View
227
Download
2
Category
Preview:
Citation preview
Néstor Duch-Brown and Bertin Martens
2015
Institute for Prospective Technological Studies Digital Economy Working Paper (2015-07)
Barriers to Cross-border eCommerce in the EU Digital Single Market
European Commission
Joint Research Centre
Institute for Prospective Technological Studies
Contact information
Address: Edificio Expo. c/ Inca Garcilaso, 3. E-41092 Seville (Spain)
E-mail: jrc-ipts-secretariat@ec.europa.eu
Tel.: +34 954488318
Fax: +34 954488300
JRC Science Hub
https://ec.europa.eu/jrc
This publication is a Working Paper by the Joint Research Centre of the European Commission. It results from the Digital
Economy Research Programme at the JRC Institute for Prospective Technological Studies, which carries out economic
research on information society and EU Digital Agenda policy issues, with a focus on growth, jobs and innovation in the
Single Market. The Digital Economy Research Programme is co-financed by the Directorate General Communications
Networks, Content and Technology.
Legal Notice
This publication is a Technical Report by the Joint Research Centre, the European Commission’s in-house science service.
It aims to provide evidence-based scientific support to the European policy-making process. The scientific output
expressed does not imply a policy position of the European Commission. Neither the European Commission nor any person
acting on behalf of the Commission is responsible for the use which might be made of this publication.
All images © European Union 2015
JRC96872
ISSN 1831-9408 (online)
Spain: European Commission, Joint Research Centre, 2015
© European Union, 2015
Reproduction is authorised provided the source is acknowledged.
Abstract
This report presents empirical evidence about the internationalisation of European online firms. After describing some
characteristics that define the internationalisation of firms engaged in online activities, it mainly focuses on the obstacles
that these firms face when trying to sell their products and/or services to other EU Member States. It relies on data from a
firm survey carried out in January-February 2015 in four different sectors (manufacturing, wholesale and retail,
accommodation and food and information and communication) and is complemented with additional sources of
information. We find that a limited number of barriers really matter for online trade. These include settling the costs of
cross-border disputes, suppliers’ restrictions to selling cross-border, delivery costs, taxation rules, and knowledge of the
rules abroad. In line with the offline trade literature, the data confirm that they matter mostly for small firms, which find
it harder to overcome the fixed trade costs associated with these barriers.
Acknowledgments
The authors1 would like to thank their colleagues Lucilla Sciolli, Martin Ulbrich, Gianluca Papa, Andrea
Martens, Ivan Breskovic, Claire Whitaker, Konstantinos Zisis, and Dan Dionisie from DGs CNECT, GROW
and JUST for their comments on earlier versions of this report and contributions to the debate.
1 Both authors are researchers at the JRC/IPTS in Seville (Spain). The views and opinions expressed in this
report are the authors’ and do not necessarily reflect those of the JRC or the European Commission. This research was co-financed by DG CNECT.
4
Summary
This report presents empirical evidence about the internationalisation of European online firms. After
describing some characteristics that define the internationalisation of firms engaged in online activities,
it mainly focuses on the obstacles that these firms face when trying to sell their products and/or
services to other EU Member States. It relies on data from a firm survey carried out in January-February
2015 in four different sectors (manufacturing, wholesale and retail, accommodation and food and
information and communication) and is complemented with additional sources of information. We find
that a limited number of barriers really matter for online trade. These include settling the costs of cross-
border disputes, suppliers’ restrictions to selling cross-border, delivery costs, taxation rules, and
knowledge of the rules abroad. In line with the offline trade literature, the data confirm that they matter
mostly for small firms, which find it harder to overcome the fixed trade costs associated with these
barriers.
5
Table of Contents
Acknowledgments ............................................................................................................................................ 3
Summary ............................................................................................................................................................... 4
1. Introduction .................................................................................................................................................... 6
2. Data and methodology ............................................................................................................................ 8
2.1 Data structure and potential sample bias ................................................................... 8
2.2 Firm characteristics ................................................................................................................... 9
2.3 Firms' online activities .......................................................................................................... 10
2.4 Methodology ............................................................................................................................... 13
3. Online internationalisation: a small numbers phenomenon ............................................ 14
4. The structure of target markets in online trade..................................................................... 17
5. E-commerce premia ............................................................................................................................... 20
6. Productivity and specialisation ......................................................................................................... 26
7. Barriers to cross-border e-commerce .......................................................................................... 30
7.1 Barriers to engage in cross-border e-commerce (extensive margin) ........ 32
7.2 Barriers that limit the volume of cross-border e-commerce (intensive margin) ................................................................................................................................ 34
8. Conclusions .................................................................................................................................................. 36
References ......................................................................................................................................................... 38
Annex .................................................................................................................................................................... 39
6
1. Introduction
E-commerce plays an important role in the EU economy. It has been growing at impressive rates during
the past 15 years. In 2014, it represented on average 7% of total retail sales in the EU-28. However,
growth is concentrated mainly in domestic markets while cross-border e-commerce seems to be lagging
behind. The Digital Agenda Scoreboard (2014) reports that more than 50% of all consumers buy online
but only 15% buy online across the border. On the other hand, in 2013, 15% of all firms sold online and
nearly half of them (7%) sold across the border2. This looks substantial when compared to the share of
exporting firms in offline trade (Berthou et al., 2015) and might indicate that the most important
barriers are on the consumer side. Still, the rapid rise of the Internet in the last two decades has
nurtured the idea of the "death of distance" (Cairncross, 1997) where all online services, both domestic
and cross-border, would only be a click away. We know today that the concept was largely overstated
and geographical distance and national borders remain important factors in online trade in goods and
services (Blum & Goldfarb, 2006; Lendle et al, 2012; Gomez et al, 2014; Alaveras & Martens, 2015).
Gomez et al (2014) report that in 2011 only 18% of all B2C e-commerce spending in the EU was cross-
border spending between Member States (MS). These studies seem to confirm the importance of
barriers on the consumer side, such as consumer preference for home markets (home bias) and cultural
and language differences, as the main sources of cross-border online trade costs. However, there may
still be a margin for regulatory and supply-side barriers in holding back online trade flows.
While policy makers cannot change geographical distance or language, they may be in a position to
reduce the regulatory trade costs associated with crossing national borders. Indeed, this has been the
main preoccupation of traditional offline trade policy. Since many offline sources of trade costs continue
to apply online, the reduction or elimination of regulatory barriers that induce trade costs remains
relevant online as well. Moreover, new barriers may appear in online trade. Many potential online
barriers were identified years ago (Coppel, 2000) and they have become more relevant now that e-
commerce has emerged as an important distribution channel. For instance, diverse tax regimes,
complications with payments systems, heterogeneous consumer protection rules, cross country legal
and regulatory barriers or vertical restrictions to selling online, among others, may stand in the way of a
fully-integrated digital single market in the EU. The EU’s Digital Single Market policy package targets
many barriers that are believed to hinder cross-border e-commerce. The removal of these barriers is
crucial in boosting economic integration in both the digital economy and the more traditional offline
Single Market. Both consumers and firms would greatly benefit from the reduction of these trade costs.
The main objective of this study is to identify the barriers to cross-border e-commerce in the EU that
really matter from a supply-side or firm perspective. For this purpose, the European Commission
2 Data from Eurostat's survey on the usage of information and communication technologies in enterprises,
which refers to companies with at least 10 employees and does not consider the financial sector.
7
launched a Eurobarometer firm-level survey in early 2015 with a specific focus on the barriers to cross-
border e-commerce. The survey combines evidence on firms’ perceptions of barriers to cross-border e-
commerce with information on actual cross-border e-commerce activity of these firms. The
Eurobarometer report (Flash Eurobarometer 413, 2015) shows that several barriers are frequently
mentioned as relevant by firms of all sizes, sectors and distribution channels. In this study we go a step
further and compare the perceived barriers with observed cross-border activity in order to check
consistency while controlling for a number of factors that might affect the relationship between these
two. In line with standard trade models, we split the analysis of cross-border e-commerce between an
extensive margin (whether a firm engages in cross-border e-commerce or not) and an intensive margin
(cross-border e-commerce as a share of total e-commerce).
We find that several perceived barriers in the survey have a statistically significant negative impact on
cross-border e-commerce, especially for small firms. These include the trade costs associated with
cross-border complaints and disputes, suppliers’ restrictions to selling abroad, high delivery costs,
dealing with foreign tax regulation and the lack of knowledge of the rules in other countries. These
findings provide empirical evidence for some of the barriers targeted in the Digital Single Market
Strategy paper and mainly for small firms. The findings are in line with offline cross-border trade
research where smaller firms find it harder to overcome the fixed costs associated with cross-border
trade.
The next section describes the survey data, the characteristics of firms in the survey sample and
explains the methodology used. Section 3 describes the online internationalisation status of EU firms
that derives from the survey. Section 4 analyses the markets targeted by firms selling and buying online.
Section 5 addresses the issue of both e-commerce premia and cross-border e-commerce premia.
Section 6 describes the productivity and specialisation patterns related to online activities. Section 7
discusses the results related to the barriers to cross-border e-commerce in the EU, both in terms of
engagement (extensive margin) and sales (intensive margin). In addition, a distinction is made in terms
of firms selling online and firms buying online. The last section offers conclusions and some policy
recommendations.
8
2. Data and methodology
2.1 Data structure and potential sample bias
The data used in this report were collected by TNS on the basis of a questionnaire applied to a sample
of 8,705 firms in 26 Member States3 in early 2015. The data are unique: there is no comparable data
source on perceived barriers to cross-border e-commerce by firms in the EU. The data were first reported
in the Flash Eurobarometer 413 (2015). Survey data are inevitably subject to difficulties. The sample
has to be representative of the population. Bias can come from difficulties in defining the appropriate
population to be sampled; from under-coverage, when parts of the population are inadequately
represented; from non-response, when the survey is full of missing entries; or also from measurement
error due to the presence of leading questions or the respondents' behaviour.
There is no inventory of the population of online firms in the EU. Hence, sampling had to start from the
universe of EU firms. TNS used the Orbis database on EU firms to construct the sample for all the
countries except the UK and Ireland. Since the actual sampling procedure applied by TNS is not known,
we can only verify the outcome of the sampling procedure by comparing sample composition with other
sources of information, in this case Eurostat's statistical surveys on firms, business registers or from
various administrative source (as compiled in the Structural Business Statistics –SBS- section of
Eurostat's database).
The sample includes 400 firms for the larger Member States, 300 firms for Croatia and Slovenia, 200
firms for Latvia, Hungary, Bulgaria and 100 firms in the case of Luxemburg, Estonia and Slovakia. The
data covers four sectors: manufacturing, wholesale and retail trade, accommodation and food and
information and communication. The data can be appropriately weighted to better represent the
universe of European firms4. We compare the weighted distribution of firms by country with firm survey
data from Eurostat (SBS). We find that Spain, Greece, the Netherlands, Sweden and Germany are
slightly overrepresented; while Italy, Poland, Romania, Portugal and Bulgaria are somewhat
underrepresented. This sample bias may affect the analysis since these countries differ in their
proportions of firms doing e-commerce and cross-border e-commerce and hence these differences may
not be captured by the survey data.
Additionally, we checked for potential sample bias at the sector level. Again, comparing the weighted
distribution of firms in the sample with that from Eurostat, we noticed that manufacturing, wholesale
and retail trade and accommodation and food sectors are under-represented whereas the information
3 The survey does not include data for Cyprus and Malta. 4 Sample bias can also be introduced by differences in firms' size distribution, the proportions of the types
of firms or the channels they use to sell/buy online. However, checking for sample bias in these cases is much more difficult because of the lack of appropriate data.
9
and communication sector is over-represented in the sample. Accommodation and food has one of the
highest proportions of firms engaged in e-commerce according to Eurostat.
The structure of the database is represented in Figure A1 in the Annex. The figure shows strong attrition
in the number of replies as the questionnaire progresses towards the barrier questions. This may present
a difficulty in adequately identifying the effects of these barriers. Another potential source of attrition
bias comes from segmentation. A simple sample truncation by firm size and sector for firms selling
online leaves us with a small number of observations in some of the partition cells. Moreover, these few
firms often share the same attributes and hence drop out of the estimations because they are
considered constants. For instance, this is a problem with large firms since they only represent 6% of the
total number of firms (see below). As a result, a partition by firm size may create problems. However,
despite the database’s minor limitations, it does help us to understand the behaviour of European firms
in terms of their online and cross-border activities.
2.2 Firm characteristics
Some basic features of the sample of firms are presented in Table 1. Apart from country and sector,
firm characteristics include age, size, type, activity and sales trend. Firm size is defined in terms of
employment. We have defined four categories: micro firms with 1 to 9 employees (56%)5; small firms
with 10 to 49 employees (25%); medium-sized companies employing between 50 and 249 individuals
(13%) and large firms with 250 or more employees (6%). We also split the sample into two age groups,
young firms created after 2009 (15%) and old firms already in operation before that date (85%). Firms
can also be characterised by type: independent (82%), part of a national group (8%) or part of an
international corporation (10%). The dynamics of firm performance is captured by sales trends. Firms
were asked about the trend of their turnover from the moment of the interview back to January of
2012: sales growing by more than 25% (12%); between 5 and 25% (32%); remained roughly the same
(35%); decreased between 5 and 25% (16%); and decreased by more than 25% (5%). Finally, firms are
classified by the nature of the online markets in which they operate. Firms can operate in several
markets simultaneously, hence the sum of the shares does not add up to 100%: firms selling goods to
consumers (62%) or other firms (75%); selling digital services online to consumers (9%) or other firms
(13%); or providing traditional services offline to consumers (29%) or other firms (39%). More firms are
apparently involved in Business-to-Business (B2B) than in Business-to-consumer (B2C) trade.
Table 1 shows that firms are more likely to buy online than to sell online. However, firms selling online
are more likely to do so cross-border than firms purchasing online. The larger the firm, the more likely it
is that it sells or buys online across the border. A firm is more likely to do cross-border e-commerce if it
5 The figure in parenthesis is the un-weighted share of firms in each category.
10
has expanded in the last two years –looking for new markets- and also if it has experienced difficulties
with a declining turnover, in which case exports may be seen as a new source of revenues.
Table 1: Characteristics of firms doing e-commerce
Selling Buying
Total Cross-border* Total Cross-border*
Age
Old 41.6% 46.7% 85.6% 41.0%
Young 38.9% 53.2% 88.1% 47.6%
Size
Micro 39.4% 47.2% 87.1% 41.1%
Small 44.9% 49.0% 82.5% 47.7%
Medium 56.1% 50.8% 77.3% 48.7%
Large 76.4% 60.3% 80.0% 59.8%
Type
Independent 40.1% 47.5% 86.2% 42.0%
National group 47.7% 47.5% 85.9% 41.3%
International group 50.8% 52.8% 84.6% 51.2%
Activity
Goods to consumers 48.5% 45.7% 83.9% 39.8%
Goods to firms 42.0% 46.3% 87.2% 45.1%
Digital services to consumers 67.7% 55.4% 83.1% 50.0%
Digital services to firms 53.5% 53.2% 87.9% 56.4%
Services to consumers 48.1% 53.1% 86.9% 42.3%
Services to firms 39.3% 53.2% 90.8% 47.2%
Sales trend
Fall by more than 25% 45.8% 42.5% 88.2% 39.4%
Fall between 5% and 25% 43.0% 44.0% 83.7% 39.2%
Remained the same 36.5% 39.5% 87.0% 38.9%
Rise between 5% and 25% 42.4% 55.7% 85.5% 44.9%
Rise by more than 25% 44.2% 58.4% 89.8% 54.7% * Cross-border figures are calculated with respect to the number of firms selling or buying online. Note: Figures are computed using weights. Source: own calculations with data from Eurobarometer 413.
2.3 Firms' online activities
The questionnaire tackled two main blocks of online activities by firms: sales and purchases. A
schematic representation of the data is shown in Figure A1 in the Annex. Questions about online sales
included the channel used (own website, small or large third party platforms, EDI-type transactions) and
whether the firm sells domestically, across the border to other EU countries or to third countries (US,
China, Japan, etc.). Firms engaged in e-commerce also responded to questions addressing the
importance of some pre-defined barriers to cross-border sales. Questions about online purchases
followed a similar structure, asking for the main channels, the geographic origin and the barriers faced.
11
In the sample, 3,945 firms (45%) declared they use e-commerce to sell their products and/or services.
Within this subset of firms, the most frequent channel used is the firm's own website (79%), followed by
small platforms, large platforms and EDI-type transactions, used by 28%, 26% and 23% of firms selling
online, respectively. Multi-channel strategies are used by 40% of the firms, and the remaining 60% only
use one channel for their e-commerce sales.
The average share of online sales over total turnover (excluding firms with null share) is 25%, but the
median is 10%. However, 7% of firms selling online declare their turnover from online sales to be zero.
In contrast, 5% of firms (165) are pure players, with their online sales representing 100% of their total
turnover. Firms are also asked about the geographic destination of their online sales. While 98% of
firms selling online do so domestically, 50% sell their products online across the border to other EU
Member States and 26% also to third countries. The breakdown of the turnover from e-commerce is as
follows: on average, 81% comes from online sales in the domestic market, 14% from sales to other EU
MS and the remaining 5% from third countries. In this last group, the US, Switzerland, Norway and
Iceland are the most frequent destinations for online sales made by EU firms. In the case of cross-
border e-commerce with other EU Member States, the average firm sells online to 4 different countries,
being Germany, the UK and Italy the most frequent destination markets..
Looking at online purchases, 7,156 firms (83%) answered that they purchase online. In this group, firms
basically use the provider's website to buy online (68%). While the proportion of firms that uses small
platforms (39%) and large platforms (37%) are quite similar, the proportion of firms using EDI-type
electronic transactions is rather low (20%). The proportion of firms using just one channel to buy online
is 57% -similar to that for selling, meaning that the multi-channel strategy is used by the remaining
43% of firms.
On average, the share of purchases online (excluding observations with zero value) is 23%, and the
median is again 10%. In this case, 4% of firms, which state that they buy online, indicate that their
share of online purchases is 0%, while some 3% of firms declare that their share of online purchases is
100%. Similarly, we know whether the firms purchase online from their domestic providers or go across
the border to procure items. On average, the share of purchases coming from the domestic market
accounts for 77% of purchases online, while 18% comes from other EU Member States and 5% from
third countries. In the case of online purchases, 93% of firms purchase online domestically, 49%
purchase across the border from other EU Member States and 21% buy from third countries.
The data shows that there are some significant differences between selling and purchasing online. For
all the Member States, firms are more likely to purchase online than sell online (Figure 1). As a matter of
fact, in the majority of countries the proportion of firms purchasing online is above 80%. However, if we
12
look at the cross-border dimension, this difference is no longer valid. First, for some countries the
proportion of firms selling online across the border is higher than the proportion of firms buying online
across the border. In addition, for the remaining countries the differences are much less pronounced
than in the previous case (Figure 2). This indicates that even if purchasing online is a more frequent
activity domestically, purchasing cross-border is as likely to face barriers as selling cross-border.
Figure 1: Firms selling and buying through e-commerce, by country
Source: own calculations with data from Eurobarometer 413
Figure 2: Firms selling and buying through E-commerce across the border, by country
Source: own calculations with data from Eurobarometer 413
0.2
.4.6
.81
BE DK DE EL ES FI FR IE IT LU NL AT PT SE UK BG CZ EE HU LV LT PL RO SK SL HR
Selling Buying
0.2
.4.6
.81
BE DK DE EL ES FI FR IE IT LU NL AT PT SE UK BG CZ EE HU LV LT PL RO SK SL HR
Selling Buying
13
Table 2: Cross-border e-commerce, by sector
Engagement (%) Intensity (share)
Selling Buying Selling Buying
Total 47.8% 42.2% 10.3 12.2
Manufacturing 45.6% 40.7% 9.1 11.6
Wholesale and retail trade 36.7% 38.9% 5.9 12.9
Accommodation and food 76.7% 31.3% 23.9 7.2
Information and communication 57.0% 56.0% 12.4 13.7 Note: Figures are computed using weights. Source: own calculations with data from Eurobarometer 413.
The analysis of the incidence of barriers to cross-border e-commerce needs to take into account both
the decision to engage in sales and/or purchases across the border and the intensity with which the
firms do it. Table 2 shows some statistics by sector, weighted to better represent the overall EU picture.
The accommodation and food sector has the highest engagement rate for sales, since 77% of online
firms sell across the border. The equivalent for purchases is the information and communication sector,
where 56% of firms purchase electronically across the border. The picture is the same if we look at the
intensity of the cross-border activity: the accommodation and food sector has the highest share of e-
commerce turnover from sales to other EU countries, while information and communication firms show
the highest share of electronic purchases across the border over total electronic purchases. To sum up,
this is a unique and rich dataset that provides useful evidence on the barriers to cross-border e-
commerce sales and purchases in the EU.
2.4 Methodology
To assess the real impact of many of these barriers, we move beyond the descriptive statistics
presented and discussed in the previous section6. In the following sections, we will rely on two broad
categories of methodologies. First, we compute indicators to address specific topics related to the
process of online internationalisation of European companies and compare with their corresponding
offline values or across countries or sectors. Second, we rely on econometric analysis to find robust
evidence of the process of online internationalisation of European companies. This approach allows us
to make a direct comparison of the results of this report with previous studies, both in terms of firm-
level evidence on the (offline) internationalisation process and evidence on cross-border e-commerce.
Given the nature of the survey data and the absence of relevant counterfactuals, the regression results
presented in the following sections should be seen as correlations rather than causal effects. Still, the
results should point to some factors that hinder firms’ cross-border sales and purchases. Due to the
6 Additional descriptions of the data are provided by Flash Eurobarometer 413 (2015).
14
analytical nature of this report, the results and interpretations may differ from the findings published in
the Eurobarometer (2015) report.
3. Online internationalisation: a small numbers phenomenon
This section uses micro-data to show that online internationalised firms are few and, among these few,
only a handful of firms account for the bulk of aggregate online exports.
If we focus on online trade, we can rank each country's firms in terms of their individual online exports
to other EU Member States and then we can analyse this distribution. Table 3 shows the contribution of
different segments of the ranking to aggregate exports for the total EU economy and for each sector
included in the data. In particular, the table shows the contribution to total exports of the top 1%, 5%
and 10% of cross-border exporters. With the exception of information and communication activities, the
top 1% of online exporters account for around 90% of aggregate online exports; the top 5% for more
than 97% of online exports; and the top 10% almost reaches 100%. The results for the information and
communication sector are somewhat less extreme. Overall, Table 3 shows even more concentration in
online exports than the existing evidence on traditional offline exports. For instance, the top 1% of
offline exporters in Germany carry out 59% of exports, while the top 5 carry out 81%. In Italy, the
corresponding figures are 32% and 59%, respectively (Meyer and Ottaviano, 2007).
Table 3: Share of EU cross-border exports for top online exporters
Top 1% Top 5% Top 10%
Total 89 97 98
Manufacturing 91 98 99
Wholesale and retail 94 99 99
Accommodation and food 93 98 99
Information and communication 35 90 97 Source: own calculations with data from Eurobarometer 413
Next, we look at different country cases where we detect various degrees of concentration. Figure 3
shows four selected cases, while the full picture with all the countries in the sample can be found in the
Annex (Figure A2). For comparisons, a line representing the case in which all firms export the same value
would be a diagonal connecting the left-bottom and right-upper vertex. Hence, the further away the
curve is from this imaginary diagonal, the more concentrated the aggregate exports are in the hands of
a few firms. The figure shows a very high concentration of online exports for most countries. A
particularly striking feature is that as we move from Greece to Slovakia, the initial point also moves up,
meaning that the top online exporter represents a higher share of aggregate online exports.
15
Figure 3: Online exports concentration, selected countries
Source: own calculations with data from Eurobarometer 413
An interesting alternative is to look at the online export intensity. Table 4 does precisely that, and also
compares online exports with offline exports. The table indicates that almost 2% of firms in Belgium
that are selling online across the border have a share of online exports over turnover at least equal to
75%. In the case of offline trade, that share is 1.4%. The countries with the highest online export
intensity are Luxembourg (9.7%), Slovenia (7.4%) and Austria (6.2%). These countries are also among
those with the highest offline export intensity, to which we should add Slovakia (5.2%). At the other end
of the distribution, the countries with the lowest online export intensity are Estonia, Finland and Czech
Republic. In the offline case, the countries with the lowest export intensity are the UK, France and
Germany. These results show that, even though only a small subset of firms exports a major share of
their turnover, they still account for a large fraction of total exports. In addition, this happens in both
online and offline exports.
What Table 4 shows is that only a handful of firms export a large fraction of their turnover, both online
and offline. On aggregate, only around 2% of firms export more than 75% of their turnover online.
Similarly, around 3% of firms export more than 75% of their turnover offline. Comparing these
percentages with the ones reported in Table 3 reveals that the fraction of firms with top export intensity
is larger than the fraction of top exporters. Hence, top online exporters do not necessarily exhibit top
export intensity.
In summary, aggregate online exports are determined by a few top online exporters that are relatively
big. This points to the existence of a process, similar to what has already been found offline (Meyer and
Ottaviano, 2007) through which only firms that are large enough and have other attributes that make
.2.4
.6.8
1.2
.4.6
.81
0 .5 1 0 .5 1
Greece Hungary
UK Slovakia
Cu
mu
lative
sh
are
of
on
line
exp
ort
s
Cumulative share of firmsGraphs by coun
16
them different from the average firm can survive international competition. We will explore some of
these characteristics in Section 5. Before that, in Section 4, we will have a look at the different markets
(domestic, EU, or Rest of the World –RoW-) online firms' target.
In Section 2.2, we showed that the average EU firm sells on average to 4 export markets. The larger the
number of markets a firm serves, the larger their average distance from the firm's country of origin.
Hence, distance affects aggregate trade flows mostly by reducing the number of exporters rather than
by reducing the average exports per firm. Unfortunately, the available data do not allow us to compare
these effects. Hopefully, future research will target these very relevant issues.
Table 4: Distribution of the sample of online exporters by cross-border export intensity
Share of cross-border sales over total turnover greater than
Online Offline
75% 50% 25% 75% 50% 25%
Luxemburg 9.7 13.1 16.4 15.5 20.1 24.2
Slovenia 7.4 8.1 11.0 3.7 8.2 13.2
Austria 6.2 14.5 17.8 6.3 11.1 17.6
Ireland 3.7 9.9 16.7 2.6 4.5 12.8
Portugal 3.0 5.9 10.4 2.4 5.8 10.5
Lithuania 2.9 5.9 13.3 2.1 5.2 9.4
Belgium 1.9 4.5 6.5 1.4 3.7 12.7
Greece 1.8 4.2 9.0 2.1 3.3 7.2
Slovakia 1.8 8.4 10.4 5.2 10.1 16.2
Latvia 1.5 3.0 7.5 2.9 7.6 14.5
Denmark 1.4 2.5 4.9 3.9 9.5 15.8
Spain 1.2 2.1 5.9 2.3 4.2 9.2
Sweden 1.2 1.4 3.4 2.2 5.5 10.3
Germany 1.1 2.2 6.1 0.7 3.1 7.9
Romania 0.9 2.5 3.5 3.4 5.9 7.6
Croatia 0.8 4.7 7.9 2.4 5.7 10.4
Bulgaria 0.7 2.2 3.9 2.3 3.6 5.7
Poland 0.5 0.9 1.9 3.6 5.6 9.2
Italy 0.5 5.0 6.9 1.3 4.5 6.9
France 0.4 2.3 5.5 0.4 2.9 7.8
Hungary 0.2 1.2 3.6 1.8 6.6 11.8
UK 0.1 1.5 2.0 0.2 2.4 5.5
Netherlands 0.1 1.0 4.2 1.1 5.7 8.9
Czech Republic 0.1 3.6 7.6 1.6 4.5 8.6
Finland 0.0 1.6 3.2 1.5 2.7 6.0
Estonia 0.0 2.5 8.2 3.1 7.9 17.1
EU average 1.9 4.4 7.6 2.9 6.2 11.0 Source: own calculations with data from Eurobarometer 413
17
4 The structure of target markets in online trade
This section looks at the distribution of online sales and purchases in different target markets for firms.
For this purpose, we define three target markets: domestic, cross-border EU and rest of the world. It also
compares online sales with offline sales.
The survey supplies information about the markets that firms target. Figures A4a and A4b show the
target markets for online sales by firm size and sector, respectively. In order to be able to compare this
distribution with the one that corresponds to offline sales, Figures A5a and A5b show the target markets
for offline sales, by firm size and sector, respectively. Finally, we also look at the distribution of target
markets for firms buying online. In this case, Figures A6a and A6b show the corresponding plots for firm
size and sector. All these tables are in the statistical Annex.
Each figure portrays two types of information. First, it gives an account of the number of firms in each
segment (number of circles in each triangle). Additionally, each triangle also gives the location of each
one of these firms in the target market space. The figures show that most points are concentrated in or
around the top vertex, meaning that the majority of firms are mostly targeting their domestic markets.
This result is pervasive: it happens within the different segmentations by size and by sector and also
occurs for the different activities that firms perform - online sales (Figures A4a and A4b), offline sales
(Figures A5a and A5b) and online purchases (Figures A6a and A6b).
In order to compute a summary statistic that allow us to obtain more information about the importance
of the different target markets for the firms in the sample, we compute the distances with respect to an
hypothetical centroid defined as the top vertex. Then, we computed the average distance of each firm
from this hypothetical centroid. The results are shown in Table 5 by firm size and by sector. This variable
measures how far on average firms in each segment are from this point, where all firms would be
concentrating exclusively on the domestic market. This distance has been normalized and lies between
0, when the firm is located exactly in the hypothetical centroid, and 100 when it is located at the
maximum possible distance (i.e. in another vertex).
18
Table 5: Average distance to hypothesised Gravitational vertex, by size and sector
Online Offline
By size: Micro 10.9 7.5
Small 12.2 12.8
Medium 15.1 19.2
Large 12.7 20.8
By sector: Manufacturing 12.8 16.9
Wholesale and retail 7.5 7.3
Accommodation and food 25.5 13.0
Information and communication 11.7 9.8 Note: distance has been normalised and lies between 0 and 100. In this case, 0 means average location at the gravitational vertex, defined as 100% sales to the domestic market and 0% sales to other EU and RoW markets. A value of 100 would indicate average location at the maximum possible distance (i.e. in one of the alternative vertex).
The results from Table 5 are interesting. First, we should note that the average distances from the
hypothetical centroid are similar for online and offline sales in the two cases portrayed, i.e. by firm size
and by sector. Second, note that these distances are small and close to zero in the majority of cases.
This means that the mass of firms selling online and offline is located quite close to the top vertex. In
other words, the propensity to sell in the domestic market is very high whereas the propensity to export
–to other EU markets or to the RoW markets- is very low.
Looking at these results in more detail, we detect several interesting issues. First, looking at the values
of this indicator by firm size, we see that average distance is higher in online sales than offline sales for
micro firms, while it tends to be equal for small firms. However, it tends to be lower for medium-sized
and large firms. This indicates that micro firms rely more on online sales to be more active in
international markets whereas large firms tend to be more internationalised offline than online. When
looking at the results by sector, we see that manufacturing tends to be more open to international
markets offline than online, but that the reverse is true for both the accommodation and food and
information and communication sectors.
Finally, we have plotted the distribution of these distances comparing between online and offline. Figure
4a shows the plots by firm size and Figure 4b by sector. As can be seen from the different panels of the
figures, the majority of firms are concentrated in low values of the distance index, meaning that
domestic markets are more important than foreign markets. The proportion of firms lying far away from
the hypothetical centroid as defined previously is very small, corroborating the previous results. Finally,
19
the figures also show that the differences between the online and offline distributions are modest. With
a very few exceptions given by specific ups and downs, both distributions generally have the same
shape.
The decisions to target the domestic market instead of foreign markets, or to export online to other EU
or RoW countries instead of selling on the domestic market are not independent. A firm will decide its
strategy based on the expected profits to be obtained by each decision. To model these choices
simultaneously, a multivariate discrete choice model is needed. So, in order to account for possible
systematic correlations between decisions to sell online in different target markets, we estimate a
trivariate probit model with binary equations for each outcome (Cappellari and Jenkins, 2006). The
estimate for the cross equation correlation –presented in Table A6 in the Annex- is positive, indicating
complementarities between the three decisions (bottom lines of the table). The results from the
estimation indicate that a small number of factors affect the probability of selling on foreign markets.
Hence, most channels for selling online –with the exception of EDI-type transactions- significantly
increase the likelihood of selling online in all three target markets. Firms in the Accommodation and
food sector are more likely to be targeting all three markets. Being in the wholesale and retail sector has
a negative effect on the probability of targeting foreign markets, either other EU or RoW. Finally, the fact
that firms sell digital services to consumers also impacts positively on targeting the domestic and RoW
markets, although it has no impact on cross-border online trade with other EU Member States.
Figure 4a: Agglomeration in target market space, by size
Source: own elaboration with data from Eurobarometer 413.
0.1
.20
.1.2
0 20 40 60 80 0 20 40 60 80
Micro Small
Medium Large
Online Offline
x
Graphs by size_cat
20
Figure 4b: Agglomeration in target market space, by sector
Source: own elaboration with data from Eurobarometer 413.
5. E-commerce premia
This section shows that firms doing e-commerce score better than offline firms in various performance
measures. However, this distinction is less clear between online and offline exporters.
One important dimension of the growing use of e-commerce is that it is a form of technological change
which raises the firm's value added per unit of input, i.e. its productivity. Business-to-Business (B2B) e-
commerce can improve productivity, for instance, by permitting better inventory control, by allowing the
procurement of more suitable and cheaper inputs, and lowering transaction and information costs,
among many other reasons. Similarly, Business-to-Consumers (B2C) e-commerce can also boost
productivity by reducing time and costs in making transactions, by reducing handling costs and
facilitating better customer relationships. Hence, in this section, we look at some of the differences that
these transformations may have made in firms doing e-commerce. We also look at the productivity
differences for firms doing online trade.
Table 6a reports employment, turnover, wage and value added7 e-commerce premia8, defined as ratios
of firms doing e-commerce over firms not doing e-commerce. The message from this table is that in the
great majority of cases, e-commerce firms perform better than non-e-commerce firms. These
7 These variables were not originally included in the survey. The inclusion of wages and value added was
made possible by matching the survey with data from the Orbis database (Bureau van Dijk, 2015). For this reason, the number of firms in these cases is lower than the total number of available firms.
8 Strictly speaking, employment and turnover cannot be considered performance indicators. We deal with this issue below.
0.1
.20
.1.2
0 20 40 60 80 0 20 40 60 80
Manufacturing W holesale and retail
Accommodation and food Information and communication
Online Offline
x
Graphs by nace
21
differences are particularly acute in the case of employment and wages, for which the EU average
premia are the highest. However, as expected, there is huge cross-country variation in all items.
The overall e-commerce premia are particularly high for Bulgaria, Germany, Denmark and Croatia while
they are quite disadvantageous for Greece, Latvia, Lithuania and Slovakia. Looking at particular cases,
the employment premium is quite high in Austria while unfavourable in Greece; the turnover premium is
highest in Bulgaria and lowest in Slovakia; and the wage premium is also most unfavourable for e-
commerce firms in Slovakia while it turns out to be quite high for e-commerce companies in Poland.
Finally, an important variable is the value added premium, quite high in Hungary, Germany and Bulgaria,
and quite unfavourable in Greece, Ireland and Luxemburg. In addition, Table 6b shows that there is a
strong (and statistically significant) correlation between these variables, especially those more closely
related to performance (added value and wages).
Table 6a: Comparative performance: e-commerce
Employment Turnover Wage Added value
Austria 5.4 5.1 1.5 1.7
Ireland 5.4 1.1 2.3 0.5
Denmark 5.3 2.8 5.0
Netherlands 3.4 0.8 2.1
Croatia 3.3 6.5 3.2
Hungary 3.2 2.0 3.3 4.6
Germany 3.2 4.4 4.3 5.6
Sweden 2.5 2.1 2.0 2.1
Belgium 2.3 2.7 2.3 1.1
Spain 2.3 1.2 1.7 1.7
France 2.2 3.0 1.9 1.9
Bulgaria 2.0 9.1 2.6 5.8
Romania 2.0 6.5 3.1 3.0
Portugal 1.9 1.0 2.3 1.9
Poland 1.8 0.6 5.9 3.5
Italy 1.6 0.9 1.5 1.5
Finland 1.5 3.0 1.5 1.0
Slovenia 1.2 0.6 1.3 1.1
UK 1.2 0.9
Estonia 1.0 1.9 3.6
Luxemburg 1.0 1.3 0.7 0.6
Slovakia 0.8 0.4 0.7 1.5
Czech Republic 0.8 4.9 0.9 0.8
Lithuania 0.7 0.8
Latvia 0.7 0.7
Greece 0.5 0.5 0.9 0.3
EU average 2.3 2.0 2.3 2.2 Source: own calculations with data from Eurobarometer 413
22
Table 6b: Rank correlation of performance indicators
Employment Turnover Wage Added value
Employment 1
Turnover 0.471** 1
Wage 0.555*** 0.316 1
Added value 0.442* 0.358 0.740*** 1 Note: * indicate significant at the 10% level, ** significant at the 5% level and *** significant at the 1% level.
Additional computations to those shown in table 6a can be performed by comparing firms that are
doing cross-border e-commerce with firms that are exporting offline to other EU Member States. The
results of this exercise are presented in Table 7. In this case, the picture is not as clear as it was with e-
commerce. Here, 15 out of the 26 countries show overall positive premia, the values are in general
lower than those observed in the previous case, and the country rankings in each category are more
variable than in the previous case.
Table 7: Comparative performance: cross-border e-commerce
Employment Turnover Wage Added value
Ireland 2.0 0.6 1.4 0.3
Spain 1.6 0.9 1.5 1.5
Finland 1.5 2.2 1.6 1.0
Denmark 1.5 1.3 1.6 1.0
Netherlands 1.4 0.7 1.8
Poland 1.4 0.3 2.6 1.2
Austria 1.4 2.0 1.3 1.4
Romania 1.2 0.2 1.9 3.4
France 1.2 0.9 1.1 1.0
Slovenia 1.2 0.6 1.1 1.1
Slovakia 1.2 0.3 1.2 3.5
Croatia 1.1 6.7 1.2 1.0
Italy 1.0 0.4 0.9 0.8
Sweden 1.0 1.2 1.4 1.6
Belgium 1.0 2.0 1.0 0.7
Hungary 0.9 3.2 0.9 1.5
Portugal 0.8 0.2 0.8 0.7
Germany 0.8 0.4 0.3 0.3
Latvia 0.8 0.8
Estonia 0.8 0.7 1.2
Lithuania 0.6 0.9
Luxemburg 0.6 0.7 0.5 0.6
Bulgaria 0.5 1.2 0.6 0.6
Greece 0.3 0.2 1.5 0.6
Czech Republic 0.3 0.3 0.5 0.4
UK
EU average 1.0 1.0 1.1 1.1 Source: own calculations with data from Eurobarometer 413
23
Before entering into a more robust analysis, we computed the average performance of different groups
of firms and for a set of indicators more related to performance at the aggregate EU level. The results
are shown in Table 8. There, we defined four comparison groups: i) exporters vs. non-exporters; ii) firms
doing e-commerce vs. the rest; iii) firms doing cross-border e-commerce against the rest, and; iv) firms
doing cross-border e-commerce vs. firms doing cross-border traditional commerce. The table shows that
on average, exporters perform better than the rest. Firms doing e-commerce perform on average better
in terms of labour productivity and investment per employee, while they perform worse than other firms
in terms of R&D intensity and also show a lower average profit margin. Finally, firms doing cross-border
e-commerce perform better than the rest of firms in value added per employee, R&D intensity and
investment per employee. This group also performs better than firms doing traditional trade to other EU
Member States in terms of investment and profit rate. These indicators are simply weighted means.
Many factors – such as the size distribution of firms - can affect these ratios. A more robust
methodology would be required to effectively assess the premia associated to superior performance by
exporting firms.
Table 8: Relative performance between relevant firm groups
exporters vs. others
e-commerce vs. others
cross-border e-commerce
vs. others
cross-border e-commerce
vs. cross-border traditional commerce
Labour productivity 1.01 1.04 0.68 0.61
Value added per employee 1.21 1.09 1.04 0.96
R&D intensity 1.69 0.49 1.23 0.87
Investment per employee 1.57 1.09 1.42 1.25
Profit margin 1.03 0.91 0.97 1.11 Source: own calculations with data from Eurobarometer 413 and Orbis.
Since the contribution of Melitz (2003), literature on firm heterogeneity and international trade has
focused on the main determinants of exports at the firm level, based on productivity differentials. There
is a large amount of empirical literature assessing these different approaches, with an equivalently
large number of different methodologies, covering many industries and a great deal of countries.
However, the analysis of the factors explaining export performance for e-commerce firms has received
very little attention. This could be due to different factors such as the difficulties in accessing data on
this industry at the firm level, the high heterogeneity of the firms that form this industry, the
peculiarities of the products they produce, distribute -and eventually export- among many others.
As a second exercise, we take a closer look at the relationship between exports and productivity. Recent
empirical studies about exporting firms have accumulated a great deal of evidence showing that
exporters are more productive than non-exporters. The exporter productivity premium has become a
24
stylized fact, present in almost every country and industry for which there is firm-level data,
independently of the productivity measure used and, in some cases, it turns out to be robust when
controlling for unobserved firm characteristics.
In the literature on international trade and firm heterogeneity, the exporter productivity premium is
defined as the relative productivity differential between exporting and non-exporting firms of the same
size and from the same industry. Econometrically, it is estimated by regressing the log of productivity to
a dummy variable equal to one when the firm is an exporter and zero otherwise. In addition, the number
of employees and its squared value, and dummy variables for the industries (and years when panel
data at firm level is available) are included to control for firm size, industry affiliation and time trend.
In what follows, we apply this approach first to investigate the "e-commerce productivity premia",
comparing firms that are doing e-commerce with firms that are not doing e-commerce. Secondly, we
focus on cross-border e-commerce productivity premia, comparing in this case the sub-sample of firms
that are exporting online to other EU Member States with the rest. As a first step, the e-commerce
productivity premium is estimated using pooled data for the complete sample including dummy
variables to control for time and industry effects. The results obtained for the e-commerce productivity
premia with an OLS estimator are shown in column 1 of Table 10. Then, using a similar specification,
and in order to deal with some of the problems derived from having a dependent variable with extreme
values, we use quantile regressions to assess the validity of the results in different positions of the
distribution of the dependent variable. We run regressions at the 0.25, 0.5, 0.75 and 0.9 quantiles and
the results are shown in columns 2, 3, 4 and 5 of Table 10. In a further step, we look at robust
estimation9 of the e-commerce productivity premium, by means of which one can give a more precise
account of outliers in the database10.
Results are reported in Table 9. The first column of the table shows the standard OLS estimation and
indicates that the estimated export premium is positive, statistically significant and relevant from an
economic point of view –e-commerce firms are on average (all things being equal) 3.4% more
9 Full robustness in a regression based on cross-section data can be achieved by using the so-called MM-
estimator that can resist contamination of the data set of up to 50% of outliers (i.e., that has a breakdown point of 50 % compared to 0% for OLS (ordinary least squares)). A more detailed description of the method is beyond the scope of this report, but the interested reader can consult Verardi and Wagner (2011).
10 Productivity differences between firms can be related to other variables besides firm size and industry affiliation that are not included in the empirical model defined above to estimate the exporter productivity premium. Not controlling for these frequently unobserved factors can lead to biased estimates for the exporter premium. A standard solution for this problem that is widely used in the literature on the micro-econometrics of international firm activities is the estimation of fixed effects models for panel data. For obvious reasons, this approach is impossible for our analysis (Verardi and Croux, 2009).
25
productive than non-e-commerce firms11. When we move up in the productivity distribution, the
differences widen as well, passing from a premium of 0.7% in the 0.25 quantile to a 4% premium in the
0.9 quantile. When we use the robust estimator to avoid contamination from outliers, the estimated
premium is still positive and statistically significant, though much lower in value terms (0.9%).
Table 9: Productivity premia (in %)
OLS Quantile
MM Robust (0.25) (0.5) (0.75) (0.9) E-commerce 3.4 0.7 1.3 2.3 4.0 0.9 Online exporter 3.6 0.9 1.5 3.8 7.5 1.1 Source: own estimations with data from Flash Eurobarometer 413.
For the analysis of cross-border e-commerce premia, we follow a similar strategy. First, the online
exporter's productivity premium is estimated using OLS with the cross-section of firms. Second, we
estimate quantile regressions in order to assess the differential effects by means of parameter
estimates at different positions of the distribution of the dependent variable. Finally, we use robust
estimation to take into account more accurately the problem of outliers.
Estimation results for the online exporter productivity premium are shown in the second row of Table 9.
Results from the OLS estimator are in column 1 while results using quantile regressions are reported in
columns 2, 3, 4 and 5. Overall, the picture is similar to the results already presented. The estimated
exporter premium is statistically different from zero, positive, and large from an economic point of view
both for the OLS results (3.6%) and for the four considered quantiles, ranging from 0.9% for the lowest
quantile (0.25) to 7.5% for the largest one (0.9). Column 6 of Table 9 gives the estimated results from
the robust estimator. The estimated exporter productivity premium is again statistically highly
significant and very large from an economic point of view. When controlling for outliers, the point
estimate is smaller than the point estimate reported for the application of the non-robust standard
approach using OLS (1.1%) but still significant in economic terms. These results are in line with previous
findings concerning the exporter productivity premium in different countries and industries (Wagner,
2012; Vogel and Wagner, 2011; and Temouri et al., 2010).
Finally, Table 10 shows the results of the online exporter productivity premium by sector using the
robust estimator to control for the presence of outliers in the data. From the table we can see that the
11 The estimated coefficient for the e-commerce (cross-border e-commerce) dummy has been transformed
by 100(exp(ß)-1). The transformation shows the average percentage difference in labour productivity (all things being equal) between e-commerce (cross-border e-commerce) and non-e-commerce (non-cross-border e-commerce).
26
premium is positive and statistically significant for three of the four sectors considered. The only sector
where there are no productivity differences between online exporters and other firms is in
accommodation and food. For the other three, the premium amounts to 1% in the case of
manufacturing, 1.4% for wholesale and retail, and 1.1% in the information and communication
activities.
Table 10: Online exporters’ productivity premia, by sector (in %)
Manufacturing Wholesale and retail
Accommodation And food
Information and communication
Online exporter 1.0 1.4 0.3* 1.1 * Not statistically different from zero. Source: own estimations with data from Flash Eurobarometer 413.
An important limitation of this analysis is not being able to control for firm fixed effects. We would need
panel data to be able to take these into account. However, the findings described in this report are
relevant and novel. The productivity premium for e-commerce firms and, more interestingly, for online
exporters, is relevant. In sum, online exporters, according to these results, tend to be approximately 2%
more productive than offline-exporters. Although this figure seems plausible, productivity differences
between firms could be related to other variables besides firm size and industry affiliation that are not
included in the empirical model to estimate the exporter productivity premium, either because
information is missing or because they are unobservable to a researcher. More research would be
needed to address this issue.
6 Productivity and specialisation
In this section, we look into some sector specificities that have not been properly analysed in previous
sections. In particular, we look at the relationship between online revealed and estimated comparative
advantages at the sector and country levels. Previous empirical evidence on the internationalisation of
firms shows that countries generate more highly productive firms (and therefore internationalised firms)
in some industries than in others. In order to elucidate these differences, explanations rely on national
specificities of the entry and exit process at the industry level as a key driver of international
competitiveness.
So far, firm-level evidence (Melitz and Ottaviano, 2007) has shown that, while low-performing firms are
active in their own domestic markets only, better performing ones are also active in foreign markets. The
more efficient these firms are, the more they sell abroad thanks to richer product lines and more
numerous destinations. A country’s penetration of foreign markets is thus mainly driven by those
27
extensive margins. How do we reconcile this recent firm-level view with the traditional macro-view of
aggregate export performance as determined by countries’ inter-industry cost differentials?
In the traditional view, a country specialises in the production of those goods that its firms are able to
supply at a relatively low cost compared with their competitors in other countries. This pattern of
specialisation in production then implies a corresponding pattern of specialisation in exports: a country is
a (net) exporter of the products in which it exhibits a relative (‘comparative’) cost advantage.
Accordingly, the observed pattern of trade can be used to infer a country’s pattern of comparative
advantage (and disadvantage) across industries. This is the idea behind the ‘index of revealed
comparative advantage’ (henceforth, simply RCA) defined as:
(1)
where X refers to exports, c is the country label, s is the industry12 label and w is the label for the group
of countries under consideration (in our case the EU-26). The index is larger (smaller) than one if the
exports of country c are more (less) specialised in industry s than the exports of the other countries. In
this case, country c is said to exhibit a revealed comparative advantage (disadvantage) in industry s. In
addition, since this index is not bounded from above, it is common to normalise it to restrict its values to
between -1 and 1 as follows:
(2)
where SRCA stands for Symmetric RCA. Figure A6 in the Annex plots the SRCA calculated for online and
offline exports for the EU-26 countries included in the survey. Looking at the different panels of the
graph, some patterns emerge. For instance, one group of countries (Germany, Finland, Netherlands, and
Sweden, among others) show many sectors in the upper-left part of the graph. This means that its
online export specialisation is on average higher than the corresponding offline export specialisation in
those sectors. The opposite is true for the group of countries including Spain, France, Latvia, Lithuania,
and others. In these countries, the majority of sectors lie in the bottom-right area of the plot, meaning
that their offline export specialisation is higher than the corresponding online specialisation in those
sectors. The other EU countries included in the survey fall between these two extremes.
12 In this section, we use the NACE classification at two digit level, which in our case gives us information for
35 different sectors. A list of these sectors is included in Table A10 in the annex.
28
On the other hand, Figure A7 in the Annex plots the SRCA of selected countries against their index of
‘estimated comparative advantage’ (ECA) across several sectors included in the survey. The ECA is
defined as:
(3)
where P is productivity, c is the country label, s is the industry label and w is the label for the group of
the countries under consideration (the EU-26 as in the previous case). The index is larger (smaller) than
one if country c is relatively more (less) productive in industry s than the other countries. In this case,
country c is said to exhibit an estimated comparative advantage (disadvantage) in industry s. As in the
previous case, the index was normalised to restrict its values between -1 and 1 (see formula 2), and is
termed Symmetric ECA (SECA).
The different panels of the figure reveal a somewhat positive correlation between the online SRCA and
SECA. The latter is obviously only a crude measure of comparative advantage in online trade, as it does
not take into account important determinants of international competitiveness such as differences in
factor prices and accessibility across countries. Nevertheless, the figure creates a bridge between the
micro and the macro perspectives.
We have already established that countries generate larger numbers of highly productive firms in some
industries than in others. As these are the firms that will be able to compete in international markets,
the aggregate online export performance of a country is therefore better in some industries than in
others. This confirms that national specificities of the entry and exit process at the industry level also
constitute the key driver of international competitiveness in online internationalisation. Hence, we have
provided evidence that the relative online export performance of countries at the macro level is
positively correlated with the relative productivity of their firms measured at the micro level.
But what is the relationship between online and offline specialisation? In order to answer this question,
we rely on a regression methodology that relates both measures13. The degree of specialisation
similarity can be tested by means of the following regression equation (country by country):
13 The methodology will only briefly be presented in this paper, while the reader can consult Dalum, Laursen
and Villumsen (1998) in particular (or Cantwell, 1989), for further details.
29
The superscripts Online and Offline refer to the activities carried out by the firms, respectively. The
dependent variable, SRCA of online activities for firm i in sector j, is tested against the independent
variable which is the value of the offline SRCA. Here, α and β are standard linear regression parameters
and ϵ is a residual term. Basically, the size of β measures how similar the online specialisation pattern
of a country is with respect to its offline specialisation pattern. If β is low, one can talk about a small
degree of specialisation overlap, while the specialisation pattern can be said to be identical if β is not
significantly different from one. In addition, β/R, where R is the correlation coefficient from the
regression, would measure whether the level of specialisation is higher online than offline. If β/R>1,
online specialisation is higher, while offline specialisation is higher if β/R<1. Figure 5 shows the results
(which are also presented in Table A5 in the Annex).
Figure 5: Relationship between online and offline specialisation.
Source: own elaboration with data from Eurobarometer 413 and results from table A5.
The figure shows that there are three groups of countries:
i) those that show a degree of online specialisation similar to their offline specialisation (values of β
close to one);
ii) those that show slightly different specialisation patterns from their offline or traditional
specialisation (values of β between 0 and 0.5) and;
iii) those that have totally different specialisation configurations online than they do offline (negative
values of β).
The downward sloping relationship between the similarity of specialisation patterns and the overall
intensity of specialisation shows that those countries more specialised in online activities are those
countries where the similarity with the offline specialisation is lower. However, the majority of countries
0,000
0,500
1,000
1,500
2,000
2,500
-1,000 -0,800 -0,600 -0,400 -0,200 0,000 0,200 0,400 0,600 0,800 1,000
|β|/
R
β
30
tend to be more specialised online than offline, a result given by the fact that the scope of online
activities still tends to be more limited than that of traditional trade activities.
7. Barriers to cross-border e-commerce
In this section, we turn our attention to the effects of the barriers on cross-border e-commerce between
the different Member States. We follow a two-step strategy, as is typically done in traditional
international trade models (Mayer and Ottaviano, 2007). First, we estimate the impact of the barriers on
a firm's decision to sell across the border. This is equivalent to the number of firms that sell cross-
border– the extensive margin. The decision to sell implies a medium to long-term strategic decision by
the firm to be present in export markets. Clearly, perceived barriers will play an important role in this
decision. The decision is a binary variable, taking the value 1 if the firm is selling online cross border and
0 otherwise. The appropriate estimation methodology is a logit or probit regression model that assesses
the impact of a series of independent explanatory variables which include the respondents’ perceptions
of the barriers to the probability of engaging in online trade. Formally, we assume that there is an
auxiliary random variable, so that:
where . X is a vector that contains variables representing firms' characteristics and the
perceived barriers. Y can then be viewed as an indicator for whether this latent variable is positive:
{
Given data on these observations, the model is estimated by Maximum Likelihood. We expect to find
negative signs for the estimated coefficients for the barrier dummy variables in the regression analysis:
a barrier would normally reduce the propensity to sell across the border.
The second step seeks to explain the impact of these perceived barriers on the volume of cross border
e-commerce -the intensive margin of cross-border trade. Volume is measured as the share of total
cross-border e-commerce; hence it is a variable that can take any value in the interval [0-100]. Here we
also expect to see negative signs on the coefficients for the barrier dummy variables in the regression
analysis. We use a generalised linear regression model (Papke and Wooldridge, 1996) to allow non-
normally distributed errors, as is the case for shares. In a generalized linear model (GLM), each outcome
of the dependent variables, Y, is assumed to be generated from a particular distribution in the
exponential family (a large range of probability distributions that includes the normal, binomial, Poisson
31
and gamma distributions, among others). The mean, μ, of the distribution depends on the independent
variables, X, through:
where E(Y) is the expected value of Y; Xβ is the linear predictor, a linear combination of unknown
parameters β; g is the link function. As before, this model is usually estimated by Maximum Likelihood.
While both the extensive and intensive margin models use the same explanatory variables, we expect to
see differences in the coefficients between the two.
The right-hand side dependent variables in the regression models are more objectively measurable
factors while the explanatory variables include more subjective perceptions of the relative importance
of barriers to cross-border trade. By combining the two in a single regression, we check which perceived
barriers have a statistically significant impact on actual firm behaviour. Respondents may have flagged
some barriers as relevant but if this does not transpire in firms’ actual behavioural outcomes the
regression analysis will demonstrate the statistical insignificance of these barriers. We also run the
regressions separately for firms’ cross-border sales and purchases.
We run the regression analysis by firm size group because we expect to see differences in the relative
importance of barriers between small and large firms, based on the findings from traditional offline
trade. It is important to highlight that when the regressions are run on the weighted sample data, all
statistical significance disappears. We discard these results because we have no information on the
weightings carried out by TNS and prefer the original un-weighted data.
Small firms have a harder time coping with fixed cost barriers because they cannot write off these fixed
costs over a large market size. The traditional trade literature also finds that there are few exporting
firms and their distribution is highly skewed: exporters are bigger, generate higher value added, pay
higher wages, employ more capital per worker and more skilled workers and have higher productivity
than non-exporting firms (Mayer and Ottaviano, 2007). In theory, e-commerce should bring trade costs
down, in particular those related to transport (in terms of time), search costs, information costs, and
distribution costs. To what extent this expected reduction in trade costs suffices to engage more and
smaller firms in international trade or deepen their internationalisation is an empirical research question
that we address in this study.
32
7.1 Barriers to engage in cross-border e-commerce (extensive margin)
In this section, we will first refer to the results associated with the characteristics of firms, which are
common for the analysis of selling and buying online across the border. Secondly, we discuss the
specific results regarding the barriers firms face vis-à-vis their decision to engage in cross-border sales
or to carry out cross-border purchases. Tables A6 and A7 in the Annex, respectively, show the full set of
results. The first column displays the overall (aggregated) results; the following columns show the
results by firm size. Stars next to the coefficients show the level of statistical significance. As can be
observed, in general many coefficients are not significant and should be considered as zero (i.e., having
no effect on the decision to sell or buy across the border). The table shows the marginal effects or the
impact of each of the independent variables on the probability of doing cross-border e-commerce.14
Contrary to our expectations, Tables A6 and A7 show that size and age have no impact on the
companies' decisions to sell or buy cross-border. This is a somewhat strange result since the importance
of firm size in online exports has been confirmed by other studies (Alaveras & Martens, 2015). In this
respect, little is known about the role of size for online imports. Since we are dealing with marginal
effects, an alternative interpretation is that size and age do not affect the probability of engaging in
cross-border sales and purchases. Hence, factors other than size and age should have a more relevant
role in these decisions.
For instance, distribution channels do have an impact. Firms using their own websites or third-party
platforms are more likely to sell cross-border, particularly micro and small firms. This effect disappears
for medium-sized firms and even turns negative for large firms. Firms using their providers' websites are
more likely to be purchasing across the border. This factor is driven exclusively by micro firms. Firms
using EDI-type channels to purchase electronically are also more likely to buy across the border, a result
mainly driven by small firms. The use of small platforms seems to be correlated with the probability of
large firms buying from other EU countries online, consistent with anecdotal evidence on the role of B2B
platforms for intermediate goods transactions.
The type of products sold by firms also has no significant impact on cross-border e-commerce. This
result is surprising since we are considering specifically two categories in which firms sell digital services
online. Apparently, these firms are focused exclusively on their domestic markets, or other factors like
copyright laws restrict them from selling across the border. A possible alternative explanation is that
most firms declare they are engaged in several activities at the same time, and these multiple answers
could have an effect on the estimated coefficients. Here it would be interesting to explore the effects of
diversification on the decision to sell across the border. On the other hand, firms selling goods to
14 More detailed results are available from the authors upon request.
33
consumers are less likely to buy cross-border while firms selling goods to firms are more prone to cross-
border electronic purchases, again consistent with a B2B interpretation of these data. This last result is
driven mainly by micro firms.
Results by sector show that taking the manufacturing sector as a reference, firms in the accommodation
and food sector, basically related to tourism activities, are more likely to place their services across the
border. This result is consistent for any size class within this sector. No significant differences are found
for the remaining sectors, except for the case of large firms in the wholesale and retail trade sector,
which are less likely than their manufacturing counterparts to engage in cross-border online sales.
In terms of the barriers to cross-border online sales, the results including the full sample show that the
strategic decision to sell across the border is related to three barriers: high delivery costs; suppliers'
restrictions to selling abroad; and the cost of resolving complaints and disputes (see Table 11 for a
summary by firm size). According to these results, lowering the delivery costs could represent an
increase of 7.5% in the number of firms engaged in selling online across the border. Similarly, getting rid
of the vertical restrictions imposed by suppliers would boost the number of firms selling online by
10.6%. Finally, an additional 9% in the number of firms selling online cross-border could be achieved by
lowering the costs of resolving cross-border complaints and disputes. However, the incidence of these
barriers differs across firm size. For instance, delivery costs are more relevant for larger firms. Suppliers'
restrictions and dispute resolution costs matter only for smaller firms. Other statistically significant
barriers include copyright in the case of micro firms; and taxation and product labelling costs for small
firms. It should be noted that the effects by firm size are more important than the results at the
aggregated level. For instance, removing the restrictions imposed by suppliers could increase the
number of small firms selling online across the border by 40%.
Table 11: Statistically relevant barriers to cross-border e-commerce sales by firm size: extensive margin Firm size Barrier Micro Small Medium Large Delivery costs are too high -15.3% -17.8% Copyright prevents you from selling abroad or is too expensive to sell abroad
-14.6%
Dealing with foreign taxation is too complicated or too costly -21.1% Your product labelling has to be adapted -13.3% Your suppliers restrict or forbid you to sell abroad -40.2% Resolving complaints and disputes cross-border is too expensive -9.8% -10.5%
Source: own estimations with data from Flash Eurobarometer 413.
Turning to purchases, the three main relevant barriers for cross-border purchases are related to security
of payments, language skills and the cost of resolving complaints and disputes. The elimination of these
barriers would increase the number of firms engaging in online purchases across the border by 5%, 7%
34
and 12%, respectively. As before, these barriers are more relevant for smaller than for larger firms, with
the exception of dispute resolution, which was also found to be significant for large firms (see table 12).
Table 12: Statistically relevant barriers to cross-border e-commerce purchases by firm size: extensive margin Firm size Micro Small Medium Large Payments to other countries are not secure enough -7.7% You lack the language skills for dealing with foreign countries -6.7% Resolving complaints and disputes cross-border is too expensive -14.0% -9.6% -21.5% Source: own estimations with data from Flash Eurobarometer 413.
These results indicate that some barriers are statistically significant but with a puzzling positive sign. In
the aggregated results for online sales these are data protection concerns and firms' slow internet
connection; while in the case of online purchases this only occurs for product labelling. This also
happens for some other barriers with the results disaggregated by firm size. However, a positive sign
suggests that a barrier actually increases cross-border trade. This is implausible. We therefore discard
these results.15
7.2 Barriers that limit the volume of cross-border e-commerce (intensive
margin)
In this section, we look at the effects of barriers on the volume of cross-border sales and purchases. As
we did in the previous section, we first briefly offer an overview of the effects of the different firm's
characteristics and then we proceed to a more detailed analysis of the incidence of the different barriers
on cross-border volumes of sales and purchases, respectively. The main results are presented in Tables
A8 and A9 in the annex. As before, the first column shows the aggregated results, while the following
columns display the results by firm size. The table shows the marginal effects or the impact of each of
the independent variables on the share of cross-border e-commerce over total trade. Statistically
significant coefficients have stars attached to them.
Results from Tables A8 and A9 show that firm size is positively correlated with online export intensity.
Hence, larger firms generate a larger share of revenue from online exports than smaller firms. However,
this is not the case with online imports. In the case of the intensive margin, age is not significant in both
equations. In addition, in the overall results the different distribution channels have no effect. However,
medium-sized firms using large platforms are likely to have higher shares of cross-border e-commerce
sales. The same correlation occurs with EDI-type transactions and large firms.
15 This result could also arise from problems related to the appropriate definition of the sample.
Unfortunately, we are not able to appropriately control for biases included in the sample.
35
The different activities carried out by firms are seldom relevant. In the overall sample, only those firms
that sell goods to consumers are less likely to sell online across the border, a result that is mainly driven
by the smaller size classes (micro and small). In addition, medium-sized firms selling goods to firms are
also less likely to go online across the border. The rest of the activities do not show significant
correlations with the decision to engage in cross-border e-commerce. However, if we look at purchases,
the type of market matters more, since firms selling goods to consumers and firms selling traditional
services to firms show a significantly lower intensity of cross-border purchases than firms performing
other activities, and particularly traditional services to consumers. In terms of the differences by size-
class, these results are again mainly driven by micro and small firms, whereas medium-sized firms
selling services to consumers carry out more online cross-border purchases.
Taking the manufacturing sector as a reference, firms in the wholesale and retail trade and in the
information and communication industries are less likely to sell online across the border. This result
applies to all size classes except micro firms in the case of wholesale and retail and only to medium-
sized firms in the information and communication sector. However, firms in the accommodation and
food industry are more likely to do cross-border e-commerce, a result driven particularly by micro firms.
In terms of purchases, both accommodation and food and information and communication sectors are
less likely to purchase across the border (with respect to the baseline of the manufacturing sector),
while the opposite is true for wholesale and retail activities.
Looking at the effects of barriers on export intensity, the results indicate that five barriers are
statistically significant: i) delivery costs; ii) guarantees and returns; iii) foreign taxation; iv) suppliers'
restrictions to selling abroad; and v) product and/or services specificity. In this case too, these barriers
have more impact on smaller firms than on larger ones (See Table 13). Of these five relevant barriers,
only guarantees and returns are correlated with the export intensity of large firms. In the case of large
firms, there are two additional barriers: knowledge about the rules to be followed and interoperability
issues. Security of payments is a relevant barrier for medium-sized firms and language skills for small
firms.
By removing any of these barriers at the aggregate level, the volume of cross-border sales could
increase: from 3.2% in the case of lowering the costs of guarantees and returns, to 6% if the restrictions
imposed by suppliers were effectively eliminated.
36
Table 13: Relevant barriers to cross-border e-commerce sales by firm size: intensive margin Firm size Micro Small Medium Large Delivery costs are too high -4.2% -13.5% Guarantees and returns are too expensive -25.1% You don’t know the rules which have to be followed -45.1% Payments from other countries are not secured enough -15.1% Dealing with foreign taxation is too complicated or too costly -5.4% You lack the language skills to deal with foreign countries -7.1% Your suppliers restrict or forbid you to sell abroad -10.7% -21.8% For reasons of interoperability, you cannot provide your products/services
-17.2%
Your products and/or services are specific to your local market -11.4% Source: own estimations with data from Flash Eurobarometer 413.
The results from the analysis of cross-border online purchases show that in aggregate terms, language
skills and the cost of disputes are correlated with a low intensity of online cross-border purchases.
Language skills affect all size classes negatively except large firms. The costs of complaints and
disputes resolution have an effect on the intensity of cross-border electronic purchases of both micro
and large firms. As can be seen in Table 14 –a summary of the results by firm size- large firms also
face the additional barrier of high delivery costs and medium-sized firms' concerns about protection of
their data seem to be also negatively affecting their share of online purchases across the border.
Table 14: Relevant barriers to cross-border e-commerce purchases by firm size: intensive margin Firm size Micro Small Medium Large Delivery costs are too high -14.8% Payments to other countries are not secure enough -3.6% You lack the language skills for dealing with foreign countries -5.3% -7.1% -8.7% Resolving complaints and disputes cross-border is too expensive -9.9% -18.1% You are concerned your data is not well protected when purchasing abroad
-6.8%
Source: own estimations with data from Flash Eurobarometer 413.
8 Conclusions
The main objective of this study was to identify the barriers to cross-border e-commerce in the EU that
really matter from a firm’s perspective. We find that the most relevant barriers to cross-border sales,
including the firm’s decision to export and the resulting volume of online sales, are delivery costs and
suppliers' restrictions. The most important barriers to a firm’s cross-border online purchases are the lack
of language skills and the costs associated with resolving complaints and disputes across the border.
Other barriers appear to be less important or only significant for specific firm sizes, such as dealing with
foreign tax regulation, or the security of payments, among others. A fully-integrated Digital Single
Market would be facilitated by removing these barriers.
37
We have also examined several dimensions of the process of internationalisation of European online
firms. We have detected some stylised facts that correspond with evidence from the literature on offline
firms. We find that aggregate online exports are driven by a few top exporters, that the share of exports
in turnover remains limited and that the number of online exporters is one of the main determinants of
aggregate online exports. We have also provided original evidence related to online export productivity
premiums and the relationship between online and offline specialisation patterns. Many of these
findings are in line with those from offline cross-border trade research (Mayer and Ottaviano, 2007)
where smaller firms find it harder to meet the fixed costs associated with cross-border trade. According
to the present survey data, channelling sales through an established online platform does not really
seem to make it any easier for small firms to overcome these costs: they still need to deal with many
issues.
The results presented in this report help to identify relevant Digital Single Market policy implications:
What matters most for a country's online trade performance is the number of firms that engage in
exporting. Hence, policies should focus on expanding the online export base, particularly by making
it easier for small and medium-sized firms to export online.
Even though they seem to be few and small, the costs associated with online internationalisation
matter because they reduce the number of exporters significantly. Hence, measures that efficiently
remove these barriers would help to foster cross-border e-commerce.
Some sectors have greater unexploited export potential than others. They are characterised by a
larger presence of small, low-productivity firms. They are more likely to react to online import
competition through the exit of the worst-performing firms and therefore have greater potential for
unexploited productivity gains.
38
References
Alaveras, G. and B. Martens (2015) “Online trade in services in the EU Digital Single Market”, JRC/IPTS Digital Economy Working Papers, forthcoming.
Anderson, J. E. and E. van Wincoop (2002), “Borders, Trade and Welfare”, Brookings Trade Forum 2001, Susan Collins and Dani Rodrik, eds., Washington: The Brookings Institution, 207-244.
Anderson, J. E. and E. van Wincoop (2004), "Trade costs". Journal of Economic Literature 42(3), pp. 691-751.
Berthou, A., E. Dhyne, M. Bugamelli, A.M. Cazacu, C.V. Demian, P. Harasztosi, T. Lalinsky, J. Meriküll, F. Oropallo and A.C. Soares (2015) Assessing European firms' exports and productivity distributions: the CompNet trade module. European Central Bank Working Paper 1788.
Blum, B. and A. Goldfarb (2006) Does the Internet Defy the Law of Gravity? Journal of International Economics 70(2), 384–405.
Cairncross, F. (1997) The death of distance; how the communications revolution will change our lives.
Orion books: London.
Cantwell, J. (1989) Technological Innovation and Multinational Corporations. Blackwell: Oxford.
Cappellari. L. and S. Jenkins (2006) Calculation of multivariate normal probabilities by simulation, with
applications to maximum simulated likelihood estimation. The Stata Journal 6(2), 156-189.
Coppel, J. (2000), “E-Commerce: Impacts and Policy Challenges”, OECD Economics Department Working Papers, No. 252, OECD Publishing.
Dalum, B. K. Laursen and G. Villumsen (1998), Strctural change in OECD export specialisation patterns:
de-specialisation and "stickiness", International Review of Applied Economics 12, 447-467.
Digital Agenda Scoreboard (2015) Digital Agenda Targets. Progress report. European Commission.
Duch-Brown, N. and B. Martens (2015), "The European Digital single Market", JRC/IPTS Digital Economy Working Papers, forthcoming.
Flash Eurobarometer 413 (2015), Companies engaged in online activities. European Commission. Gomez, E. B. Martens and G. Turlea (2014) ”The drivers and impediments to cross-border e-commerce in
the EU”; JRC/IPTS Digital Economy Working Paper. Lendle, A., M. Olarreaga, S. Schropp and P. L. Vezina (2012), There Goes Gravity: How eBay Reduces
Trade Costs, The World Bank Policy Research Working Paper 6253. Mayer, T. and G. I. P. Ottaviano (2007), "The Happy Few: The internationalisation of European firms. New
facts based on firm-level evidence". Bruegel Blueprint Series. Melitz, M.J. (2003) The Impact of Trade on Intra-Industry Reallocations and Aggregate Industry
Productivity. Econometrica 71(6), 1695-1725.
Papke, L. E. and J. M. Wooldridge (1996), "Econometric methods for fractional response variables with applications to 401(K) participation rates". Journal of Applied Econometrics 11, pp. 619-632.
Temouri, Y., Vogel, A. and Wagner, J. (2010) Self-selection into export markets by business services
firms. Evidence from France, Germany and the United Kingdom. IZA Discussion Paper, 5147.
Verardi, V. and Wagner, J. (2011) Robust Estimation of Linear Fixed Effects Panel Data Models with an
Application to the Exporter Productivity Premium. Jahrbücher für Nationalökonomie und Statistik
231(4), 546-557.
Verardi, V.and Croux, C. (2009) Robust Regression in Stata. The Stata Journal, 9(3), 439-453.
Vogel, A. and Wagner, J. (2011) Robust estimates of exporter productivity premia in German business
services enterprises. Economic and Business Review, 13(1-2), 7–26.
Wagner, J. (2012) New Methods for the Analysis of Links between International Firm Activities and Firm
Performance: A Practitioner’s Guide. University of Lüneburg Working Paper Series in Economics
No. 227.
39
Annex:
Table A1: Triprobit estimations of e-commerce target markets
E-commerce Domestic Cross-border Rest world Size (baseline: micro firms) Small 0.380*** 0.0354 -0.00717 (0.0766) (0.0534) (0.0581) Medium 0.382*** -0.0226 0.0325 (0.114) (0.0671) (0.0697) Large 0.572*** 0.109 0.0543 (0.188) (0.0948) (0.0990) Young 0.146 0.0938 -0.0917 (0.0934) (0.0610) (0.0684) Own website 0.259*** 0.236*** 0.196*** (0.0851) (0.0566) (0.0613) Small platform 0.200*** 0.182*** 0.169*** (0.0761) (0.0498) (0.0516) Large platform 0.579*** 0.272*** 0.185*** (0.0935) (0.0506) (0.0537) EDI type 0.335*** 0.0656 0.0597 (0.102) (0.0530) (0.0554) Goods to consumers 0.415*** -0.227*** -0.168*** (0.0708) (0.0525) (0.0591) Goods to firms -0.0659 0.0380 -0.00704 (0.0644) (0.0543) (0.0578) Digital services to consumers 0.268** 0.108 0.167* (0.123) (0.0838) (0.0908) Digital services to firms -0.159 0.113 0.0199 (0.110) (0.0836) (0.0910) Services to consumers -0.196*** 0.0939 0.00883 (0.0687) (0.0625) (0.0709) Services to firms 0.153** -0.0671 -0.00199 (0.0647) (0.0602) (0.0690) Structure (baseline: independent) National group 0.456** -0.0780 -0.107 (0.202) (0.0798) (0.0856) International group -0.113 0.0298 0.167** (0.116) (0.0754) (0.0781) Sector (baseline: manufacturing) Wholesale & retail 0.313*** -0.289*** -0.302*** (0.0649) (0.0568) (0.0627) Accommodation and food 0.558*** 0.570*** 0.651*** (0.140) (0.0797) (0.0811) Information and communication 0.303*** -0.0375 0.0804 (0.0954) (0.0805) (0.0829) Constant 0.661*** -0.233*** -0.798*** (0.110) (0.0870) (0.0956) Observations 3,489 3,489 3,489
-0.291*** (0.0538) -0.351*** (0.0475) 0.800*** (0.0366)
Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
40
Table A2: E-commerce productivity premia
OLS Quantile
MM Robust (0.25) (0.5) (0.75) (0.9) E-commerce 0.0332*** 0.00697*** 0.0127*** 0.0230*** 0.0388* 0.00852*** (0.0113) (0.00191) (0.00334) (0.00736) (0.0202) (0.00212) Size -1.07e-05 2.49e-05*** 5.33e-05*** 0.000105*** 2.32e-05 0.000113*** (3.37e-05) (5.30e-06) (1.02e-05) (2.34e-05) (4.86e-05) (2.24e-05) Size2 -5.76e-10 -3.48e-09*** -6.91e-09*** -1.25e-08*** -6.54e-09 -3.53e-08*** (4.63e-09) (6.14e-10) (1.32e-09) (2.82e-09) (5.42e-09) (6.79e-09) Constant 0.187*** 0.0355*** 0.0837*** 0.163*** 0.344*** 0.0698*** (0.0119) (0.00205) (0.00356) (0.00737) (0.0190) (0.00215) Observations 6,922 6,922 6,922 6,922 6,922 6,922 Note: all regressions include sector fixed effects. Standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1
Table A3: Online exporters’ productivity premia
OLS Quantile
MM Robust VARIABLES (0.25) (0.5) (0.75) (0.9) Online exporter 0.0352** 0.00885*** 0.0152*** 0.0377*** 0.0727*** 0.0110*** (0.0141) (0.00232) (0.00410) (0.00934) (0.0262) (0.00284) Size -5.30e-06 2.64e-05*** 5.65e-05*** 0.000109*** 2.38e-05 4.74e-05*** (3.36e-05) (5.26e-06) (1.03e-05) (2.33e-05) (4.84e-05) (1.66e-05) Size2 -1.09e-09 -3.63e-09*** -7.25e-09*** -1.29e-08*** -6.61e-09 -5.27e-09*** (4.63e-09) (6.07e-10) (1.33e-09) (2.81e-09) (5.36e-09) (1.63e-09) Constant 0.194*** 0.0367*** 0.0859*** 0.165*** 0.346*** 0.0744*** (0.0114) (0.00196) (0.00341) (0.00702) (0.0181) (0.00200) Observations 6,933 6,933 6,933 6,933 6,933 6,933
Note: all regressions include sector fixed effects. Standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1
Table A4: Online exporters’ productivity premia, by sector
Manufacturing Wholesale and retail
Accommodation And food
Information and communication
Online exporter 0.0101*** 0.0140*** 0.00321 0.0110** (0.00380) (0.00509) (0.00199) (0.00466) Size 0.000200*** 0.000162*** 5.57e-05*** 4.16e-05 (2.72e-05) (4.26e-05) (1.63e-05) (2.78e-05) Size2 -1.94e-07*** -7.11e-08** -4.42e-08*** -6.70e-09 (3.39e-08) (2.91e-08) (1.01e-08) (4.44e-09) Constant 0.159*** 0.235*** 0.104*** 0.160*** (0.0122) (0.0259) (0.00636) (0.0159) Observations 1,877 3,099 810 1,147 Note: all estimations performed with the MM estimator. Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
41
Table A5: Online vs. offline specialisation, by country
β |β|/R
Czech Republic 0.969 1.772
Portugal 0.841 1.409
Ireland 0.832 1.149
Netherlands 0.762 1.244
UK 0.697 1.240
Spain 0.612 1.068
Croatia 0.592 1.081
Latvia 0.562 1.362
Estonia 0.547 1.121
Belgium 0.515 1.019
France 0.507 1.506
Bulgaria 0.505 1.216
Slovenia 0.502 0.900
Greece 0.493 1.298
Austria 0.470 1.121
Italy 0.300 1.219
Denmark 0.291 0.586
Slovakia 0.250 0.657
Romania 0.187 0.753
Lithuania 0.115 1.183
Finland 0.090 0.639
Hungary -0.022 1.567
Poland -0.127 0.897
Germany -0.269 2.174
Sweden -0.399 1.630
Luxemburg -0.718 1.570 Note: All coefficients statistically different from zero and from one at the 90% except bold coefficients, which are not statistically different from zero and underlined coefficients, which are not statistically different from one. Source: own estimations with data from Eurobarometer 413.
42
Table A6. Barriers to cross-border e-commerce in the EU: decision to sell cross-border (1)Total (2)Micro (3)Small (4)Medium (5)Large Log(Size) 0.00458 Young -0.0354 -0.0539 0.0343 0.142 Channel: Own website 0.0645** 0.0773* 0.235*** 0.0238 -0.341** Small platform 0.0379 0.0632 0.140*** -0.0966 -0.0966 Large platform 0.0872*** 0.148*** 0.110* 0.0184 -0.0130 EDI type 0.0399 0.00997 0.0456 0.00384 0.0728 Market: Goods to consumers -0.0383 -0.0403 0.0113 -0.0635 -0.0153 Goods to firms -0.00599 0.0194 -0.120* -0.0260 0.0133 Digital services to consumers -0.0413 -0.102* -0.0374 0.124 -0.542 Digital services to firms 0.0391 0.0461 0.0701 -0.0275 0.754 Services to consumers 0.0178 0.0665 0.0128 0.0463 -0.200* Services to firms -0.0153 -0.0135 0.0235 -0.0356 -0.157* Sector (wrt manufacturing): Wholesale and retail trade -0.0121 0.0379 -0.00751 -0.00349 -0.235*** Accommodation and food 0.270*** 0.246*** 0.429*** 0.278** Information and communication -0.0387 -0.0290 0.00787 -0.0274 0.0863 Barriers: Delivery costs are too high -0.0747*** -0.0668 -0.0204 -0.153** -0.178* Guarantees and returns are too expensive -0.0157 -0.0564 0.113* -0.0844 0.0298 You don’t know the rules which have to be followed
-0.0260 -0.0303 0.0886 -0.155 0.0371
Payments from other countries are not secured enough
-0.0450 -0.0518 -0.00790 -0.161
Copyright prevents you from selling abroad or is too expensive to sell abroad
-0.0746 -0.146** 0.120 0.0318 -0.00921
Dealing with foreign taxation is too complicated or too costly
-0.0352 0.00892 -0.211*** -0.0473 -0.125
Your product labelling has to be adapted -0.0271 -0.0648 -0.133* 0.0361 0.0906 You lack the language skills to deal with foreign countries
-0.000585 -0.00504 -0.0340 0.0582 0.301
Your suppliers restrict or forbid you to sell abroad
-0.106** -0.0239 -0.402*** -0.139 0.171
Your suppliers do not allow you to use third platform to sell your products and/
-0.0229 0.00307 -0.109 0.144
Your suppliers request you to sell abroad at a different price
0.0804 0.00217 0.226* 0.396*** -0.0757
You are concerned your data is not well protected when selling abroad
0.120** 0.143** 0.382*** -0.0531 -0.153
For reasons of interoperability, you cannot provide your products and/or service
-0.0398 -0.0796 0.0129 -0.0867 -0.0925
Your products and/or services are specific to your local market
-0.0202 -0.0186 -0.0139 0.371* -0.167
Your company’s Internet connection is not fast enough
0.105** 0.198*** -0.0773 0.0541 -0.0251
Clients abroad do not have a fast enough Internet connection
-0.0100 -0.0163 -0.101 -0.120
Resolving complaints and disputes cross-border is too expensive
-0.0904*** -0.0980** -0.105* 0.00198 -0.0622
Observations 1,376 701 344 223 91
43
Table A7. Barriers to cross-border e-commerce in the EU: decision to buy cross-border (1)Total (2) Micro (3) Small (4)Medium (5) Large Log(Size) 0.00751 Young 0.00762 0.00833 0.0357 -0.0208 Channel: Provider website 0.0489*** 0.0497** 0.0369 0.0681 -0.0696 Small platform -0.00192 0.0127 -0.0590** 0.0301 0.126* Large platform 0.0291* 0.00647 0.108*** 0.0155 -0.0441 EDI type 0.0543*** 0.0396 0.0663* 0.0694 -0.0875 Market: Goods to consumers -0.0401** -0.0305 -0.0390 -0.0672 -0.0995 Goods to firms 0.0460** 0.0530** 0.0570 -0.0424 0.136 Digital services to consumers 0.0285 0.0232 0.129* -0.123 -0.181 Digital services to firms 0.0220 0.0216 -0.0581 0.141* 0.0352 Services to consumers 0.0209 0.0173 0.00892 0.0750 0.0264 Services to firms 0.00527 -0.00721 0.0723* -0.0275 -0.0393 Sector (wrt manufacturing): Wholesale and retail trade -0.00933 -0.0142 0.0348 -0.0206 0.0501 Accommodation and food -0.0185 -0.0108 0.0278 -0.0751 -0.00169 Information and communication -0.0240 -0.00898 -0.0281 -0.0611 -0.0111 Barriers: Delivery costs are too high 0.00151 0.00924 -0.0308 -0.00233 0.0503 Payments to other countries are not secure enough
-0.0508** -0.0773*** -0.0351 -0.0364 0.178**
The product labelling has to be adapted 0.129*** 0.144*** 0.105** 0.104* 0.106 You lack the language skills for dealing with foreign countries
-0.0670*** -0.0675*** -0.0562 -0.0495 -0.112
Copyright prevents foreign suppliers from delivering to your country or makes it
-0.0150 -0.0334 -0.00248 0.0665 0.0777
Foreign suppliers refuse to deliver to your country
-0.00981 0.0172 -0.0505 -0.0525 -0.142
Resolving complaints and disputes cross-border is too expensive
-0.118*** -0.140*** -0.0964*** -0.0565 -0.215***
You are concerned your data is not well protected when purchasing abroad
0.0305 0.0389 0.0651* -0.0537 0.0353
For reasons of interoperability, you cannot use foreign products and/or services
0.00668 -0.0107 0.0364 0.0305 0.0809
Observations 3,325 1,941 841 400 142
44
Table A8. Barriers to cross-border e-commerce in the EU: Volume of sales (1)Total (2)Micro (3)Small (4)Medium (5)Large Log(Size) 0.00851** Young 0.0116 0.0179 0.000929 -0.0312 Channel: Own website -0.0150 -0.0171 0.0651 -0.0351 -0.0435 Small platform -0.0149 0.00466 -0.00100 -0.0478 0.0122 Large platform 0.00219 -0.00358 -0.00898 0.0969*** 0.00297 EDI type 0.00770 -0.0188 0.00968 -0.00780 0.174*** Market: Goods to consumers -0.0603*** -0.0762*** -0.0334* -0.0248 0.0269 Goods to firms -0.00598 0.0174 -0.0259 -0.0976** -0.0132 Digital services to consumers -0.0311 -0.0343 -0.0292 -0.0839 0.0533 Digital services to firms 0.0260 0.0114 0.0673 -0.0184 0.0514 Services to consumers 0.0108 0.0436 -0.0339 -0.0441 -0.165 Services to firms -0.0169 -0.0254 0.0276 0.0240 -0.0560 Sector (wrt manufacturing): Wholesale and retail trade -0.0660*** -0.00784 -0.118*** -0.103** -0.140** Accommodation and food 0.0597* 0.0774** 0.0650 -0.0324 Information and communication -0.0727** -0.0464 -0.0568 -0.171** 0.0230 Barriers: Delivery costs are too high -0.0486** -0.0420** -0.00595 -0.135** -0.0390 Guarantees and returns are too expensive -0.0321* -0.0117 0.0104 -0.0995 -0.251*** You don’t know the rules which have to be followed
0.0116 0.000625 0.0631 -0.000154 -0.451***
Payments from other countries are not secured enough
-0.0256 -0.0253 0.0376 -0.151*
Copyright prevents you from selling abroad or is too expensive to sell abroad
0.00710 -0.0553 0.0735 0.181 0.312**
Dealing with foreign taxation is too complicated or too costly
-0.0497** -0.0536** -0.0784 -0.0376 0.138**
Your product labelling has to be adapted -0.0172 -0.0449 0.000166 -0.185 0.387*** You lack the language skills to deal with foreign countries
-0.0161 -0.0114 -0.0709* 0.0288 0.193***
Your suppliers restrict or forbid you to sell abroad
-0.0595** -0.0234 -0.107* -0.218*** 0.0815
Your suppliers do not allow you to use third platform to sell your products and/
0.0441 0.0261 0.00625 0.161**
Your suppliers request you to sell abroad at a different price
0.0497* 0.0216 0.0554 0.101* 0.253**
You are concerned your data is not well protected when selling abroad
0.0494** 0.0897*** 0.0445 0.0491 -0.00874
For reasons of interoperability, you cannot provide your products and/or service
0.0178 0.00930 0.00956 0.115 -0.172*
Your products and/or services are specific to your local market
-0.0517* -0.0255 -0.114*** -0.122 -0.204
Your company’s Internet connection is not fast enough
0.0488** 0.0676** 0.0361 0.0439 -0.175
Clients abroad do not have a fast enough Internet connection
-0.0194 -0.0339 -0.0401 0.0466
Resolving complaints and disputes cross-border is too expensive
-0.000105 0.00820 0.0107 -0.0288 -0.255
Observations 1,376 701 344 223 91
45
Table A9. Barriers to cross-border e-commerce in the EU: Volume of purchases (1)Total (2)Micro (3)Small (4)Medium (5)Large Log(Size) -0.00134 Young 0.0262 0.0240 0.0377 0.000779 Channel: Provider website 0.00235 0.0119 0.000419 0.00234 -0.0639 Small platform -0.0198 -0.0104 -0.0205 -0.0401 0.00173 Large platform -0.0131 -0.0138 0.00147 -0.0135 -0.0375 EDI type 0.00999 -0.0163 0.0313 0.0224 -0.0320 Market: Goods to consumers -0.0371** -0.0307 -0.0677*** -0.0329 -0.0375 Goods to firms 0.0218 0.0149 0.0740*** -0.00739 -0.0419 Digital services to consumers -0.0260 -0.0165 -0.0183 -0.0880 -0.0873 Digital services to firms 0.0213 -0.00433 0.0442 0.0487 0.0805 Services to consumers 0.0325* 0.0265 0.0236 0.146*** 0.0611 Services to firms -0.0282** -0.0458** 0.00876 -0.0366 0.0252 Sector (wrt manufacturing): Wholesale and retail trade 0.0330* 0.0376* 0.0841** -0.0405 0.00508 Accommodation and food -0.0516* -0.0206 -0.0316 -0.212*** -0.154 Information and communication -0.0521*** -0.0381 -0.0137 -0.138*** -0.172* Barriers: Delivery costs are too high -0.0133 -0.0143 0.00507 -0.00984 -0.148** Payments to other countries are not secure enough
-0.0175 -0.0365* -0.0201 0.0427 0.0663
The product labelling has to be adapted 0.0201 0.0181 0.0331 -0.00701 0.0789 You lack the language skills for dealing with foreign countries
-0.0620*** -0.0534*** -0.0713*** -0.0874** -0.0345
Copyright prevents foreign suppliers from delivering to your country or makes it
-0.0162 -0.0222 -0.0270 0.0486 -0.0240
Foreign suppliers refuse to deliver to your country
0.0282 0.0280 0.0162 0.0156 0.0955
Resolving complaints and disputes cross-border is too expensive
-0.0743*** -0.0987*** -0.00560 -0.0577 -0.181***
You are concerned your data is not well protected when purchasing abroad
0.00691 0.0103 0.0359 -0.0680** 0.0666
For reasons of interoperability, you cannot use foreign products and/or services
-0.00542 -0.00968 -0.0130 0.0113 0.0313
Observations 3,325 1,941 841 400 142
46
Table A10. Sectors at the 2 digit NACE classification
Code Sector
10 Manufacture of food products
11 Manufacture of beverages
12 Manufacture of tobacco products
13 Manufacture of textiles
14 Manufacture of wearing apparel
15 Manufacture of leather and related products
16 Manufacture of wood and of products of wood and cork
17 Manufacture of paper and paper products
18 Printing and reproduction of recorded media
19 Manufacture of coke and refined petroleum products
20 Manufacture of chemicals and chemical products
21 Manufacture of basic pharmaceutical products
22 Manufacture of rubber and plastic products
23 Manufacture of other non-metallic mineral products
24 Manufacture of basic metals
25 Manufacture of fabricated metal products
26 Manufacture of computer, electronic and optical products
27 Manufacture of electrical equipment
28 Manufacture of machinery and equipment n.e.c.
29 Manufacture of motor vehicles, trailers and semi-trailers
30 Manufacture of other transport equipment
31 Manufacture of furniture
32 Other manufacturing
33 Repair and installation of machinery and equipment
45 Wholesale and retail trade and repair of motor vehicles and motorcycles
46 Wholesale trade, except of motor vehicles and motorcycles
47 Retail trade, except of motor vehicles and motorcycles
55 Accommodation
56 Food and beverage service activities
58 Publishing activities
59 Motion picture, video and television production, sound recording and music publishing activities
60 Programming and broadcasting activities
61 Telecommunications
62 Computer programming, consultancy and related activities
63 Information service activities
Figure A1: Structure of the survey
A1
Sell
online
A2
Buy
online
Firms
(8705)
- country
- sector
- size (empl, turn)
- age
- type (org)
- category (mkt)
- sales trend
Yes
(3945)
No
(4743)
Yes
(7156)
No
(1503)
Missing (17)
Missing (46)
Own website (3107)
Small platform (1093)
Large platform (1038)
EDI (906)
Q1
Channels
(3827)
D7 Share online
sales (3504)
D7
Share of offline sales (7729)
Q2a: Online domestic (3445)
Q2c: Online RoW (918)
Q2b: Online other EU (1739)
Q4a: Offline domestic (7601)
Q4b: Offline other EU (3586)
Q4c: Offline RoW (1763)
Barriers to EU cross-border sales:
Q6a1-Q6a17 - Delivery (507)
- Guarantees (349)
- Rules (259)
- Payments (239)
- Copyright (136)
- Taxation (243)
- Labelling (153)
- Language (222)
- Suppliers forbid (125)
- Suppliers restrict (115)
- Suppliers dif price (135)
- Data (220)
- Interoperability (171)
- Product specificity (147)
-Own connection (279)
- Clients connection (148)
- Disputes (351)
Q3
In which countries do you sell?
Q5
Motives (1665)
Obstacles Q6c1-Q6c10
- Digital skills (945)
- Rules (1086)
- Delivery (1343)
- Guarantees (1271)
- Suppliers forbid (689)
- Higher price (952)
- Suppliers restrict (628)
- Risk prices down (979)
- Damaged image (763)
- Connection (1080)
0% (1787)
≠ 0% (1816)
Used to sell (56)
Tried but gave up (52)
Considering (438)
Trying (152) Barriers to EU
cross-border sales:
Q6b1-Q6b17 - Delivery (200)
- Guarantees (148)
- Rules (114)
- Payments (78)
- Copyright (63)
- Taxation (100)
- Labelling (59)
- Language (81)
- Suppliers forbid (61)
- Suppliers restrict (49)
- Suppliers dif price (33)
- Data (59)
- Interoperability (64)
- Product specificity (47)
-Own connection (63)
- Clients connection (45)
- Disputes (158)
Never (967)
Own website (4875)
Small platform (2793)
Large platform (2644)
EDI (1404)
Q7
Channels
(6807)
D8 Share online
purchases (6361)
Q8a: Online domestic (5936)
Q8c: Online RoW (1326)
Q8b: Online other EU (3129)
Barriers to EU
cross-border purchases
Q10a1-Q10a9 - Delivery (979)
- Payments (573)
- Labelling (457)
- Language (496)
- Copyright (537)
- Suppliers refuse (534)
- Disputes (996)
- Data (743)
- Interoperability (425)
Q9
Motives
(2916) 0% (3221)
≠ 0% (3354)
Used to buy (207)
Tried but gave up (145)
Considering (717)
Trying (112)
Barriers to EU cross-border
purchases
Q10b1-Q10b9 - Delivery (305)
- Payments (170)
- Labelling (165)
- Language (172)
- Copyright (192)
- Suppliers refuse (202)
- Disputes (339)
- Data (211)
- Interoperability (130)
Never (1735)
N of firms using: 1 channel: 2265 2 channels: 965 3 channels: 439 4 channels: 158
Provided D7 ≠ 100% (3339)
DSM Firms Survey Structure of the data Number of firms in parenthesis
N of firms using: 1 channel: 3881 2 channels: 1422 3 channels: 1025 4 channels: 479
Q11
Homogeneous rules
(2614)
61% answers favourably
48
Figure A2: Online export concentration
.2.4
.6.8
1.2
.4.6
.81
.2.4
.6.8
1.2
.4.6
.81
.2.4
.6.8
1
0 .5 1 0 . 5 1 0 .5 1 0 .5 1
0 . 5 1 0 .5 1
BELG IQUE DANM ARK D EU TSCHL AND EL LADA ESPANA SU OM I
FRANCE IRELAND ITALIA LU XEM BOURG NEDERLAND ÖSTERREICH
PO RTUG AL SVERIG E UK BALGARIJ A CESKA REPUBLIKA EESTI
M AG YARO RSZAG LATVI A LIETUVA POLSKA RO MANIA SL OVENSKA REPUBLIC
SL OVENIJA H RVATSKACu
mu
lative
sh
are
of
on
line
exp
ort
s
Cumulative share of firmsGraphs by B Country
49
Figure A3a: Structure of online sales by target markets and firm size
Figure A3b: Structure of online sales by target markets and sector
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
Micro Small
Medium Large
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
Manufacturing Wholesale and retai l
Accommodation and food Information and communication
50
Figure A4a: Structure of offline sales by target markets and firm size
Figure A4b: Structure of offline sales by target markets and sector
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
Micro Small
Medium Large
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
Manufacturing Wholesale and retai l
Accommodation and food Information and communication
51
Figure A5a: Structure of online purchases by target markets and firm size
Figure A5b: Structure of online purchases by target markets and sector
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
Micro Small
Medium Large
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
0
20
40
60
80
100 0
20
40
60
80
100
020406080100
Domestic Cross-border
Rest of the world
Manufacturing Wholesale and retai l
Accommodation and food Information and communication
52
Figure A6: Revealed comparative advantage: online vs. offline
56 10
46
25
55
4518 32
58
56 3218 10
13
25
23
10
58
25 59254528
46
47
28 10
46
10
46
55
45
4646
2547
56
46
10
46
47
4646
28
46
25
46
55
46
55
46
4747
102562
46
18
55
47
464646
28
46
47
33
23
4747
1018
46
18
4646
1445
46
59
55
59
23
46
27
46
476247
10
4646
45
55
62 58
4545 27
5523
4646
6247
45
46
6262
45
46
10
47
4545
46
4756 10
47
46
23
464646
45
5862
55
14
46
55
62 58
46
20
4762
23
10
46
4747
1056 5959
46
5625
46
5555
46
6247
4646
23
45 59
55
4525
23
47
10
58
46
4756
46
32
13
626247
4646
4756
58
5645
5862624756
625656
46
58
45
62
45
23
474756
46
62
28
47
46
62
18
62
46
62
46
2528 56 595662
56
4646
56
46
62
4646
475625 56
23 55
25 32
47475614
47565645
4756
47
14
55
45
55
47
45 5647
56
46
5647
56564747
23
565647
32
46
62
46
1845
62
45
475656
4756
4725 27 5656
46
13
4646
475656
475656
6256 3345
55
6256
4756
624756
58476256
46
18
23
45452562 58
45
6247
46
565662
45
4756 3256
6225 59
5558
46
56
55
62
46
47
55
4756
46
562562
45
5862
55
6262
46
45
475625
46
6262
46
55
46
566247
45
475645 5656 5945
6262626262
46
56 4747
46
22
47 62
4546
47
46
10
63
45
25
4646464646
47
46464646
25
4646
32
46
10
4646
56
4646
13
46
25
28
4716
46464646
58
62
46
10
4646
1058
2527
45
62
33
28
32
4646464628
3162
2846
2846
10
25
28
23 4723
46
4747
4646 45
10
4646
6247
46
62
45
626263
464646
17
28
27
45
4747
45
4747
46
6356
58
47
22
47
28
4747
46 45
62
10
28
6347 6262
45
47
45
632527
46
62
28
61
22
5547 63 624755
58
56 4747
58
6247
4646
47 62
33
6259 62616247
46
31
4646
58
62
46 4546
31
46
62
46 454646
47
4646 4546 4546
26
56 47
45
6259 62
4646
626247
28
47
4646
33
626262
33
6262
46
474747
45
56 47 6262
4646 4546
56
2846
6259 6247474747 62475956
45
47 63 6263
46
32 4759
46
27 4747 6247
46 454646
63 62
46
47
46
4747 6262
46
3256 625556 62
46
4723 59
26
4725 6255
46
10
47296262
46
5656
46
62474747 6256 6262474747 6262
28
47
46
25
46 45
5956 4755
28296262
46
33
6262
10
46
6262
46
58
47
46
47
284546
47474725
464646
25 47
46
4747
4646
21
4747
4646
47 6225 47
46
5845454646
47
464628
4628
4646
62
464646 454646 45
47
4646
33
32
46
16 47
26
284546
2846
58
62
45
58
59
4762
55
46
5932
59
56
62
58
63
6262 47
27
6262 47
58
47
45
4747
27
17
62
55
56
3232
47
464645
58
46
58
46
6262
46
26
61
62
56
62 47
46
1020
46
6262
28
33
62
46
4747
22
58
59
25
62
28
4762
4646
20 10
56
32
46
58
47
18
47
32
62
58
33
47
1846
33
62
27
6262
55
62
45
4747
553055
5858
62
4545 25
28
46
32
45
58
1846 25
10
46
58
55
25
28
4645
62
46
63
4762 47
46
28
45
282822
62
25
24
47
464646
4762 47
56
47
46
62
28
1825
28
46 25
47
46
5858
47
59
46
55
33
46
62
61
4762 47
4646
62
45
20
18
59
56
60
25
58
46
59
47
6361
454646
6262
2525
61
46
62
4646
4762 47
45
62
58
47
25
33
22
626262 47
58
62
59
4747
27
24
47
464545
47
61 32
47
28
46454646
47
28
4618
63
56
46
62 4762
46
2832
27
46
62 476262
27
285555
47
20
62
28
62 4747
56
47
25
626226
4762
46
56
62
28
464518
26
10
27
25
22
46
62 4747
46
47
28
47
3228
46
47
28
62
58
62
58
13
58
25
56
13
56
4646
10
45
62
4645
47
631422
58
46
47
454546
47
4646
62 47
22
45
10
46
55
46
62
18
6262
59
56
55
476226
476262
58
4646
55
4762 47
4646
47
22
62
63
33
62
17 24
4646
2647
46
58
62
28
4646
4747
28
56
4646
4726
476262
27
60
46464646
4747
25
62
5825
464645
32
62
46
47
62
46
25
55
11
18
28
5555
4646
1111
10
47
23
46
25
45
55
24
5517 55
45
14
464646
5555
4546
62
31
5555
46
55
1425
55
23
10
26
555555
59
55
4645
11
10
11
46
5555
18
46
14
55
4610
5555
11
5547
25
4646
56
46
14
18
20
4546
2527
4747
20
46
62
45
47
56
62
55
46
25
46
47
62
45
2862
25
46
47
4646
59
10
58
464646
62
21 59
46
55
60 56
4646
58
1046
47
464546
62
45
5631
46
17
25
62
5660
46
14
46
55
46
32
46
47
11
46
11
55
4646
47
46
31
46
63
23
1847
2558
46
62
46
2658
6363
46
56
4645
58
4646464646
21
46
62
59
62
58
104646
63
58
46
47
4610
464646
11
4646
23
46
55
46
32
47
58
46
11
14
10
25
62
1046
25
47
1046
4747
464546
10
47
62
47
46
47
45
6262
4546
10
47
4646
28
4646
17 55
2358
47
464646464646
62
45
58
47
59
62
1046
47
62
46
62
46
63
4645
55
4646464646
47
464646
554747
46
32
46
47
6158
55
45
47
46
1420
46
63
4646
62
25
47
62
21
104546
62
46
62
10
58
46
59
464646
55
62
4645
55
46
5826 25
4646
47
25
55
11
4646
626262
45
6262
5658
6262
55
2527
4646
62
18
1046
5555
62
58
46
6262
18
62
4546
28
46
62
104546
18
46
62
47
46
62
58
62
10
58
46
2628
45
25
222217
46
202745
55
46
28
46
28 56
2546
10
55
182828
6228
224526
4646
58
46
4710
46
56
55
46
31
5555553046
32 58
55
46462525
10
46
104524
46
1745 56
292925
22
46
47
46
26
46
18
2555
4646
55
46
55
4545
55
634762 4758 186262
46
59
25
11 6345 1845
46
56
2546
15 1862
46
1845
46
47
55
563147
55
27
46
18
4646
62
4646
4747
4646
4762 47
25
18
55
62
46
17 56
46
1445 56
46
62
25
62
61
46
1013
5529
10
46
6245
55
62
46
471047626259
46
2726 5610
61
56
46
58 4745 474747
46
5862 314562
46
62 5628
46
626245 4745 58 4747
46
476245
46
47 224545 4762 4745 47 56
4625
186310
4646
55
314710
5555
58 1045 56
25
24 47
46
1845 47
25
4745 4727 472833 1045
46
15 4762
55
47
46
55
46
45 1462 20
4646
5628 56
2546
26453245 47
46
6228 47
46
56
55
6262 56476245
46
5627 10
46
47
46
56
3046
62
55
46
29
46
45 20
55
10
464646
6232
46
6262 56
46
31 56
46
6262
55
63
55
58
4646
45 47474759 4747
46
6217
46
564758
46 30
6262624528
6262
61
62
46
58 56635862 5845 5656561847 5662 4762285862
46
4513 18 564762 10
46
4745
4646
58 4731
55
47
4646
29
4646
11
184746
22
10
25
2061 3229
17
18
61 27 4661
58
47
28
10 4647 184747
28
47
28
474747
2845
46
45
25
4746
45
1032
62
4647
25
22
56
45
46
58
465547
25
46
58
46
6262
28
23
58
2761
25
47
45
46
22
1847 1332 56
4545454528
634656
28
18
23
55
58
474633
62
47
22
47
62
23
20 46
62
59
58
31
46
25
45
2562
56465646
62
4756
55
62
4620 4647
5646
62
47
58
2245
62
46464647
62
61
25
47
6262
13
45
4756
13
45
47
45
59 56595547
4659
58
55
62
56
62
47
45
55
28
55
45
1047
5627
62
10
62
5546
47
25
21
6262
56
45
46
2562
46
23
13
62
47
62
59
62
5556
18474656
47
62
59
62
4746
25
58
4746
62
47
6262
5556
22
6262
18
454545
474747
2845
10 4633
28
294747
104747
6225
18
4558
46
4528
31
33 46474763
4659
45
626262
564646
2562
3347
59
6262
475559
47564646
62
4756
47474747
17
63
45
4646
58
554733
45
62
474727 46
55474646
4746
47
58
3247
6262
4747
6262
28
62
45
5646
62
45
5959
62
59 5647
62
634746
62
5627 4647
4545
4647
45
46
17
4746
4746
62
47565633
474733 462063
4663
46
62
56336347
564747
5632 5646
45
4756
4759
62
58
62
46
6262
5647
62
47
47
58
25
22
474646
555555
46
59
22
10 452847
46
14
55
58
55
46
55
58
4747
55
47474746
56
59
46
58
4647
55
59
55
4725
4745
46
4547
62
4747
55
4647474747
4545
55
45
55
2525
5656
25
5555
45
46
56
46
56
63
555562
47
6214
56
6311
56
47
22
46
55
4510
55
25
4547
5562
29
5555
4747
55
45
5562
63
55
59
56
4747
55
46
56
55
47
62
47
5562
46
4547
46
6255
45
464646
59
464647
4647
45
55
4547
29
62
45
55
4746 25
47
62
47
5555
47
58
47
62
58
474747
58
56
55
56
4745
55
47
58
1010
62
46 25
63
6262
47
58
6255
29
47
55
22
5555
28
46
4547
55
45
46
6255
47
5858
45
29
4547
56
4746
47
56
47
62
47
5656
1047
4547
55
63
464747
10 4547
5555
4725
56
2847
59
5656
62
47
55
58
284747
4547
56
62
4759
62
47
58
474747
55
26
46
62
59
25
55
56
62
4747
56
46
58
55
47
56
4547
5562
555555
63
55
58
46
45
58
56
62
56
55
1047
55
56
47
295656
47
62
56
254747
6255
56
47
58
25
45
56
6262
1047
56
4759
55
474746
62
4528 6347
25
5656
55
47
55
56
4547
55
45
5555
22 5629
46
55
26
55
45
56
28
56
62
10
56
4710
62
11
26
46
55
4746
4710
56
46
58
55
26
45
55
18
464625
18
46
626262
5647
5647
5647 2546
626262
4756
6132 45 4630 4547 45584747 46
62
46
62
61
60 62
47
62
6147 2156
62
5646
55
62
465856
18
47
60
47 4632 4647 46
62
4547
62
47
62
56
55
181818
58585656
15
4732 4647474747
2525
56 474727 47
55
46 4746 476228 131346
32
25
27 62625645
46
14
131646
25
4546 6246 62564646 47
452246 4756
45454616 104747
55
31
25
4745
25
46
32
22 4717 624756
2525
2746 28
55
4645
55
25
47464646
33
46 4728 224645
46
55
31 1346 27
3333
46
5555
2828
25
22
14
46
55
62154546
25
474746 28 47155646 4745
62
25
46 28 6247
55
47624762
14
46 28 22
14
46 62
55
464645
62
25
6245
10
25
4646
25
28
25
31
32
28
55
5646 28
25
31 6227 56
33
46
25
4527 62
45
55
464646 6210
25
27
25
28 22
25
476246 47
55
46 4762
25
55
56 47624646
55
101010
25
4715
5555
4745
4727 28 3147 1045
62
32
10472846 622846 6245
10
25
451746 22 473146
4546 28
25
22 4727 28
25
28
25
55
62
33
28
1414
2827 10154645
46 4745
4747
45
47
46
4545
47474747
45
46
474747 6247
4545
46
45
46
624747
45
6247
4646
16
46
47
464646
45
1647
45
46464646
45
62
46
47 62
46
47
4646
62
45
11
46
45
16
46
62
46
62
464646
6262624747 62
46
47
45
47
45
47
4646
464646
47
10
45
25
45
46
56
4547
10
4646
11636256 58
4633
27
46
28
33 4646
16
47
46
11
46
55
47
22
33
28
62
47
32
47
46
45
10
55
47
6262
46464646
56
46
56
47
22
25
46
45
2524
47
62
46
45
59
47
46
4747
46
32
25
46
47 45
10
46
47 45
25
30
45
10
47
5659
46
47
46
58
25
4646
47
28
46
4545
28
46
4545
18
55
45
46
27
33
4747
4646
62
45
46
45
62
46
47
56
2518
464646
4747
18
55
46
45
46
62
45
32
23
454547
28
46
47
46464646
47
46
47
55
47
22
4747
58
4646
18
46
47 45
23
4747
18
4646
1162 27
46
28
62
4545
5956
33
55
47
46
32
47
46
6262
47
5662
18
62
33
62
47474747 45
46
25
464646
62
4633
4545
3032
46
10
25
55
47
46
18
46
32
4747
46
4747
46
6262
46
10
47
56
47
18
46
62
47
46
625962
47
565956
55
4747
55
10
4547
10
46
56
33
47
62
4646
47
46
47
55
626256
46
47
59
46
47
5963
47
62 58
47
626232
55
46
47
46
45
2758
28
454747
18
2963
46
56
46
28
47
62
4545
25
62
46
29
46
47
18
46
63
474747
58
47 454747
46
47
10
55
46
4747 454747
2962
55
47
56
4747
59
46
3262
4747
5662
45
1056
4646
5555
47
56
46
4747
5662
47 4547 45
46
56
46
5662
4547
4646
47
46
47
13
46
2525
47
10
4646
62
47
13
47
46
63
46
55
47
46
22
47
55
25
46
4747
46
45
24
4646
47
25
4626
55
46
47
5555
4747
1025
4763
6262
47
6262
46
4747
25
47
62
58
4747
62
474747
5662
264747 45
56
63
55
47
62
4747 45
4616
55
10
2647
555555
6256
555555
2862
5555
4747
55
56
5555
2862
5555
63
24
47
5555555555
63
55
13
45
5555
56
55
47
46
63
55
46
62 10
46
4726
62
46
25
32
47
2825
26
55
46
47
32
22
47
58
47
25 2856
16
25
47
13
55
47
55
47
4646
621725
2645
10
47
282562
46
5555
47
32
47
46
10
63
29
46
25
4763
25 22
6363 4563
46
4763 474747
13
46
47
46
2225
46
6256
454763
22
47
32
4547
56
5555
63 474763
55
56
47
55
4763 45
46
10
464646
4747
464646
62
46
55
46
55
46
4547
46
63 47
46
62
4626
47
62
474747
62
55
46
62
46
63
62
47
55
32
47
46
63
46
45
58
63 47
46
6262
47
62
47 4563 4547
46
4763 47
566262
55
47
56
4547
464646
55
454747
46
55
45
25
464646
47
10
55
62
4626
47
46
55
62
46
4747
46
55
46
13
4646
24
4547
46
10
164747
5662 22
47
46
55
47
2862
47
1062
47
22
55
25
45
25
63 4747
626262
46
4747
55
62
4626
47
462222
4747
58
47
55
10
4645
46
47
10
46
5656
47
46
56
4546
22 1545
56
5555
46
10
15
55
45
56
55
4545
47
55
4645
62
4747
10
46
18
4645
56
47
10
47
4545
4747
55
45
47
5562
4747
5531 18
46
5562
47
464646464658 45
46
56
45
62
45
5656
4646
55 18
46
5555
46
5562
55
10
464646
10
46
4747
6255
47
55
56
4645
47
46
62
47
5555
464545
47
46
5555
4747
62
46
55 18
47
62
46
47
46
5555
46
55
46
47
45
55
56
47
4646
62
46
18
46
55
46
6262626255
224646
3162
47
18
47
45
47
45
626255
56
46
626262
47
45
55
46
56
62
47
464558 45
46
565656
4646
55
4747
4546
31 1855
46
47
55555562
46
62
56
47
6218
62
46
47
46
6255
45
47
56
62
58
55
45
56
47
6255
62
58
56
6231
45
47
56
47
626255
4646
62
46
4747
5562
55555562
45
47
62
4645
56
47
56
47
55
47
626262
47
6255
62
58
62626262
56
62
45
62
56
47
46
626255
62
56
18
47
46
47
56
4546
47
56
47
56
45
626262
56
6255
58
62
4546
62
4747
58
625555
47
31 55
47
55
47
6262
45
56
46
55
56
55
45
55
56
31
47
55
45
10
47
55555562
46
55
56
46464647
17
46
11
2846
20
58
23
28
17
2728
24
55
20
46
20
4628
46
25
17
46
2322
17
24
46
2529
22
46
16
28
4522
10
4545
46
10 27
45
5546
47 5846
29
47
17
24
45
47
29
46
10
4646
20
47 2846
10
2525
20
46
22
25
4646
2747
16
281010
25
55
28
17
25
10
46
22
4647
3222
45
47
24 25
4618
16
46
22
46
22
25
47
55
2429
32
29
28
24
4646464647
25
4646
22
4618
25
2846
47
1656
13
4646
56
28
45
46
45
23
45
46
4516
4746
10
46464655
47
22
104746
58
29
47 28
45
4746
33
10
25
56
4746
45
4646
13
59
27
4655
6262
4628
25
4545
2846
62
47
20
55
62
61
27
464647
4647
6225
22
47
5555
28
62
45
55
62
46
27
4658
2525
47
25
46
20
55
2846
56
1358
45
184646
474746
62
2847
6262
45
46
25
10
22
4647
62
4746
56
18
47
29
46
14
58
17
10
62
46
62
47
33
58
59
62
1846
20
55
45
6262
55
47
5656
62
16
62
10 27
46
22
282847
11 45
55
62
47
20
10474746
20
33
5546
25
58
45
2525
17
46
56
47 10
62
45 59
554646
62
474747
56
25
5810
45
58
56
474646
62
5656 59
464647
6229
45
5546
56
25
55
47
56
47
55
10
4646
45
62
47474646
16
46
13
5959
46
56
62
46
33
46
61
25
47
626222
16 454516
47 5846
47
56
1058
59
47
56 45 62
56
47474747
565645
62
10
565656
47
4556
55
4646
455656
46
5656
474747
56
28
47
56
4755
56
62
56
55
62
4545
62
56
5547
45
32
56
5547
56
47
5645
46
55
56565645
4747
56
47
56
47475547
56
5547
5656
27
5645
56
47
4646
55
56
475547
32
4545
4755
56
46
4545
47
10
45
47
32
47
27
47
45
47
56
47
5656
32
20 27
10
45
47
62
4747
45
28
56
27
5656
55
27
47474747
32
565645
564545
56
1328
32
455656
47
45
62
32
28 45
62
464625 474645 47 5610
55
3345 4646
62
46
55
33 4747 28
62
55
221058
55
62
47 5647 284745 4711 31464645
62
464745
62
5633 47 56464658
62
56
62
4646
55
10 474625
62
2845 47 222545 47 56
55
4658 464746
55
61 4747 5646 56
59
47 56
62
45 46
6262
4646 56
62
25
55
11 46474745
62
58 46
62
33 4747
62
3145 47
55 59
474746 315663 4647
55
63 4747
59
46 56
55
56474525 3110 464710 47
62
314726 4647
62
474647
62
56
59
47
62
55
47
6262
58 4746 5647
62
464747
59
564647 28
55
5826 46 564747
6262
6161 47
62
3147
13
25
32
47
46
14
4646 3346
45
46
4747475656
14
47
33
56
16
47
25
47
14
10
46
161620 2345
33
1028 23
5656
11
5647 27
23
31
46
47
232317
47
46
16
55
2525
46
56
25
13
62
46 3346
45
464646
27
45
464646
56
2845 23
4756
47475656
3228
46
27
464646
25
47474747
45
47
17 3245 61
31
45
27
62
3346
20 26
4646
45 61
46464646
13
46
27
4646
27
62
464646
10
31
464646
45
46
25
22
46
4747
32
47
46
16
47
62
25
23
4747
46
47
25
46
56
294520
46
45
31
56
4646
47
62
18
46
47
45
25
23
47
62
45
46
29
46
28
47
45
6262
27
4646
63
46
25
46
56
4646
22
63
4646
55
4646
10
47
25
47
464646
56
4646
17
46
252525
55
5858
13
47
25
27
23
4646 33
23
62
16
5656
464646
6324
46 3346
47
18
46
11
47
46
61
25
56
46
27
6214
45
56
16
46
45
46
45
46
4747
15
46
2445 26
56
23
31
2747
334646
28 2623
33
11
25
11
47
464646
10
46
23
2525
4646
47
46
10
47
28
47
18
47
23286120
46
4524
25
33
4756
1828
14
47
3162
5647
46
25
14
33
31
10
56
55
46
29
25
33
47
28
47
32
56
33
474747
4545
47
4646
56
14
46
16
47
58
55
47
464646
4756
14
47
22
46464646
56
13
28
47
4516
4756
22
47
45 23
4756
46
45
5647
5647
46
4756
3255
47
45
25
61
464646
3216
47
22
6247 13
31
2546
452958
4625
454510 45
4646
16
46
16
25
47 55
46
47474716
31
1316
6210
4622
4547
2546
25
18
45
46
47
25
4533 45
22
33474747
18
46
1347
4625
4646
47 45
46
1347 624747 55624747 6247 4510473315
46
4747
46276246
47
33
2645
27
47 45
46 2546
26
46
45
2717
47
46
47 45
2517274622
47
62
474747
23
45
46
47
2246
47
2528
47
46 25
4545
4620
474747
4646
55
47
25
4747
10
47
25
47
1046
47
10
23
47
23
22
3346
4726
45
55
46
4747
4662
2546466218
4547
10
47
63
55
2647 454747
2546
47
46
47
46
47
106246
45
4618 25466262
26
46
47
46
47
254646
47
25
47 454747
62464646
47
25
47
25
4547
46464622
47
25
2262
45
55
4628
45
46464620464646
45
33461346
47 45454545
4622
45
63
55
45
46
47474747
46
4747
46
4747 454747 10 16
62
55
47
1610
56
4647
16
55
4747
45
464647
63
47
45
621062
10
33
4646
20
464646
55
56
47
62
45
4647
45
46
31
58
46
62
45
4646
58
31
45 63
46464647
46
62
46464647
46
62
45
4647
59
63
4647
33
6345
56
33
46
62
46
45
46
55
62
58 63
47
62
47
45
47
62
464647
58
47
59
46
20
47
62
26
6262
4646
26
4646
626262
46
55
4647
46
45
46
2513
26
62
1631
58
4747
5533
47
55
1010 16
47
31
55
56
47
16
47474746
474746
45
4647
55
47
55
16
45
4747
56
25
47
45
10
46
10
47
55
47
5525
33
4746
1610
4746
4545
47
62
16
47 4646
14
47
46
32
58 16
18
47
4646
47
32
16
4646
10
62
13
46
55
32
4747
55
46
45
4758
56
4646
14
46
45
55
46
1658
62 46
58
46
59
4662
56
58
46
55
46
45
62 46
59
46
59
3145
62
55
626261
4646
45
55
6146
45
20
4556
46
45
32
46
45
47
46
45
47
20
4646
47474747
46
47
4646
45
4646
47
3215
62
47
45
55
31
10
45
61
31
16
46
47
23
47
46
47
56
4758
46
4716
4646
47
46
10
4646
474747
45
5555
56
13
4556
4758
10
4646
45
10
47
4545
47
45
4716
31
4747
45
14
46
4556 4531
4661
58
56
1414
47
62
31
55
46
16
46
161647
45
58
4556 45
4747
63
4758
4663
56
4646
23
47
45
58
31
4747
10
45
14
59
62 4646
47
46
56
47
56
47
62 4662
47
46
47
62 46
45
62
56
4663
62 46
1320
4759
455656 45
61
47474758
62
58 47
4545
47
62
55
46
59
56
47
46
47
62
1647
46
4545
4662 46
47
14
62
31
59
62
47
62
47
46
47
464646
45
474747
56
55
4663
46
56
464646
47
62
47
62
16
4662
47
62
4758
63
6262 4646
20
18
47
46
56
4758
45
4646
45
47
6262
56
4662
4747
62 46
47
62 46
47
46
45
63
47
46
56
13
47
18
46
4556
59
56 45
466262
63
45
62
4747
45
47
46
4747
45
4747
46
45
474747
62 464662 46
10
4747
46
45
46
472314 22 4710
55
10
46
452213
46
22
6245
4758
464646
475933
55
33 4759
62
46
6262
46
45
46
56
46
10 22
62
564545
474763 28
46
4525 56 4725
31
4545
46
61 47
46
22
46
45
4646
47
62
283363
46
61
62
475625 47
46
4561
5555
2816
31
474545
55
4747
46
14
46
62
4747
62
25
46
33 22 47
6262
14 4745
33
24
46
3363 47
62
33
6262
25104545
1623 4725
46
62
46
4758
62
46
454514
24
56
55
16
55
46
6123
24
562559
55
4556
62
46
47475825
46
5645
46
56
62
5945452563 5625 285623 472523 4747
464646
5656 47
46
56
46
142356 162256
46
61
55
25 56
46
5861
31
59
6262
2825 56 58 4733
46
23
62
55
46
61
62
31
63
46
45
31
456262
55
566110
55
45
24
56 4713 56
62
28 4761 47
31
56 47
55
46
62
5555
47101045
6262
4756
4646
565656
55
13 1645
2222 4714 565614 22 5816
62
474756
46
33 56 474759
31
562356
62
474725
31
2245
46
1445
23 61
46
56 47
4646
45
46
25 47
31
25 56 4756
4646
1310 2825
46
4533 22 475656 28
46
3313
46
1656
46
24
4547
46
25
31
16
46
3345
46
62
14
46
55
46
56 474714
55
46
62
46
1025 5610
55
33
464646
4763
46
25
55
47
45
474646 4756
46
10
46 47
31
474646
28
47
1325
6246
55
47
18
47
22
4747
1856
4746
1313
46 4747
1425
4531
4746
16
31
46
33
47624646
31
62
1445
474762
14
47
45
474646
1163 3156 25
46 4746 4747
55
61
47
55
47
23
4746 47
45
466247
31
62
11
22
46
55
46466262
4662
466262
3223
4756
4747624646 476262
45
4746
63
624747
63
55
476246 476246
2562
5846 476262474646 4746 47472518
58
61
4746
31
46 47
45
4646
2533
46
4545
46 47
55
46
31
474762
13
6256 6246 4746
33 10336156 25
45
46
453261
46
61 32
474747474746
56
4662
10
55
47
55
5847
18
474747475656
13
4747
45
46 4747464646 474758
47476247
61
47
3245 14
55
624746 475656
584747
56
16
5611
4746
252563
4746 47
10
47474758
16
4756
16
4646466262
63 4562
10 14
46
61
55
474756
58
10
6262
63 18
47
454556
474747
63
4747474758
6247
16
47
10
46
16
45
46
45
4747474646
25
22
47
14
46
13
47474725
624646
4510
16
46
56
46
2545
464662
46
1010
47
55
10
472518
22
55
28
55
22
2514
55
47
31
55
474746
45
46
45
55
62
3245
4747
23 45
4746
4510
476247
32
46
62
5846 4516 1046 45
25
464647
28 634646 314747
1447
25 62
3226 451010474747
62
26
29
4728 31
25
1047
45585623 2246
25
284647
28473146 32 58464747
464647
29
25
471316 56
4746
25
46 453247
25
46 4522321047
561347
5647
46 6310 5647
25
4545
47
16
4546
62
46
4747
46
23
46
32
4747
46
18
62
46
33
46
47
22
14
33
62
45
61
4747
56
58
45
4747
25
45
18
46 56
55
46
47
46 14
1029
46464646
47
4645454646
25
4646
25
46
58
16
28
25
59
47
59
5646
62
25
62
46
47
4546
47
4646
22
46
25
4747
46
25
58
4546
47
46
2229
4747
464545
4747
62
18
62
46
33
62
22
45
62
27
46
474747
32
47
45 56
25
46
47
14
25
46
18
16
46
47
22
4718
464646
22
55555555
47
23
45
25
14
25
1616
25
32 58
25
55
46
16
45
16
2710
25
47
56
62 1662
46
18
454546
2525
4646
22
25
62
454646
18
28
46 5646
32
25
4646
22
25
47
33
46
47
10
46
33
46
25
464546
25
5614
4718 47
45464546
25
46
1511
4645
47
46
25
47
23
46
32
454646 144646
4747
464646
474718
25
4646
13
28
55
25
58
45
2525
18
4546
62
4646
47
4645
25
47
4646
61
4546
27
46
1015
464646
25
18
28
47
46
62
464645
25
16
2525
14
25
4662
47
46 25464646
58
56
10
45
28
6246
62
28
46
47
63
58
47
56
45
63
46
45
62
4747
45
47
45
464646464662
46
47
5845
464646
28
62254646
45
47
5662
46
47
4646
28
45
47
55
464646
63
6246
11
46
4558
464662
46
47
624646
63
23
63
17 46466222
47
4662
55
62
454545
4662
55
28
56
45
4747
626256
45
62
58 55
46
55
6246
55
4662
47
46
55
62
555861
4646
55
47
28
47
17
555561
5662
2646
55
25
58 55
56
47
46
4747
46
47
10
5858
47
55
46
63
4646
55
22
47
4646
4747
62
47
10
15
55585858
18
61 5555
25
55
2546
55
46
5545
46
5858
4646565662
45
18
46 2662
45
464646
4558 55
4646
45
62
58
47
62
47
6262
5555
4662
47
58
47
46
58
464646
47
4646
47
46
47
46
4558
62
58
63
46
5545
2526
4718
4646
10
464646
584545
4632
45
62
55
46464662
461346
47
2546 252562
47
55
-1-.
50
.51
-1-.
50
.51
-1-.
50
.51
-1-.
50
.51
-1-.
50
.51
-1 -.5 0 .5 1 -1 -.5 0 .5 1 -1 -.5 0 .5 1 -1 -.5 0 .5 1
-1 -.5 0 .5 1 -1 -.5 0 .5 1
BELGIQUE DANMARK DEUTSCHLAND ELLADA ESPANA SUOMI
FRANCE IRELAND ITALIA LUXEMBOURG NEDERLAND ÖSTERREICH
PORTUGAL SVERIGE UK BALGARIJA CESKA REPUBLIKA EESTI
MAGYARORSZAG LATVIA LIETUVA POLSKA ROMANIA SLOVENSKA REPUBLIC
SLOVENIJA HRVATSKA
Sym
metr
ic R
eveale
d C
om
para
tive
Ad
van
tage
- O
nlin
e
Symmetric Revealed Comparative Advantage - Offline
53
Figure A7: Revealed vs. estimated comparative advantage
5610
46
25
55
4518 32
58
56 321810
13
25
23
10
58
25 5925 4528
46
47
28 10
46
10
46
55
45
4646
2547
56
46
10
46
47
4646
28
46
25
46
55
46
55
46
4747
102562
46
18
55
47
464646
28
46
47
33
23
4747
1018
46
18
4646
14 45
46
59
55
59
23
46
27
46
47 6247
10
4646
45
55
6258
454527
5523
4646
6247
45
46
6262
45
46
10
47
4545
46
475610
47
46
23
464646
45
5862
55
14
46
55
6258
46
20
47 62
23
10
46
4747
10565959
46
5625
46
5555
46
6247
4646
23
45 59
55
4525
23
47
10
58
46
4756
46
32
13
626247
4646
4756
58
5645
5862624756
625656
46
58
45
62
45
23
474756
46
62
28
47
46
62
18
62
46
62
46
2528 56595662
56
4646
56
46
62
4646
475625 56
23 55
25 32
47475614
47565645
4756
47
14
55
45
55
47
45 5647
56
46
5647
56564747
23
565647
32
46
62
46
1845
62
45
475656
4756
472527 5656
46
13
4646
475656
475656
62563345
55
6256
4756
624756
5847 6256
46
18
23
4545256258
45
6247
46
565662
45
4756 3256
6225 59
5558
46
56
55
62
46
47
55
4756
46
562562
45
5862
55
6262
46
45
475625
46
6262
46
55
46
566247
45
475645 56565945
6262626262
46
564747
46
22
47 62
4546
47
46
10
63
45
25
4646464646
47
46464646
25
4646
32
46
10
4646
56
4646
13
46
25
28
47 16
46464646
58
62
46
10
4646
1058
2527
45
62
33
28
32
4646464628
31 62
2846
2846
10
25
28
23 4723
46
4747
4646 45
10
4646
6247
46
62
45
626263
464646
17
28
27
45
4747
45
4747
46
63 56
58
47
22
47
28
4747
46 45
62
10
28
63 47 6262
45
47
45
63 2527
46
62
28
61
22
554763 6247 55
58
564747
58
6247
4646
47 62
33
6259 6261 6247
46
31
4646
58
62
46 4546
31
46
62
46 454646
47
4646 4546 4546
26
5647
45
6259 62
4646
626247
28
47
4646
33
626262
33
6262
46
474747
45
5647 6262
4646 4546
56
2846
6259 6247474747 624759 56
45
4763 6263
46
324759
46
27 4747 6247
46 454646
63 62
46
47
46
4747 6262
46
3256 625556 62
46
4723 59
26
4725 6255
46
10
4729
6262
46
5656
46
62474747 6256 6262474747 6262
28
47
46
25
46 45
59 5647 55
2829
6262
46
33
6262
10
46
6262
46
58
47
46
47
284546
47474725
464646
25 47
46
4747
4646
21
4747
4646
47 6225 47
46
5845454646
47
464628
4628
4646
62
464646 454646 45
47
4646
33
32
46
1647
26
284546
2846
58
62
45
58
59
47 62
55
46
5932
59
56
62
58
63
626247
27
626247
58
47
45
4747
27
17
62
55
56
3232
47
464645
58
46
58
46
6262
46
26
61
62
56
6247
46
1020
46
6262
28
33
62
46
4747
22
58
59
25
62
28
47 62
4646
20 10
56
32
46
58
47
18
47
32
62
58
33
47
1846
33
62
27
6262
55
62
45
4747
55 3055
5858
62
454525
28
46
32
45
58
184625
10
46
58
55
25
28
4645
62
46
63
47 6247
46
28
45
282822
62
25
24
47
464646
47 6247
56
47
46
62
28
1825
28
4625
47
46
5858
47
59
46
55
33
46
62
61
47 6247
4646
62
45
20
18
59
56
60
25
58
46
59
47
6361
454646
6262
2525
61
46
62
4646
47 6247
45
62
58
47
25
33
22
62626247
58
62
59
4747
27
24
47
464545
47
61 32
47
28
46454646
47
28
4618
63
56
46
6247 62
46
2832
27
46
6247 6262
27
28 5555
47
20
62
28
624747
56
47
25
626226
47 62
46
56
62
28
464518
26
10
27
25
22
46
624747
46
47
28
47
3228
46
47
28
62
58
62
58
13
58
25
56
13
56
4646
10
45
62
4645
47
63 1422
58
46
47
454546
47
4646
6247
22
45
10
46
55
46
62
18
6262
59
56
55
47 6226
47 6262
58
4646
55
47 6247
4646
47
22
62
63
33
62
1724
4646
2647
46
58
62
28
4646
4747
28
56
4646
4726
47 6262
27
60
46464646
4747
25
62
5825
464645
32
62
46
47
62
46
25
55
11
18
28
5555
4646
1111
10
47
23
46
25
45
55
24
5517 55
45
14
464646
5555
4546
62
31
5555
46
55
1425
55
23
10
26
555555
59
55
4645
11
10
11
46
5555
18
46
14
55
4610
5555
11
5547
25
4646
56
46
14
18
20
4546
2527
4747
20
46
62
45
47
56
62
55
46
25
46
47
62
45
28 62
25
46
47
4646
59
10
58
464646
62
2159
46
55
60 56
4646
58
1046
47
4645
46
62
45
56 31
46
17
25
62
5660
46
14
46
55
46
32
46
47
11
46
11
55
4646
47
46
31
46
63
23
1847
2558
46
62
46
2658
6363
46
56
4645
58
4646464646
21
46
62
59
62
58
104646
63
58
46
47
4610
464646
11
4646
23
46
55
46
32
47
58
46
11
14
10
25
62
1046
25
47
1046
4747
4645
4610
47
62
47
46
47
45
6262
4546
10
47
4646
28
4646
17 55
2358
47
464646464646
62
45
58
47
59
62
1046
47
62
46
62
46
63
4645
55
4646464646
47
464646
554747
46
32
46
47
6158
55
45
47
46
14 20
46
63
4646
62
25
47
62
21
104546
62
46
62
10
58
46
59
464646
55
62
4645
55
46
582625
4646
47
25
55
11
4646
626262
45
6262
5658
6262
55
2527
4646
62
18
1046
5555
62
58
46
6262
18
62
4546
28
46
62
104546
18
46
62
47
46
62
58
62
10
58
46
262845
25
222217
46
202745
55
46
28
46
28 56
25 46
10
55
182828
6228
224526
4646
58
46
47 10
46
56
55
46
31
55555530 46
3258
55
46462525
10
46
1045 24
46
1745 56
292925
22
46
47
46
26
46
18
2555
4646
55
46
55
4545
55
63 4762 4758186262
46
59
25
1163 451845
46
56
25 46
151862
46
1845
46
47
55
563147
55
27
46
18
4646
62
4646
4747
4646
4762 47
25
18
55
62
46
17 56
46
1445 56
46
62
25
62
61
46
1013
5529
10
46
6245
55
62
46
47 1047626259
46
2726 5610
61
56
46
58 4745 474747
46
5862 314562
46
62 5628
46
626245 474558 4747
46
476245
46
47224545 4762 4745 47 56
4625
1863 10
4646
55
3147 10
5555
58 1045 56
25
2447
46
1845 47
25
4745 4727 472833 1045
46
154762
55
47
46
55
46
4514 6220
4646
5628 56
25 46
2645 3245 47
46
6228 47
46
56
55
6262 56476245
46
5627 10
46
47
46
56
30 46
62
55
46
29
46
4520
55
10
464646
62 32
46
6262 56
46
31 56
46
6262
55
63
55
58
4646
45 474747594747
46
62 17
46
564758
4630
6262624528
6262
61
62
46
58 5663 58625845 56565618 47 5662 4762285862
46
45 1318 564762 10
46
4745
4646
58 4731
55
47
4646
29
4646
11
184746
22
10
25
206132 29
17
18
6127 4661
58
47
28
104647184747
28
47
28
474747
2845
46
45
25
4746
45
1032
62
4647
25
22
56
45
46
58
465547
25
46
58
46
6262
28
23
58
2761
25
47
45
46
22
1847 1332 56
4545454528
6346 56
28
18
23
55
58
474633
62
47
22
47
62
23
2046
62
59
58
31
46
25
45
2562
5646 5646
62
475655
62
46 204647
5646
62
47
58
2245
62
46464647
62
61
25
47
6262
13
45
4756
13
45
47
45
59 56595547
4659
58
55
62
56
62
47
45
55
28
55
45
1047
5627
62
10
62
5546
47
25
21
6262
56
45
46
2562
46
23
13
62
47
62
59
62
5556
184746 56
47
62
59
62
4746
25
58
4746
62
47
6262
5556
22
6262
18
454545
474747
2845
104633
28
294747
104747
6225
18
4558
46
4528
31
33 46474763
4659
45
626262
564646
2562
3347
59
6262
47 5559
47564646
62
4756
47474747
17
63
45
4646
58
554733
45
62
474727 46
55474646
4746
47
58
3247
6262
4747
6262
28
62
45
5646
62
45
5959
62
59 5647
62
63 4746
62
5627 4647
4545
4647
45
46
17
4746
4746
62
47565633
474733 46 20
6346
6346
62
563363 47
564747
5632 5646
45
4756
4759
62
58
62
46
6262
5647
62
47
47
58
25
22
474646
555555
46
59
22
10 452847
46
14
55
58
55
46
55
58
4747
55
47474746
56
59
46
58
4647
55
59
55
4725
4745
46
4547
62
4747
55
4647474747
4545
55
45
55
2525
5656
25
5555
45
46
56
46
56
63
555562
47
6214
56
63 11
56
47
22
46
55
4510
55
25
4547
5562
29
5555
4747
55
45
5562
63
55
59
56
4747
55
46
56
55
47
62
47
5562
46
4547
46
6255
45
464646
59
464647
4647
45
55
4547
29
62
45
55
474625
47
62
47
5555
47
58
47
62
58
474747
58
56
55
56
4745
55
47
58
1010
62
4625
63
6262
47
58
6255
29
47
55
22
5555
28
46
4547
55
45
46
6255
47
5858
45
29
4547
56
4746
47
56
47
62
47
5656
1047
4547
55
63
464747
10 4547
5555
4725
56
2847
59
5656
62
47
55
58
284747
4547
56
62
4759
62
47
58
474747
55
26
46
62
59
25
55
56
62
4747
56
46
58
55
47
56
4547
5562
555555
63
55
58
46
45
58
56
62
56
55
1047
55
56
47
295656
47
62
56
254747
6255
56
47
58
25
45
56
6262
1047
56
4759
55
474746
62
45286347
25
5656
55
47
55
56
4547
55
45
5555
22 5629
46
55
26
55
45
56
28
56
62
10
56
4710
62
11
26
46
55
4746
4710
56
46
58
55
26
45
55
18
4624 4625
18
46
626262
5647
5647
56472524 46
626262
4756
61 3245 463045 474558 474746
62
46
62
61
60 62
47
62
61 47 2156
62
5646
55
62
4658 56
18
47
60
474632 464746
62
45 47
62
47
62
56
55
181818
5858 5656
15
473224 4647474747
2525
56 474727 47
55
46 4746 47 6228131346
32
25
27 62625645
46
14
131646
25
4546 6246 62564646 47
452246 4756
45454616 10 4747
55
31
25
4745
25
46
32
224717 624756
2525
2746 28
55
4645
55
25
47464646
33
46 47 28 224645
46
55
311346 27
3333
46
5555
2828
25
22
14
46
55
62154546
25
474746 2847155646 4745
62
25
46 286247
55
47 6247 62
14
46 28 22
14
46 62
55
464645
62
25
6245
10
25
4646
25
28
25
31
32
28
55
5646 28
25
31 6227 56
33
46
25
4527 62
45
55
464646 6210
25
27
25
28 22
25
47 6246 47
55
46 47 62
25
55
56 47 624646
55
101010
25
4715
5555
4745
4727 2831 471045
62
32
10 47 2846 622846 6245
10
25
4517 46 22473146
4546 28
25
224727 28
25
28
25
55
62
33
28
1414
282710 154645
46 4745
4747
45
5547
46
4545
47474747 55
45
46
474747 6247
4545
46
45
46
624747
45
6247
4646
16
46
47
464646
45
1647
45
46464646
45
62
46
47 62
46
47
4646
5562
45
11
46
45
16
46
62
46
62
464646
6262624747 62
46
47
45
47
45
47
4646
464646
47
10
45
25
45
46
56
4547
10
4646
1163 62 5658
4633
27
46
28
33 4646
16
47
46
11
46
55
47
22
33
28
62
47
32
47
46
45
10
55
47
6262
46464646
56
46
56
47
22
25
46
45
2524
47
62
46
45
59
47
46
4747
46
32
25
46
4745
10
46
4745
25
30
45
10
47
5659
46
47
46
58
25
4646
47
28
46
4545
28
46
4545
18
55
45
46
27
33
4747
4646
62
45
46
45
62
46
47
56
2518
464646
4747
18
55
46
45
46
62
45
32
23
454547
28
46
47
46464646
47
46
47
55
47
22
4747
58
4646
18
46
4745
23
4747
18
4646
11 6227
46
28
62
4545
59 56
33
55
47
46
32
47
46
6262
47
5662
18
62
33
62
4747474745
46
25
464646
62
4633
4545
3032
46
10
25
55
47
46
18
46
32
4747
46
4747
46
6262
46
10
47
56
47
18
46
62
47
46
62 5962
47
5659 56
55
4747
55
10
4547
10
46
56
33
47
62
4646
47
46
47
55
6262 56
46
47
59
46
47
5963
47
6258
47
626232
55
46
47
46
45
2758
28
454747
18
2963
46
56
46
28
47
62
4545
25
62
46
29
46
47
18
46
63
474747
58
47454747
46
47
10
55
46
4747454747
2962
55
47
56
4747
59
46
3262
4747
5662
45
10 56
4646
5555
47
56
46
4747
5662
47454745
46
56
46
5662
4547
4646
47
46
47
13
46
2525
47
10
4646
62
47
13
47
46
63
46
55
47
46
22
47
55
25
46
4747
46
45
24 15
4646
47
25
4626
55
46
47
5555
4747
1025
4763
6262
47
6262
46
4747
25
47
62
58
4747
62
474747
5662
264747 45
56
63
55
47
62
4747 45
4616
55
10
2647
555555
62 56
555555
28 62
5555
4747
55
56
5555
28 62
5555
63
24
47
5555555555
63
55
13
45
5555
56
55
47
46
63
55
46
62 1510
46
4726
62
46
25
32
47
2825
26
55
46
47
32
22
47
58
47
25 28 56
16
25
47
13
55
47
55
47
4646
621725
2645
10
47
2825 62
46
5555
47
32
47
46
10
63
29
46
25
4763
25 22
6363 4563
46
4763 474747
13
46
47
46
2225
46
62 56
454763
22
47
32
4547
56
5555
63 474763
55
56
47
55
4763 45
46
10
464646
4747
464646
62 18
46
55
46
55
46
4547
46
63 47
46
62
4626
47
62
474747
62
55
46
62
46
63
62
47
55
32
47
46
63
46
45
58
63 47
46
6262
47
62
47 4563 4547
46
4763 47
566262
55
47
56
4547
464646
55
454747
46
55
45
25
464646
47
10
55
62
4626
47
46
55
62
46
4747
46
55
46
13
4646
24
4547
46
10
164747
5662 22
47
46
55
47
28 62
47
181062
47
22
55
25
45
25
63 4747
626262
46
4747
55
62
4626
47
462222
4747
58
47
55
10
4645
46
47
10
46
5656
47
46
56
4546
221545
56
5555
46
10
15
55
45
56
55
4545
47
55
4645
62
4747
10
46
18
4645
56
47
10
47
4545
4747
55
45
47
5562
4747
5531 18
46
5562
47
46464646465845
46
56
45
62
45
5656
4646
55 18
46
5555
46
5562
55
10
464646
10
46
4747
6255
47
55
56
4645
47
46
62
47
5555
464545
47
46
5555
4747
62
46
55 18
47
62
46
47
46
5555
46
55
46
47
45
55
56
47
4646
62
46
18
46
55
46
6262626255
224646
3162
47
18
47
45
47
45
626255
56
46
626262
47
45
55
46
56
62
47
46455845
46
565656
4646
55
4747
4546
31 1855
46
47
55555562
46
62
56
47
6218
62
46
47
46
6255
45
47
56
62
58
55
45
56
47
6255
62
58
56
6231
45
47
56
47
626255
4646
62
46
4747
5562
55555562
45
47
62
4645
56
47
56
47
55
47
626262
47
6255
62
58
62626262
56
62
45
62
56
47
46
626255
62
56
18
47
46
47
56
4546
47
56
47
56
45
626262
56
6255
58
62
4546
62
4747
58
625555
47
3155
47
55
47
6262
45
56
46
55
56
55
45
55
56
31
47
55
45
10
47
55555562
46
55
56
46464647
17
46
11
2846
20
58
23
28
17
2728
24
55
20
46
20
4628
46
25
17
46
2322
17
24
46
25 29
22
46
16
28
4522
10
4545
46
10 27
45
5546
475846
29
47
17
24
45
47
29
46
10
4646
20
47 2846
10
2525
20
46
22
25
4646
2747
16
28 1010
25
55
28
17
25
10
46
22
4647
322245
47
2425
4618
16
46
22
46
22
25
47
55
2429
32
29
28
24
4646464647
25
4646
22
4618
25
2846
47
1656
13
4646
56
28
45
46
45
23
45
46
45 16
4746
10
46464655
47
22
104746
58
29
47 28
45
4746
33
10
25
56
4746
45
4646
13
59
27
4655
6262
4628
25
4545
2846
62
47
20
55
62
61
27
464647
4647
6225
22
47
5555
28
62
45
55
62
46
27
4658
2525
47
25
46
20
55
2846
56
1358
45
184646
474746
62
2847
6262
45
46
25
10
22
4647
62
4746
56
18
47
29
46
14
58
17
10
62
46
62
47
33
58
59
62
1846
20
55
45
6262
55
47
5656
62
16
62
10 27
46
22
282847
11 45
55
62
47
20
10474746
20
33
5546
25
58
45
2525
17
46
56
47 10
62
45 59
554646
62
63 474747
56
25
5810
45
58
56
474646
62
565659
464647
62 29
45
5546
56
25
55
47
56
47
55
10
4646
45
62
47474646
16
46
13
5959
46
56
62
46
33
46
61
25
47
626222
164545 16
475846
47
56
1058
59
47
5645 62
56
47474747
565645
62
10
565656
47
4556
55
4646
455656
46
5656
474747
56
28
47
56
4755
56
62
56
55
62
4545
62
56
5547
45
32
56
5547
56
47
5645
46
55
56565645
4747
56
47
56
474755
47
56
5547
5656
27
5645
56
47
4646
55
56
4755
47
32
4545
4755
56
46
4545
47
10
45
47
32
47
27
47
45
47
56
47
5656
32
2027
10
45
47
62
4747
45
28
56
27
5656
55
27
47474747
32
565645
564545
56
1328
32
455656
47
45
62
32
28 45
62
464625 47 4645 475610
55
33 45 4646
62
46
55
33 474728
62
55
221058
55
62
4756 4728 4745 4711 31 464645
62
464745
62
5633 4756 464658
62
56
62
4646
55
10 47 4625
62
2845 47222545 4756
55
4658 4647 46
55
61474756 4656
59
4756
62
45 46
6262
464656
62
25
55
11 46474745
62
58 46
62
33 4747
62
3145 47
5559
4747 46315663 4647
55
63 4747
59
4656
55
56 474525 3110 464710 47
62
314726 4647
62
47 4647
62
56
59
47
62
55
47
6262
58 47 4656 47
62
464747
59
56 464728
55
5826 4656 4747
6262
616147
62
3147
13
25
32
47
46
14
464633 46
45
46
4747475656
14
47
33
56
16
47
25
47
14
10
46
16162023 45
33
102823
5656
11
564727
23
31
46
47
232317
47
46
16
55
2525
46
56
25
13
62
4633 46
45
464646
27
45
464646
56
28 4523
4756
47475656
3228
46
27
464646
25
47474747
45
47
1732 4561
31
45
27
62
33 46
2026
4646
4561
46464646
13
46
27
4646
27
62
464646
10
31
464646
45
46
25
22
46
4747
32
47
46
16
47
62
25
23
4747
46
47
25
46
56
29 4520
46
45
31
56
4646
47
62
18
46
47
45
25
23
47
62
45
46
29
46
28
47
45
6262
27
4646
63
46
25
46
56
4646
22
63
4646
55
4646
10
47
25
47
464646
56
4646
17
46
252525
55
5858
13
47
25
27
23
464633
23
62
16
5656
464646
6324
4633 46
47
18
46
11
47
46
61
25
56
46
27
6214
45
56
16
46
45
46
45
46
4747
15
46
24 4526
56
23
31
27 47
33 4646
282623
33
11
25
11
47
464646
10
46
23
2525
4646
47
46
10
47
28
47
18
47
23286120
46
4524
25
33
4756
1828
14
47
3162
5647
46
25
14
33
31
10
56
55
46
29
25
33
47
28
47
32
56
33
474747
4545
47
4646
56
14
46
16
47
58
55
47
464646
4756
14
47
22
46464646
56
13
28
47
4516
4756
22
47
4523
4756
46
45
5647
5647
46
4756
3255
47
45
25
61
464646
3216
47
22
6247 13
31
2546
45 2958
4625
4545 1045
4646
16
46
16
25
4755
46
47474716
31
1316
6210
4622
45 47
2546
25
18
45
46
47
25
4533 45
22
33 474747
18
46
1347
4625
4646
4745
46
134762474755624747624745 104733 15
46
4747
4627 6246
47
33
2645
27
4745
4625 46
26
46
45
27 17
47
46
4745
251727 4622
47
62
474747
23
45
46
47
2246
47
2528
47
4625
4545
4620
474747
4646
55
47
25
4747
10
47
25
47
10 46
47
10
23
47
23
22
33 46
4726
45
55
46
4747
4662
25 464662
18
45 47
10
47
63
55
264745 4747
25 46
47
46
47
46
47
1062
46
45
461825 466262
26
46
47
46
47
25 4646
47
25
4745 4747
62464646
47
25
47
25
45 47
46464622
47
25
2262
45
55
4628
45
46464620 464646
45
33 4613 46
4745454545
4622
45
63
55
45
46
47474747
46
4747
46
474745 4747 10 16
62
55
47
1610
56
4647
16
55
4747
45
464647
63
47
45
6210
6210
33
4646
20
464646
55
56
47
62
45
4647
45
46
31
58
46
62
45
4646
58
31
4563
46464647
46
62
46464647
46
62
45
4647
59
63
4647
33
63 45
56
33
46
62
46
45
46
55
62
5863
47
62
47
45
47
62
464647
58
47
59
46
20
47
62
26
6262
4646
26
4646
626262
46
55
4647
46
45
46
25 13
26
62
1631
58
4747
5533
47
55
1010 16
47
31
55
56
47
16
47474746
474746
45
4647
55
47
55
16
45
4747
56
25
47
45
10
46
10
47
55
47
5525
33
4746
1610
4746
4545
47
62
16
47 4646
14
47
46
32
58 16
18
47
4646
47
32
16
4646
10
62
13
46
55
32
4747
55
46
45
4758
56
4646
14
46
45
55
46
1658
6246
58
46
59
46 62
56
58
46
55
46
45
6246
59
46
59
3145
62
55
626261
4646
45
55
6146
45
20
4556
46
45
32
46
45
47
46
45
47
20
4646
47474747
46
47
4646
45
4646
47
3215
62
47
45
55
31
10
45
61
31
16
46
47
23
47
46
47
56
4758
46
4716
4646
47
46
10
4646
474747
45
5555
56
13
4556
4758
10
4646
45
10
47
4545
47
45
4716
31
4747
45
14
46
4556 4531
4661
58
56
1414
47
62
31
55
46
16
46
161647
45
58
4556 45
4747
63
4758
4663
56
4646
23
47
45
58
31
4747
10
45
14
59
624646
47
46
56
47
56
47
6246 62
47
46
47
6246
45
62
56
4663
6246
13 20
4759
455656 45
61
47474758
62
5847
4545
47
62
55
46
59
56
47
46
47
62
1647
46
4545
46 6246
47
14
62
31
59
62
47
62
47
46
47
464646
45
474747
56
55
4663
46
56
464646
47
62
47
62
16
46 62
47
62
4758
63
62624646
20
18
47
46
56
4758
45
4646
45
47
6262
56
46 62
4747
6246
47
6246
47
46
45
63
47
46
56
13
47
18
46
4556
59
56 45
46 6262
63
45
62
4747
45
47
46
4747
45
4747
46
45
474747
624646 6246
10
4747
46
45
46
472314 22 4710
55
10
46
452213
46
22
6245
4758
464646
475933
55
33 4759
62
46
6262
46
45
46
56
46
1022
62
564545
47476328
46
4525 56 4725
31
4545
46
61 47
46
22
46
45
4646
47
62
283363
46
61
62
475625 47
46
4561
5555
28 16
31
474545
55
4747
46
14
46
62
4747
62
25
46
33 22 47
6262
14 4745
33
24
46
3363 47
62
33
6262
25 104545
1623 4725
46
62
46
4758
62
46
454514
24
56
55
16
55
46
6123
24
5625 59
55
4556
62
46
47475825
46
5645
46
56
62
5945452563 562528 5623 4725 23 4747
464646
5656 47
46
56
46
14 23 561622 56
46
61
55
25 56
46
5861
31
59
6262
2825 56 584733
46
23
62
55
46
61
62
31
63
46
45
31
456262
55
5661 10
55
45
24
56 4713 56
62
28 4761 47
31
56 47
55
46
62
5555
47101045
6262
4756
4646
565656
55
13 1645
2222 4714 565614 22 5816
62
474756
46
33 56 474759
31
5623 56
62
474725
31
2245
46
1445
2361
46
56 47
4646
45
46
25 47
31
25 56 4756
4646
13 102825
46
4533 22 47565628
46
33 13
46
1656
46
24
4547
46
25
31
16
46
3345
46
62
14
46
55
46
56 474714
55
46
62
46
1025 56 10
55
33
464646
4763
46
25
55
47
45
474646 4756
46
10
46 47
31
474646
28
47
1325
6246
55
47
18
47
22
4747
1856
4746
1313
46 4747
1425
45 31
4746
16
31
46
33
47624646
31
62
1445
474762
14
47
45
474646
1163 315625
46 4746 4747
55
61
47
55
47
23
4746 47
45
4662 47
31
62
11
22
46
55
46466262
4662
466262
3223
4756
4747624646 476262
45
4746
63
62 4747
63
55
476246 476246
2562
5846 476262 474646 4746 474725 18
58
61
4746
31
46 47
45
4646
2533
46
4545
46 47
55
46
31
474762
13
62 566246 4746
33 1033615625
45
46
45 3261
46
61 32
474747474746
56
4662
10
55
47
55
5847
18
474747475656
13
4747
45
46 4747464646 474758
474762 47
61
47
324514
55
62 4746 475656
584747
56
16
5611
4746
252563
4746 47
10
47474758
16
4756
16
4646466262
63 4562
1014
46
61
55
474756
58
10
6262
63 18
47
454556
474747
63
4747474758
62 47
16
47
10
46
16
45
46
45
4747474646
25
22
47
14
46
13
47474725
624646
4510
16
46
56
46
2545
464662
46
1010
47
55
10
4725 18
22
55
28
55
22
2514
55
47
31
15
55
474746
45
46
45
55
62
3245
4747
2345
4746
4510
4762 47
32
46
62
584645 16104645
25
464647
2863 4646 314747
1447
25 62
3226 45 1010474747
62
26
29
4728 31
25
1047
45 58 562322 46
25
28 4647
2847
3146 3258464747
464647
29
25
471316 56
4746
25
4645 3247
25
464522 32104756 13
4756
474663 10 56
47
25
4545
47
16
4546
62
46
4747
46
23
46
32
4747
46
18
62
46
33
46
47
22
14
33
62
45
61
4747
56
58
45
4747
25
45
18
46 56
55
46
47
46
24
14
1029
46464646
47
4645454646
25
4646
25
46
58
16
28
25
59
47
59
5646
62
25
62
46
47
4546
47
4646
22
46
63
25
4747
46
25
58
4546
47
46
22 29
4747
464545
4747
62
18
62
46
33
62
22
45
62
27
46
474747
32
47
45 56
25
46
47
14
25
46
18
16
46
47
22
4718
464646
22
55555555
47
23
45
25
14
25
1616
25
3258
25
55
46
16
45
16
27 10
25
47
56
621662
46
18
454546
2525
4646
22
25
62
454646
18
28
46 5646
32
25
4646
22
25
47
33
46
47
10
46
33
46
25
464546
25
56 14
4718 47
45464546
25
46
1511
4645
47
46
25
47
23
46
32
454646 144646
4747
464646
474718
25
4646
13
28
55
25
58
45
2525
18
4546
62
4646
47
4645
25
47
4646
61
4546
27
46
1015
464646
25
18
28
47
46
62
464645
25
16
2525
14
25
4662
47
4625 464646
58
56
10
45
28
624662
28
46
47
63
58
47
56
45
63
46
45
62
4747
45
47
45
46464646466246
47
5845
464646
28
6225 4646
45
47
566246
47
4646
28
45
47
55
464646
63
6246
11
46
4558
46466246
47
624646
63
23
63
1746466222
47
4662
55
62
454545
4662
55
28
56
45
4747
626256
45
62
58 55
46
55
6246
55
4662
47
46
55
62
555861
4646
55
47
28
47
17
555561
5662
2646
55
25
58 55
56
47
46
4747
46
47
10
5858
47
55
46
63
4646
55
22
47
4646
4747
62
47
10
15
55585858
18
61 5555
25
55
25 46
55
46
5545
46
5858
4646 565662
45
18
46 2662
45
464646
4558 55
4646
45
62
58
47
62
47
6262
5555
4662
47
58
47
46
58
464646
47
4646
47
46
47
46
4558
62
58
63
46
5545
25 26
4718
4646
10
464646
584545
4632
45
62
55
4646466246 1346
47
25 46252562
47
55
-1-.
50
.51
-1-.
50
.51
-1-.
50
.51
-1-.
50
.51
-1-.
50
.51
-1 -.5 0 .5 1 -1 -.5 0 .5 1 -1 -.5 0 .5 1 -1 -.5 0 .5 1
-1 -.5 0 .5 1 -1 -.5 0 .5 1
BELGIQUE DANMARK DEUTSCHLAND ELLADA ESPANA SUOMI
FRANCE IRELAND ITALIA LUXEMBOURG NEDERLAND ÖSTERREICH
PORTUGAL SVERIGE UK BALGARIJA CESKA REPUBLIKA EESTI
MAGYARORSZAG LATVIA LIETUVA POLSKA ROMANIA SLOVENSKA REPUBLIC
SLOVENIJA HRVATSKA
Sym
metr
ic R
eveale
d C
om
para
tive
Ad
van
tage
- O
nlin
e
Symmetric Estimated Comparative Advantage
54
Europe Direct is a service to help you find answers to your questions about the European Union
Freephone number (*): 00 800 6 7 8 9 10 11
(*) Certain mobile telephone operators do not allow access to 00 800 numbers or these calls may be billed.
A great deal of additional information on the European Union is available on the Internet.
It can be accessed through the Europa server http://europa.eu.
How to obtain EU publications
Our publications are available from EU Bookshop (http://bookshop.europa.eu),
where you can place an order with the sales agent of your choice.
The Publications Office has a worldwide network of sales agents.
You can obtain their contact details by sending a fax to (352) 29 29-42758.
European Commission
Joint Research Centre – Institute for Prospective Technological Studies
Title: Barriers to Cross-border eCommerce in the EU Digital Single Market
Authors: Néstor Duch-Brown and Bertin Martens
Spain: European Commission, Joint Research Centre
2015 – 53 pp. – 21.0 x 29.7 cm
EUR – Scientific and Technical Research series – ISSN 1831-9408 (online)
JRC Mission As the Commission’s in-house science service, the Joint Research Centre’s mission is to provide EU policies with independent, evidence-based scientific and technical support throughout the whole policy cycle. Working in close cooperation with policy Directorates-General, the JRC addresses key societal challenges while stimulating innovation through developing new methods, tools and standards, and sharing its know-how with the Member States, the scientific community and international partners.
Serving society Stimulating innovation Supporting legislation
Recommended