Working Paper on International Trade in Healthcare Services

Description
Growing evidence indicates that international trade in healthcare services is growing. Nevertheless, a major literature gap exists with regard to the nature of international healthcare trade and its extent

Assessing International Trade in
Healthcare Services
Lior Herman
1

Lior Herman ([email protected]) is a Research Associate at ECIPE
ABSTRACT
Growing evidence indicates that international trade in healthcare services is growing. Nevertheless, a major
literature gap exists with regard to the nature of international healthcare trade and its extent.
Taking a comprehensive approach, this research examines the magnitude, directions, patterns of specialisation,
growth and other aspects related to international trade in healthcare services. Within this framework, trade is
analysed with regard to cross border trade, consumption of healthcare by foreign nationals, commercial presence
of healthcare services providers, as well as the movement of healthcare professionals across borders.
ECIPE WorkIng PaPEr • no. 03/2009
www.ecipe.org
[email protected] Rue Belliard 4-6, 1040 Brussels, Belgium Phone +32 (0)2 289 1350
JEL Code: F14, F15, F20, F22, F23, I11
Keywords: healthcare services, international trade, modes of supply, cross border trade, consumption
abroad, commercial presence, movement of natural persons
2
ECIPE WORKING PAPER
No. 03/2009
INTRODUCTION
Much has been written in recent years about the growth of international trade in healthcare
services. An abundance of stories and anecdotal evidence document patients travelling across the
globe to receive healthcare services, such as medical and cosmetic surgeries, dental care and many
other kinds of treatments. A great deal has also been written on the movement and relocation
of health professionals, such as doctors, nurses or physiotherapists. In some countries, foreign
health professionals constitute a major part of the health workforce. Furthermore, technological
advancements enabling remote servicing of healthcare, offer new opportunities to provide com-
plicated and often expensive treatments, such as surgeries, radiological examinations and even
psychiatric consultancy, in places where such expertise is lacking.
The rise of international trade in healthcare services has largely been addressed in the medical
and healthcare professional discourse as well as in the popular media. At the same time, it was
largely neglected in academic literature, particularly with regard to analysis based on holistic and
comprehensive approaches. It is indeed striking how very little is actually known on the extent
to which health services are traded internationally. While some anecdotes exist with regard to
the trade of speci?c narrow sub-sectors in healthcare, very modest knowledge exists on the
magnitude of international trade in health services. A few scholars have attempted to provide a
comprehensive analysis, however a literature gap remains (See Chanda 2001; Waeger 2007).
In an attempt to ?ll this void, this paper maps and analyses trends in international trade in health
services, drawing on a wide range of sources to provide a comprehensive and systematic picture.
In order to do so, it uses data covering developed and developing economies, although largely
focusing on developed economies, notably OECD and EU member states.
2
The paper initially
discusses in brief the general theme of measurement and classi?cation of international trade in
services. The remaining sections of the paper measure international trade in healthcare services
through modes of service supply, concluding with an overall analysis of the ?ndings.
1. MEASURING TRADE IN HEALTH SERVICES
Services are internationally traded in different dimensions which relate to the geographical
location and proximity between consumers and producers, as well as factors of production (work-
ers). These dimensions inhibit our ability to provide a single measurement which will capture the
magnitude of international trade in services, as is often the case with regard to trade in goods.
3

Stern and Hoekman de?ne three dimensions: separated services, demander-located services and
provider-located services (Stern and Hoekman 1987). While the ?rst category relates to the
trading of services across borders in the same manner in which goods are traded, the latter two
categories relate to the speci?c location where exchange is conducted. Demander-located serv-
ices refer to the mode of trade that require the presence of the supplier in close proximity to
demand, while provider-located services necessitate the movement of consumers to the location
of the suppliers. This de?nition has been widely adopted in the literature and provides also the
conceptual and legally binding framework of the General Agreement on Trade in Services (GATS)
of the World Trade Organisation (WTO) through the four modes of supply categorisation. Table
1 provides de?nitions, explanations and examples for each mode of supply.
* The ECIPE Working Paper series presents ongoing research and work in progress. These Working Papers
might therefore present preliminary results that have not been subject to the usual review process for ECIPE
publications. We welcome feedback and recommend you to send comments directly to the author(s).
3
ECIPE WORKING PAPER
No. 03/2009
TABLE 1: THE FOUR MODES OF SERVICES SUPPLY
MODE DEFINITION EXPLANATION EXAMPLE
1
Cross border
supply
The service crosses the border, while
the supplier and consumer remain in
different territories.
Sale of translation services from
country A to country B via the Internet
or fax.
2
Consumption
abroad
The consumer crosses the border to the
territory of the supplier and consumes
the service there.
The purchase of hotel accommodation
(tourism services) by a tourist from
country A when travelling in country B.
3
Commercial
presence
The supplier crosses the border to the
territory of consumption and establishes
a commercial presence.
The local establishment of a branch of
a bank from country A in country B.
4
Presence of
natural persons
Temporary movement of labour to the
consumer’s territory. This movement can
be either as an intra-corporate transfe-
ree, self-employment or salaried labour.
The employment of a person from
country A as an engineer in country B.
Services are almost always supplied or traded through more than one mode. Technology renders
feasible the supply of almost all services through cross border supply (mode 1) with very few
exceptions (World Trade Organisation 1996). The distinction between modes 3 and 4 (i.e. de-
mander-located services) is that while the supply of services through commercial presence is
more focused on the local establishment of foreign legal entities, supply of services through the
presence of natural persons
4
is concerned with the country of origin of the person supplying the
service.
An assessment of international trade in services in general and of healthcare services in particular,
must address all possible avenues through which it is actually conducted. Practically, this approach
has three main advantages. First, it offers a holistic analysis, which does not consider only cross-
border trade or FDI. Second, in the absence of border measures such as tariffs, trade in services
statistics account only partly for the degree to which trade has actually been internationalised.
Third, such an observation can shed light on the linkages and tradeoffs that exist between modes
of supply, enabling better understanding of the determinants and motivations of trade, as well as
identi?cation of barriers and impediments to trade in services.
Within this framework, measurements of international trade in services can be carried out with
the use of proxy indicators that can reduce the statistical obstacles inherent in the quanti?cation
of trade in services.
5
Its main limitation however stems from the fact that it pieces together dif-
ferent measurements which do not necessarily provide for cross modes of supply comparisons.
Nevertheless, in the absence of a uni?ed statistical approach, this framework improves on existing
measurements, which only partly capture the level of international trade in services. It allows for
a comprehensive examination as to the magnitude of internationalisation of services, using the
best available data.
Health services include both health and medical services. Health services activities broadly cor-
respond with the categorisation of health services as de?ned by Division 93 of the United Nations
Central Product Classi?cation (CPC). These services include human health services (CPC 931),
veterinary services (CPC 932) and social services related to health (CPC 933). Since statistical
classi?cations focus on the core of activity in each category, certain trade-related services are
usually left out of the de?nition of speci?c trade in services categories. In the case of trade in
health services, these include health education services and health insurance services. In order to
4
ECIPE WORKING PAPER
No. 03/2009
provide a wide and comprehensive perspective of the health sector, this study also includes health
education services. Health insurance services are not included due to the lack of cross country
comparable data.
Each of the next sections evaluates international trade through a different mode of service trade:
cross-border trade, consumption abroad, commercial presence and the movement of natural
persons. Several measurements are applied in each section to quantify the level of trade interna-
tionalisation. The ?nal section concludes the chapter.
2 MODE 1: CROSS BORDER TRADE
2.1 What is it?
Cross border trade in services occurs when the service supplier and the service consumer
remain in their respective countries, and only the service travels across the border as part of the
transaction. This mode of trade somewhat represents the typical export-import mode common
to trade in goods.
Trade in health services through mode 1 is in fact not a recent phenomenon. Traditionally, cross
border trade of health services included services such as clinical consultation and shipment of
laboratory samples. These were provided using mail, telephony and fax machines (Chanda 2001).
Nevertheless, the development of modern Information and Communication Technologies (ICT)
has enabled and increased the tradability of numerous health services, which in the past necessitat-
ed close proximity between the service provider and the patient (or the health consumer).
6
Thus,
while trade through mode 1 has been an old feature of healthcare services, it is considered to have
risen signi?cantly over the past 20 years. A US Federal report estimated in 2004 the telemedicine
market in the US at 380 million dollars, with an annual growth rate of more than 15%.
Examples of cross border trade in health services using ICT include, telemedicine, telepathol-
ogy, telesurgery, telepsychiatry, teleradiology and other analysis and diagnosis of laboratory tests,
remote consultations and surveillance, as well as remote education and the purchase of health
insurance. Such trade allows greater healthcare availability from at least two perspectives. First,
specialised treatments can be performed even in places where specialised medical professionals
are not present. This has great potential for better delivery of healthcare services in developing
countries, but also within developed countries, where specialists concentrate in larger hospitals,
often located in big cities. Second, telemedicine enables provision of healthcare on a 24/7 basis
all year round, and minimises the congestion for treatments, where the growth of demand has
increased faster than the number of medical professionals.
Box 1: International Trade in Teleradiology
Teleradiology is the electronic transmission from one place to another of radiological images and
data. Examples of teleradiology are X-Ray scans, Magnetic Resonance Imaging (MRI) and Computed
Tomographies (CT). Teleradiology can have a key role in the provision of specialised radiological treat-
ments, where specialists are scarce, such as neurology and paediatric radiology. Medical studies have
reported that technical problems are rare and that cross border teleradiology services are rapid (often
provided within 30 to 60 minutes) and precise (Wachter 2006; Steinbrook 2007).Companies have
also been offering virtual medical record repositories, which enable on the one hand, patients to store
their medical records, and on the other hand, medical facilities to transmit across secure networks,
patients medical records and results (Boland 2008).
5
ECIPE WORKING PAPER
No. 03/2009
Studies have found that the demand for imaging services has signi?cantly increased and that from
1999 to 2004, imaging services growth was 62%. Graph 1 summarises these ?ndings (Ebbert,
Meghea et al. 2007).
GRAPH 1: CUMULATIVE GROWTH IN VOLUME PER MEDICARE BENEFICIARY (%), 1999-2004
Source: (Steinbrook 2007)
However, even in markets where teleradiology is widely used, international trade in teleradiology
remains very low. A 2003 study which included 78% of all radiologists in the US found that 67%
of all radiology activities within the US were carried out using teleradiology (Ebbert, Meghea et
al. 2007). At the same time, cross border teleradiology trade has been considerably low, despite
growing supply of such services in places like India (Ebbert, Meghea et al. 2007; Boland 2008).
Various barriers may account for this lack of international trade including legal provisions prohib-
iting such trade across the border, as well as licensing and quali?cations requirements.
2.2 Cross border trade in services
Cross border trade in healthcare services is minimal and rather insigni?cant in absolute and
relative terms, particularly with regard to the share of healthcare in countries’ economies. The
level of trade ?ows are low even in countries where appropriate infrastructure for these kinds of
transactions exists. Furthermore, trade directions are often unpredictable and countries are at
times net exporters and at others net importers. Trade is also low even among countries that are
highly economically integrated, such as EU member states.
Table 2 summarises cross border export and import patterns for 16 OECD member states. The
availability of data varies considerably between countries and across years, which makes it dif?cult
to provide long term assessment. Nevertheless, it is possible to draw key ?ndings concerning the
internationalisation of healthcare services. First, trade is volatile and rather unpredictable, which
makes it dif?cult to establish the directions of trade for individual countries. The Czech Republic
and Slovenia are the only countries that can be regarded as net exporters of health services. At the
same time, Australia is the only clear net importer of health services. Other countries are at times,
in trade surplus and at other times in trade de?cit, with an unclear trade orientation.
Second, trade volatility is not only a case of the direction of trade. For all countries, including both
those who are net exporters and importers, the pattern of change in the levels of trade from one
6
ECIPE WORKING PAPER
No. 03/2009
year to another is highly unpredictable. For example, the volume of both exports and imports of
Australia, Czech Republic, Italy, Luxembourg and Slovenia, the countries for which data is best
available, are constantly changing. Thus, even if a country is a net exporter of health services, the
degree to which it is exporting seems to randomly surge or decline. This ?nding is evident when
growth of trade is calculated.
7
In Italy, for example, exports rose in 2002 by 233.8% and then
declined in the next years by 3%, 40.3% and 6.1% respectively. Similarly, Italian imports declined
in 2002 by 2.2%, then in 2003 by 13% and then rose in 2004 by 48% and declined again in the
following year by 10.1%.
Third, the level of trade for both exports and imports is signi?cantly low in absolute terms.
Exports and imports combined, as an index of trade activity, are marginal in terms of economic
activity. Trade activity is highest in Italy ($86.7 million) and Denmark ($49.2 million), and is the
lowest in Luxembourg ($2.3 million), Lithuania ($2.9 million) and Hungary ($3.9 million).Even
in large economies, such as the United Kingdom, Australia and Poland, trade activity reaches only
$29 million, $25.1 million and $20.7 million respectively.

TABLE 2: EXPORTS AND IMPORTS OF HEALTHCARE SERVICES: MODE 1 (MILLION USD)
The low levels of international trade in healthcare services through cross border trade are strik-
ing when trade is measured in relation to several other parameters, as indicated in table 3. A ?rst
indication of the relative signi?cant low trade in health services is the average ratio of total trade
to GDP.
8
On average trade in health services is as little as 0.01% of total GDP. The highest shares
COUNTRY
2000 2001 2002 2003 2004 2005
EX IM EX IM EX IM EX IM EX IM EX IM
Australia n.a. 11.02 5.172 9.83 3.803 9.78 10.38 12.32 11.77 24.27 6.109 32.07
Cyprus .. .. .. .. .. .. 0.291 0.097 3.248 3.717 7.58 8.466
Czech
Republic
15.195 12.36 23.52 11.4 28.27 18.02 .. .. 30.02 22.09 25.66 15.52
Denmark .. .. .. .. .. .. .. .. .. .. 25.51 23.68
Hungary .. .. .. .. .. .. .. .. 1.465 2.905 2.2 1.317
Italy .. .. 20.58 38.5 68.7 37.65 66.59 32.73 39.73 48.43 37.31 43.53
Korea * .. .. .. .. .. .. .. .. .. .. 2.8 3.6
Lithuania .. .. .. .. .. .. .. .. 2.877 0.04 7.316 ..
Luxembourg .. .. .. .. 1.185 0.352 1.331 1.084 0.848 1.642 1.337 1.456
Malta .. .. .. .. .. .. .. .. 4.732 1.874 3.094 3.512
Poland .. .. .. .. .. .. .. .. 10.39 9.843 12.36 8.964
Portugal .. .. 4.734 11.8 5.647 11.29 7.9 7.9 .. .. .. ..
Romania .. .. .. .. .. .. .. .. .. .. 8.72 4.983
Slovakia .. .. .. .. .. .. .. .. .. .. 11.12 5.448
Slovenia .. .. .. .. 6.673 3.839 6.386 4.083 10.41 5.529 10.09 6.284
United
Kingdom
.. .. .. .. .. .. .. .. .. .. 21.82 7.273
* Data for Korea is for 2006
Source: OECD Stat, UN
7
ECIPE WORKING PAPER
No. 03/2009
of trade to GDP are found in the Czech Republic (0.33%) and Slovakia (0.35%), while the ?gures
are much lower for other countries, such as the United Kingdom (0.001%), Italy (0.005%) and
Hungary (0.003%).
When these ?ndings are benchmarked against national healthcare expenditures, the assumption
that cross border trade in health services should mirror the activity in this sector or at least fol-
low its main trend, is not supported. Among the 16 countries examined, the proportion of total
health expenditure as a percentage of GDP is on average 7.38%, and in several states reaches
almost 10%. Furthermore, with private expenditure on healthcare services on the rise (well
above 20% of total expenditure on healthcare), the potential for greater international trade is far
from being ful?lled.
The ?ndings in table 3 also show that compared with output in the healthcare services sector,
trade is extremely minimal. Commonly used as a measurement of trade internationalisation or
trade openness, the trade-to-output index provides an insight into the relative degree to which
trade is conducted in terms of the overall production activity in a given sector (Krugman and Ob-
stfeld 2006).
9
With the exception of Slovakia, where the trade-output ratio is 4.48%, and Malta,
where it is 1.71%, the index level is below 1% for all countries. In some economies, this ratio is
as low as 0.06% (Australia, Hungary) and 0.02% (UK). The average for all countries is 0.71%.
TABLE 3: TRADE IN HEALTHCARE SERVICES IN MODE 1 AND HEALTHCARE ECONOMIC ACTIVITY
SHARE OF
AVERAGE TRADE IN
HEALTH SERVICES
OF GDP
SHARE OF TOTAL TRADE
IN HEALTH SERVICES
(EXPORTS AND IMPORTS)
OF GROSS OUTPUT OF
HEALTH SERVICES
SHARE OF TOTAL
EXPENDITURE
ON HEALTH OF
GDP
SHARE OF PRIVATE
EXPENDITURE ON
HEALTH OF TOTAL
EXPENDITURE ON
HEALTH
Australia 0.00354% 0.06% 9.18% 32.31%
Cyprus … 0.93% 5.88% 56.45%
Czech Republic 0.03304% 0.84% 7.03% 10.31%
Denmark 0.01902% 0.15% 8.84% 16.55%
Hungary 0.00361% 0.06% 7.78% 29.34%
Italy 0.00492% 0.08% 8.43% 25.22%
Korea 0.00081% … 5.34% 47.94%
Lithuania … 0.52% 6.24% 28.64%
Luxembourg 0.00632% 0.13% 7.12% 9.81%
Malta … 1.71% 8.55% 24.39%
Poland 0.00717% 0.15% 6.07% 29.83%
Portugal 0.00896% 0.11% 9.38% 27.67%
Romania … … 5.15% 29.22%
Slovakia 0.03571% 4.48% 6.52% 21.26%
Slovenia … … 8.75% 23.83%
United Kingdom 0.00132% 0.02% 7.76% 15.64%
Source: Author’s calculations based on data from OECD, UN, EU KLEMS, WHO NHA
The minimal role of trade in healthcare services in mode 1 is also evident with regard to closely
integrated economies, such as European Union (EU) member states. Proxy variables on the usage
8
ECIPE WORKING PAPER
No. 03/2009
of ICT among general practitioners and physicians show that only a fraction of patient data stored
is being cross-border transferred.
10
The data shows that advanced e-health infrastructure widely
exists throughout EU member states. Furthermore, the data also reveals that the vast majority
of general practitioners in Europe (80%) are using this infrastructure for record and storage of
individual administrative patient data, and that the lion’s share of them are also using e-health
infrastructure to record and store key medical data, such as medical history, basic medical param-
eters, symptoms and reasons for encounters, diagnoses, medications, laboratory results, ordered
examinations and results, other measurements, treatment outcomes and to a lesser extent also
radiological images (European Commission 2008).
Speci?c ?gures on EU patients’ data reveal that on average only 0.7% of stored data is being cross
border exchanged. Since this data represents trade and trade-like activity, it is striking that this
number is signi?cantly low, compared with existing high levels of e-health infrastructure and
data storage. The Netherlands (4.7%), Malta (3.3%), Cyprus (2.8%), Denmark (1.9%), France
(1.7%) and Sweden (1.5%) are the only countries where medical data exchanges across the bor-
der out of stored data are higher than 1% (European Commission 2008).
11
Table 4 measures the geographical concentration of EU member states’ trade, using the Hir-
schmann-Her?ndahl Index.
12
The ?ndings show low degree of trade orientation towards the EU
region. On average only less than 10% of member states’ trade (exports and imports) is done
within the EU. Italy and Denmark are exceptions with higher than average levels of imports from
the EU, 39% and 29% respectively.
Finally, the low intensity of trade and the lack of specialisation among the member states are also
re?ected in the measurement of their revealed comparative advantages, as indicated by table 4.
The index of Revealed Comparative Advantage (RCA)
13
shows that specialisation is particularly
low. On a scale of 1 to -1, whereby 1 indicates full comparative advantage and -1 indicates com-
plete lack of it, Cyprus had an RCA score of 0.35, the highest among the member states. Other
member states with positive RCA scores were Romania (0.33), Czech Republic (0.29), Poland
(0.28), Slovakia (0.25) and Slovenia (0.16).
TABLE 4: HEALTHCARE CROSS-BORDER TRADE CONCENTRATION AND SPECIALISATION
IN EU MEMBER STATES, 2005
HIRSCHMANN-HERFINDAHL INDEX RCA
EX IM EX+IM
Cyprus 0.06 0.03 0.05 0.35
Czech Republic 0.12 0.07 0.09 0.29
Denmark 0.11 0.27 0.19 -0.38
Hungary 0.01 0.01 0.01 -0.35
Italy 0.29 0.39 0.34 -0.12
Lithuania 0.05 n.a. n.a. n.a.
Luxembourg 0.02 0.02 0.02 -0.06
Poland 0.10 0.06 0.08 0.28
Romania 0.09 0.05 0.07 0.33
Slovakia 0.09 0.06 0.08 0.25
Slovenia 0.06 0.05 0.05 0.16
Source: Author’s calculations based on United Nations Service Trade Statistics Database
9
ECIPE WORKING PAPER
No. 03/2009
3. MODE 2: CONSUMPTION ABROAD
3.1 What is it?
Trade in health services through consumption abroad takes place when the consumer crosses
the border and consumes the service in the territory of the service provider. In contrast to cross-
border trade in services, feasibility of mode 2 trade is not subject to the availability of appropri-
ate enabling technology. In fact, ancient examples exist of people travelling to spa towns across
Europe for what they believed were the healing effects of mineral water.
Mode 2 is best exempli?ed by consumption of tourism services abroad. Within this context,
health tourism has been a common feature, though not necessarily well documented. Typical
health tourism services include cardiac surgeries, plastic and cosmetic surgeries, dental treat-
ments and fertility treatments.
Chief motivations for healthcare tourism are associated with rising costs of domestic healthcare,
in particular for specialised services, long waiting times for treatment, and lack of public health
insurance in certain countries (Ramesh 2005). Table 5 shows differences in the costs of several
specialised medical treatments between the United States (an important source of health tourists)
and three Asian countries. These differences, which at times are over 30% lower, are an important
incentive in patients’ decisions to receive treatment abroad. Other surveys comparing prices
found for example that treatments such as hip replacement can be 70% lower for a treatment
package including actual treatment, as well as travel and hotel lodging costs (Treatment Abroad
2006). Among the countries considered to be hubs of health tourism, are India, Thailand, Costa
Rica, Columbia, Hungary, Poland, Lithuania, Malaysia, Jordan and Tunisia (Burne 2008; Einhorn
2008).
TABLE 5: MEDICAL COSTS: SPECIALISED TREATMENTS
COSTS (IN USD) COSTS COMPARED TO THE USA
USA SINGAPORE THAILAND INDIA SINGAPORE THAILAND INDIA
Heart Bypass 130000 18500 11000 10000 14.23% 8.46% 7.69%
Heart Valve
Replacement
160000 12500 10000 9000 7.81% 6.25% 5.63%
Angioplasty 57000 13000 13000 11000 22.81% 22.81% 19.30%
Hip Replacement 43000 12000 12000 9000 27.91% 27.91% 20.93%
Hysterectomy 20000 6000 4500 3000 30.00% 22.50% 15.00%
Knee Replacement 40000 13000 10000 8500 32.50% 25.00% 21.25%
Spinal Fusion 62000 9000 7000 5500 14.52% 11.29% 8.87%
Source: Einhorn, 2008, Authors’ calculations
Another important facet of consumption abroad of health services is health education. Certain
countries have been a hub for international medical students from both developed and developing
countries. Driving factors include language af?nity, post-colonial ties, future migration incen-
tives, as well as shortages of training infrastructure (hospitals), lack of knowledge and technical
and technological capacity in the foreign student‘s home country (Khadria 2004; Gluszynski and
Peters 2005).
14
Data for Canada shows that 17.5% of all students in Canada studying life sciences
15
were foreign
or visa students (Gluszynski and Peters 2005). Similarly, the share of foreign students studying
10
ECIPE WORKING PAPER
No. 03/2009
health professions in the USA between 2005 and 2006 was almost 5%, representing an increase of
3.1% from the previous year (Institute of International Education 2006). In the United Kingdom
the share of foreign students studying clinical medicine was 1.6% (HESA 2007).
3.2 CONSUMPTION ABROAD TRADE
Consumption of healthcare services abroad is captured in trade statistics as the “Health-
related expenditure” within the “Travel” category of EBOPS.Data on exports and imports of the
consumption abroad of health services is summarised in Table 6. Trade data for this mode of supply
is more readily available, though still lacking for several economies, such as the United Kingdom,
where data is available only for 2005, or the United States, where data is not available at all.
Trade directions are much clearer for mode 2 than they are for mode 1. Table 6 shows that coun-
tries that are net exporters are Belgium, Croatia, Czech Republic, Estonia, Greece, Hungary, Italy
and Turkey. Bulgaria has been a net importer until 2002 and since then became a net exporter.
Net importing countries are Canada, Cyprus, Iceland and Luxembourg. These ?ndings largely
correspond with the assumption suggested above that price and currency differences incentivise
the consumption abroad of health services.
16
Average growth of trade in this mode of supply is relatively high, particularly when compared
with trade in mode 1. Thus, while cross-border trade growth has been volatile and with no clear
patterns, exports and imports combined in mode 2 for each country have been constantly rising.
The average growth rate for the countries covered is 23.5%. On the whole, Bulgaria and Czech
Republic have experienced exceptionally high growth rates, with an average of 32.86% and
43.24% respectively. Italy is the only exception, where average growth rate has been negative at
-0.31%.
Table 6 also shows that in absolute and relative numbers, the magnitude of trade in health services
is far more signi?cant through consumption abroad than that of cross-border trade. Italy’s and
the United Kingdom’s volumes of trade for the same year were $238 million and $233 million
respectively, compared with $87 million and $25 million in cross-border trade. Trade through
mode 2 is higher than trade in mode 1 by 20 to 30 times in several instances, and even higher in
some cases, such as in Hungary.
11
ECIPE WORKING PAPER
No. 03/2009
TABLE 6: TOTAL EXPORTS AND IMPORTS OF HEALTH SERVICES (MODE 2, $MILLION)
Nevertheless, trade in health services in mode 2 is still signi?cantly low when measured against
output in the health sector. Table 10 shows that the annual average for the countries examined
is 1.69%, with Greece and Luxembourg having the highest trade-to-output ratio of 3.8%. Italy
and the United Kingdom are at the bottom levels of trade-to-output measurement with 0.24%
and 0.09% respectively.
On average, consumption abroad of health services represents a small share of the total consump-
tion of travel services, averaging 2.06%. Croatia is an exception and 7.75% of its travel services
are attributed to health-related travel. The ?gures are exceptionally high with regard to the con-
sumption of health services by Croatians abroad, which are 17.73% of travel imports. Iceland
is also an exception in the opposite direction as 58.05% of its travel exports are attributed to
consumption of health by foreigners. These ?gures are summarised in table 7.
As in mode 1, levels of trade in health services in mode 2 are also very low compared with total
expenditure on health as a share of GDP and as a share of private expenditure. The low levels of
2000 2001 2002 2003 2004 2005
COUNTRY EX IM EX IM EX IM EX IM EX IM EX IM
Belgium 272.9 159.6 370.9 183.2 446.3 302.1 534.4 296.5
Bulgaria 1.7 1.7 2.4 3.2 2.1 2.9 3.7 1.9 3.9 3.9 7.1 5.3
Canada 63.3 213.4 63.3 237.7 63.7 231.8 73.5 249.0 81.4 263.5 90.8 283.1
Croatia 65.9 23.6 88.4 22.6
Cyprus 6.4 6.4 2.3 6.6 2.2 7.5 4.7 10.9 4.9 10.8
Czech Republic 28.7 12.6 83.4 13.7 94.5 19.0 117.5 27.1 146.2 34.1 166.2 36.2
Estonia 3.3 1.0 4.1 0.8 4.8 1.9
Germany 641.3 556.3 914.8 876.5 1035.2
Greece 15.9 56.6 42.5 17.0 58.7 27.0 73.1 19.0 61.1 20.0
Hungary 230.4 21.2 221.4 36.7
Iceland 0.5 7.4 0.4 6.3 0.2 5.1 0.8 7.1 0.2 8.6 0.1 9.1
Ireland 10.0
Italy 155.2 87.8 188.1 62.7 172.8 68.0 171.9 74.6 144.2 83.3 157.0 81.0
Latvia 1.6 0.0 1.6 1.8 1.8 0.0 3.7 0.0 1.8
Luxembourg 6.8 44.5 9.7 58.3 11.7 67.2 12.9 70.5
Korea* 50.9 98.5
Romania 2.0 4.0 2.0 1.1 1.2 2.5
Slovenia 6.0 10.3 12.1 10.2 12.8 12.4 15.8 13.3 11.2 13.9
Switzerland 569.8 585.5 689.5 696.4 889.7 897.9
Turkey 168.9 45.0 215.9 54.0 402.5 146.2
United
Kingdom
125.7 107.5

* Korea ?gures are for 2006
Source: OECD Stat, UN
12
ECIPE WORKING PAPER
No. 03/2009
trade are particularly striking compared with the relatively high degree of the share of private
expenditure on health of total expenditure on health, which is on average almost 30% in the
countries under review. Table 7 benchmarks trade in mode 2 against the above indicators.
TABLE 7: TRADE IN HEALTH SERVICES IN MODE 2 AND SELECTED INDICATORS

TOTAL TRADE IN
HEALTH SERVICES
(EXPORTS AND
IMPORTS) AS A SHARE
OF GROSS OUTPUT OF
HEALTH SERVICES
AVERAGE SHARE OF HEALTH SERVI-
CES (241) OF TRAVEL SERVICES (236)
TOTAL
EXPENDITURE
ON HEALTH AS
% OF GDP
PRIVATE
EXPENDITURE
ON HEALTH AS
% OF TOTAL
EXPENDITURE
ON HEALTH
EX IM
TRADE
VOLUME

Belgium 1.93% 4.69% 1.81% 2.96% 9.79% 28.30%
Bulgaria 0.44% 0.91% 0.60% 7.50% 41.28%
Canada 0.63% 1.79% 1.26% 9.50% 29.90%
Croatia 6.62% 17.73% 7.75% 8.13% 18.90%
Cyprus 1.79% 0.08% 0.76% 0.22% 5.88% 56.45%
Czech Republic 2.42% 2.88% 1.26% 2.33% 7.03% 10.31%
Estonia 0.97% 6.28% 4.06% 5.58% 5.12% 22.80%
Germany 0.17% 1.28% 10.58% 21.56%
Greece 0.76% 1.70% 0.90% 0.55% 9.77% 54.59%
Hungary 3.79% 0.01% 1.00% 3.62% 7.78% 29.34%
Iceland 58.06% 1.36% 0.92% 9.71% 17.66%
Ireland 0.05% 0.16% 7.12% 24.16%
Italy 0.24% 0.03% 0.42% 0.51% 8.43% 25.22%
Latvia 5.88% 0.38% 6.39% 47.05%
Luxembourg 3.83% 0.42% 2.35% 1.23% 7.12% 9.81%
Korea 12.04% 5.34% 47.94%
Romania 1.28% 5.15% 29.22%
Slovenia 1.06% 5.88% 1.57% 1.07% 8.75% 23.83%
Switzerland 0.93% 11.19% 42.11%
Turkey 3.74% 2.04% 7.40% 30.54%
United Kingdom 0.09% 0.26% 7.76% 15.64%
Source: Author’s calculations based on data from OECD, UN, EU KLEMS, WHO NHA
Contrary to the ?ndings in mode 1, closely economically integrated economies such as EU mem-
ber states trade more with each other and develop specialisation patterns. These ?ndings can be
partially attributed to the existence of several directives facilitating the movement of patients in
Europe as well as a growing body of case law against member states restricting the right of move-
ment for European patients (Hazopoulos 2006). A strong indication towards market integration in
Europe for healthcare services provided through mode 2 is given by the Hirschmann-Her?ndahl
index. As seen in table 8, Ireland had the lowest score in the index of 0.5 which indicates that half
of its trade is oriented towards Europe. This score is higher than any score reported from cross-
border trade. Other member states reported signi?cantly higher scores, sometimes beyond 0.9,
like in the case of both Belgium (0.94) and Luxembourg (0.93).
Finally, RCA scores (table 11) also point towards a clearer pattern of specialisation among the
member states. Three countries achieved relatively high RCA scores: Hungary (0.81), Greece
13
ECIPE WORKING PAPER
No. 03/2009
(0.77) and the Czech Republic (0.72). Luxembourg shows a comparative disadvantage of (-0.67).
The data concerning Italy is surprising because it suggests that Italy (0.32) has a mild compara-
tive advantage in exporting healthcare services through mode 2 but a comparative disadvantage
in mode 1.
TABLE 8: HEALTHCARE CONSUMPTION ABROAD TRADE WITHIN THE EU, 2005
4. MODE 3: COMMERCIAL PRESENCE
4.1 What is it?
Trade through commercial presence involves the movement of the service supplier to the
territory of the consumer. Most commonly, this is carried out through the establishment of
some sort of legal entity, such as subsidiaries, branches, representative of?ces, joint ventures,
partnerships and acquisitions of local companies. It overlaps to a large extent with foreign direct
investment in services.
Foreign commercial presence in the healthcare service sector has not been signi?cantly re-
searched. The vast majority of the literature has focused on speci?c case studies, rather than ac-
counting for the actual magnitude of internationalisation taking place through this mode of supply.
In these studies, the United States has been regarded as an important source country for health
care service ?rms establishing abroad, in particular in Latin America and the United Kingdom
(Holden 2002; Jasso-Aguilar, Waitzkin et al. 2004). One particular study of the United Kingdom
found that 22% of all independent hospital beds were owned by the United States (Mohan 1991:
857 cited in Holden 2002). According to a study using the Fortune Global 500 list for 2002 as a
single year, direct health services providers were the least internationalised, while producers of
goods were the most internationalised (Holden 2005).
Traditionally, data is not available for the trade of foreign companies within a country’s domestic
market. Foreign companies established are usually regarded as local entities and treated as such
in national accounts and statistics. In recent years however, countries have begun to produce For-
eign Af?liates Trade in Services Statistics (FATS) which cover a variety of indicators regarding the
TRADE (MILLIONS USD) TRADE GROWTH
HIRSCHMANN-HERFIN-
DAHL INDEX1
RCA
EX IM BALANCE EX IM EX IM EX+IM
Belgium 512.01 269.08 242.92 18.34% -1.55% 0.96 0.91 0.94 0.31
Cyprus 2.59 7.58 -4.99 -9.50% 15.58% 0.53 0.70 0.65 -0.49
Czech Republic 110.00 17.97 92.03 11.10% 39.81% 0.66 0.50 0.63 0.72
Estonia 4.16 1.76 2.41 17.71% 176.18% 0.87 0.91 0.88 0.41
Germany n.a. 802.27 -802.27 n.a. 28.80% n.a. 0.77 0.77 n.a.
Greece 39.42 5.10 34.32 -25.49% 128.64% 0.65 0.25 0.55 0.77
Hungary 193.75 20.25 173.49 -4.45% 65.20% 0.88 0.55 0.83 0.81
Ireland n.a. 4.98 -4.98 n.a. n.a. 0.50 0.50 n.a.
Italy 107.13 54.81 52.32 -9.30% 33.60% 0.68 0.68 0.68 0.32
Luxembourg 12.87 65.34 -52.47 9.84% 13.29% 1.00 0.93 0.94 -0.67
Romania n.a. 2.49 -2.49 n.a. n.a. n.a. 1.00 1.00 n.a.
Slovenia 9.14 3.92 5.22 -22.39% 288.47% 0.82 0.28 0.52 0.40
United Kingdom 58.30 65.59 -7.29 n.a. n.a. 0.46 0.61 0.53 -0.06
1
Romania’s score of 1 should not be interpreted as complete EU trade orientation since it only reports trade statistics towards the EU (hence,
EU=World)
Source: Author’s calculations based on United Nations Service Trade Statistics Database
14
ECIPE WORKING PAPER
No. 03/2009
activities of foreign companies established in the host country, including export, import, sales,
turnover and employment. This data is still incomplete, yet taken together with other sources
of information, can provide good knowledge as to the actual magnitude and patterns of trade in
services through commercial presence. Other useful sources of data are the AMADEUS database,
foreign direct investment statistics, UNCTAD’s Transnationality Index and list of cross-border
mergers and acquisitions, as well as the Fortune Global 500 index.
4.2 Commercial Presence
Using company data extracted from the AMADEUS
17
database, it is evident that commercial
presence constitutes a signi?cant mode of international trade in healthcare services.
18
Analysis
of AMADEUS data provides cross country information regarding the level and nature of activity
of foreign companies, and can thus serve as a good estimation for trade in services through com-
mercial presence.
The ?ndings, presented in table 9, indicate that international trade in Europe takes place through
establishment and ownership of companies throughout the EU and the EEA. The share of total
foreign companies within the healthcare service sector is 16.51%. This share rises to 18.21%
when social work activities and veterinary activities are ignored. These ?gures are higher by far
than the ratios of trade statistics reported for modes 1 and 2. These ?gures ought to be compared
with similar ratios in other sub-sectors in the service industry, to allow a better understanding
as to how far integration takes place through commercial presence. Nevertheless, even in the
absence of comparable data, this ratio represents a high degree of foreign ownership, particularly
as private healthcare provision is in competition with public provision and is restricted by it in
many of the member states.
Foreign ownership in medical practice activities comprises 24.39% of all ownership, while for-
eign ownership in hospital activities is 10.38%. These ?gures seem to correlate with rising private
expenditure on healthcare services, as well as with the growing tendency towards privatisation
and outsourcing that takes place in public health provision. Table 9 shows disaggregated data on
the share of foreign and domestic companies by sub-sector.
TABLE 9: DOMESTIC AND FOREIGN COMPANIES WITH ULTIMATE OWNERSHIP IN THE
HEALTH SERVICES SECTOR
SUB-SECTOR
NACE 1.1
CLASSIFICATION
DOMESTIC FOREIGN
Human health activities 8510 90.00% 10.00%
Hospital activities 8511 89.62% 10.38%
Medical practice activities 8512 75.61% 24.39%
Dental practice activities 8513 90.00% 10.00%
Other human health activities 8514 65.45% 34.55%
Veterinary activities 8520 50.00% 50.00%
Social work activities 8530 91.30% 7.97%
Total 83.49% 16.51%
Source: AMADEUS, Author’s calculations
15
ECIPE WORKING PAPER
No. 03/2009
While evidence suggests that signi?cant international trade in healthcare services takes place in
Europe through commercial presence, less evidence supports that this is indeed a global trend.
The data for both stocks and ?ows of FDI in health services suggest that they play a marginal
role. The share of inward FDI stocks in health services of total FDI in services is constantly low
at around 0.2% for developed economies. The ?gures are even lower for outward FDI stocks,
where developed countries’ position is 0.02%. Lower shares of inward and outward FDI in health
services out of total FDI in services exist for FDI ?ows. Developing countries’ FDI shares do not
exceed 0.1% when measured in three different time intervals over the past two decades. How-
ever, while the share of FDI in health services is relatively low when compared with total FDI in
services, it has been growing considerably over the last years. From 1990 to 2005, inward FDI
stocks grew by 762% and outward stock by 380% in developed economies.
Healthcare companies are also absent from major internationalisation indices, such as the Transna-
tionality and Internationalisation indices,
19
which indicate further evidence for low international
activity. Not a single healthcare company is listed in these indices for the years, 1993, 1994 and
1999 to 2006.
20

Limited indication with regard to key countries involved in commercial presence international
trade in healthcare services can be found with data on cross-border mergers and acquisitions
(M&A). M&A examine the degree to which foreign ownership of companies is spreading. The
data shown in table 10 on global M&A have been extracted from UNCTAD’s World Investment
Report for the years 2004-2006, and details cross border M&A whose value exceeds $1 billion.
Five M&A are found between the years 2004 to 2006, with no M&A taking place in 2003. The
share of those M&A out of total M&A is low, yet somewhat surprising given the lack of health
services companies within the Transnationality Index. The yearly average value of M&A in health
services for 2004-2006 is $3.9 billion.
Two of the M&A have been in the nursing and personal care facilities. Other M&A took place in
the surgical hospital industry, kidney analysis centres and drug stores and proprietary stores. The
M&A in the drug store industry has been included in this survey due to its proximity to health
services, though should not be viewed as part of the health services industry analysed here. All
acquired companies were either US or British companies, with acquiring companies spread over
three continents. With the exclusion of the drug stores M&A, all M&A were not concluded in the
same sector, and none of the acquiring companies are health services companies.
TABLE 10: CROSS-BORDER M&A DEALS IN HEALTH SERVICES WITH VALUES OF OVER $1 BILLION
COMPLETED IN 2003-2006
TOTAL NUMBER OF
M&A
M&A IN HEALTH SERVICES

NUMBER SHARE VALUE (BILLION USD)
2003 56 0 0.00% 0
2004 75 2 2.67% 6.9
2005 141 1 0.71% 1.2
2006 172 2 1.16% 3.6
2003-2006 444 5 1.13% 11.7
Source: UNCTAD World Investment Reports 2004, 2005, 2006, 2007
16
ECIPE WORKING PAPER
No. 03/2009
Holden (2005) studied the internationalisation of health service ?rms by using Fortune’s Global
500 Index for the year 2002. Fortune’s Global 500 annually lists the world’s biggest companies,
taking revenues as the indicator of ?rm size.
21
His inconclusive ?ndings showed that health service
?rms’ internationalisation is still low, though internationalisation is more prominent, to varying
extents, in industries with proximity to health services. Such industries included insurance com-
panies, pharmaceutical corporations and catering ?rms (Holden 2005).
While Fortune’s Global 500 Index provides a good estimation for ?rm size, it is less attractive
for the examination of companies’ internationalisation into foreign markets. Hence, rather than
measuring international activity, the index looks at ?rm size in terms of whether it is operating
exclusively in a single market or not. Overcoming this problem and revisiting Holden’s work,
Fortune’s Global 500 list has been analysed for the years 2005-2007 with an independent exami-
nation of each relevant company’s pro?le to assess whether it is internationally spread in foreign
markets or not.
22
The results are detailed in table 11. Ten health services companies were on the
Global 500 List in 2005 and nine companies were ranked in the following two years. The average
ranking of health services companies was 298, 262 and 245 respectively for each year, positioning
them around the middle of the index. Nine of the companies listed in the Index have appeared in
all three years, with only one company leaving the Index after 2005. The highest rank in the Index
(66) was achieved by UnitedHealth Group in 2007. However, a close examination reveals that
only ?ve of these companies are operating beyond a single market (United States). Three of them
operate in several different markets, while two companies are established in the United Kingdom
and Canada. This evidence suggests that internationalisation of large health ?rms is still at a low
level. Table 14 summarises the ?ndings from the Fortune Global 500 List.
17
ECIPE WORKING PAPER
No. 03/2009
TABLE 11: HEALTH SERVICES COMPANIES IN THE FORTUNE GLOBAL 500 LIST
YEAR
RANK IN
YEAR
COMPANY
GLOBAL 500
RANK
REVENUES
($MILLIONS)
PROFITS
($MILLIONS)
INTERNATIONAL
ORIENTATION
2007 1 UnitedHealth Group 66 71542 4159 International
2007 2 WellPoint 103 56953 3094.9 US only
2007 3 Medco Health Solutions 148 42543.7 630.2 US only
2007 4 Caremark Rx 172 36750.2 1074 US only
2007 5 Aetna 263 25568.6 1701.7 International
2007 6 HCA 265 25477 1036 US, UK
2007 7 Humana 332 21416.5 487.4 US only
2007 8 Express Scripts 411 17660 474.4 US, Canada
2007 9 Cigna 446 16547 1155 International
2006 1 UnitedHealth Group 116 45365 3300 International
2006 2 WellPoint 117 45136 2464 US only
2006 3 Medco Health Solutions 148 37871 602 US only
2006 4 Caremark Rx 173 32991 932 US only
2006 5 HCA 244 24455 1424 US, UK
2006 6 Aetna 271 22885 1635 International
2006 7 Cigna 399 16684 1625 International
2006 8 Express Scripts 413 16266 400 US, Canada
2006 9 Humana 473 14418 308 US only
2005 1 UnitedHealth Group 123 37218 2587 International
2005 2 Medco Health Solutions 137 35352 482 US only
2005 3 Caremark Rx 204 25801 600 US only
2005 4 HCA 228 23502 1246 US, UK
2005 5 WellPoint 280 20815 960 US only
2005 6 Aetna 298 19904 2245 International
2005 7 Cigna 333 18176 1438 International
2005 8 Express Scripts 405 15115 278 US, Canada
2005 9 Humana 474 13104 280 US only
2005 10 Tenet Healthcare 495 12496 -2640 US only
Source: Fortune Magazine, Individual companies’ pro?les.
5 MODE 4: MOVEMENT OF NATURAL PERSONS
5.1 What is it?
The ?nal mode of services supply takes place when labour moves between countries and pro-
duces the service in the consumer’s home territory. The movement of natural persons can take
place in various ways. First can be the movement of intra-corporate transfers, whereby employ-
ees of a certain company move between countries but are still employed within the same com-
pany.
23
Another can be the movement across the border of independent persons seeking work
independently.
Health professionals can move permanently, or temporarily, for purposes such as working holi-
days (sabbatical), study visits for the acquisition of knowledge and techniques, as well as ?xed-
term contracts. Various push and pull factors have been surveyed in the health sector to explain
18
ECIPE WORKING PAPER
No. 03/2009
this. Push factors include low wages in the home country, poor working conditions, scarcity of
resources and career development limitations. Among the pull factors are higher absolute and
relative wages, better working conditions, career opportunities, greater availability of resources
for work,the shortage of medical staff in many OECD countries and various policies enacted
by OECD countries to attract physicians and nurses (Buchan 2006; Simoens and Hurst 2006;
Buchan 2007).
5.2 Movement of Natural Persons
Generally, statistics on the movement of natural persons leave a lot to be desired. Where they
exist they are often incomplete and lack comparability between countries and sectors. Never-
theless, a growing body of literature in the medical ?eld has been examining in recent years the
magnitude and directions in the employment of health professionals and International Medical
Graduates (IMG) outside of their home country.
Several conclusions can be drawn from OECD data on the composition of foreign-trained physi-
cians in the workforce of several OECD member states (Simoens and Hurst 2006). First, great
variety exists among the source countries of physicians working abroad in OECD countries.
Source countries include most OECD member states, as well as other countries in Europe, Latin
America, Asia and Africa. But, while source countries vary to a large extent, the magnitude of
diversity is different across OECD countries. For example, while the United Kingdom and the
United States attract physicians from numerous countries in different regions, some countries,
like Denmark and Austria, are sourcing physicians from a limited range of countries.
Second, developing countries are an important source of physicians who are open to moving
abroad. Physicians are moving to OECD economies from developing nations in Africa, Eastern
Europe, the Middle East, Asia and more. India, Pakistan and South Africa play a signi?cant role as
source countries, particularly with regard to the Anglo-Saxon countries: Canada, United King-
dom and the United States.
Finally, despite the noticeable direction of exports from developing countries to developed coun-
tries, developed countries are often themselves source countries for exporting physician services
through mode 4. Germany, United Kingdom and New Zealand are examples. The direction of the
?ow of physicians between countries is not one way. Several OECD countries are at the same time
source and host countries to each other. Ireland and the United Kingdom, and Australia and the
United Kingdom are examples.
24
Analysis of data gathered in several EU member states leads to
key ?ndings concerning movement of healthcare professionals.
25
First, the numbers and shares of
foreign health professionals employed in many member states are growing. The United Kingdom
reported in 2004 that over 9.37% of its healthcare labour force was staffed with foreign nationals.
Speci?cally, 18.13% of its medical doctors were foreign nationals.These numbers are considered
to be even higher today following the 2004 and 2007 enlargements of the EU and the abolition of
barriers to cross-border movement of people within the EU (Blitz 2005; Research and Statistics
Service 2006). In the same way, 13.93% of the Netherlands’ healthcare professionals were foreign
nationals (Ministry of Justice 2006).
Second, while foreign health professionals have a growing role in the provision of healthcare
services in Europe, most of them are from non-EU countries. This evidence supports past ?ndings
indicating the signi?cance of developing countries as a supply source for healthcare professionals
employed in the EU (Simoens and Hurst 2006). For example, the share of health professionals
employed in Germany from both the EU-25 and the EEA is only 1.42% compared with 2.35%
19
ECIPE WORKING PAPER
No. 03/2009
coming from outside the EU. This difference is much higher in the Netherlands, where only
4.04% of the health professionals came from the EU, compared with 9.89% who came from
outside the EU. Similarly, only 2.36% of health professionals in the UK came from other EU
countries, in contrast to almost 7% coming from outside the EU (Derst, Heß et al. 2006; Ministry
of Justice 2006; Research and Statistics Service 2006).
Third, in some of the member states that are a destination for EU health professionals, these EU
health professionals take precedent over non-EU health professionals in specialised areas.
Fourth, in some specialised areas, EU health professionals moving to other member states rep-
resent a relatively large share of total professionals working in these ?elds, as well as signi?cantly
exceeding the share of non-EU professionals in these areas. In Austria, EU foreign nationals con-
stitute 8.17% of all physiotherapists, 7.54% of occupational therapists, 6.27% of speech thera-
pists, 6.94% of paediatric nurses and more.
26
In Belgium, 7.18% of medical doctors and 4.75%
of physiotherapists were EU foreign nationals. 8.7% of the pharmacists and 8% of psychologists
in Ireland came from other member states. In Sweden, 7.03% and 5.17% of medical specialists
and nurses respectively were from other member states. Finally, 10.53% of all psychologists in
the United Kingdom were foreign EU-nationals (Pacolet and Merckx 2006; Quinn 2006; Schutz
2006; Swedish EMN NCP 2006).
Fifth, somewhat surprisingly new member states’ share of healthcare professionals moving to
other member states is very low. In many cases, such as in Belgium, Germany, the Netherlands,
Sweden and the United Kingdom, their share is below one percent. Since the data reported ad-
dresses 2004, the year in which the EU-10 acceded to the EU, there is a possibility that a bias
exists in the data and that their actual share today is much higher. Some of the above ?ndings are
summarised in graph 2.
The data concerning the movement of health professionals suggests that rather than being in?u-
enced by legal and institutional developments at EU level to allow greater mobility for health-
care (and other) professionals, the mobility of EU health professionals to other member states
is in?uenced by a broader international trend. This general trend in the EU is considered to be
in?uenced by both shortages of healthcare professionals in many member states, as well as active
recruitment policies of some of the latter (European Migration Network 2006).
20
ECIPE WORKING PAPER
No. 03/2009
GRAPH 2: OVERVIEW OF HEALTHCARE WORKERS IN SELECTED MEMBER STATES, 2004
Further data on medical graduates in the United States, the United Kingdom, Australia and Canada
suggest that the magnitude of IMGs in physician workforce is signi?cantly high. According to the
data presented in table 12, in the United Kingdom in 2004, 28.3% of employed physicians were
IMGs. The ratio of foreign physicians has considerably increased in the past two years, particularly
for the United Kingdom, with the vast majority coming from developing countries. Developing
countries contributed 75.2%, 60.2%, 43.4% and 40% to the United Kingdom, United States,
Canada and Australia, respectively. In contrast, IMGs from these four countries accounted for
2.5%, 6.5%, 22.3% and 33.5% of the workforce (not counting the home country).
27
Table 12
reports the distribution and magnitude of IMGs in the physician workforce of those four OECD
countries disaggregated to main source countries.
Data for the United States also shows the share of IMGs within the physician workforce accord-
ing to specialisation areas. 36% of internal medicine physicians are IMGs. IMGs also account for
31.4% in psychiatry, 29% in anaesthesiology, 28%in paediatrics,20% in general surgery, 18.8%
in radiology, and 17.8% in both family medicine and obstetrics/gynaecology (American Medical
Association 2007).
Although these ?ndings should be interpreted with caution they suggest that, in contrast to other
modes of supply, the magnitude of mode 4 trade in health services is high and signi?cantly inter-
nationalised.
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
A
u
s
t
r
i
a
B
e
l
g
i
u
m
E
s
t
o
n
i
a
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
l
y
L
a
t
v
i
a
S
w
e
d
e
n
U
n
i
t
e
d

K
i
n
g
d
o
m
A
u
s
t
r
i
a
B
e
l
g
i
u
m
E
s
t
o
n
i
a
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
l
y
L
a
t
v
i
a
S
w
e
d
e
n
U
n
i
t
e
d

K
i
n
g
d
o
m
Physiotherapists Pharmacists Dentists
Non-EU Nationals Other EU Nationals Nationals
Nurses/Midwives Doctors/Physicians
A
u
s
t
r
i
a
B
e
l
g
i
u
m
E
s
t
o
n
i
a
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
l
y
L
a
t
v
i
a
S
w
e
d
e
n
U
n
i
t
e
d

K
i
n
g
d
o
m
A
u
s
t
r
i
a
B
e
l
g
i
u
m
E
s
t
o
n
i
a
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
l
y
L
a
t
v
i
a
S
w
e
d
e
n
U
n
i
t
e
d

K
i
n
g
d
o
m
A
u
s
t
r
i
a
B
e
l
g
i
u
m
E
s
t
o
n
i
a
G
e
r
m
a
n
y
I
r
e
l
a
n
d
I
t
a
l
y
L
a
t
v
i
a
S
w
e
d
e
n
U
n
i
t
e
d

K
i
n
g
d
o
m
21
ECIPE WORKING PAPER
No. 03/2009
TABLE 12: IMGS IN THE PHYSICIAN WORKFORCES OF SELECTED OECD STATES
SOURCE COUN-
TRY
IMGS FROM SC1
(NO OF WORK-
FORCE)
IMGS FROM SC1
(%WORKFORCE)
SOURCE COUNTRY
IMGS FROM
SC1 (NO OF
WORKFORCE)
IMGS FROM SC1
(%WORKFORCE)
CANADA AUSTRALIA
United Kingdom 2,735 4 United Kingdom 4,664 8.6
South Africa 1,754 2.6 India 2,143 4
India 1,449 2.1 New Zealand 1,742 3.2
Ireland 1,164 1.7 South Africa 1,253 2.3
Saudi Arabia 658 1 Sri Lanka 627 1.2
Egypt 558 0.8 Egypt 545 1
United States 519 0.8 Singapore 438 0.8
Poland 441 0.6 Ireland 424 0.8
France 432 0.6 Hong Kong 312 0.6
Pakistan 320 0.5 Poland 189 0.3
Philippines 261 0.4 Philippines 157 0.3
Australia 247 0.4 Malaysia 152 0.3
Hong Kong 224 0.3 Pakistan 133 0.2
Vietnam 223 0.3 China 112 0.2
Taiwan 189 0.3 Vietnam 108 0.2
Romania 187 0.3 Germany 101 0.2
Jamaica 179 0.3 Myanmar 93 0.2
Sri Lanka 163 0.2 Hungary 85 0.2
Lebanon 161 0.2 Serbia & Montenegro 78 0.1
Kuwait 154 0.2 Slovakia 76 0.1
SOURCE
COUNTRY
IMGS FROM SC1
(NO OF WORK-
FORCE)
IMGS FROM SC1
(%WORKFORCE)
SOURCE COUNTRY
IMGS FROM
SC1 (NO OF
WORKFORCE)
IMGS FROM SC1
(%WORKFORCE)
UNITED STATES UNITED KINGDOM
India 40,838 4.9 India 15,093 10.9
United States* 25,380 3 Ireland 2,845 2.1
Philippines 17,873 2.1 Pakistan 2,693 1.9
Pakistan 9,667 1.2 South Africa 1,980 1.4
Canada 8,990 1.1 Egypt 1,592 1.1
China 6,687 0.8 Nigeria 1,529 1.1
Former USSR 5,060 0.6 Germany 1,523 1.1
Egypt 4,593 0.5 Sri Lanka 1,422 1
Mexico 4,578 0.5 Iraq 1,248 0.9
South Korea 4,401 0.5 Australia 872 0.6
Iran 4,002 0.5 Spain 657 0.5
United Kingdom 3,439 0.4 Greece 596 0.4
Dominican
Republic
3,232 0.4 Myanmar 487 0.4
Syria 3,219 0.4 Jamaica 472 0.3
Germany 3,071 0.4 Italy 464 0.3
Lebanon 2,556 0.3 Bangladesh 464 0.3
Nigeria 2,392 0.3 Netherlands 419 0.3
Argentina 2,374 0.3 Sudan 395 0.3
Poland 2,365 0.3 Libya 394 0.3
Colombia 2,362 0.3 New Zealand 305 0.2
1 Source Country
2 U.S. IMGs are US citizens who have gone abroad for medical education and returned to the United States to practice.
Source (Mullan 2005)
22
ECIPE WORKING PAPER
No. 03/2009
6. OVERALL PATTERNS OF HEALTHCARE TRADE
Graph 3 provides a graphical illustration of the overall magnitude of international trade in
healthcare services, based on the data generated in this research. Each axis corresponds to a dif-
ferent mode of service supply and is a positive scale (modes 3 and 4 do not represent negative
measurement). Since comparable data between different modes of supply is not available (with
the exclusion of mode 1 and 2), each axis uses a different measurement for the level of interna-
tional trade. The further the area covered is from the intersection of the axes, the higher the level
of trade. The graph shows that most international trade in healthcare is conducted through the
cross border movement of healthcare professionals and the commercial establishment of foreign
?rms. The graph also illustrates the trade-off often associated between cross border trade and
commercial presence.
GRAPH 3: INTERNATIONAL TRADE IN HEALTHCARE SERVICES BY MODES OF SUPPLY
Table 16 provides a ranking for which countries are most active in these trading patterns, accord-
ing to different modes of provision
28
.
TABLE 16: LEADING COUNTRIES IN INTERNATIONAL HEALTHCARE TRADE
1
RANKING CROSS-BORDER TRADE CONSUMPTION ABROAD COMMERCIAL PRESENCE
MOVEMENT OF
PROFESSIONALS
2
1 Italy Germany USA Australia
2 Denmark Switzerland United Kingdom Germany
3 Czech Republic Belgium Canada India
4 Australia Turkey Ireland
5 United Kingdom Canada Norway
6 Poland Hungary Pakistan
7 Slovakia Italy Philippines
8 Slovenia United Kingdom South Africa
9 Cyprus Czech Republic United Kingdom
10 Romania Korea USA
1
based on the availability of data
2
Countries listed under the movement of professionals are not ranked against each other, but constitute key countries in this mode of
trade, either exporting, importing or both. The listing is in alphabetical order.
High level
High level
High level
Mode 3
(% of fereign ?rms in the sector) (Ex+Im)
(Ex+Im)
(% of health professionals in the workforce)
Mode 2
Mode 1
Mode 4
High level
23
ECIPE WORKING PAPER
No. 03/2009
CONCLUSION
On the basis of the availability of data this paper ?nds that a great deal of variation exists within
different segments of international trade in healthcare services. For the most part, international
trade is conducted through the movement of foreign health professionals between countries,
as well as the presence of foreign healthcare ?rms in local markets. International trade through
cross-border activity, whereby healthcare services are provided and consumed in different ter-
ritories, remains very low. Trade based on the travel of healthcare consumers to foreign markets
is also very low but is signi?cantly higher than cross border trade.
With regard to the movement of healthcare professionals in the EU, the results are somewhat
surprising. The high numbers of foreign healthcare professionals hosted in many member states
come mainly from outside the EU. On the one hand the mobility of EU healthcare professionals is
very low, particularly when compared with extra-EU health professionals. Even more surprising
is that low levels of movement have been noted with regard to the new member states. However,
EU healthcare professionals’ mobility rates are still higher than the overall levels of cross-border
labour mobility within the EU (Heinz and Ward-Warmedinger 2006). Furthermore, the data
shows that concentration of EU healthcare professionals takes place in some specialised healthcare
professions in several member states.
The directions of trade and specialisation patterns vary between modes of service provision and
supply. Countries are at times net exporters of healthcare through one mode of provision and at
the same time are net importers in another. Given that comparative advantage for each provision
mode is in?uenced by different elements such as labour costs, technology, transportation costs,
regulatory and legal frameworks (mobility, recognition of quali?cations, etc.) and more, it is not
surprising that specialisation patterns for the same country differ considerably between modes
of supply.
Given the economic signi?cance of the healthcare sector in overall economic activity, and in
particular the high levels of both total and private expenditure on healthcare, it seems that there
is scope and unexploited potential for greater international trade in healthcare services. Trade
is growing in cross border trade, consumption abroad, commercial presence and movement of
professionals. But the greatest potential lies in cross border provision and consumption abroad
of services.
While not attempting to address these issues, this paper opens up several questions. First, what
are the enabling market factors needed to support this trade? It is evident from the analysis of
cross border trade that the availability of technological infrastructure is not a suf?cient condi-
tion for such a provision. If technology is not enough, perhaps focus should be given to other
elements in the market, such as the existence of economies of scale, level of education, language,
labour unit costs and more. Furthermore, the issue of complementarities and trade-off between
modes of supply merits further research. Second, what is the role of government policies in the
provision of healthcare services? Government regulation and policy can have great effects on the
ability of consumers and producers to move between countries, but even if physical mobility is
unrestricted, it does not necessarily enable ?exibility and mobility of social bene?ts, contributions
or insurance, which are closely linked with consumer choice. Data con?dentiality and transfer-
ability is another issue which may in?uence cross border trade and consumption abroad. Lastly,
if trade negotiations have an effect on trade, the ?ndings of this paper, particularly with regard to
different aspects of healthcare trade, should be used to assess the current focus of negotiations,
whether bilaterally or multilaterally.
24
ECIPE WORKING PAPER
No. 03/2009
2.7 BIBLIOGRAPHY
American Medical Association. (2007). International Medical Graduates in the U.S. Work-
force: Discussion Paper. [Online]. Available at:http://www.ama-assn.org/ama1/pub/upload/
mm/18/img-workforce-paper.pdf [Accessed: March 11
th
2009]
Blitz, B., K. (2005). ‘Brain circulation’: the Spanish medical profession and international medical
recruitment in the United Kingdom.Journal of European Social Policy, 15(4), 363-379.
Boland, G. W. L. (2008). Teleradiology Coming of Age: Winners and Losers. AJR, 190(5), 1161-
1162.
Buchan, J. (2006). Migration of health workers in Europe: policy problem or policy solution?
In C.A. Dubois, M. McKee and E. Nolte (Eds.), Human resources for health in Europe, 41-62..
UK: McGraw-Hill.
Buchan, J. (2007). Health Workers Migration in Europe: Assessing the Policy Option. Euro health,
13(1), 6-8.
Burne, J. (2008). A World of Medical Opportunity: Special Report on Health Tourism. [Online].
Available at:http://www.treatmentabroad.net/medical-tourism/a-world-of-medical-opportu-
nity/ [Accessed: March 11
th
2009]
Chanda, R. (2001). Trade in Health Services. CMH Working Paper Series, Commission on Mac-
roeconomics and Health, World Health Organisation: Paper No. WG 4: 5.
Derst, P., B. Heß, et al. (2006). Labour Market Participation of Non-Germans in the Health Sec-
tor in Germany. [Online]. Available at:http://emn.sarenet.es/Downloads/prepareShowFiles.
do?directoryID=71 [Accessed: March 11
th
2009]
Dorrenbacher, C. (2000). Measuring Corporate Internationalisation. [Online]. Available at:http://skylla.wzb.eu/pdf/2000/i00-101.pdf [Accessed: March 11th 2009]
Ebbert, T. L., C. Meghea, et al. (2007). The State of Teleradiology in 2003 and Changes since
1999. AJR 188, 103-112.
Einhorn, B. (2008). Top Medical Tourism Destinations. Business Week. [Online]. Avail-
able at:http://images.businessweek.com/ss/08/03/0317_hospitals/index_01.
htm?technology+slideshows [Accessed: March 11
th
2009]
EurActiv. (2004). eHealth. [Online]. Available at:http://www.euractiv.com/en/health/ehealth/
article-117474 [Accessed: March 11
th
2009].
EurActiv. (2008). Survey: EU doctors slow to adopt eHealth practices.[Online]. Available at:http://www.euractiv.com/en/health/survey-eu-doctors-slow-adopt-ehealth-practices/article-
171978 [Accessed: March 11th 2009]
European Commission. (2004). e-Health - making healthcare better for European citizens: An
action plan for a European e-Health Area. COM (2004) 0356.
European Commission. (2008). Benchmarking ICT use among General Practitioners in Europe:
Final Report. [Online]. Available at:http://ec.europa.eu/information_society/eeurope/i2010/
docs/benchmarking/gp_survey_?nal_report.pdf [Accessed: March 11th 2009]
European Migration Network (2006). Managed Migration and the Labour Market - the
25
ECIPE WORKING PAPER
No. 03/2009
Health Sector. [Online]. Available at:http://emn.sarenet.es/Downloads/prepareShowFiles.
do?directoryID=71 [Accessed: March 11
th
2009]
Gluszynski, T. and V. Peters (2005). Survey of Earned Doctorates: A Pro?le of Doctoral Degree
Recipients. Statistics Canada. [Online]. Available at:http://www.statcan.gc.ca/pub/81-595-
m/81-595-m2005032-eng.pdf [Accessed: March 11
th
2009]
Hazopoulos, V. (2006). The ECJ Case Law on Cross Border Aspects of Health Services: Brie?ng
Note Prepared for DG Internal Policies of the Union, Policy Department Economic and Scienti?c
Policy. [Online]. Available at:http://www.europarl.europa.eu/comparl/imco/studies/0701_
healthserv_ecj_en.pdf [Accessed: March 11
th
2009]
Holden, C. (2002). The Internationalization of Long Term Care Provision. Global Social Policy,
2(1), 47-67.
Holden, C. (2005). The Internationalization of Corporate Healthcare: Extent and Emerging
Trends. Competition & Change, 9(2), 185-203.
Ietto-Gilles, G. (1998). Different Conceptual Frameworks for the Assessment of the Degree of
Internationalization: an Empirical Analysis of Various Indices for the Top 100 Transnational Cor-
porations. Transnational Corporations, 7(1), 17-40.
Institute of International Education (2006). Open Doors 2006: Report on International Educa-
tional Exchange. New York: Institute of International Education.
Jasso-Aguilar, R., H. Waitzkin, et al. (2004). Multinational corporations and health care in the
United States and Latin America: strategies, actions, and effects. Journal of Health and Social
Behavior, 45(Extra Issue), 136-157.
Khadria, B. (2004). Human Resources in Science and Techonology in India and the International
Mobility of Highly Skilled Indians. STI Working Paper 2004/7. Paris: OECD.
Krugman, P. R. and M. Obstfeld (2006). International Economics : theory and policy. Boston:
Addison Wesley.
Ministry of Justice (2006). Small Scale Study II: Managed Migration and the Labour Market –
The Health Sector, the Netherlands. [Online]. Available at:http://emn.sarenet.es/Downloads/
prepareShowFiles.do?directoryID=71 [Accessed March 11
th
2009]
Of?ce of Citizenship and Migration Affairs (2006). Managed Migration and the Labour Market
- the Health Sector, Latvia. [Online]. Available at:http://emn.sarenet.es/Downloads/prepare-
ShowFiles.do?directoryID=71 [Accessed March 11
th
2009]
Pacolet, J. and S. Merckx (2006). Managed Migration and the Labour Market - The Health Sector,
The Belgian case. [Online]. Available at:http://emn.sarenet.es/Downloads/prepareShowFiles.
do?directoryID=71 [Accessed March 11
th
2009
Quinn, E. (2006). Managed Migration and the Labour Market - the Health Sector in Ireland. [On-
line]. Available at:http://emn.sarenet.es/Downloads/prepareShowFiles.do?directoryID=71
[Accessed March 11
th
2009]
Ramesh, R. (01/02/2005). Is Health Tourism the Future? The Guardian. London.
Research and Statistics Service (2006). EMN Small Scale Study II: Managed Migration and the
26
ECIPE WORKING PAPER
No. 03/2009
Labour Market – the Health Sector, United Kingdom.[Online]. Available at:http://emn.sarenet.
es/Downloads/prepareShowFiles.do?directoryID=71 [Accessed March 11
th
2009]
Schutz, B. (2006). Managed Migration and the Labour Market - the Health Sector, Austrian
Report. [Online]. Available at:http://emn.sarenet.es/Downloads/prepareShowFiles.
do?directoryID=71 [Accessed March 11
th
2009]
Simoens, S. and J. Hurst. (2006). The Supply of Physician Services in OECD Countries. OECD
Health Working Papers. Paris: OECD.
Steinbrook, R. (2007). The Age of Teleradiology. New England Journal of Medicine, 357(1),
5-7.
Stern, R. M. and B. M. Hoekman. (1987). Issues and Data Needs for GATS Negotiations on
Services. The World Economy, 10(1), 39-60.
Swedish EMN NCP (2006). Managed Migration and the Labour Market – the Health Sec-
tor, Sweden. [Online]. Available at:http://emn.sarenet.es/Downloads/prepareShowFiles.
do?directoryID=71 [Accessed March 11
th
2009
Treatment Abroad. (2006). Treatment Abroad Price Watch 2006. [Online]. Available at:http://
www.treatmentabroad.com/about/pricewatch-survey-2006/ [Accessed: March 11
th
2009]
Wachter, R. M. (2006). International Teleradiology. North England Journal of Medicine, 354(7),
662-663.
Waeger, P. (2007). Trade in Health Services: An Analytical Framework. Advanced Studies in In-
ternational Economic Policy Research. Kiel, Kiel Institute for the World Economy: Working
Paper no. 441.
World Trade Organisation. (2006). The General Agreement on Trade in Services: An Introduc-
tion. [Online]. Available at: www.wto.org/english/tratop_e/serv_e/gsintr_e.doc [Accessed:
March 11th 2009]
FOOTNOTES
I would like to thank Lucy Davis, Fredrik Erixon and Razeen Sally for their thoughtful comments.The usual 1.
disclaimer applies.
The normative question, whether healthcare should be traded in the ?rst place, falls outside the scope of 2.
this paper.
International trade in goods is usually measured in a single number which relates to the value of units of 3.
goods sold. A good is usually ?rst produced and then sold, locally or internationally, autonomously from its
production process.
The usage of the term “natural persons” is a legal convention in describing the above movement of labour. 4.
It is used to differentiate between people who are natural and unnatural entities such as corporations.
Measurement of services activity is by far more dif?cult than that of goods, and suffers from numerous 5.
statistical ?aws, that mainly derive from the intangible nature of many services. A great deal of services
transactions are not measured since they are cross-border traded without any inspection or counting,
such as in the case of e-commerce. Another reason is that transactions which bundle together goods and
services are usually measured solely as goods transactions, thus many companies whose core activity is
in manufacturing, perform services activities, but are statistically regarded as being in the manufacturing
sector (Porter, 1998). For a comprehensive discussion see: Lipsey, R. E. (2006). Measuring International
Trade in Services, NBER Working Paper No. 12271, Cambridge: NBER
27
ECIPE WORKING PAPER
No. 03/2009
Many health service activities form part of what is referred to as e-health: 6. “the application of information
and communications technologies across a whole range of functions that affect the health sector, from
the doctor to the hospital manager and from data processing to social security administrators and the
patient” (EurActiv 2004). While e-Health is part of the general modernisation of the health sector, it is
considered to be an important infrastructure for cross border trade of services.
Growth of trade is calculated as either exports in a given year over the exports of the previous year: 7.
, or as imports in a given year over imports of the previous year

Calculated as 8. , whereby EX and IM respectively denote total exports and total
imports, i represents country and n the number of years calculated.
The index is calculated for each individual years and countries as 9. whereby EX and IM respectively
represent total exports and imports, GO indicates gross output of health services, i denotes country and
n represents year.
Although these proxy variables do not cover the whole span of activities within the health sector, they 10.
nevertheless represent an important part of it. General Practitioners are in most instances the ?rst stop
for patients seeking health and serve as a “service junction” between patients and health professionals.
Nevertheless, this data should be taken as indicative and complementary to the above analysis of trade
statistics.
The data is based on a research commissioned by the European Commission on the usage of ICT among 11.
general practitioners in Europe. The survey covers 6,789 observations obtained from comprehensive
interviews, conducted in all 27 EU member states, as well as in Norway and Iceland.

The Hirschmann-Her?ndahl Index (HHI) is given 12. by and measures the geographical
concentration of trade (exports, imports or a combination of both) by reporting the degree to which a
country’s or a region’s trade is dispersed across various destinations. The index takes values between
0 to 1, whereby higher values indicate greater concentration. In the index, d is the destination, s is the
source country or region, w is the set of countries in the world and X is the bilateral ?ow of exports from
source to destination. According to the direction of trade measured, X can be substituted by I (imports) or
TT (total trade).
Revealed Comparative Advantage is calculated as 13. whereby it is the difference between
imports (I) of country a from country b in sector j and the exports (X) of country a from country b in sector
j, over the sum of imports (I) of country a from country b in sector j and the exports (X) of country a from
country b in sector j.
The actual magnitude of consumption abroad of health education services has not been well quanti?ed in 14.
the literature and consists of anecdotes rather than systematic measurement.
Life sciences comprising of health, agriculture and biology. 15.
Although reported for only a single year, the ?gures for the UK seem to at least partially contradict the 16.
views that the UK is an importer of health tourism. Various ?gures in the media report that 50,000 United
Kingdom citizens travelled overseas for medical treatment in 2007 and that 75,000 are expected to travel
in 2008, reaching an expected ?gure of 200,000 people travelling out of the United Kingdom for health
consumption by 2010 (Ramesh 2005; Burne 2008)
Analyse Major Databases from European Sources (AMADEUS). AMADEUS database covers in-depth 17.
?nancial information for some 10 million companies in Europe. The data contains information for both
private and public companies, across countries and industries. Companies’ ?nancial data is accompanied
with ?gures and records of ownership and subsidiary data, stock prices for listed companies, mergers
and acquisitions information as well as market research and news.
The survey conducted was carried out on 3,974 companies within the EU, the European Economic 18.
Area (EEA) and Croatia. A sample size of 773 companies is used for the study following data cleanup
EX
n+1
-1
EX
n
M
abj
–X
abj
M
abj
+X
abj
RCA
baj
=
IM
n+1
-1
IM
n
n
GDP
n
IM EX
n
i
n
i
n
i ? ?
+
2006
2000
2006
2000
n
GDP
n
IM EX
n
i
n
i
n
i ? ?
+
2006
2000
2006
2000
EX
n
+IM
n
GDP
n
n n
i i i
(EX
n
+IM
n
)
GO
n
i
i
i
n
GDP
n
IM EX
n
i
n
i
n
i ? ?
+
2006
2000
2006
2000
n
GDP
n
IM EX
n
i
n
i
n
i ? ?
+
2006
2000
2006
2000
n
GDP
n
IM EX
n
i
n
i
n
i ? ?
+
2006
2000
2006
2000
s
X
sd
HHI=
d
2
)
)
sw
X
sw
28
ECIPE WORKING PAPER
No. 03/2009
to ensure accuracy and comparability. The parameters examined included company’s name, industry
sub-sector, headcount (number of employees), annual turnover, annual balance sheet total, ultimate
ownership, ultimate ownership’s country, and percentage of ultimate ownership out of total ownership.
Ultimate ownership is regarded as a single entity holding 25% or more of total direct or indirect
ownership.
The UNCTAD Transnationality Index is a scale for measuring internationalisation of transnational 19.
companies. The Index focuses on ?rms’ foreign assets and is calculated as the average of three ratios:
foreign assets to total assets; foreign sales to total sales; and foreign employment to total employment.
The Transnationality Index ranks the top 100 non-?nancial transnational companies. In conjunction with
the Index UNCTAD also provides the Internationalisation Index, which calculates the number of foreign
af?liates divided by the number of all af?liates (Ietto-Gilles 1998) and (Dorrenbacher 2000).
The absence of healthcare service companies from the indices should not be interpreted as a lack of 20.
commercial presence by these companies in international trade since the indices measure the extent to
which internationalisation takes place, rather than its actual occurrence.
Firms are categorised according to industries, and the index reports various ?nancial parameters for each 21.
company.
Assessment of Fortune’s Global 500 was motivated by the fact that it covers 500 companies annually 22.
and thus has greater coverage than provided by the Transnationality and Internationalisation indices.
Intra-corporate transference is also popularly referred to as ”relocation”. 23.
It is also noticeable that language af?nity plays an important role and physicians tend to move between 24.
countries with similar languages.
The data were assembled in 11 case studies conducted under the European Migration Network. Member 25.
states that participated in the studies include: Austria, Belgium, Estonia, Germany, Greece, Ireland, Italy,
Latvia, the Netherlands, Sweden and the United Kingdom. In almost all cases the year of reference is
2004. For the ?nal report, see: European Migration Network, 2006.
The data for Austria does not include medical doctors. 26.
For each country, the combined share of the other three reporting countries is calculated, omitting the 27.
host country itself.
Data limitations might explain why the United Kingdom is the only country that appears in all modes of 28.
supply.

doc_754406523.pdf