign-controlled enterprises also vary in relation to services imports (Figure 5). In Luxembourg, Hungary and Ireland, the former are responsible for smaller proportions of services imports. The situation is more balanced in Finland, Sweden, Iceland and Estonia, where they account for a slightly higher percentage. As with exports, the highest proportion of services imports by domestically controlled enterprises (75 %) is in Denmark.
Smaller services traders tend to be domestically controlled, while large traders are more likely to be foreign controlled (multinationals).
Figure 6 shows the positive correlation (r=0.44) between the value of services exported from a country by small traders and by domestic traders. Luxembourg is an exception: small traders there are more often foreign controlled.
Trade in services by enterprises’ economic activity
Traders in transport and storage activities tend to export the biggest volumes of services, followed by those in manufacturing.
Traders’ economic activity indicates their main field of operation. Enterprises in some activities (e.g. manufacturing) export mainly goods, while others (NACE sectors H-N) are mostly involved in services. Figure 7 shows the proportions of services exports attributable to enterprises in different activities, as defined by their main economic activity. Transport and storage enterprises are responsible for over half of the services exports from Denmark and Lithuania, and for the biggest proportion from Iceland, Norway and Estonia. Traders in the information and communication sector were responsible for half of the services exports from Ireland. In the Netherlands, Belgium, Sweden and Norway, a large proportion of services traders operate in the field of professional, scientific and technical activities.
On the import side (Figure 8), transport enterprises account for around a third of services imported by Denmark, Iceland and Lithuania. In Finland, enterprises active in manufacturing account for the biggest proportion (a third of the total).
The STEC approach focuses on traders in services, similarly as goods traders are covered by ‘trade in goods by enterprise characteristics’ (TEC) statistics. With the globalisation of international trade and enterprises exporting goods and services together, we need to combine the TEC and STEC approaches to get a comprehensive overview of traders in both goods and services. In the case of Estonia (Figure 9), traders export mainly goods in industrial activities and services in the rest. The proportion of goods was highest in exports in the manufacturing (C), agriculture, forestry and mining (A-B), wholesale and retail trade (G) and electricity and gas and water supply (D-E) sectors. Services dominated exports in the information and communication (J), financial and insurance (K) and administrative and support activities (N). Trade in goods is primarily located in goods producing activities, while services trade is carried out mainly by enterprises classified as being engaged in services activities with the only exception in wholesale and retail trade, which is classified as a services activity but exports predominantly goods.
Services exports intensity
Services exports intensity is relatively high among transport and storage enterprises and lower (only 1-6 %) among wholesale and retail trade enterprises.
We express exports intensity as the share of exports in total turnover. It is an indication of whether traders are more open to globalisation (high export intensity) or more involved in local markets (low export intensity).
In transport services, export intensity varies widely, from 5 %, 4 % and 4 % respectively in Finland, Netherlands and Sweden to 62 %, 57 % and 46 % in Lithuania, Iceland and Denmark (Figure 10). In wholesale and retail trade, enterprises sell services primarily for domestic markets, leading to export intensity of only 1-6 %.
Source data for tables and graphs
For better comparability, the ‘unknown’ part was removed from the analysis. This includes travel and some financial services, which cannot be linked with the business register due to different data collection methodologies (travellers’ data, administrative sources, etc.). The total services breakdown (including the non-matched share) is presented in Annexes 1 and 2.
The development of STEC statistics represents a major step forward in integrating services trade data into business statistics. STEC statistics make the link between the level of trade in services and the characteristics of the enterprises involved. They provide information on particular types of enterprise, broken down by enterprise size, the economic activity in which they operate and their ownership (domestically or foreign-controlled).
STEC data are produced by combining statistical business register information with data on international trade in services at enterprise level. This enables us to link data on the value of each enterprise’s exports and imports to its characteristics, as provided in the business register. The resulting dataset provides a breakdown of the trader population according to size (small, medium and large), type of ownership and category of main economic activity. Linking different datasets adds value to data, without increasing the collection burden on enterprises and at limited cost to data compilers.
Statistics on international trade in services currently provide the monetary value of trade in services, broken down by category (e.g. computer or legal services) and partner country. These are produced from the transactions recorded under countries’ balance of payments, which captures all transactions that take place between residents and non-residents. By linking the two datasets (business register and services trade data from the balance of payments) at enterprise level using a common identifier (the enterprise’s register code), we can create a cross classification providing us with new data on enterprises involved in services trade.
STEC statistics provide users with new indicators on traders operating in the international services market. They give us additional insight into trade statistics, by allowing the profiling of the enterprises according to selected characteristics such as number of employees, type of ownership and economic activity.
Policymakers, trade analysts and researchers have called for more comprehensive and integrated data on international trade and globalisation. This would help to understand the effect of international trade on growth, economic development, employment and countries’ economic interdependency, which is evidently deepened by economic globalisation. Also, the increasing importance of international trade in services — accelerated by digitalisation and the development of new services — has led to growing demand for data on this aspect of STEC.
International trade in services statistics offer a picture of services flows between countries, broken down by type of service. While this constitutes important input for trade analyses, such data do not tell us anything at micro level about the enterprises that actually trade in services. The profiling of traders is crucial if we are to gain an understanding of the drivers and consequences of services trade at micro (enterprise) level to complement the big picture provided by traditional services trade statistics.
The development of STEC statistics has been endorsed by the European Statistical System Committee. The main organisations active in this area at international level (including Eurostat, the UN, the OECD, the World Trade Organisation and the United Nations Conference on Trade and Development) have produced a paper, International trade information system in 2020: a vision for the future, in which they call for an integrated approach to statistics on trade in goods and services and on enterprises and multinationals.
- The reference year for Austria, the Czech Republic and Norway is 2013, for Poland 2011.
- Enterprise in this article means the "legal unit".
- i.e. the relationship between the value of the services an enterprise exports and its turnover
- For better comparability, the ‘unknown’ part was removed from the analysis. This includes travel and some financial services, which cannot be linked with the business register due to different data collection methodologies (travellers’ data, administrative sources, etc.). The total services breakdown (including the non-matched share) is presented in Annexes 1 and 2.
- If one excludes Luxembourg from the sample, the correlation coefficient is 0.89.
- All sectors are included; 100 % equals total service exports.