International sourcing statistics
From Statistics Explained
- Data from September 2008. Most recent data: Further Eurostat information, Main tables and Database.
This article is based on the results of an ad hoc survey of international sourcing carried out in 11 European Union countries and one non-EU country (Norway). The survey investigated the magnitude and impact of international sourcing by enterprises of both their core business and support functions. Although sourcing abroad can result in the loss of domestic jobs, it can also improve an enterprise's competitiveness, and hence both secure existing jobs and create new ones.
Main statistical findings
The continuous globalization of the economy has pushed many enterprises to adopt international sourcing as a business model, i.e. to move certain business functions that were performed in-house or domestically sourced by the resident enterprise to either non-affiliated (outsourcing to external suppliers) or affiliated enterprises abroad (insourcing). This article is based on the results of an ad hoc survey of international sourcing in 12 European countries.
Essentially, the survey results reveal that:
- Most international sourcing of core business, or support functions, or both, remain intra-EU.
- It is most common among Irish, UK, Danish, Finnish and Slovenian enterprises.
- Manufacturing enterprises source far more than those active in other sectors: well over 50 % of Irish and UK enterprises and over one third of Danish enterprises do so.
- In Ireland, Italy, Sweden and the UK, more core business functions than support business functions are sourced.
- Among the support business functions, the international sourcing of ‘distribution and logistics’ as well as ‘marketing and (after-) sales’ is most widespread.
Graph 1 shows the shares of enterprises which, during 2001-2006, sourced internationally, and those that did not but were planning to do so during 2007-2009.
During 2001-2006, 16 % of the surveyed enterprises had moved certain business functions abroad. More than twice as many enterprises in Ireland and the United Kingdom did so (38.3 % and 34.7 % respectively). The two small and open Nordic economies, Denmark (25.4 %) and Finland (22 %), were also significantly above the average.
With the exception of Slovenia (19.7 %), the remaining countries participating in the survey were either at the average level (Norway 16 %), or under it, the lowest share being registered in the Czech Republic (4.5 %).
When enterprises that did not source internationally were asked about their plans in the near future (2007-2009), relatively few Irish (3.1 %)and UK enterprises (2.4 %) intended to do so, as these countries already had high sourcing levels. Conversely, 12.0 % of Swedish and 7.2 % of Danish and Finnish enterprises not currently sourcing certain business functions abroad were planning to do so. The remaining countries had shares ranging from 1.5 % to 5.5 %.
Limited to 2001-2006, Graph 2 outlines the share of enterprises that had sourced internationally, according to their main activity. Without exception, manufacturing enterprises had adopted it to a far larger extent (often three or four times more), than those active in other sectors. Indeed, the majority of Irish (57.1 %) and UK manufacturing enterprises (57.3 %), and over one third of the manufacturing enterprises in Denmark were sourcing internationally. At the other end of the scale, only 6.3 % of Czech manufacturing enterprises surveyed had adopted it, which is three times more than Czech enterprises active in other sectors (2.1%). These rather low levels may indicate that the Czech Republic is still more of a receiver of international sourcing than active in it.
Core functions have surprisingly high shares
One might expect that enterprises would more often source support functions than core ones. Support functions are those not directly linked to core functions, such as the production of final goods or services which are intended for the market and produce income. However, as can be seen in Graph 3, relatively significant shares of enterprises have sourced core business functions internationally. In addition, some countries have a larger share of enterprises that are sourcing core functions than support ones.
The latter aspect is particularly clear in the United Kingdom where 27.2 % of the enterprises surveyed had sourced core functions, whereas 23.2 % did so for support functions. The picture is more balanced for Ireland where enterprises sourced both core business and support functions to a fairly large extent (29.9 % versus 28.7 %). In Italy and Sweden, a larger proportion sourced core business functions as well but, generally, at a lower level. In Graph 3, the displayed shares are not mutually exclusive, i.e. a surveyed enterprise may have sourced both function types.
The rather large difference noted for UK enterprises has had an impact on the overall average of the participating countries. At the aggregated level, a marginally larger proportion of enterprises sourced core business functions abroad.
Table 1 provides further details on this aspect and looks at the international sourcing of core business and support functions by sector. The highest shares were noted for Ireland and the UK. However, whereas the former also displays similar high shares for support functions, the UK enterprises source these functions to a lesser extent, although they are still sourced at more than double the average value of all the participating countries.
Unsurprisingly, considering the nature of their business, non-manufacturing enterprises source their core business functions internationally to a far lesser extent. Among the 12 countries, once again Ireland and the UK come top, having the highest share of non-manufacturing enterprises that source abroad. However, these sourcing levels are only about half (or even less) as high as the manufacturing ones. In addition, in Ireland and in the UK there is a fairly small difference between the share of enterprises sourcing core business functions and those sourcing support ones.
Focusing on support business functions only, regardless of the sector, Table 2 reveals that ‘Distribution and logistics’ and ‘Marketing, sales and after-sales services’ are the functions most frequently sourced internationally. Certain particularities appear: Danish, Dutch and Finnish enterprises source ICT services more than other functions, and Ireland and the UK source significant shares of engineering and related technical service functions.
European destination for international sourcing
The choice of destination for international sourcing is influenced by a multitude of factors, such as:
- availability of appropriate labour;
- infrastructure in the country selected;
- a possible network of affiliated enterprises; or even
- language and cultural barriers.
The destination for the international sourcing of all functions, either core business or support, or both, is shown in Graph 4. A first glance reveals that for those countries for which data are available, at least 40 % of the internationally sourced functions remain ‘intra-EU’. The exception is Slovenia where the proportion amounts to just over 30 %. Intra-EU sourcing can reach values as high as 69 % in Sweden, 64 % in the Netherlands, 63 % in the Czech Republic and 66 % in the non-EU Norway (in the latter, the term ‘EU sourcing’ is more appropriate than ‘intra-EU’ sourcing).
For Slovenia, ‘other European countries’ are the main destination of either core or support functions, or both. The destination is most likely to be their neighbour, Croatia. Sourcing to the USA and Canada had the highest shares in the UK and Ireland (19.5 % and 12.6 % of all enterprises respectively), followed by Italy, (10.0 %). Among the former countries, this is presumably due to the lack of a language barrier.
International sourcing to India, traditionally linked with ICT services, is not as common as might be expected. With the exception of the Netherlands and the UK, sourcing to China is more widespread. Here, too, enterprises' declarations of their sourcing destinations are not mutually exclusive. This means that overlaps can occur as enterprises source functions to various countries/country groups.
Types of sourcing business patterns
It is normal for an enterprise that its choice of sourcing destination will be influenced by the location of its international network of branches or affiliated companies. This has to be taken into account when looking at the results in the previous sections. Comparing international insourcing (sourcing within the enterprise group) and outsourcing (sourcing outside the enterprise group), Graph 5 shows that, regardless of the business sector, insourcing is generally more common than outsourcing. At the aggregated level, close to 70 % of all those surveyed enterprises which stated they were sourcing internationally did so within their own group of affiliated enterprises - insourcing. It should also be taken into account that among these groups an enterprise's degree of ownership may vary. Overall, just over 36 % of enterprises were outsourcing. But here again, enterprises might use both in- and out-sourcing, which explains why the cumulated proportions frequently exceed 100 %. Although insourcing is often twice as common as outsourcing, in Denmark and Ireland they both have similar proportions. Conversely, the gap between them is particularly wide in the Czech Republic, Germany and Sweden.
Focus on sourcing ICT services
During 2001-2006 in the 12 countries, among the surveyed enterprises which internationally sourced services effectively that were linked to information and communication technologies, an average of three quarters (75.2 %) did so within their own enterprise group - insourcing (see Graph 6). The outsourcing of ICT services was performed by just over 30 % of enterprises.
Among the participating countries, Danish, Dutch and Finnish enterprises sourced ICT services more often than other support functions (see Table 2).
Insourcing proportions of over 80 % were found in:
- Czech Republic
- the Netherlands
In Germany, a value of close to 97 % was registered. At the other end of the scale, the lowest share for insourcing was seen in Denmark (61.5 %).
Outsourcing ICT services, which always carries more business risk than insourcing, was less widespread, displaying shares ranging between 16.2 % in Slovenia and 41.3 % in Denmark. In order to maintain competitiveness, and often limited to highly-specialised tasks, enterprises sometimes have to hire the services of specialised IT companies that are not part of their enterprise group (outsourcing).
Impact of international sourcing activities
International sourcing is essentially driven by cost-efficiency concerns or to gain access to new markets (see next section).
When a manufacturing enterprise decides to source core business functions internationally (i.e. the manufacturing of goods), it is feared that jobs will be lost in the enterprise's [home] country. International sourcing has received and continues to receive a lot of policy and media attention. The word ‘outsourcing’ often has a negative perception in public opinion. Indeed, international sourcing can result in a loss of domestic jobs, but it can also improve an enterprise's competitiveness and hence secure existing jobs, as well as create new and higher-value-added jobs. These latter aspects are far less visible as it is a more indirect and long-term process.
Both aspects are reflected in the survey results on this issue, although data are not shown in detail because of data quality and confidentiality problems. It appears that the Czech Republic and Slovenia have taken advantage of the international sourcing process, as the majority of their enterprises (51 % and 60 % respectively) have had employment creations, and far more highly skilled jobs were created than lost. In the high-wage countries of Denmark, Finland and Sweden, more enterprises have had losses of highly skilled jobs than creations. In these three countries, between 20 % and 28 % of the enterprises had created jobs through international sourcing. In Germany and the United Kingdom, the situation appears more balanced between job losses and creations. In general, in all participating countries, available job loss and creation figures show that overall employment effects remain at low levels.
Main motivation is reduction of labour costs
During 2001-2006 in the 12 countries, the main motivation among the surveyed enterprises that internationally sourced services was the 'reduction of labour costs' (45 % of the enterprises – see Graph 7, although multiple answers were possible). ‘Access to new markets’ and ‘strategic decisions taken by the Group head’ followed, both with shares of around 36 %. ‘Reduction of costs other than labour costs’ was the reason in over 30 % of the enterprises.
Globally, the least-motivating reasons were ‘Following the behaviour/example of competitors/clients’(14 %), and ‘Tax or other financial incentives’ (9 %).
Investigating the same motives nationally, it appears that the picture is quite mixed. Among the 12 countries, five feature the highest percentages for the ‘reduction of labour costs’ but only one (Slovenia) ranks ‘access to new markets’ as the most important motivation factor, although it ranks second at the aggregate level.
‘Strategic decisions taken by the Group head’ came first in the Czech Republic, Portugal and the United Kingdom, whereas it was ‘other motivations’ in Germany, the Netherlands and Sweden.
Turning from the motivation for international sourcing to the perception of its actual impact, Graph 8 outlines the impacts of international sourcing considered as 'positive' at aggregate level.
Most enterprises in the surveyed countries considered 'competitiveness’ as a positive impact (65 %). Almost 60 % of the enterprises surveyed in the 12 countries (except for Portugal whose data is confidential), found the reduction of labour costs to be a positive consequence of international sourcing (not shown in Graph 8). ‘Access to new markets’ was mentioned by another 46 %. The remaining four impacts stayed close together, all being mentioned by roughly a quarter of the enterprises.
Furthermore, the survey revealed that relatively few barriers were met during the process of international sourcing (Graph 9). Between 2001 and 2006, among the enterprises that did so, the obstacle that was most often given as being 'very important' was that of the ‘Legal and administrative’ kind, which was given by 12 % of the enterprises. All other barriers had values of 10 % or less, including that of ‘linguistic or cultural barriers’ which was given by 8 % of the enterprises. ‘Uncertainty of international standards’ was the least problem: only five out of 100 enterprises considered that as a disturbing or blocking factor.
International sourcing plans for 2007-2009
Graph 10 shows the proportion of enterprises which intend to internationally source business functions between 2007 and 2009. One question referred to the destination of future sourcing activities, and another the business support functions that would likely be concerned. Compared to 2001-2006, the destination of sourcing activities appears to be slightly different. Sourcing within the EU still shows the highest shares in most of the countries for which data are available, but there are clear particularities: Slovenian enterprises will most likely continue to source mainly to ‘other European countries’ as many enterprises have done in the past. As a destination, India is of interest for Denmark and Portugal. The highest proportion of enterprises intending to source to China were in Italy (around a quarter). China was also an important destination of planned sourcing for Germany and Denmark (shares close to 20 %).
Asked about what support business functions would likely be sourced abroad in the next years (2007-2009), manufacturing enterprises generally continued to display higher proportions of planned international sourcing than those active in other sectors (data not shown). Enterprises in the United Kingdom had proportions between 20 % and 25 % in several categories (see the category labels displayed in Table 2), with a particularly high 35 % for ‘distribution and logistics’. Ireland followed with lower shares overall, but nevertheless 29 % for ‘distribution and logistics’ and 23 % for ‘engineering and related technical services’. Manufacturing enterprises in Slovenia and Germany were mainly interested in the future sourcing of ‘marketing, sales and after sales services’ (17 % for both countries).
Non-manufacturing enterprises intend to internationally source far less: The only countries where these enterprises scored proportions of over 10 % were Germany and Ireland, and this was only for ‘Marketing, sales and after sales services’ (12 % and 11 % respectively).
Data sources and availability
The data presented in this article are from a survey on the relocation of the domestic production of goods and services to producers abroad. This results from a decision taken by a resident producer to stop their domestic production. The data, which cover the period 2001-2006, has been collected on a voluntary basis in 13 countries:
- Czech Republic
- the Netherlands
- United Kingdom
The international sourcing statistics cover NACE Rev.1.1 (Statistical Classification of Economic Activities in the European Community) sections C to I and K which, broadly speaking, cover Non-financial market activities. The data refers to enterprises with 100 or more employees.
Further Eurostat information
- International Sourcing in Europe - Statistics in focus 4/2009
- Features of International Sourcing in Europe in 2001-2006 - Statistics in focus 73/2009
- Plans for International Sourcing in Europe in 2007-2009 Statistics in focus 74/2009
Methodology / Metadata
- International sourcing statistics - all activities (ESMS metadata file - iss_esms)
- Structural Business Statistics (ESMS metadata file - sbs_esms)
- ↑ The survey was economy-wide and covered enterprises with 100 or more employees.
- ↑ 2.0 2.1 Results from Sweden and the UK are affected by the special set-up used when determining target populations. In Sweden, international sourcing is probably underestimated, while in the UK it may be overestimated.
- ↑ The average has a somewhat limited reliability as the data from certain countries are provisional (Czech Republic and Portugal) or are unreliable (the Netherlands).
- ↑ As the survey covered only service activities, the Spanish results are not analysed.