52016SC0173

EUROPEAN COMMISSION

Brussels, 25.5.2016

SWD(2016) 173 final

COMMISSION STAFF WORKING DOCUMENT

IMPACT ASSESSMENT

Accompanying the document

proposal for a REGULATION OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL

on adressing geo-blocking and other forms of discriminiation based on place of residence or establishment or nationality within the Single Market

{COM(2016) 289 final}
{SWD(2016) 174 final}

1.Context and scope

Policy context

The Single Market and the Digital Single Market represent two of the Commission's top-ten priorities. On 6 May 2015, the Commission presented the Digital Single Market Strategy 1 to make the Single Market fit for the digital age. On 28 October 2015, the Commission presented the Single Market Strategy 2 to create more cross-border opportunities for business and consumers.

Both strategies consider that Europe's growth is restrained because businesses and consumers cannot fully exploit the economies of scale which are offered in principle by the Single Market. Where buying online has become normal for consumers, buying online cross-border remains the exception. Only half of the companies selling online do so cross-border. When consumers face restrictions linked to nationality or residence in over the counter transactions, this creates mistrust in the Single Market. The two strategies thus announced legislative action to put an end to and prevent unjustified "geo-blocking", i.e. practices used for commercial reasons by online sellers that result in the denial of access to websites based in other Member States, or to the products or services sought for; or where different prices or conditions are automatically applied on the basis of geographic location.

These strategies also announced several legislative actions to address the major remaining barriers which still restrict cross-border transactions so as to make it easier for businesses to sell goods or services to nationals or residents from other Member States:

the two proposed Directives on certain aspects concerning contracts for the supply of digital content 3 and on certain aspects concerning contracts for the online and other distance sales of goods 4 will remove barriers due to consumer contract law differences by fully harmonising the law applicable to defective goods and digital content .

an initiative to extend the Single Electronic Mechanism for VAT registration, declaration and payment to all ecommerce supplies, allowing companies to handle all VAT affairs with their home country administration when selling to multiple Member states. 5

the VAT Action Plan adopted on 7 April 2016 6 envisages ways to simplify VAT rules for e-commerce in the context of the Digital Single Market (DSM) Strategy .

a proposed Regulation on cross-border parcel delivery, which will include measures to improve price transparency for European deliveries, including for prices of small shipments and to enhance regulatory oversight of the cross-border parcel markets. 7

a copyright reform that includes inter alia a proposal for a Regulation on ensuring cross-border portability of online content services 8 , which allows users to access the services they have subscribed to in their MS of residence while temporarily abroad, and an upcoming update of the copyright framework in order to facilitate online distribution of TV and radio programmes across the EU.

a reform of the Consumer Protection Cooperation Regulation 9 in order to strengthen enforcement for problems regarding cross-border services.

Nevertheless, companies may also refuse to sell cross-border for other reasons. In some cases, geo-blocking appears to be linked to agreements between suppliers and distributors. Such agreements may restrict competition in the Single Market in breach of EU competition rules. The Commission presented initial findings on geo-blocking of a competition sector inquiry into e-commerce 10 relating to the online trade of goods and the online provision of services, identifying practices where companies geo-block as a result of (vertical) agreements. For those cases where companies unilaterally and not based on the occurrence of additional complexities and cost refuse customers' access to commercial offers based on the residence of the customer and otherwise discriminate based on nationality or place of residence or establishment ("unjustified geo-blocking and other geo-discrimination"), the two strategies announced specific legislative action, because it will not be reduced by any of the other initiatives. It is also foreseen in the Commission Work Programme 2016. The present impact assessment relates to this legislative action.

Scope of this impact assessment

This impact assessment will examine geo-blocking and discrimination based on nationality or place of residence or establishment.

Its scope is driven by two factors: a) in which services sectors do companies already benefit from a more integrated market, but nevertheless geoblocking occurs frequently and b) which cross-border transactions are already facilitated in the EU for the benefit of customers?

1. The choice of the sectors:

Major steps have already been taken to remove regulatory and administrative barriers for service providers across Europe, moving towards a truly integrated Single Market. The Services Directive 11 and the E-Commerce Directive 12 are the key instruments at European level facilitating cross-border trade for service providers in a large number of services sectors. The Directives cover online and, as regard the Services Directive, offline retail of goods and services, tourism services, entertainment, business services, construction services, etc. Both Directives contain lists of activities which are excluded from the scope, mainly because they are covered by sectors-specific legislation or are either non-economic or economic activities, which are not easily traded across borders. 13 The E-Commerce Directive introduced the country of origin principle for providers of information society service 14 . The implementation of the Services Directive has removed thousands of disproportionate requirements imposed on service providers and it aims to prevent the introduction of new ones if discriminatory or disproportionate. As a result, for the services covered by those two Directives cross-border transactions are already facilitated in the EU. The scope of these two Directives is the result of a carefully reached political balance between Member States and the European Parliament.

However, facilitating the provision of services across national borders is not enough for the establishment of a genuine Single Market. It is equally important to ensure that those who purchase such services as customers for their final use (end-users) can make full use of the opportunities of the Single Market. Article 20 of the Services Directive expressly imposed the respect of the principle of non-discrimination on companies to the benefit of service recipients.

To complete the aim of the two Directives the initiative will cover the issue of discrimination based on nationality, place of residence or place of establishment inter alia for services covered by the scope of these Directives. The analysis of the complaints indicates that the majority of the reported situations of alleged discrimination occur within a few sectors covered by the scope of the Services Directive and, to some extent the e-Commerce Directive - retail of tangible goods (39%) and accommodation and leisure services (16%). 15

The current EU passenger transport legislation already explicitly prohibits discrimination based on nationality and place of residence for three modes of transport (air tickets, bus and coach transport and waterborne transport). The Rail Passenger Rights Regulation does not contain such a provision, but the Commission intends to insert in it a new provision on the prohibition of discrimination with the similar content as the non-discrimination clauses in the other passenger rights regulations, in the context of the on-going revision. Thus the national authorities in charge of passenger rights which are already operational for all modes of transport will be able to prevent carriers and travel agents from resorting to unjustified geo-blocking and discriminatory payment.

Other sectors are too specific or too sensitive in order to be included into this initiative. This concerns in particular health for which payment of health care services is subject to the 2011 Patients' Rights Directive clarifying under which conditions social security institutions would pay. This also concerns financial services and audiovisual services (media) for which supervision by public authorities in Member States is foreseen and such supervision set a different context compared to services which can freely be provided under the Services Directive and the E-commerce Directive . As for gambling, Member States are only bound by the Treaty (freedom to provide services) and might under certain circumstances even prohibit consumers to access such service. These examples explain that the rationale for excluding sectors and activities from the Services Directive should be the same as for excluding those from the initiative for which this impact assessment is prepared

For provision of retail financial services, where the location of collateral and the assessment of risks create important national specificities, issues of discrimination when accessing the service (mortgage, opening of bank account, cross-border insurance) will be addressed in the follow-up to the Green Paper on retail financial services from December 2015. However, payment difficulties related to the purchase of goods and services are addressed under this impact assessment.

Regarding audiovisual services, for which the issue of territoriality of copyright is particularly relevant, a specific framework exists in the form of the Audiovisual Media Services Directive and the Satellite and Cable Directive, both of which are currently under review. Non-audiovisual content services (e-books, music, software, games etc.) also raise some questions. On the one hand, there is evidence that geo-blocking of these services occurs. Including these services within the scope of the measure could therefore enhance consumer benefits. Moreover, the different treatment of the sale of the same content on physical carriers (physical books, CD, DVD, etc.) and their digital equivalent (i.e. e-books, music, etc.) would mean that the geo-blocking initiative covers traditional physical carriers, while not addressing similar problems existing with their digital equivalent. On the other hand, their coverage might raise some complex legal issues that need to be considered. For music, for example, providers will typically have obtained multi-territorial licenses which allow them to serve the European market. Nonetheless, for some of the music in their catalogues, they may only have national licenses, in which case they would not be able to give full access to their repertoire to consumers from other Member States. In the case of e-books, a number of Member States 16 have implemented fixed e-book price laws. Providers of e-books, when selling to residents in those Member States, would have to respect the regulated e-book prices. In order to take these specific aspects into account, while such services should be included in the scope the proposed legislation, their inclusion should not apply to the equal treatment obligation at this stage. The possibility of its extension to these services will be subject to a review.

As a conclusion, the impact assessment is limited to the intersections of the sectorial scopes of the Services Directive and the e-commerce Directive.

2. The geographical and thematic scope of the transactions between companies and end users

In terms of potentially discriminated parties, this impact assessment covers both private individuals (consumers) and businesses when they act as end-users of a given product or service. The impact assessment excludes, however, transactions where goods and services are purchased by businesses for resale. Including agreements that cover sales of goods and services for resale in this initiative would effectively touch upon widely used distribution schemes between companies in a B2B context, such as selective and exclusive distribution, which generally allow for manufacturers to select their retailers and to incentivise retailers to investment. These distribution systems are implemented through vertical agreements between undertakings which fall under EU competition law 17 . However, in order to avoid circumvention by means of vertical agreements the initiative addresses restrictions in agreements which prevent traders from responding to unsolicited customer requests (passive sales) within the scope of the application of this Regulation, building on existing EU competition law concepts.

The initiative applies to all traders operating within the EU, in parallel with existing consumer protection laws such as the 2011 Consumer Rights Directive and the 2006 Consumer Protection Cooperation Regulation (CPC Regulation). Remuneration for the purchase of goods (retail) or services where the payments are already regulated under EU law for the benefit of payees (traders) and customers is also covered by this initiative.

This implies that traders established in non-EU countries but providing their services within the Single Market would be equally concerned. This is in line with legislation concerning the protection of consumers and payment services, which apply irrespective of where the trader is based, and creates a level playing-field between EU companies and non-EU companies. Furthermore, from a customer's perspective it can be difficult to determine whether a company is established and excluding non-EU trader could result in relocation of activities in order to circumvent the rules proposed under this initiative. To avoid an extra-territorial application of EU law, the legislative initiative would clarify that it only applies to transactions for goods and services offered and supplied and paid within the EU: consumers must receive the service or the good in the EU. For instance, the initiative would not apply to an EU consumer renting a car in the US, or to an EU consumer ordering a product to be picked up in Morocco.

Legal context

Direct discrimination based on nationality, as well as indirect discrimination, such as discrimination based on the place of residence or establishment or similar criteria are, in principle, contrary to the principles of the EU Treaty, as recognised by the settled case-law of the Court of Justice 18 . The same goes, therefore, for unjustified geo-blocking and other forms of geographically-based discrimination.

The principle of non-discrimination has been implemented on cross-sector level by Article 20(2) of the Services Directive. It obliges Member States to ensure that companies do not treat service recipients differently based on their place of residence or establishment or nationality, unless justified by objective criteria. Recital 95 of the Directive refers to a long list of generic justifications which companies may be able to invoke. 19 In 2012, the Commission issued guidance on the application of Article 20(2) of the Services Directive

The e-Commerce Directive (2000/31/EC) lays down the country of origin principle for information society services and providers of such services. 20 Member States may not create barriers for the taking up and pursuit of information society services by companies if provided from another Member State. It does not include a non-discrimination provision creating obligations for service providers themselves. The concept of an information society service, under the e-Commerce Directive, and as interpreted by the Court of Justice 21 means any service normally provided for remuneration, at a distance, by electronic means and at the individual request of a recipient of services. Therefore the concept of information society service covers a considerable part of today's e-commerce. The e-commerce Directive however does not include a non-discrimination provision creating obligations for service providers themselves.

The Consumer Rights Directive 22 has fully harmonised certain aspects of consumer and contract law applicable to online sales by traders to consumers. According to Article 8(3) of the Directive, the trader selling through websites shall inform the consumer at the latest at the beginning of the ordering process about delivery restrictions and which means of payment that are accepted. In general, European consumer law covers traders established in third countries directing their activities to the EU.

Under the current EU competition law framework with regard to vertical agreements passive sales restrictions (i.e. provisions in agreements which restrict a retailer from responding to unsolicited requests from individual customers) are generally considered restrictive of competition pursuant to Article 101(1) TFEU and cannot normally be exempted under Article 101(3) TFEU. Accordingly the Vertical Block Exemption Regulation 23 qualifies passive sales restrictions as "hardcore" restrictions which do not benefit from a block exemption on the basis of the nature of the restrictions and the fact that those restrictions are likely to produce negative effects on the market. Those so called "hardcore" restrictions are generally restrictions "by object" when assessed in an individual case under Article 101 TFEU. Practice shows that restrictions by object are unlikely to fulfil the four conditions set out in Article 101(3) TFEU.

As a result, the vast majority of passive sales restrictions in vertical agreements are unlawful under EU competition law. Only in very exceptional circumstances can passive sales be lawfully restricted in an agreement under EU competition law. 24

2.Problem definition

2.1.The overall challenges of a Single Market without borders

The creation of the Single Market is one of the cornerstones of European integration. In it, more than 500 million citizens and more than 20 million companies of the EU should benefit from free access to goods and services from anywhere in the Union, with increased choice and lower prices. To date, efforts to complete the Single Market have been mainly focused on ensuring that companies are free to sell goods and services cross-border. Thus the Single Market has widened the choice for customers who are increasingly searching cross-border for goods and services.

Many buyers, for language, cultural and proximity reasons, still mostly buy goods and services in their country of residence. Despite this, they make use of the Single Market by travelling cross-border or by ordering from distance sellers, nowadays mostly online. As far as travelling cross-border is concerned, 35% of EEA citizens travel at least once a year abroad 25 , either for tourism 26 , business or (if living close to a border 27 ) in search of better offers. However, thanks to the Internet, purchasing at a distance has rapidly increased, with online sales reaching EUR 200 bn for goods, with consumer electronics and appliances, clothing and footwear and media products the most popular items, and another EUR 70 bn for travel and tourism in 2014. Of this, on average 20% cross a border. 28 53% of EU citizens buy online, but only 16% do so cross-border.

When travelling abroad, customers would be surprised if they walked into a shop only to find that the trader refuses to sell to them or only at a different price because they are foreigners 29 . However, in the online world, a trader refusing to sell to foreigners or only doing so at a different price is relatively common. IP-tracking allows websites to identify the provenance of the visitor, as do delivery addresses or payment system information, since there is no cash payment. As shown in the evidence below, customers frequently experience that traders based in other countries simply refuse to sell to them or change their price because of the country where they live or from where they want to access the service.

Most recently, a large mystery shopping survey (MSS) analysed approximately 10,500 websites in the EU, which were visited first by mystery shoppers as domestic users and subsequently as cross-border users from another Member State. It found that only 37% of websites actually allowed cross-border EU visitors to reach the stage of successfully entering payment card details, i.e. the final step before completing the purchase (see annex 6). Equally in the second half of 2015, an e-commerce sector inquiry conducted within the framework of EU competition law found that 36% of online retailers (out of more than 1000 online retailers in the EU who participated in the inquiry) stated that they do not sell cross border in at least one product category in which they are active. This inquiry also showed that geo-blocking was practiced by companies of all sizes, although larger ones were more likely to do so.

Regarding consumer experiences, in 2014 a representative survey of final consumers found that 19% of cross-border online shoppers had experienced restrictions at one stage or another of the purchasing process. 30 Similarly, 2015 Eurostat figures show that 10% of cross-border online shoppers were restricted based on their nationality or place of residence in the last twelve months 31 . These orders of magnitude of the problem are confirmed by a number of other sources (see annex 5). Moreover, the Commission services established and analysed a database of over 1500 discrimination-related complaints or enquiries received between 2008 and 2015 through internal channels 32 . They reflect complaints and enquiries online and over the counter and mostly concern the sectors of retail (39%) as well as accommodation and leisure services (16%).

Problems have also been found with regard to firms purchasing goods and services cross-border. A 2016 B2B Eurobarometer survey of 4.200 companies across 15 Member States 33 reveals that firms face restrictions as well, which is why 68% of firms interviewed are not purchasing from abroad 34 . Among the companies having purchased goods and services cross-border (online and offline) as end-users during the last 12 months on average 10% have experienced problems carrying out these purchases. Smaller firms have experienced more problems than larger firms and microenterprises are the most affected.

Geo-blocking and other restrictions based on nationality, place of residence or place of establishment should however not be confused with the emerging trend of dynamic pricing. In the age of Big Data, firms set their online prices based on a wealth of customer attributes, including browsing history, operating system, and previous purchases/visits to the website. This impact assessment does not examine variables other than the country of residence or establishment and nationality of the customer used by companies in their pricing decisions. The reason is that only the latter is a point of reference assessing the equal treatment principle under the Treaty. Information about place of residence or establishment and nationality is either collected by directly asking customers to reveal it (e.g. choosing their country from a drop-down menu) or deducing it from proxy variables such as the IP address.

2.2.Unjustified geo-blocking and discrimination based on nationality, place of residence or place of establishment

2.2.1.Situations concerned by unjustified restrictions

Geo-blocking and other restrictions based on nationality, place of residence or place of establishment can be justified or unjustified. If they are based on businesses incurring significant extra complications and costs, for example because of the existence of additional regulatory or other obstacles for cross-border sales, they are likely to be considered justified where ther are proportionate. Otherwise, they constitute "unjustified geo-blocking and other types of geo-discrimination".

In many instances, it is impossible without a case-by-case investigation into the specific circumstances whether such restrictions are justified or not. However, there are several situations where local and foreign customers are in a very similar, if not identical, position (i.e. the trader does not bear any specific cost linked to the nationality, place of residence or place of establishment of the customer; he does not have to arrange for cross-border delivery, pay VAT in another country or apply another Member State's consumer legislation):

(1) Foreign customers are sometimes prevented from buying a physical good even if they are willing to pick up the product in the country of the trader or arrange themselves for the cross-border part of the delivery. For example, according to a recent survey, 32% of Western European 35 international online buyers have – for a variety of reasons including but not limited to unavailability of shipping to home address - used a forwarding address to get goods delivered 36 , indicating significant demand. Yet according to the Mystery Shopping Survey 26% of websites do not allow foreign customers to proceed to the stage where they could organise such forwarding.

(2) Foreign customers are sometimes prevented from buying electronically supplied services, such as cloud services 37 , even though delivery costs cannot be an argument for not selling cross-border and the taxation implications of cross-border sales have already been subject to the facilitation mechanism of the VAT mini-one stop since the beginning of 2015.

(3) Despite the trader having the required rights for the relevant territories, foreign customers are sometimes prevented from accessing electronically supplied non-audiovisual copyright protected content services. A study of the largest provider of downloadable music revealed that cross-border access was not or only to a very limited extent possible 38 . In the recent Commission's competition inquiry, 44% of music providers respondents answering stated that they are required to geo-block 39 . The reasons behind these restrictions (e.g. availability of licences) are yet to be analysed by the inquiry. According to a recent study on the geographical market segmentation in the EU e-books market, European consumers face different access restrictions based on their residence. Overall, these differences in access mainly seem not to be driven by objective legal reasons related to the EU copyright regime 40 .

(4) Customers moving to another country to receive the service at the location where the trader operates are sometimes charged prices which differ from those applied to domestic customers, such as when staying at a hotel, renting a car, going to a sport event or shopping in a cross-border region. Large-scale web-scraping studies showed that European consumers continue to face price discrimination when booking rental cars across the EU. 41 For example, consumers from Germany accessing the non-country specific EU website (.com or .eu) of the car rental company got offered in 13% of cases higher price than consumers from Italy with no possibility to see the lower price available to other consumers. On average, the difference in price is 11€, but can get up to 134€. Similar effects can be observed on the same website when comparing consumers in the UK and Netherlands to those in Germany: 16% of prices are higher for the UK consumers, with the average price difference being 8€, and 13% of prices are higher for the Dutch users, with the average difference being 32€, although the exchange rate is the same in both situations. Overall, the study detected price differences on the non-country specific EU website against users in all Member States, ranging from 6% of prices for Austrian users to 16% for UK users, with most Member States falling around 10% of prices. 42 Similarly, a citizen who moves temporarily to another Member State (without taking up residence there) may face higher prices or different conditions compared to local residents.

In these cases, differences of treatment can be considered in principle as being unjustified.

2.2.2.Unjustified Restrictions at different stages of the purchasing process

Unjustified restrictions based on nationality or place of residence or establishment of a customer take place at various stages of a purchasing process, presented below.

The results of the public consultation conducted in autumn 2015, discussions organised by the Netherlands Presidency at the Informal Competitiveness Council at the end of January 2016, and the outcome of a stakeholder workshop held in Amsterdam on 18 February 2016 with more than 80 key stakeholders and representative organisations from consumers and from the business community (see annex 2) and many other stakeholder contacts underlined that it is important to assess each of these stages in order to fully comprehend the problem.

2.2.2.1.Lack of transparency

Customers are particularly frustrated if they receive no information about restrictive practices. Under the Consumer Rights Directive, consumers should receive at the beginning of the purchasing process information about delivery restrictions and the accepted means of payment. However, the investigations regularly show that a large number of websites does not comply 43 with this obligation. Furthermore, there is no obligation for traders to provide an explanation for such restrictions to consumers. The outcome of the public consultation shows that 27% of respondents who asked for an explanation have never received one from the trader about the reasons why they refused to sell or why they charged higher prices based on the place where the consumer lives (see annex 2). In all the complaints received directly or indirectly by the Commission services (see annex 8), 92% of cases do not mention any type of justification given by traders.

2.2.2.2.Lack of access to websites and rerouting

In the three situations mentioned above (see section 2.2.1), as well as in cases where geoblocking may be justified, customers are also confronted with the situation that access to the website or to an application is blocked or that they get automatically rerouted to another website. Geo-blocking of this type is often based on technological ways of using the location of the customer as a proxy, such as IP-tracking.

The 2015 online mystery shopping survey found that only 2% of websites completely block access or automatically reroute visitors from another Member State, which however account for 7.5% of cross-border online traffic in the survey sample (see also annex 6 for more details). This hinders in particular the development of cross-border price comparison sites, as they are not able to scrape information on prices in other Member States.

2.2.2.3.Unjustified refusals to sell or selling under different conditions

There are many examples where traders refuse to sell or charge higher prices depending on the customer's place of residence or establishment or nationality even though local and foreign customers are in a very similar, if not identical, situation (i.e. the trader does not bear any specific cost linked to the nationality or place of residence or establishment of the consumer; they do not have to arrange for cross-border delivery, pay VAT in another country or, if there is no directing of activities to consumers in that Member State, the trader does not need to comply with more stringent mandatory consumer contract rules of the consumer's Member State) 44 . Traders use different techniques to identify the place of residence or nationality of the customer through information on the delivery/billing address, but also information about passport, ID card or company registration number (VAT number), and the country of issuance of the payment card, the credit card number, the country where a bank account is based or again by the IP-address. It may simply be impossible for the customer to enter his complete details online – the postcode format may only accept a limited range of postcodes, or the country field may be a scroll-down list with a limited choice, etc. The 2015 online mystery shopping survey (see annex 6) found that 27,6% of websites do not allow foreign visitors to reach the stage of having successfully registered in order to initiate a purchase (including both justified and unjustified refusals).

In three specific situations described below, differences of treatment based on residence or nationality can be considered to be unjustified:

(1) When foreign customers are prevented from buying a physical good even if they are willing to pick up the product in the country of the trader or arrange for the cross-border part of the delivery.

(2) When foreign customers are prevented from buying electronically supplied services, such as cloud services, even though delivery costs cannot be an argument for not selling cross-border and the taxation implications of cross-border sales have already been subject to the facilitation mechanism of the VAT mini-one stop since the beginning of 2015.

(3) When customers move to another country to receive the service at the location where the trader operates, they are sometimes charged prices which differ from those applied to domestic customers, such as when staying at a hotel, going to a sport event or renting a car in a cross-border region.

2.2.2.4.Refusal of foreign means of payment

The Mystery Shopping survey shows that for cross border shoppers who reached the payment stage, approximately a quarter of websites (26%) did not accept/offer their means of payment or it was not possible to successfully enter their payment card details. According to the 2016 B2B Eurobarometer companies facing problems did so when they indicated that they wanted to pay with a bank account located in another EU country (19%) or with a credit/debit card issued in another EU country (15%).

This suggests that the payment element of a transaction is probably a regular source for transaction refusals. Internet retailers may refuse transactions based on considerations connected with the payment itself. However, they may use elements of the payment (country of issuance of payment card, IBAN number) as an 'indicator' of the location of the buyer to refuse a transaction on other grounds.

2.2.2.5.Refusal to deliver

Even if consumers have managed to proceed in the purchasing process, they often fall at the last hurdle. The 2015 online mystery shopping survey showed that out of the 72% of websites which allowed visitors to register, approximately one third (32%) did not provide delivery options to country of the visitor(see also annex 6 for more details). Finally, the analysis of complaints collected by the Commission shows that 9% of consumers complained about non-delivery to their member state (see also annex 7 for details).

The 2012 Guidance stated that "The lack of alternatives for delivery can rarely be invoked by a service provider to refuse supply to a given Member State. For parcel deliveries up to 20 kg, the Postal Services Directive imposes an obligation on Member States to ensure the provision of universal postal services, including cross-border. This obligation means that at least one delivery option in a cross-border context should be available in all Member States."

2.2.2.6.Conclusion

Overall, two cross-border shopping attempts out of three fail (the following table synthetises the problems identified above). This is substantiated by the mystery shopping survey (see Figure 1), by the B2B Eurobarometer survey (see Figure 2) and by the responses to the public consultation.

Figure 1: Failure rate (%) in a 2015 online cross-border mystery shopping survey

Source: GfK Mystery Shopping Survey and JRC/IPTS calculations (forthcoming)

In theory, one would have to identify the motivation for each geoblocking situation in order to quantify how much it is based on objective barriers and how much is unjustified. This is not possible in practice. It implies a case-by-case analysis by enforcement authorities under the current legal framework (article 20). This could not be done with the studies carried out and the complaints received. It is not possible to conduct such additional studies, because, contrary to competition sector-specific enquiries, the Commission has no power to demand that companies explain in detail their pricing strategies. This could only be done in targeted cases attracting media attention (car hire, 45 for instance). The complaints received, the responses to the public consultation and Eurobarometer surveys demonstrate that there is a significant degree of frustration on the side of consumers.

However, other than in very exceptional circumstances, there are no costs or complications associated with permitting cross-border access to websites. As a general rule, automatic redirecting should thus be considered unjustified. On the other hand, while a refusal to deliver could be either justified or unjustified, depending on the motivation for such a refusal (see section 2.2), they are likely to be justified, given the objective complications of cross-border delivery, tax laws and, if the trader directs its activities to the consumer's Member State, differences in applicable consumer protection. Therefore, it is possible to estimate that unjustified geoblocking is practiced by between 2% and 27.6% of websites, keeping in mind that the 2% account for 7.5% of cross-border traffic.

Even if to a smaller extent than for consumers, problems are also encountered at various stages of the purchasing process in B2B transactions. Among companies experiencing problems when purchasing goods and services cross-border issues related to VAT or payment where among the most cited (Figure 2). 46

Figure 2: Step of the purchasing process during which the problems were experienced (B2B)

Source: European Commission, Flash Eurobarometer 434, 'B2B Geoblocking', 2016 47

The significance of the issue is also reflected by the results of the public consultation on geo-blocking (annex 2). A vast majority (98%) of consumer respondents consider that consumers should be able to purchase and access services everywhere in the EU. They also think that geo-blocking and other geographically based restrictions create significant obstacles to the single market. Companies largely share the view that geo-blocking/discrimination is an issue and that consumer choice and competition are important. A large majority of consumer respondents (70-94% depending on the option) supports a legislative solution to tackle unjustified geo-blocking, while companies agree in principle (up to 50% agreement depending on the option) but stress respect for contractual freedom.

2.3.The drivers of the problem (underlying causes)

There are different reasons why traders may not sell goods or services to customers in other Member States. They may not sell or provide access to customers from other Member States because the regulatory complications or language issues outweigh the benefits from additional sales, for example because the company would have to register for VAT in different Member States. This may even be the case if the company does not invest in commercial promotion in a foreign market but is approached by a customer from abroad. When the differences in treatment are thus based on objective and actual additionnal complications and extra costs for the seller and they are proportionate thereto, they may be justified. Otherwise, they are unjustified.

Based on the evidence gathered it is not possible to provide an assesment of the relative importance of each driver, given that the prevalence will differ depending various factors. Taking the exampel of VAT differences, the relative importance will depend on the type of product and the Member State concerned.

2.3.1.Barriers leading to geoblocking and other types of discrimination

Barriers which create significant additional complications and extra costs for the seller include the following:

The complications of having to deal with many different national legal and tax systems represent a real obstacle for companies trying to trade cross-border both on and offline. The complexity of VAT for making intra-EU B2C supplies is cited by business as one of the top three barriers to cross-border e-commerce. 48

Differences in national consumer laws may be a reason for refusing cross-border sales in a B2C context (this would not be relevant in a B2B relationship),.

Technical specifications or rules on labelling and selling arrangements may also differ depending on where in the EU the trader has expressly chosen to supply products. They may require the trader to adapt their products and packaging accordingly, in particular concerning linguistic versioning.

Beyond these objective differences, even the uncertainty as to whether the trader should apply foreign law or not may be a driver. Given the case-by-case assessment required under the Rome I regulation 49 pursuant to the case law of the Court of Justice of as regards the applicability of national law, a company may very well prefer to forego the additional revenue rather than to face a badly understood legal situation.

The lack of possibilities to arrange for good after-sales services can be a reason for traders to deny cross-border sales.

Lack of affordable, high-quality delivery services is consistently cited amongst the top reasons mentioned by both e-retailers and consumers 50 for not engaging in cross-border e-commerce.

Until the implementation of the Payment Servuices Directive 2 51 in 2018, some operators may prefer not to sell to other Member States because their fraud prevention system is based on access to informational resources such as public registries which may not be available in other countries.

Companies may refuse to sell to residents of other Member States because of legislation forcing them to do so (for instance a measure adopted by a local public authority which restricts access to coffee-shops to Netherlands residents 52 ).

Companies may in specific circumstances refuse to sell to residents of other Member States on the basis of lawful (vertical) agreements. For instance, paragraph 61 of the Vertical Guidelines 53 acknowledges that there are situations in which territorial restrictions may be objectively necessary for an agreement of a particular type and therefore fall outside Article 101 (1) TFEU altogether. Agreements with territorial restrictions may also fulfil the conditions under Article 101 (3) and therefore be compatible with the internal market.

For online content services, companies may in some cases only hold the rights (licences) for the use of copyright-protected content for a specific territory, which prevents them from providing such services to all Member States.

Geoblocking or different treatment based on these issues can be considered justified (see annex 11). Some of these issues are addressed in the legislative proposals put forward as part of the (Digital) Single Market Strategy (see section 1) and not in this initiative. As a consequence, justified geoblocking can be expected to decrease in the future.

2.3.2.Drivers for unjustified differences of treatment:

2.3.2.1.Market segmentation along national borders

Companies may prevent cross-border sales in order to enforce market segmentation. Such action may be based on a unilateral decision of businesses, or on contractual arrangements preventing resale of products (in a B2B context).

If a supplier forbids a retailer from servicing an unsollicited request from another Member State ("passive sale"), is in principle against EU competition law 54 . Other than in cases of abuse of dominance, EU competition law does not, however, address restrictions stemming from unilateral business decisions of non-dominant companies, including decisions taken and applied within a group of companies.

In order to systematically analyse restrictions of competition that create barriers to cross-border e-commerce, the Commission launched a sector inquiry into the e-commerce sector in all Member States. The inquiry with respect to goods showed that most restrictions of cross-border sales are linked to unilateral decisions by traders, not by contractual arrangements with suppliers. 55 Moreover, as the enforcement practice of national competition authorities demonstrates, such restrictions do not only take place for sales of goods, but also in other services sector, such as hotel bookings or package travel. 56

Individual companies might find it profitable to segment the market. Market segmentation is often an issue where the trader has distinct operations in or targeting several Member States. This is typically the case of multinational companies, although it is also possible that SMEs can apply such restrictions. In certain cases it could be even beneficial for some customers i.e. depending on national demand and supply, certain goods and services may be sold for higher price in some countries than in others (see also section 2.3). The Study on business practices applying different condition of access based on the nationality or the place of residence of service recipients from 2009 57 highlighted a number of reasons for differences of treatment, which the Commission considers do not constitute objective criteria justifying a difference of treatment. Market segmentation in itself is a legitimate way to take into account different markets in a commercial strategy and can take many forms, for instance, special rebate add appearing only on French websites, a different presentation, a special price "train from Belgium+hotel" available on the Belgian version of a website, or even different prices of the packages/options on different national versions of the website etc. It is only a problem when the company refuses requests from consumers based in another country than the one targeted in order to enforce this market segmentation, when the foreign consumer is effectively in the same situation as the national consumer.

Companies also sometimes rely on alleged legal uncertainty concerning the applicability of foreign consumer law, VAT, or labelling requirements etc, to apply different conditions on customers, when in fact these problems do not arise in the three cases mentioned above.

2.3.2.2.Ineffective enforcement of the existing legal framework

Although it has been in force since 2009 in all Member States, the effectiveness of the current legal framework (Article 20 of the Services Directive) is hampered by a lack of proper application and enforcement on national level. One of the main reasons for the lack of enforcement is the legal uncertainty regarding justifications for differences in treatment based on customer's nationality or place of residence or establishment. Differences in treatment do not always constitute discrimination. According to Article 20(2) of the Services Directive, differences in the conditions of access are legitimate if directly justified by objective criteria. To determine whether a difference in treatment adds up to discrimination or not, it has to be assessed if the reasons invoked by the trader indeed constitute objective criteria justifying the difference in treatment. Recital 95 of the Services Directive provide for a broad list of possible objective criteria which makes it difficult to apply the provision in practice. At present, the assessment must always be made on a case by case basis, thus creating uncertainty for traders, customers and national enforcement authorities. 58

Furthermore, some Member States even failed to identify clear enforcement authorities. 59

The situations covered by Article 20(2) may require Member States's authorities to cooperate with each other in order to ensure an effective enforcement. 60 Regulation (EC) No 2006/2004 on consumer protection cooperation (CPC Regulation) lays down the general conditions and framework for cooperation between national enforcement authorities in the EU. The cooperation is applicable to consumer rules in various areas, such as unfair commercial practices, distance selling and passenger rights which are listed in its Annex. However, Article 20(2) of the Services Directive is to date not included in the Annex. It means that there is currently no cooperation in terms of coordinated enforcement of Article 20 (2) by national authorities. The Commission has proposed to include Article 20(2) of the Services Directive in the Annex of the CPC Regulation. Adoption is planned for May 2016.

Guidance on this Article 20 was issued by the Commission in 2012, 61 but it did not lead to any clear improvement in terms of application and enforcement on this point.

Although it is often stressed the need for a case-by-case analysis, the Guidance provided examples of situations (including those identified above) where it was difficult to find any justification for differences of treatment based on nationality or residence.

-For instance, it said that "techniques allowing service providers to identify the location of the recipient and thus to direct the consumer to the offer adapted to the territory where he is resident are not per se indicators of discrimination" but that if "recipients in each of these countries are completely barred from accessing information on the conditions of access offered to recipients resident in other Member States", the situation may be very different.

-It also stated that "the lack of alternatives for delivery can rarely be invoked by a service provider to refuse supply to a given Member State." All the more, if the consumer is prepared to organise delivery itself, this makes a difference of treatment even less justifiable.

-Similarly, for services received on the premises of the trader (or outside of the country of residence of the consumer), such as hotels or car hire, the Guidance recognised that different pricing and marketing policies could be set up towards the various Member States, but that the objective is for consumers to "become more active in seeking out such favourable conditions offered online by service providers established in other Member States." Such an objective can be jeopardized, not only when consumers are prevented from accessing information, but also when they applied a higher price than the one announced on the website visited.

-As far as payments are concerned, the Guidance considered that refusals of foreign means of payment would become unjustified when the relevant barriers have been addressed, which they have.

Despite this, the recurring complaints received from consumers, and the fact that the Guidance clearly encouraged Member State authorities to "take into account progress made in the completion of the internal market and the further elimination of barriers", the Commission is not aware of any trader having been sanctioned for infringing the non-discrimination provision of article 20 (2) of the Services Directive. Equally, no Member States took action and, based on the guidance, amended their national provisions transposing Article 20(2). For instance, the progressive entry into force of the Single Euro Payment area 62 and of the Payment Services Directive 2 should have led to a more stringent application of rules against discriminations linked to means of payment. This was not the case. In specific cases, only the intervention of the Commission prompted the trader to give up its refusal of foreign credit cards or direct debits from SEPA-based bank accounts.

This lack of enforcement action can partially be explained with reference to the low value involved in consumer disputes which deters consumers from pursuing legal action, but also by the low degree of priority given by national authorities to enforcement of article 20.

Article 8(3) of the Consumer Rights Directive requires traders to indicate clearly and legibly at the beginning of the ordering process whether any delivery restrictions apply and which means of payment are accepted. The 2015 online mystery shopping survey showed that in 37% of all websites assessed, no information on delivery restrictions was clearly displayed on the starting page or during the ordering process. This was as high as 44% for those websites that were found to practice some kind of restrictions. If Article 8(3) was respected by all traders, consumers would at least be aware of delivery restrictions and the payment means accepted by trader before going through the ordering process.

2.4.Who is affected, in what ways and to what extent?

Customers are often prevented from taking full advantage of the opportunities of the (Digital) Single Market. This has two main effects. Firstly, they are prevented from accessing a greater variety of products. For example, evidence shows that less than 7% of all smartphone models are available in all Member States; almost 50% are available in one Member State only 63 . Secondly, customers face higher prices as competition is less intense than it would otherwise be; for some electronic goods, such as laptops, tablets, desk top computers or smart phones, customers in the most expensive Member State are paying between 30% and 60 % more than if they would be able to purchase from the cheapest Member State. 64 In the previously mentioned web-scraping of car rental tariffs, customers were charged up to 53% more for the same car during the same period and on the same location due to the fact that they came from specific Member States. 65

Whether customers realise that they are being discriminated or not depends on the technological means used to identify their residence: if IP-tracking or another technology of similar effect is used, customers will from the start only see the higher prices; if they suddenly face higher prices after entering address or credit card, they will notice. In 30% of the complaints received by the Commission, customers complained about such price discrimination (see annex 7 for details). This results in customer frustration, as they are largely unable to enforce their rights. Lodging a complaint or finding the appropriate enforcement body when facing different conditions either online or over the counter is often too burdensome compared to the detriment endured.

Moreover, as a direct result of unjustified geoblocking, e-commerce and cross-border trade are lower than they would otherwise be, resulting in a loss of consumer surplus.

For businesses as end-users, the B2B Eurobarometer undertaken in February 2016 gives evidence on problems experienced when purchasing goods and services cross-border. The results also showed that smaller companies, and in particular microenterprises, were most impacted. For companies buying goods and services cross-border as end-users, especially small ones who do not have a legal department, finding the right assistance body is a challenge, since it varies from country to country (sometimes it is the ministry of economy, the local Chambers of commerce, or the competition authority). For companies, the only enforcement bodies available are often national courts.

Businesses may apply unjustified geoblocking through unilateral measures. If they are using this practice to maintain price segmentation between different national markets, geoblocking increases their benefits. Larger firms with presence in more than one Member States may target different markets with different offers and prices However, as shown by the mystery survey, 2% of websites, particularly those of companies active in several Member States, automatically send the customer to a different website (or block access) which in turn accounts for 7.5% of cross-border traffic.

If companies do not sell to foreigners because they are afraid of ending up having to deal with the complications of selling cross-border (especially in the case of microenterprises) they are not benefitting; on the contrary, they would benefit from being able to sell more easily to a larger customer base.

2.5.How would the problem evolve, all things being equal?

In case of no EU action unjustified geo-blocking practices would continue to exist segmenting the Single Market. Consumers will be prevented from making cross-border purchases or will be (unfairly) charged higher prices for the same good or service and consequently would not benefitting from increased availability of product and services and increased price competition. Traders could continue to deny access to certain services cross-border by applying territorial restrictions limiting the free movement of services. Companies would also continue to apply different conditions on the basis of residence without objective reasons. As described in previous sections, there are not only divergences in national legislation, such as consumer protection legislation and product labelling rules, having an impact on traders' choice of limiting cross-border trade. There is also pure market fragmentation based on unilateral decisions taken by companies without objective reason.

The adoption and entry into force of the proposed legislation on digital contracts 66 and parcel delivery will contribute (although not totally) to reducing the impact of some the drivers for justified restrictions, which could take away part of the justifications often invoked by companies. So would the announced initiative on extending the VAT Mini One-Stop-Shop to B2C supplies of tangible goods. Finally, following the entry into force of the revised Payments Services Directive, perceived risk of fraud should no longer be an objectively justifiable argument to geo-block. However, while due to these new legislation there will be less justified geo-discrimination they will not reduce unjustified geo-discrimination which is driven by the intent of business to artificially segment markets along national borders. So, as such, these legislative initiatives do not have an impact on unjustified geo-discrimination. On the contrary, for cases where traders currently use specific drivers as mere pretexts to explain plain market segmentation strategies, the removal of those objective barriers may even lead to an increase of cases of unjustified restrictions.

The e-commerce sector inquiry launched by the Commission in 2015 unveils more in-depth information with respect to agreements between undertakings that impose territorial restrictions. These restrictions may be tackled by competition law enforcement. Restrictions based on agreements are not part of this impact assessment.

Moreover, with the advancing technological sophistication of e-commerce providers, it is possible that unjustified geo-blocking might actually increase in the future. As the software used for tracking and identifying online customers gets better and cheaper, geo-blocking will become easier to implement.

Furthermore, the rapid development of mobile e-commerce in conjunction with the abolition of roaming charges in 2017 will further exacerbate the problem, as consumers may well be geoblocked because of the IP-number of their device whilst they are in the country of the trader.

Consumers would therefore remain unable to enforce their rights despite the revised Regulation on consumer protection cooperation. There also would be no solution for B2B transactions. While the inclusion of Article 20 of the Services Directive within the scope of the CPC Regulation would be beneficial to ensure coordinated enforcement across the EU, it would not be sufficient on its own to resolve the problems described in this impact assessment. In fact, while the revised CPC Regulation would provide additional powers for authorities to cooperate and would strengthen cooperation for widespread infringements, the fact would remain that the principle of non-discrimination laid down in Article 20(2) of the Services Directive is to be understood in light of the (numerous and not always very precise) possible justifications listed in its recital.

2.6.Evaluation of existing policy framework

The assessment of Article 20(2) of the Services Directive has shown that it is at present not an effective or efficient EU intervention to achieve the pursued objectives. On 8 June 2012, the Commission published guidelines on the on the application of Article 20(2) Service Directive. 67 The guidelines provide clarification on the interpretation of the relevant article, in particular its scope, and demonstrate typical situations of discriminations.

Awareness-raising activities are regularly conducted, at national level by consumer associations and European Consumer Centres, and at EU-level by the Commission (for instance through the issuance of a Practical Guide for consumers in 2014 68 and communication activities such as press releases in the car hire sector 69 ). As a follow up, no Member State started changing their national rules or practices. To the Commission's knowledge, only Malta issued its own Guidance on discrimination, which essentially reproduces word by word the 2012 Guidance. This did not have any noticeable result, since the Guidance can only be based on the broad and vague wording of article 20.

An approach based on voluntary agreements had been attempted with respect to the car hire sector, with a publicity campaign by the Commission in 2014 leading to commitments by some companies. The most recent web-scraping study of March 2016 demonstrated the several companies nevertheless continued to apply unjustified differences based on the place of residence or establishment of consumers and the Commission still receives complaints.

One of the main reasons seem to be the broad list of possible objective criteria justifications mentioned in Recital 95 of the Services Directive. However, the objectives of the non-discrimination principle remain highly relevant, and coherent with other EU policy actions, such as the Digital Single Market Strategy and the Single Market Strategy for goods and services. The introduction of Article 20(2) of that Directive has not brought any substantive added value due its lack of effectiveness, but the evaluation shows that EU intervention to address the issues concerned is necessary to reach the objectives.

3.Why should the EU act

3.1.Does the EU have the right to act?

Traders are free to make contracts with any customer in the EU. Nevertheless, Article 18 of the TFEU establishes the general prohibition of discrimination between persons on the grounds of nationality. Articles 34 and 56 of the TFEU provide for the free circulation of goods and the freedom to provide services. The Court of Justice further clarified the implications of the non-discrimination provision in Article 56. It said that Article 56 TFEU gives rights not only to the provider of services but also to the recipient. 70 It also added that the abolition of State barriers to freedom to provide cross-border services would be compromised if it could be neutralised by obstacles resulting from publicly announced decisions of private operators 71 (for instance publically available terms and conditions) or private associations. In those situations, decisions of such private entities might have effects equivalent to those of public authorities. The Court judged that equality of treatment between nationals and non-nationals prohibit not only overt discrimination based on nationality but also discriminations linked to the place of residence. 72 It held that even a measure putting at an advantage only residents of a specific region or area may be considered discriminatory; it does not have to disadvantage only nationals of other Member States.

Article 20(2) of the Services Directive does not prevent the Commission from adopting a proposal going beyond the current legislation. A new legislative initiative addressing discrimination based on residence or nationality could complement Article 20(2), which would remain applicable insofar as it is compatible with any new instrument. Even though Member States have transposed the entire Services Directive, including this provision, national laws and regulations remain little specific on this point and lack any clear criteria which justifications are acceptable or not. As a result, customers do not only lack a clear framework at European level but when it comes to exercising their rights under national law, they have no clear way forward how the principle of non-discrimination is actually applicable and effective in practice.

Geo-blocking of customers based on their nationality or country of residence applies by definition only to transactions between nationals or residents of two different Member States. The initiative complies with the subsidiarity principle as due to the cross-border nature of the problem, it can be more effectively and efficiently tackled at EU level.

At a workshop on 18 February 2016, participants discussed the possible value added of an initiative at EU level. The business community stressed that barriers to cross-border transactions (cross-border parcel delivery problems, payment issues, applicable law, VAT) should be addressed in tandem with efforts to improve the effectiveness of the non-discrimination principle. Consumer representatives stressed that failing action opportunities to serve wider markets would continue to be lost and frustration amongst citizens interested in shopping cross-border would only increase.

3.2.What would be the added-value of action at EU-level?

Differences of treatment based on residence or nationality are experienced by citizens of all EU member states. In 27 out of 28 member states, more than 10% of consumers report having been geo-blocked in the last year (see annex 5). Although some differences of treatment can also apply within one Member State (with more favourable conditions for residents of a specific town or region), the Court of Justice underlined that a measure can be discriminatory as soon as it affects residents of other Member States, even if it does not affect only them. 73

The transposition of Article 20 of the Services Directive and the subsequent guidelines issued by the Commission have not led to better application and efficient and consistent enforcement. Member States authorities cannot rely on a sufficiently clear legal framework to ensure such enforcement, as Article 20 is not sufficiently specific. Even if they were to enforce Article 20 more strictly, this would lead to an uneven application of this provision, since it would probably be interpreted in a variety of ways.

This suggests that a better and stricter application of the non-discrimination principle, where that is justified and required in certain specific situations, cannot be sufficiently achieved by Member States acting alone. By reason of the scale and expected effects, this action can be better achieved at the EU level.

EU aAction would create equal rights and obligations for businesses and consumers across the Single market, thus simplifying the current legal framework and creating a level playing field, thus allowing both customers and traders to benefit from the internal market. It would give enforcement authorities the means to act effectively against discriminatory practices.

4.Objectives

4.1. General policy objectives

Better access to goods and services for customers in the Digital Single Market

Prevent unjustified discrimination of customers in the Single Market.

4.2. Specific policy objectives

The following objectives are cumulative.

-To improve transparency for customers by enabling access to websites/applications throughout the Single Market;

-To prevent unjustified differences of treatment in access to goods and services for customers throughout the Single Market;

-To improve public enforcement in relation to unjustified geo-blocking and other discriminations based on the place of residence or establishment or nationality.

-To increase legal certainty for business for cross-border transactions.

5.Policy options

5.0 Baseline scenario - no EU policy change

The base-line scenario implies implementing Article 20 of the Services Directive, the current provisions of the Consumer Rights Directive and the 2012 Guidance document on Article 20 of the Services Directive. It also assumes the adoption of the proposed Directive on certain aspects concerning contracts for the supply of digital content 74 , the Directive on certain aspects concerning contracts for the online and other distance sales of goods 75 , the revision of the Consumer Protection Cooperation Regulation (among other changes, including Article 20 of the Services Directive in the Annex of the CPC Regulation) and the proposed regulation to improve cross-border delivery of parcels. However, Article 20 of the Services Directive has proven to be ineffective and difficult to enforce by national authorities, mainly due to the broad list of possible justifications for differences in treatment. The inclusion of Article 20 within the scope of the CPC Regulation would be beneficial to ensure coordinated enforcement across the EU in B2C-situations. However, it will not improve legal certainty or cover enforcement in B2B-situations.

5.1.Option 1: Improving transparency and enforcement

In order to improve transparency on the reasons for unjustified geo-blocking and other types of discriminations, this option would be based on the following principles:

1) transparency in cases of justified restrictions for accessing the service, applied by a trader. A trader could for instance refer to unsatisfactory delivery conditions, need to register for VAT purposes in the country of the consumer, lack of knowledge about labelling rules and consumer laws applicable in the country where the consumer resides.

2) no obligation for a trader to explain price differences (if access to the best possible price in any EU member State is no longer restricted for consumers).

The trader would be given the choice between two ways of implementing the transparency principle:

1) As a principle, traders would to provide an explanation for justified differences of treatment upon request of the customer if such differences are not otherwise covered by the legal instrument. In that case, companies should react within a reasonable timeframe to be set in a legislative proposal.

2) In order to avoid dealing with constantly incoming requests from customers, the trader could also provide justifications upfront. Pursuant to the Consumer Rights Directive, traders are under the obligation to inform consumers about only possible delivery or payment restrictions; a general disclaimer or an entry in the general sales conditions could be provided on the website about the justifications for access restrictions applied on customers from other Member States (such as cost of registering for VAT purposes in another country).

In order to limit administrative burden, the company will be obliged to provide a reply only in the language(s) announced in the general terms and conditions. If a Spanish consumer intends to buy a product on a Swedish website, written only in Swedish, he should not expect to receive a reply in Spanish.

In order to ensure enforcement, the legislative act resulting from the proposal that is now being considered would be added in the annex to the Regulation on Consumer Protection Cooperation. This would not apply to B2B issues for which Member States could designate enforcement authorities, if they consider it appropriate.

This option would apply to all companies including SMEs and micro-enterprises.

For the transparency obligation there would be no difference in treatment depending on whether the explanation is requested by a consumer or by a business (B2B), but in line with above the enforcement mechanism would vary depending on the situation. In a B2C context, enforcement would include strengthened co-ordination between national authorities under the CPC Regulation. In a B2B context, only the general enforcement rules would apply.

5.2.Option 2: Banning blocking of access

This option would make the denial of access to a website based on nationality or place of residence or establishment illegal, with a very limited set of exceptions (only if required by EU law or national law in accordance with EU law, e.g. e-commerce Directive). Rerouting would be subject to the customer's consent. Even after the expression of consent to rerouting, the original website targeted by a customer should always remain accessible and the customer should be informed about this at the time of giving his/her consent to rerouting. If blocking of access of automatic rerouting is required by EU law or national law in accordance with EU law, the customer should be informed about it.

In order to ensure enforcement, the legislative act resulting from the proposal that is now being considered would be added in the annex to the Regulation on Consumer Protection Cooperation. This would not apply to B2B issues for which Member States, in line with the principle of subsidiarity, could designate enforcement authorities, if they consider it appropriate.

This option would apply to all companies including SMEs and micro enterprises.

The option would not entail a difference in treatment based on if the goods/services are purchases by a consumer (B2C) or by a business (B2B). 76

It would also apply to websites of third country traders targeting the European Union and aiming at providing services within the Single Market.

5.3.Option 3: Option 2 + equal treatment in specific situations ("shop like a local")

This option aims at ensuring non-discrimination on the basis of nationality or place of residence or establishment in well-defined situations where no or only minimal additional costs associated to cross-border sales are created, either because the situation is akin to a national sale or because the trader has already taken the decision to incur the costs of selling abroad independently of the purchase request, or at least been able to take account of the possibility of such costs occuring (e.g. a large company operating in several Member States). Traders could recoup incurred additional costs in transparent manner to the extent to which they are objectively justified.

The option covers online and offline sales of tangible goods without obligation on the trader to organise cross-border delivery as well as services such as online digital services, tourism services, leisure and accommodation services.

Traders would not be obliged to contract under this option. For example, if a product is "out of stock" or if a certain service is unavailable at a specific point of time, traders can refuse to contract. However, traders should provide equal treatment in different steps of the sale process for foreign and for local customers.

a) Sales of goods

For sales of tangible goods, non-residents should have the possibility to contract under the same conditions as residents of the country where the trader is operating. Given the objective constraints to cross-border delivery, the trader would not be obliged to deliver the ordered tangible good to the country of the customer but the customer would be provided with the same delivery options as national customers, residing in a country in which the trader is operating. This includes delivery to an address provided by the customer within that country, as well as all other delivery options (e.g. pickup at a collection point, delivery through an intermediary, etc.) available for the consumers residing in the country. Tangible goods include inter alia books (on paper). This option would not preclude traders from applying different prices to customers in certain territories in so far as they are required to do so under the laws of Member States in accordance with Union law (e.g. book price regulations).

The customer could only rely on the after sales service in the country where the good was delivered by the trader.

In B2B situations in particular, this should exclude transactions involving products intended for resale, in order to avoid interfering with the competition law framework. For businesses as end-users, VAT is due in principle in the country where the goods are transported to.

It would also apply to third country traders already offering delivery or pick-up in at least one Member State of the European Union. If a Brazilian company offers to deliver a product to Portugal, a consumer residing in France should be able to pick it up in Portugal.

b) Sales of electronically supplied services

Customers should be entitled to purchase electronically supplied services such as cloud services, data warehousing, website hosting, remote system administration, installation of filters, firewalls, banner-blockers etc., from any trader. 77 In these situations, physical delivery does not take place. There is therefore no objective reason to advance delivery problems against foreign customers. Traders could use the VAT mini one-stop shop for VAT clearance, but would still need to find out the applicable rates in the country of destination. Traders exempted from paying VAT under national special schemes for small enterprises are not subject to the non-discrimination obligation. Traders would not be obliged to contract but only justifications unrelated to nationality or place of residence or establishment could be invoked (out of stock, suspicion of fraud based on individual circumstances etc.). In order for traders to adapt to the new situation, this part of the option will be phased in progressively.

B2B transactions for such online services are taxed at the customer's place of establishment, pursuant to VAT rules. The trader does not charge VAT; it is the customer who pays VAT on the services received at the applicable rate in their country (using the reverse charge procedure).

It would also apply to third country traders providing services in the EU.

c) Sales of electronically delivered non-audio-visual content services

Similar considerations could apply to digital content services which give access to non-audiovisual content subject to copyright protection such as music, e-books, software and games delivered online, provided that the trader has the required rights for the relevant territories. However, their potential inclusion within the scope of the non-discrimination obligation has to be analysed further.

d) Supply of goods and services in the premises of the trader or in a physical location where the trader conducts its business, outside the customer's home Member State

Traders would not be allowed to refuse to sell or differentiate prices at the same point of sale/website between customers based on the place of residence or establishment or nationality if the service is used or consumed by the customer at the place of establishment of the trader or the location where the trader provide the services outside the home Member State of the customer (e.g. buying tickets to a concert, sales of products (retail), visiting a leisure park, rental of summer accommodation, car hire, etc.). Traders would still have the freedom to set different prices across different websites/applications (including country-specific websites) but customers would be free to choose from which country website they wish to buy.

In the case of online sales, it will be up to the trader to ensure compliance so that customers can place orders on the online interface, e.g. through amendments to the lay-out of the web site if necessary or providing for taking orders by an alternative means such as e-mail.

Traders would remain free to deny sales on grounds not related to the nationality or residence of the customers, for example running out of stock or not selling below a certain age.

B2B services in respect of admission to cultural, artistic, sporting, scientific, educational, entertainment and similar events, restaurant and catering services, short-term hire of means of transport will be taxed at the place where those events actually take place.

It would also apply to third country traders providing their services within the European Union. A US company selling tickets for a large music festival in Europe would be subject to this provision. An Australian car hire company offering rental of a car in Australia to European consumers would not be subject to it.

Non-discrimination in the means of payment.

By 2018, all Member States must transpose the second Payment Services Directive. 78 As a consequence, traders will have no reason to refuse transactions in the EEA or impose additional conditions based on the country of origin of the payment provided that the retailer adopts strong customer authentication and the payments are in a currency that the trader accepts. The present proposal would not force any trader to accept any specific means of payment, but it would imply that traders who can request strong customer authentication by the payer cannot refuse payment because of the country of origin of the payment within the EEA. Traders would for instance not be allowed to reject consumer credit or debit cards issued in another country if they accept the same debit or credit card issued to customers in their country. They could, however, charge any extra costs occurred for transactions with cards of which the interchange fees have not been capped 79 .

Restraints on passive sales

As a complementary element, the option would also clarify the situation as regards vertical restraints. It would declare agreements imposing on traders obligations, in respect of passive sales, to act in violation of the proposed Regulation automatically void. This would in some case go beyond the current competition law framework and is necessary in order to avoid the rules of the proposed Regulation from being circumvented through contractual arrangements.

Under the current EU competition law framework with regard to vertical agreements passive sales restrictions (i.e. provisions in agreements which restrict a retailer from responding to unsolicited requests from individual customers) are generally considered restrictive of competition pursuant to Article 101(1) TFEU and cannot normally be exempted under Article 101(3) TFEU. As a result, the vast majority of passive sales restrictions in vertical agreements are already currently unlawful and therefore automatically void (Article 101(2) TFEU under EU competition law. Only in very exceptional circumstances can passive sales be lawfully restricted in an agreement under EU competition law. 80 The option would only bring a change for those exceptional cases where passive sales restrictions can lawfully be implemented in agreements in the scope of the application of this option. Those agreements would now be declared to be void. This change would however be justified under internal market rules to achieve the objectives of the legislation.

The legislative proposal does not affect active sales restrictions governed by competition law (in particular by the Vertical Block Exemption Regulation) 81 , according to which the supplier can designate exclusive territories to one or more retailers under certain circumstances. These exclusive distribution agreements may lead to efficiencies, especially where investments by the distributors are required to protect or build up the brand image.

This option would apply to all companies including SMEs and micro enterprises other than companies selling electronically supplied services which are exempted from paying VAT under the national special schemes for small enterprises..

5.4.Option 4: Option 2 + list of justifications deemed not to be based on objective criteria

In some cases differences of treatment based on residence or nationality can be justified by objective criteria. This option, however, would set out arguments that cannot be accepted as being based on objective criteria for traders to justify a difference in treatment, for example:

"Lack of delivery options" in the following cases:

owhen the customer arranges for delivery or takes over the risk of delivery

ofor parcels up to 20 kg

oif the trader has a physical branch or subsidiary selling the same products in the country of residence of the customer.

"Cost for shipment (for refusals to deliver/to sell)" if the customer is willing to pay the additional cost linked to delivery.

"Cost of adapting to mandatory provisions of the law of the country where the consumer has its habitual residence" when the trader does not direct its activities to this country

"Cost or difficulty of bank transfers" in the Single Euro Payment Area.

"Difficulty to recover debts from service recipients based in other countries", when the service provider can use the European Payment Order Procedure and the European Small Claims Procedure.

"Risk of fraud for electronic means of payment" if the transaction can be authenticated by using strong customer authentication pursuant to the Payment Services Directive.

The option would cover all services, on-line/off-line sales (including delivery), as well as on-line services that has a physical dimension.

This option would apply to all companies including SMEs but exempting micro enterprises.

The option would cover all purchases by customers and thus not entail a difference in treatment based on if the goods/services are purchases by a consumer (B2C) or by a business (B2B). 82 In B2B situations in particular, this should expressly exclude transactions involving products intended for resale, in order to avoid circumvention of the competition law framework.

The justifications that could or could not be invoked would depend on whether the customer is a individual consumer or a company, since the applicable VAT legislation is different in both cases, and depending on whether the transaction concerns a provision of service, a distance sale of goods, whether it involves delivery, etc.

5.5.Option 5: Option 3 + obligation to serve customers across the EU

In a Single Market with few remaining barriers for cross-border trade, customers have an expectation to be served across borders. Under this option, companies that already deliver within their own country could not, as a general rule, refuse to deliver to another Member State. They could, however, charge any extra delivery costs. This option goes thus further than option 3, where the trader would not be obliged to deliver the ordered tangible good to the country of the customer but where same delivery options as national customers would be given, for example delivery to an address in the country of the trader.

Responding to customer concerns about the availability for after-sale services, 2this option would also create an obligation for traders to provide similar after-sales services for a given product/service in all EU Member States in which the traders is present through its own branches and/or subsidiaries and through which the trader already offers the same product/services for which after-sales services are sought, regardless of the EU Member State in which the customer originally purchased the product/service. The trader in the country of the customer would however only be obliged to provide the minimum harmonised after-sale services as foreseen under EU law. For instance, if a Polish citizen living in Germany buys a product in Poland and the trader has a branch or a subsidiary in Germany where the same product is offered, the Polish customer should not have to go back to Poland to benefit from after-sales services. In order to provide for level-playing field for companies irrespective of their business model, the concept of "trader" should include all branches or subsidiaries of a given company.

Given the burdens that this option may create, SMEs and micro enterprises should in any event be exempted from this option.

The option would cover all purchases by customers as end-users and thus not entail a difference in treatment based on if the goods/services are purchased by a consumer (B2C) or by a business (B2B). 83

5.6. Other policy options

As mentioned in section 2.6 above ("Evaluation of the existing policy framework"), the published guidelines based on existing legislation did not succeed in ensuring consistency in the application of the non-discrimination clause of the Services Directive across Member States. Based on past experience, it is possible to conclude that publishing other kind of 'soft-law', such as additional guidelines or a non-binding recommendation would not be effective and not make any difference.

As mentioned in section 2.6, promoting voluntary agreements of traders/providers failed to enforce a legally binding non-discriminatory provision, although this took place in a context where the Commission managed to attract substantial media coverage, with large international market players. Such an approach does not seem likely to be effective or appropriate if it is to be extended to all players. The Commission would also be seen as intervening directly in individual enforcement cases, which are the responsibility of national authorities. Finally, it does not seem appropriate to make the respect of the fundamental principle of non-discrimination enshrined in the treaty dependent on “voluntary” agreements.

Amending the Services Directive was also considered inappropriate. The Directive is the result of very complex negotiations resulting in a delicate political balance, and both the European Parliament and the vast majority of Member States already indicated that they were opposed to any amendment to it.

6.Analysis of impacts

6.1. Option 0 – baseline scenario- see section 2.5

6.2. Option 1. Improving transparency and enforcement

Impact on the policy objectives

Requesting companies to provide an explanation to customers why they refuse them access to a particular website or redirect them automatically to another one or informing customers that they will not sell cross-border could discourage them from doing so, because it may generate negative publicity for their business model. However, the mechanism is very indirect, and by providing a valid justification, companies can avoid negative publicity. Requesting companies to provide a justification upon request would be meaningful for the concerned customer, but not necessarily induce – directly or indirectly - companies to abandon geoblocking. Thus, the positive impact could be very small. Similarly, improved enforcement of an unclear provision is unlikely to ensure non-discrimination, since enforcement authorities would not know how to enforce the rule.

On the other hand, the public consultation showed that more than eight out of ten of respondents from the consumer group are in favour of companies having the obligation to explain - either before the transaction or upon request - the reasons for the difference in treatment of customers based on residence/nationality. Between a third and half of the business respondents also agree to more transparency as a policy option in the sense that companies should explain their different treatment, either before or after the transaction.

Economic Impacts

Operating costs and conduct of business

The option would entail administrative costs.

For the explanation of access refusal or automatic redirecting, this would concern only a small share of total websites (around 2%, see annex 6), which however have a disproportionately large share of cross-border traffic (around 7.5%, see annex 6) and who are already present on several national markets through a multiplicity of websites, and already have carried out significant IT investments for that purpose.

On the other hand, the cost of providing explanations to customers if restrictions are justified would concern a very large share of companies. It could be done either upon request, or provided upfront. The cost for handling a request can be estimated to €2.05 84 . For a standard company which does not target foreign consumers and does not aim at delivering cross-border, one request a week seems a high assumption, which would result in a cost of up to 100 euros per year. Some providers would find it less profitable and more time-consuming to edit a cost estimate and an invoice than to add the information to the legal notice of the website, if it does not involve development or coding costs.

In case of billing, assuming that adding a few lines to a webpage (without coding) would involve at most 4 hours of work, the low estimate of the one-off cost for a simple operation would be around EUR 100 85 . In relation with the EU cookie notification policy, which involved higher development costs, a higher estimate of adapting of websites was of EUR 900 86 - this would then include the provision of an explanation when re-routing (as well the ban on the block of access – as under Option 2).

For the explanation of discrimination or refusal to sell/deliver when requested, this would represent a cost which is proportional to the number of websites operated. For SMEs and micro-enterprises having to explain why they will not sell/deliver to other Member States would constitute an administrative burden, the size of which will depend on the number of inquiries received.

Impact on the Single Market

Given the indirect mechanism by which this option would reduce discrimination, the impact of this option in addressing the problem will be very small. It will therefore not increase market integration or trade flows to any significant extent. However, transparency may reduce customer frustration who want to buy everywhere in the Single Market but not necessarily increase their satisfaction. Transparency, however, matters in a context where trust in the Internet relies on appropriate information of customers.

Competitiveness of business

This option is not expected to affect competitiveness of European businesses, because it applies to European and non-European companies alike.

Impacts on SMEs and microenterprises

SMEs and microenterprises would face administrative costs for providing the information (see above). These costs would occur to a large share of these enterprises and would be relatively more important for SMEs and especially microenterprises than for large companies, thus putting them at a disadvantage. Nevertheless, they could provide information in their language. SMEs, and especially microenterprises, operating in a purely national environment, with a limited number of products in catalogue, are unlikely to run multiple websites. The possibility given to companies to choose between an upfront explanation and explanation upon request would enable them to opt for the less invasive transparency measure, in line with their business model. In line with the Consumer Rights Directive, a small company selling one product on its website already has the obligation to inform consumers about delivery or payment restrictions. It could adapt its website to the new obligation by explaining, for instance, that the company does not sell in a given Member State because of the costs linked to registration for VAT purposes there. As explained above the cost of such a change would depend on whether or not the website is managed internally.

In addition, the trader would remain totally free, as it is today, to reply to requests in any language it wants. This option would not impose any translation obligation.

During the stakeholder consultation, more individual business expected a positive than a negative impact, but a clear majority of business associations expected a negative impact.

In so far as customer willingness to purchase cross-border increases thanks to better information, SMEs which are prepared sell cross-border might profit from increased sales. However, this effect is likely to be very small. Specific enforcement provisions will enable SMEs as purchasers to seek redress in case their rights have been violated.

Consumers and households

Consumers would benefit by being better informed about reasons for discrimination, assuming the reasons given by traders are correct. Nevertheless, providing information in relation to delivery restrictions and means of payment is already mandatory under the Consumer Rights Directive.

Moreover, improved enforcement mechanisms will also benefit consumers. First they will have an incentive to claim respect for their rights towards traders, knowing that they could be effectively enforced. Second, they will be encouraged to lodge a complaint, knowing precisely which the enforcement authorities (endowed with the necessary powers to act) are.

A very clear majority of consumers and of consumer associations who responded to the stakeholder consultation believe the impact of transparency requirements to be positive or very positive. One of the main conclusions of the workshop with stakeholders organised on 18/2/2016 in Amsterdam has been that addressing information shortcomings in isolation would not address the underlying problems.

Macroeconomic impact

This option does not have any macroeconomic impact.

Impact on Member States

For enforcement purposes, no new administrative arrangements are required, as the CPC network is already active. There might be an increased workload, but there is no data available to estimate this increased workload. However, the information on reasons for blocking access and rerouting concerns a very small share of websites. The additional costs due to the present proposal is, therefore, likely to be small, including with respect to third-country traders.

Social Impacts

This option does not have any social impact.

Environmental Impacts

This option does not have any environmental impact.

6.3. Option 2 Banning of blocking access to websites

Impact on the policy objectives

Requesting companies to refrain from blocking access and automatic rerouting will enable customers to compare offers on different websites of the same seller. They may well resent it if they are forced to use a website offering less advantageous conditions. As a result, some companies may start to allow visitors from other Member States to buy from any of their websites. Some may even offer identical conditions on all websites if there is a business case to do so. However, the mechanism by which discrimination in terms of access to goods and services is reduced is very indirect since most companies will likely continue discriminating when selling.

In the public consultation, four out of five respondents from the consumer and consumer organisations group agree that the forms of geoblocking such as blocking access to websites create significant obstacles to the single market, and a large majority of the respondents from the consumers and consumers' organisations group favour a ban on discriminatory blocking of access to websites. Around half of the business respondents agree.

Economic Impacts

Operating costs and conduct of business

Companies currently blocking access and practicing automatic rerouting would have either to abandon automatic rerouting or to change their websites to request approval of the customer for rerouting (except for the small number of websites for which geoblocking remains permitted because of legal obligations, such as sales restriction on Swedish tobacco "snus").

Abandoning automatic rerouting (basically, disabling the IP filtering) is inexpensive. Requesting approval from Website visitors before rerouting them entails some compliance costs but this would concern only a small share of total websites (around 2%, see above), which however are typically larger websites – they have a disproportionately large share of cross-border traffic (around 7.5%) and they are only concerned if they have at least two different websites addressing two different Member States (see annex 6) and already have carried out significant IT investments for that purpose.

Informing customes when they are rerouted for reasons of EU law or national law in accordance with EU law will concern only exceptional cases.

Impacts on the Single Market

Blocking access to websites and automatic rerouting is practiced only by 2% of websites, so even if all of them disabled blocking and accepted cross-border customers, the impact on trade would be very small. If those websites accepted cross-border customers with the same ratio as websites overall (49.2%), the impact would be smaller still (1% increase in successful cross-border shopping attempts). Since these websites are distinguished from websites in general by having actively implemented policies to segment markets already at the access level, one can assume that the actual share of such websites allowing foreign orders would be even smaller.

Competitiveness of business

This option will not affect competitiveness of European businesses, because it applies to European and non-European companies alike.

Impacts on SMEs and microenterprises

The fact that companies practicing automatic rerouting have about 4 times average cross-border traffic indicates that these are usually larger companies. Equally, from the eCommerce Sector Inquiry it emerged that retailers with a large turn-over geoblock more often than smaller retailers, since they frequently have several websites in several Member States.

Only very few of the websites practising automatic rerouting belong to SMEs, and virtually none to micro enterprises – automatic rerouting only makes sense if the company operates several websites directing sales efforts to at least two distinct national markets. Therefore the impact on micro enterprises is likely negligible.

During the public consultation a majority of businesses expected the impact of banning discriminatory blocking to be positive, but the impact of a ban on automatic rerouting to be negative, despite the fact that less than only 2% of businesses actually reroute automatically. A majority of business association expected a negative impact of both.

Consumers and households

Consumers will be able to access previously blocked websites, providing them with better information on goods and services as well as prices offered in other Member States. However, they will not necessarily be able to actually purchase these goods and services at those prices. This option is therefore of limited value to customers. It will, however, be a necessary first step for further-reaching rules on the access to goods and services.

During the public consultation an overwhelming majority of consumers and of consumer association estimated the impact of banning discriminatory blocking to be positive, and a clear majority thought the same of requesting prior approval before automatic rerouting.

Macroeconomic impact

Given the small impact on the Digital Single Market, this option does not have any measurable macroeconomic impact.

Impact on Member States

None

Social Impacts

Given the small impact on the Digital Single Market, this option does not have any measurable social impact.

Environmental Impacts

Given the small impact on the Digital Single Market, This option does not have any measurable environmental impact.

6.4. Option 3 Option 2 + equal treatment in specific situations ("shop like a local")

Impact on the policy objectives

In addition to the impacts of Option 2, for the situations concerned, geo-discrimination is likely to be reduced, although companies will only need to deliver physical goods within their normal delivery area and will keep the possibility to adapt their means of communication to specific national audiences (for instance with announcing a special offer only on one language version of their website). The specific situations are as follows: a) sale of goods when the customer picks up the good himself or arranges delivery to a national address, b) sale of electronically supplied services and c) supply of services in another country than that in which the customer has its residence,

In the public consultation, a clear majority of consumer respondents favoured rules requiring traders to accept cross-border transactions from users throughout the EU, either under conditions reflecting additional costs or in cases where users arrange the delivery themselves. Among companies, opinions are rather divided, although business associations are generally supportive of the aim to prohibit unjustified geo-blocking while at the same time highlighting the importance of contractual freedom.

This option would tend to increase legal certainty by setting out in unambiguous terms a prohibition of different treatment based on residence of nationality in three clearly identified cases. As such it would also enable national authorities to enforce the directly effective provisions, in contrast with current legislation.

Economic Impacts

Operating costs and conduct of business

For those traders which would need to implement changes in order to allow foreigners to purchase via their websites under certain circumstances, there would be the compliance costs of changing the website. Nevertheless, if they do not expect large cross-border sales volumes, companies could ensure that they accept cross-border orders temporarily by other means, such as e-mail. While in this case a single transaction is more labour-intensive, any cost would not occur without a direct benefit (a sale). In the medium-to-long run, as websites need to be regularly modernised, the acceptance of cross-border purchases can be introduced as an ancillary of a regular modernisation.

It is only when the customer organises the transport to its Member State of residence independently from the supplier that goods delivered are taxed at the origin. The assessment of this distinction has to be made with the view of avoiding the possibility of creating any tax abuse arrangements. Thus this option involves the seller delivering to an address within its normal delivery range, without interfering in any way in the transport of the good to the residence of the customer, not even by providing a list of intermediaries which could do so on behalf of the customer. The cross-border transport is entirely left for the customer to arrange.

Other compliance costs would only occur at a very limited level.

a) In the case of sale of tangible goods, for activities not directed to the country of residence of the customer, according to the Rome I Regulation the trader retains total freedom of choice over the law which should apply to the contract – and de facto, in such cases, the trader indeed usually chooses the law applicable in its country of establishment – and does not need to abide by any more stringent mandatory consumer contract rules that may apply in the consumer's Member State. In any case, mere compliance with this instrument will not be construed as implying that a trader directs his activities to the Member State where the consumer has his habitual residence or domicile.

For activities directed to the country of the customer, the trader has already accepted that it has to comply with the mandatory consumer contract rules applicable in the consumer's Member State. Not covering companies directing their activities to other Member States might have the side-effect of favouring larger operators providing their services across the Single Market through a single establishment, while subjecting operators organised with independent subsidiaries directing only their activities to their own Member States of establishment. The same rules regarding directing of activities and applicable law and jurisdiction apply in (b), (c) and (d) below.

Pursuant to the VAT directive 87 , VAT for goods is due in the Member State into which the goods are shipped; since there is no cross-border delivery (goods are only shipped within the area where the traders already deliver), and only when the supplier does not intervene directly or indirectly in the organisation of the transport of goods to the customer, no additional VAT registration will be necessary. In terms of shipping costs, this option would not require traders to ship beyond their normal delivery area, as they would do for local consumers. Additional labelling costs would not occur when the trader has not expressly chosen to supply products to EU consumers or other end-users.

b) For the sale of electronically supplied services covered by the VAT Mini-One-Stop-Shop, VAT is due in the country of the service recipient, but registration is done automatically through the one-stop-shop. It is for the supplier to decide whether he opts to account for taxes due in other Member States through the one-stop-shop or whether he prefers to register in each member State. The trader would need to identify the VAT rate in the customer's country in the case of digital sales, but under the new VAT proposal, this information will be easily available. For these services, application of the new rules would only start on 1 July 2018, giving companies sufficient time to familiarise themselves with the available information sources.

However, the facilitation provided by the VAT Mini-One-Stop-Shop is only helpful if the company is already registered for VAT. Yet under current legislation, companies falling under certain thresholds set by Member States may be exempted from VAT. For those companies, compliance with this option would create burdensome obligations. They should therefore be exempted from the obligation of equal treatment.

For the sale of electronically delivered non-audiovisual content services subject to copyright protection, such as music, e-books, software and games, if the trader has the required rights for the relevant territories, the considerations above would apply. However, given ongoing developments in the market, e.g. with respect to multi-territorial licensing, for these services the potential extension of the non-discrimination obligation requires further assessment.

c) For supply of the service in another country than that in which the customer has its residence (e.g. a hotel booking or a car rental), this would not entail any extra compliance cost linked to taxation or applicable legislation, and the VAT Directives provide that normally VAT is to be paid in the country where the service is supplied.

Traders could recoup incurred additional costs, such as relating to payment systems or other costs, in a transparent manner to the extent to which they are objectively justified.

Impact on the Single Market

The proposal to allow customers access to more choice and affordable offers should motivate consumers to shop cross-border. Price arbitrage will nevertheless be constrained by delivery costs in the case of tangible goods.

For example, an analysis of distance sale of electronic goods 88 , one of the most popular online sales categories, estimates that net effect of lifting geoblocking restrictions by companies on cross-border sales of these goods would be increase of about 1.1% of the total market size (total sales). Most of that additional trade would come at the expense of a reduction in domestic purchases, mostly in domestic offline purchases. The net trade expansion effect on these goods is estimated at 0.4% only. Nevertheless, one has to keep in mind that the baseline will be affected by the effectiveness of the related regulatory initiatives listed in section 1.

According to the previously cited survey, 32% of Western European 89 international online buyers have used a forwarding address to get delivered 90 . 42% of them did so when no shipping to the country of the customer was available, and 9% did so while travelling in the country of the seller. With 35% of citizens travelling abroad every year and 40% of citizens living in border regions, the ability to purchase without delivery to the customer's home address is likely to have a significant impact.

This option is likely to affect mostly retail of consumer goods and encourage consumers to shop cross-border, whether on a tourist, family or business trip, and in cross-border regions. It will also encourage consumers to check offers on different websites. Anecdotal evidence form companies showed that some of those who changed their practices in the leisure sector noticed an increase of cross-border sales.

Under this option, obligations on traders in agreements that violate the Regulation would be automatically void. The overall impact of this rule would not significantly affect market participants, as most passive sales restrictions are already at present prohibited and automatically void under EU competition law. In the very exceptional circumstances when they are not, it would be justified under internal market rules to go further and prohibit them to achieve the objectives of the proposed Regulation to ensure the proper functioning of the internal market and avoid circumvention.

Active sales restrictions would however not be covered by this rule, because of the major impact that this would have on exclusive distribution agreements and investment incentives of distributors. Exclusive distribution may lead to efficiencies, especially where investments by the distributors are required to protect or build up the brand image. In order to persuade a local distributor to make investments, it may be necessary to provide some territorial protection to the distributor in the form of active sales restrictions on other distributors, so that the former can recoup these investments. According to the findings of the e-commerce sector inquiry, roughly half of the manufacturers/suppliers make use of exclusivity clauses at least for one product, at least in one product category.

Competitiveness of business

Lifting geo-blocking restrictions should boost trade as on-line sellers will receive more sales orders from foreign customers. Offline price competition can be expected to increase. In the case of electronic goods, for which particularly good data is available, both online and offline prices are forecast to decline when sellers adjust to the new market conditions when geoblocking restrictions are lifted, leading to additional sales.

The net effect of lifting geoblocking restrictions by companies for tangible goods on total sales is likely to be positive. Based on an extension of the analysis of the electronics sector, producer surplus (profits) could increase by 1.0% on average across Member States, as revenue losses from those companies abandoning market segmentation to maintain higher prices in some markets are more than compensated by the additional revenues for companies which previously had not sold to customers resident in another member state.

Firms that buy goods and services online as end-users would benefit from lifting geoblocking restrictions in the same way as consumers, from lower prices and a wider variety of products.

The international competitiveness of companies will not be affected, since the obligations apply to all companies selling to EU customers, whether established in the EU or not.

Impact on SMEs and microenterprises

SMEs which already target the country of the customer will not incur any additional costs. SMEs which do not target other countries would incur only minimal additional costs. This would include virtually all microenterprises and most SMEs.

For tangible goods, by delivering within their country (and only when the supplier does not intervene directly or indirectly in the organisation of the transport of goods to the customer), they do not need to register for VAT, and no additional delivery costs occur. The risk of fraud will not be any higher following the implementation of the Payment Services Directive 2.

For digital services, the same would apply, except that they would need to pay foreign VAT, but not register for it as the Mini One-Stop Shop applies. They would need to identify the applicable foreign rate, which is already easily available and will be made even more understandable under the new VAT proposal. From the evaluation of the VAT MOSS, businesses seemed overall pleased with its introduction and level of functionality and estimated to contribute to reduced costs for business by 500 million in 2015. However, the facilitation provided by the VAT Mini-One-Stop-Shop is only helpful if the company is already registered for VAT. Yet under current legislation, companies falling under certain thresholds set by Member States may be exempted from VAT. For those companies, compliance with this option would create burdensome obligations. They should therefore be exempted from the obligation of equal treatment.

For digital non-audio visual content services which give access to content subject to copyright protection, the potential extension of the non-discrimination obligation to them requires further assessment. In any case, companies would not be obliged to acquire additional licences, but would potentially be subject to the provisions only if they already have the licenses for the territories in question.

For services provided in the premises or at the location of the trader outside the customer's Member State (such as hotels, leisure activities, festivals, car hire etc…), SMEs will not face any burden, since the place of residence or establishment of the customer cannot create any additional cost: the applicable VAT is the VAT of the country where the service is supplied, and this type of services do not involve any cross-border delivery of products. Larger traders operating a number of different websites and applying source market pricing strategies in order to maximise their profits by charging more to consumers from countries where there is more demand (or less competition) may be more affected than SMEs or micro-companies traditionally targeting a single country or without the means to develop such complex pricing strategies.

In all of the scenarios mentioned above, if the trader does not direct its activities to consumers in a Member State, according to the Rome I Regulation the trader can decide to apply the law applicable in its place of establishment and it does not need to abide by any more stringent consumer contract rules that may be applicable in the consumer's Member State.

This option only obliges traders not to treat foreign customers differently from national customer, just like they would not treat them differently if they walked into their shop.

During the public consultation, slightly more companies thought the impact to be negative than positive. Company associations had a more negative expectation.

Consumers and households

If current unjustified geoblocking practices cease, customers will have access to the catalogues of distance sales of goods catalogues in all Member States, at the same prices as domestic online shoppers in these countries (with the additional transport and handling costs to ship the goods themselves or via a third party to their country of residence, if they so wish). As a result, customers will be able to benefit from lower prices in other Member States and buy a larger variety or quality of goods available in the EU.

As mentioned above, both online and offline prices can be expected to decrease which increase consumers' purchasing power and real expenditures. In the previously cited example of the electronic goods sector, price decreases are estimated at -0.5% offline and -0.6% online, on average across the EU. The variety effect will further increase consumer welfare because they will find more appealing products than those available in the home market. Moreover, increased competition should spur innovation.

Based on the analysis of the electronics sector, the total EU28 consumer surplus increase from online purchases of tangible goods could amount to 0.8% on average for the EU28, with variations across Member States. Smaller Member States would be expected to gain more (up to 15%) because the variety of goods locally available online is more limited. The impact on consumers in larger EU economies would be more limited because larger markets offer more product variety.

Similarly, regarding electronically supplied services and services at the location of the trader outside the customer's Member State, the abolition of geoblocking will allow foreign customers access to services in all Member States, at the same prices as domestic online shoppers in these countries. The major difference compared to goods is that in these cases there are no transport costs to limit price arbitrage. Generalised access to all offers could in theory lead to progressive convergence of prices in excess of transport costs. However, customers also face information costs resulting from language and the fact that most price comparison sites are national. In fact, the following empirical evidence suggests that price arbitrage in practice might not lead to uniformisation of prices, even where transport costs are minimal.

Indeed, the existing situation in European cross-border regions points to existing large price differentials, which have not been affected by the fact that consumers regularly make use of the freedoms of the single market. In 2015, a study analysing price differentials between France, Belgium and Luxembourg pointed to a 13% price difference for the same consumer basket in large supermarkets of the same brand. 91 Already in 2009, the European Consumer Centres for France and Germany had noted very substantial price differences for kitchen elements (up to 50%) sold on both side of the border in shops of the same company. 92 These figures show that in situations characterised by a high level of cross-border mobility of customers, no real price harmonisation took place.

Partly, this is due to the fact that retailers cannot in practice have very different national prices on-line and offline. Therefore, national pricing strategies of companies are unlikely to totally change. A non-discrimination obligation would therefore not be expected to lead to price harmonisation and general increase in prices in low-price countries, Nevertheless, the Internet significantly reduces information costs, and price comparison sites will expose differences more prominently than before, enabling potential cross-border shoppers to take advantage of remaining price differentials.

During the public consultation, a clear majority of consumers and a majority of consumer associations estimated the impact of an option allowing customer to organise delivery themselves to be positive.

Macroeconomic impact

Increased cross-border trade after lifting discriminatory restrictions will tend to reduce prices in the domestic retail sector. This would benefit other sectors of the economy because it becomes cheaper to distribute goods and services through the retail sector. The overall decline in retail prices would increase consumers' purchasing power. The current levels of cross-border e-commerce already increase real household expenditure by about 1% compared to a situation without cross-border e-commerce 93 . As a result, the overall impact of the current level of cross-border e-commerce on total output (GDP) is estimated at around +0.14%, compared to a situation with no cross-border e-commerce 94 . The magnitude of the overall macro-economic impact varies across Member States, from virtually zero to +0.25% in countries that are in the vanguard of e-commerce development. Lifting geoblocking restrictions will amplify this positive impact, but a precise estimate of the macroeconomic impact of this option is not available.

Impact on Member States

For enforcement purposes, no new administrative arrangements are required beyond those mentioned in option 1. There might be an increased workload; however, there is no data available to estimate this increased workload. There were roughly 1500 complaints over 8 years (2007-2015) which have come to the Commission's attention, but many issues are under-reported.. The specificity of the rules and the limited scope of this proposal will contribute to a speedy resolution of complaints.

Whilst the possibility of a consumer arbitrage based on different VAT rates across countries cannot be completely ruled out, it is unlikely to lead to significant impacts. In so far as consumers travel themselves to collect the tangible good ordered online (as tourists, business travellers or in border regions), the impact is decisively limited by the travel costs, just as in offline transactions. If consumers use an intermediary to forward the tangible good cross-border, there are two cases. If the seller is in any way involved, VAT of the destination country is due and no VAT arbitrage is possible. If the seller is not at the slightest involved, VAT arbitrage could be possible. However, the incurred additional costs (i.e. additional shipping costs plus handling fees for the intermediary) and the complications (finding an intermediary without any assistance from the seller and engaging in a forwarding contract) are in many cases too high to be offset by the variation in VAT rates. Systematic non-compliance by businesses, e.g. by intervening in the delivery yet applying national tax rates would be fairly easy to detect and prove, since the information would be available through the company's website.

In some border areas of some Member States direct employment in retail services for certain goods particularly prone to cross-border ordering with pick-up in the country of the trader could be affected to a limited extent. Nevertheless, any savings made by the buyers due to lower prices would increase demand of other goods and services in the same area, at least partially offsetting this already limited effect.

Social Impacts

This option does not have any significant social impact

Environmental Impacts

This option does not have any significant environmental impact

6.5. Option 4 Option 2 + list of justifications deemed not to be based on objective criteria

Impact on the policy objectives

In the short term this option could create more legal certainty for traders and prevent customers from being discriminated; i.e. treated differently because of their nationality or place of residence or establishment on arbitrary grounds. However; it is doubtful whether the option is futureproof enough to impact the discipline of companies and to invite them to be more open to consumers shopping cross-border. Legislation is underway in many of the areas concerned. The option would have to be aligned with that legislation. More importantly it would not cover situations which might occur in the future. Companies frequently consider that "what is not explicitly prohibited is allowed", so in order to be comprehensive the list would need to be constantly reviewed, which politically and in terms of administrative burden requires some effort. Therefore the option in the long term is not expected to positively impact on the policy objective. Also there is a danger of dissolution (of the list) as a result of political negotiations, as happened with Article 20 and the relevant recitals in the Services Directive

Economic Impacts

Operating costs and conduct of business

It would not create an "obligation to sell": traders should still be able to refuse to sell and to apply different terms and conditions, as long as they are based on objective criteria (i.e. criteria not mentioned in the list).

The option would entail compliance costs for companies which would need to asses with the support of trade associations and law firms which justifications could be invoked. For this option to be effective, companies would annually have to assess in detail the reasons why they refuse to sell or apply different terms and conditions on grounds of nationality or place of residence or establishment, in order to rely only on justifications that are based on objective criteria. Criteria, in order to remain pertinent, would have to be adapted over time. Such adaption would require changes to the basic legal instrument for instance after the other DSM initiatives, such as on parcel and VAT, have been adopted.

Impact on the Single Market

This option would imply careful and complex crafting in order to take into account complicated and evolving situations, particularly as its effect would depend on the adoption of proposals which have just been made or are yet to be put forward, e.g. on cross-border parcel delivery or VAT. If the upcoming legislation mentioned in the baseline scenario enters into force, this option would not substantially affect the rules in place. However, depending on how the environment changes this option might paradoxically have a negative impact on the Single Market, if it encourages companies to rely on grounds to geoblock which are not explicitly prohibited.

Competitiveness of business

The increased administrative burden would negatively impact competitiveness, but European and non-European companies would be concerned in the same way On the other hand, companies might reconsider their decisions not to sell cross-border.

Impacts on SMEs and microenterprises

SMEs would face administrative costs for analysing the situation (see above). The effect of these costs would disproportionally affect micro enterprises which might not have the administrative and legal capacity to analyse in which situation justifications can be invoked. Micro enterprises should therefore be exempted from this option.

During the public consultation, companies were split on the impact of a list of unacceptable justifications, while a very clear majority of company associations thought the impact to be negative.

Consumers and households

On the one hand, consumers would benefit by being better informed about reasons for discrimination, by traders which cannot be invoked. On the other hand, consumers (and customers at large) would not immediately get access to goods and services but can only sue companies in front of national courts. It risks creating frustration among consumers and produce only long term effects.

During the public consultation, a clear majority consumers and consumer associations expected the impact to be positive.

Macroeconomic impact

This option does not have any macroeconomic impact.

Impact on Member States

This option might require particular efforts from national enforcement authorities and national courts.

Social Impacts

This option does not have any significant social impact.

Environmental Impacts

This option does not have any significant environmental impact.

6.6. Option 5 Option 3 + obligation to serve customers across the EU

Impact on the policy objectives

In addition to the impacts of Option 3, this option requires sellers to sell and deliver goods and services to a potential customer irrespective of the residence of the customer. As a consequence, geoblocking and other such forms of discrimination will be significantly reduced (with the exception of legally required geoblocking). Companies can still refuse to sell on grounds not related to the residence or nationality. Thus, the policy objective will be completely achieved.

However, requiring companies to ship to potentially all 28 Member States would be a disproportionate means to achieve the objective.

Economic Impacts

Operating costs and conduct of business

This option would imply that sellers could not refuse as a rule to sell and deliver to another Member State. Nevertheless, they could charge the extra price for delivery of the physical goods to the customer. They would thus not incur directly extra costs. However, by selling and delivering to another Member States they would incur significant indirect costs.

Firstly, they would have to organise the cross-border delivery. While the actual price charged by a courier company could be billed to the customer, the trader would have overheads in terms of finding a carrier and arranging for delivery, which are not part of the seller's normal business activity.

Secondly, if they deliver to an address in another Member State, they would have to accept returns from this address. In case of faulty products covered by the legal guarantee, there would be the additional direct costs and the indirect costs of organising the repairs or replacement in particular due to higher shipment costs.

Thirdly, for physical goods VAT is due in the Member State into which the goods were delivered. Therefore, traders would have to register for VAT in each Member State into which they deliver if they haven't done so previously. These costs have been estimated at 8 000 € per Member State 95 . Currently, there is a simplification measure in place whereby businesses do not have to register up to the distance sales thresholds for intra-EU Business to Consumer (B2C) supplies of goods (EUR 35 000 or EUR 100 000 determined by the Member State of destination) 96 . It should be noted that the Commission will be making a proposal by end 2016 to extend the cross-border VAT electronic and payment system (See 2.2.1.2) to cross-border supplies of goods which will substantially reduce these compliance costs.

Finally, when delivering goods to another member state, traders would have to respect the labelling requirement applying in this member state, e.g. in terms of language. Consequently, such sales might require companies to modify the labels on their products.

As a result of these extra costs, selling to customers in another Member States at the same price as to national customers may become unprofitable or loss-making, even if the seller is allowed to bill extra delivery costs.

Impact on the Single Market 97

The proposal to allow customers access to more offers should motivate customers to shop cross-border. Price arbitrage for physical goods will be constrained by delivery costs, which will however be lower than in option 3, as the trader will arrange cross-border delivery.

As a result, for the example of electronic goods only, the net effect of lifting geoblocking restrictions by companies could lead to an increase in cross-border sales of about 1.4% of the total market size (total sales). Most of that trade would come at the expense of a reduction in domestic purchases, mostly in domestic offline purchases.

Since for digital services consumed at the establishment of the trader no delivery is involved, the impact for these two elements would be the same as under option 3.

Competitiveness of business

The elimination of lifting geoblocking restrictions by business would boost trade and sellers will receive sales orders from foreign customers. However, offline sales could decline because online price competition increases. Both online and offline prices will decline when sellers adjust to the new market conditions when geoblocking restrictions are lifted.

Based on extension of the analysis for the popular eCommerce category electronic goods, the net effect of lifting geoblocking restrictions on total sales would be positive, as additional sales outweigh the decrease in prices. Producer surplus (profits) would increase by 1.4% on average across Member States.

Firms that buy goods and services online would benefit from lifting geoblocking restrictions in the same way as consumers. They would benefit from lower prices and a wider variety of products.

The competitiveness of European businesses will not be affected, because it applies to European and non-European companies alike

Impacts on SMEs and microenterprises

SMEs and especially micro-enterprises would find the complications of selling under unfamiliar circumstances and under unknown legal conditions prohibitive. They should therefore be exempted from this option.

During the public consultation, a slight majority of business and a very clear majority of business associations expected a negative impact.

Consumers and households

When all geo-blocking and discriminations based on nationality or place of residence or establishment, including delivery restrictions are fully lifted, customers will have access to all online product sales catalogues in all EU Member States, at the same prices as domestic shoppers in these countries (possibly with additional transport costs to ship the goods to their country of residence, which will however be lower than under option 3 since the delivery is arranged by the trader). As a result, customers would be able to price-arbitrage between online shops in all Member States and buy any variety or quality of goods available anywhere in the EU.

The combination of price arbitrage by customers and increased price competition between online sellers should reduce prices, both in online and offline retail stores, and increase consumers' purchasing power and real expenditures. For example, for electronic goods prices are expected to decline across Member States by -0.2 to -2%, depending on current price levels and market competition. The variety effect will further increase consumer welfare because they will find more appealing products than those available in the home market.

Based on an extension of the analysis of the electronic sector, EU28 consumer surplus from online purchases could amount to 1.2% on average for the EU28, with variations across Member States. Smaller Member States would be expected to gain more (up to 15%) because the variety of goods locally available online is more limited.

During the public consultation, an overwhelming majority of consumers and a clear majority of consumer associations expected a positive impact.

Macroeconomic impact

Increased cross-border trade after lifting discriminatory restrictions will tend to reduce prices in the domestic retail sector, in this option more than in option 3. This should benefit other sectors of the economy because it becomes cheaper to distribute goods and services through the retail sector. The overall decline in retail prices would increase consumers' purchasing power.

The current levels of cross-border e-commerce already increase real household expenditure by about 1% compared to a situation without cross-border e-commerce 98 . As a result, the overall impact of the current level of cross-border e-commerce on total output (GDP) is estimated at around +0.14%, compared to a situation with no cross-border e-commerce 99 . While lifting geoblocking restrictions would amplify this positive impact, a precise estimate of the macroeconomic impact of this option is not available.

Impact on Member States

For enforcement purposes, no new administrative arrangements are required beyond those mentioned in option 1. There might be an increased workload initially similar to the estimations for option 3. However, the extra costs associated with cross-border delivery might lead to low compliance rates by businesses, which would in turn lead to more numerous enforcement requests.

Social Impacts

This option does not have any significant social impact

Environmental Impacts

7.Comparison of options

7.1. Comparison in terms of effectiveness, efficiency and coherence

Compared to the baseline option, Option 1 "improving transparency and enforcement" would improve the information to websites and indirectly to goods or services. However, it would depend on the indirect mechanism of additional consumer pressure as a result of better information leading to less unjustified geoblocking, and would thus likely to be inefficient in achieving the objectives. On the other hand it would impose limited but not insignificant administrative burdens on a large share of enterprises, in particular SMEs and micro-enterprises. It would be coherent with a general policy approach in favour of providing more information (including the already existing provisions of the Consumer Rights Directive) but it would not increase legal certainty for businesses. It would however increase legal certainty for consumers and enforcement authorities, who would be better able to ascertain whether the difference of treatment is based on objective criteria. On its own, it would not have significant effects, since the underlying rules would not be clearer.

Option 2 "Banning the blocking of access" would fully achieve the objective of "ensuring access to websites" but would only indirectly improve non-discriminatory access to goods and services. It would depend on customers to increase market pressure; it would be up to the demand side to drive the market for further changes to address discriminations. Overall, costs would be quite limited, as only a small share of larger traders (with different websites) would be affected, leading to a low cost/benefits ratio, but it would increase legal certainty only with respect of access to websites is concerned, not in relation to access to the service/product. Overall, ensuring access to websites would be in line with the general policy approach in favour of eliminating restrictions linked to geographical borders.

Option 3 "Equal treatment in specific situations" (shop like a local) would fully achieve access to websites and partially achieve non-discriminatory access to goods and services, keeping in mind that it only addresses unjustified geoblocking, while other regulatory initiatives will further reduce incentives for total geoblocking. Due to the limited scope of this option (no obligation to deliver cross-border, transparency obligations only in exceptional cases), costs for businesses would be small (adaption of website, if necessary, no VAT registration abroad and no payment of foreign VAT as long as the trader does not intervene directly or indirectly in the cross-border delivery arrangements of the customer), and business would have the certainty that the specific situations do not imply additional costs. For digital services, VAT registration has been substantially simplified by the Mini-One-Stop Shop, and companies exempted from VAT according to national special schemes for small enterprises are exempted from the equal treatment obligation. For services received at the location of the trader, there is no obligation to apply foreign law and the VAT is that of the country of the trader. As a result, this option has a positive cost/benefit ratio. Moreover, it is fully in line with the overall approach of the (Digital) Single Market strategies to allow customers to take better advantage of cross-border opportunities and is not in contradiction with other pieces of legislation (it is without prejudice to competition law or to the rules on contractual obligations). It creates directly enforceable rights for consumers, thus having a positive effect on legal certainty.

Option 4 "List of unacceptable justifications" is expected to have only limited effect on non-discriminatory access to goods and services, as it would only eliminate a small number of possible justifications for companies to discriminate. Although costs would be low, so would benefits be, resulting in a less favourable cost/benefits ratio. This option would be in line with the overall approach of the (Digital) Single Market strategies to allow consumers to take better advantage of cross-border opportunities. It is unlikely that the drafting of this list would create the required legal certainty and lead to improvements in practice.

Option 5 " Including delivery" would fully achieve access to websites, goods and services, but would impose significant additional costs on businesses, as they could be required to register for VAT in another member state, need to explore delivery option in unfamiliar territory, etc., resulting in an unfavourable cost/benefit ratio. It would impose disproportionate burdens on traders whilst creating major legal complexities. It would not be fully in line with EU policies since the removal of barriers to cross-border parcel delivery is not completed. Moreover, it would interfere with the freedom of contract. This would significantly restrict the freedom to conduct business as set out in Article 16 of the Charter of Fundamental Rights, since it would prevent companies from taking the strategic decision not to deliver to certain countries and subject them to the implications that this may have.

All options would equally achieve the objective "improved enforcement", since they all provide for an inclusion of the current proposal in the annex of the Consumer Protection Cooperation Regulation, as well as specific enforcement provisions.

Table 1: comparison of options

	Effectiveness 100				Efficiency 101 (costs/ benefits)	Coherence 102
Options	Access to websites	Access to goods and services	Improve enforcement	Increase legal certainty
Baseline (0)	o	o	o	o	o	o
Transparency (1)	o	o	+	o	-	+
No blocking (2)	++	o	+	o	o	+
Equal Treatment (3+2+1)	++	+	+	+	++	++
List of justifications (4+2+1)	++	o	+	+	o	+
With delivery (5+3+2+1)	++	++	+	o	--	---

7.2.Preferred option

The option which contributes most to the achievement of the policy objectives and has the most positive overall impact is option 3 (which includes option 2). It ensures access to websites, as well as - in specific situations - to goods and services, thus improving consumer welfare to the extent that unjustified geoblocking was taking place in these situations previously. This option is proportionate with the objective to be achieved: it takes into account potential complications of cross-border selling for businesses and therefore does not require cross-border delivery by traders, hence not creating any significant burden on businesses, in particular SMEs. It applies only in obvious cases where traders are not faced with undue legal complexities. Transparency obligations are limited to exceptional cases. Traders can choose the when to provide information to consumers (upfront or upon request), and only those not covered by the three situations would be subject to it (essentially e-commerce sellers delivering their products cross-border). It addresses enforcement efficiently by mostly utilising existing enforcement structures (the CPC network) and not over-burdening Member States when enforcing the rules in B2B situations. As a result, it is the most efficient option, while scoring equally well on coherence.

The preferred option would address only those forms of discrimination and access restrictions which are caused by unilateral decisions of companies to segment market along national borders. For the other drivers, there are other relevant policy initiatives.

7.3.Choice of legal instrument

To choose a non-binding instrument does not seem appropriate. First, as regards the very specific issues at stake it would not guarantee adequate and coherent implementation at national level. Second, Member States may express the same reluctance to act as shown until now (Article 20 Services Directive), or be prevented from taking action by the existence of contravening national provisions and a lack of domestic political will to amend and/or abolish them. Third, a general non-binding instrument would also leave a very broad discretion to Member States whether and how to intervene, if at all. Already Article 20(2) of the Services Directive has proven not very effective, so a none-binding instrument could be expected to be even less effective.

A legislative instrument, preferably a Regulation, would appear to be more appropriate to address the identified problems and could guarantee that the policy options are introduced in all MS and that sufficient uniformity and legal clarity are achieved for all stakeholders to rely on the new rules.

The clear advantage of a binding legislative instrument is the fact that it can guarantee that the policy options are introduced in all 28 Member States and that the rules are enforceable, ensuring in particular: (i) a level playing field across Europe and (ii) improved certainty for consumers and traders. Binding legislation implies a certain implementation and administrative burden. However, it would be proportionate to the objectives attained. Costs faced by traders (in particular for the adaptation of websites) would be similar to those incurred under a recommendation or self-regulation, if properly applied.

Article 114 TFEU provides for the possibility to adopt measures (regulations or directives) for the approximation of the provisions laid down by law, regulation or administrative action in Member States which have as their object the establishment and functioning of the internal market.

A directive would necessarily leave some room for Member States to choose how to implement the relevant rules in their national legislation. As a result, there would be a high risk that those rules would be different in different Member States. It may also take longer to deliver the expected results in the market. Besides, a Directive would not necessarily create directly enforceable rights and obligations which customers and traders could avail to before national courts. Therefore, a directive would not be most suited to achieve the pursued objectives. The Services Directive does not address the issues at stake to an adequate degree. From a more political perspective there seems to be at this stage no stakeholder or political support to amend the Services Directive. Changes to the Directive would require a longer and comprehensive debate with all stakeholders on a wider range of topics, which would put at risk the objectives of this initiative.

A Regulation allows for quick implementation and is a particularly suitable instrument when the objective is the full harmonisation of a certain area. This type of instrument could be best used to achieve the objective of ensuring the application of non-discrimination rules across the EU. It would allow a uniform application of equal treatment provisions across Member States and would guarantee that companies from different Member States are subject to the exact same rules. Such a Regulation would take precedence over Article 20(2) of the Services Directive and the rules of national law transposing it. It would not require reopening the Services Directive.

A targeted Regulation is thus the preferred instrument.

7.4. Subsidiarity and proportionality of the preferred option

The subsidiarity principle (Article 5(3) TEU) requires the assessment of the necessity and the added value of the EU action: with respect to all issues addressed in this IA, EU intervention would be the only way to establish across Europe, common requirements necessary for the smooth functioning of the Single Market.

Geo-blocking and discrimination based on nationality or place of residence or establishment arise in the cross-border context, typically when customers intend to access a service provided by a company established in a different Member State. Due to the cross-border nature of such restrictions, it can only be tackled effectively on EU level. A ban on geo-blocking and other discriminations could only be ensured effectively if the conditions for customers’ access did not vary according to the Member States they are located in at a specific moment, or where they have their place of residence or establishment. Moreover, EU action providing for uniform rules would produce clear benefits to traders, compared to any national solutions, by creating a legally certain framework across Europe.

The preferred option does not go beyond what is necessary to solve the identified problems. Moreover, the initiative is limited to solving a problem of a cross-border nature that cannot be achieved by the Member States alone. The obligations foreseen under the preferred option are necessary to make sure that there is no unjustified discrimination of European consumers. They do not force traders to contract and deliver cross-border and therefore have a limited effect on the current trading practices and business models. Under the principle of proportionality (Article 5(4) TEU), the content and the form of EU action shall not exceed what is necessary to achieve the objectives of the Treaty. The proportionality of the different policy options considered has been assessed and described in the relevant parts of this IA.

The preferred option would cover micro enterprises both in their role as traders and end-users. The option would increase legal certainty for traders by clarifying existing obligations and specifying under which conditions customers should be treated equally in cross-border purchases. For micro enterprises with operations in a single Member state, the additional transparency requirements (on request having to inform consumers about the reasons for automatic rerouting) would de facto not apply, since automatic rerouting to another website requires operations in more than one country. For micro-enterprises operating in more than one Member State, transparency requirements would still be exceptional (only in the cases of legally justified automatic rerouting), Moreover, the transparency obligation has also been designed with the view of limiting the administrative burden by allowing traders to provide ex ante or ex post explanations, in the trader's own language and by mail, whichever is the less burdensome for the company.

For micro companies providing goods and service via a single website there will be no additional burden from the obligation not using re-routing or provide transparency on offers, unless IP-blocking techniques have already been implemented (unlikely in case of micro enterprises) which would then incur a one-off adoption costs. Despite these costs, the exclusion of micro and small businesses would not be the viable option. It would result in consumer detriment as customers would not be able to easily determine the rights they enjoy, as these would depend on the size of the company they are purchasing from which is difficult to determine by them. Furthermore, allowing weaker consumer protection depending if the trader employ 9 or 10 people would also create artificial threshold effects with the risk of de-incentivising micro companies for growing.

Micro enterprises acting as end users would clearly benefit from having non-discriminatory access to offers. Expanding cross-border for them can then be a way of finding more competitive offers from suppliers and new potential customers.

7.5. Consistency with other EU policies and with the Charter for fundamental rights

The preferred option is consistent with other initiatives of the Commission, in particular the legislative proposals on contracts for the supply of digital content, on contracts for the online sales of goods and on portability of online content services. As explained in section 2.5 ("how would the situation evolve all things being equal?"), the entry into force of these proposals will not address the issue of "unjustified" geo-blocking, so they will not have any direct impact on existing undue discrimination. They might remove some of the current objective criteria, which could turn existing justified differences of treatment into geo-discriminations. It is furthermore complementary to the forthcoming legislative initiatives on the amendment of the Regulation on Consumer Protection Cooperation, on parcel delivery, on copyright, on rail passenger rights and to modernise the VAT for cross-border e-commerce. Likewise, it is in line with the relevant results of the competition sector inquiry into e-commerce as well as the forthcoming Guidance on unfair business-to-consumer commercial practices.

The preferred option would have a limited impact on Article 16 (Freedom to conduct a business) and 17 (Right to Property) of the EU Charter of Fundamental rights. The freedom to conduct business is enshrined in Article 16 of the EU Charter of Fundamental Rights and any proposals limiting this freedom have to be carefully analysed, against the expected benefits for customers and the general interest (in light of the non-discrimination principle, which is a cornerstone of the EU). A careful balancing has to be made with the interest of protecting customers from any discrimination based on nationality or residence, taking account also of Article 38 of the Charter, and the interests associated with the EU's single market imperative as such and the fundamental freedoms of the Treaty.

An introduction of obligations on companies (in particular small companies) which may not be in-line with their business models or commercial choices may stifle innovation and have a backlash impact on consumers through higher prices. However, given that there is no obligation to contract, but simply an obligation to treat European customers in the same manner when they are in the same situation, irrespective of their nationality or place of residence or establishment, i.e. to protect them from unjustified discriminations, the concerns as regards these rights would be well addressed. Companies will not be obliged to change their business models.

According to the Court of Justice, neither the freedom to conduct a business (Article 16), nor the right to property (Article 17) are absolute rights. They must be considered in relation to their social function and to other fundamental rights and principles. Article 52(1) of the Charter allows for the restriction of these rights in certain cases. Therefore, given the importance and necessity of the proposed measures in order to achieve the policy objective of realising the EU's internal market and ensuring that European customers who are in the same situation are treated alike and the limited impact on the freedom to conduct a business and the right to property, the resulting balancing of rights leads have demonstrated that the restrictions are justified and the impact of the proposed options are proportionate.

8.Monitoring and evaluation

The Commission will ensure that the action selected in the course of this IA contribute to the achievement of the policy objectives defined in Section 4. The monitoring process could consist of two phases:

1)The first would concentrate on the short-term, starting right after the adoption of the legislative proposal. It would focus on how the Regulation is applied in the Member States by the market participants in order to ensure a consistent approach. The Commission would organise meetings with Member States representatives (e.g. group of experts) and the relevant stakeholders in particular to see how to facilitate to transition to the new rules.

2)The second would be mid- to long-term and would focus on direct effects of the rules contained in the Regulation. The table below presents the main indicators that could be used to monitor progress towards meeting the objectives pursued by this initiative, as well as the possible sources of information. Depending on the data needs, information could be gathered from Member States, customers or information society service providers. Where needed, the Commission could send questionnaires to Member States or stakeholders or organise specific surveys. The information-gathering should start 2-3 years after the start of application of the regulation.

The operational objectives for the preferred option are as follows:

1) to decrease the share of websites engaging in unjustified blocking of access to users from other EU Member States

2) to decrease the share of consumers experiencing unjustified geoblocking/geodiscrimination

A comprehensive evaluation could take place two years after the start of application of the rules in particular with a view to assessing whether the prohibition of non-discrimination should also apply to electronically supplied services, the main feature of which is the provision of access to and use of copyright protected works or other protected subject matter, provided that the trader has the requisite rights for the relevant territories, and every five years thereafter.

Table 2: Overview of monitoring arrangements

Operational objective

Indicator/Definition

Unit of measurement

Source of data

Frequency of measurement

Baseline

Target

Fewer websites practicing unjustified refusal of access and sales cross-border

Variation of websites refusing access or sales to customers based on geographical residence, analysed to identify unjustified refusals

Percentage points

Mystery shopping survey

Triannual

Multiple levels (see annex 5)

n/a

Fewer consumers experiencing unjustified refusal of access and sales cross-border

Variation of customers experiencing geo- discrimination, analysed to identify unjustified refusals

Percentage points

Consumer survey

Triannual

Multiple levels (see annex 5)

n/a

Annex 1: Procedural information

1. Agenda planning and Work Programme References

The Agenda Planning Reference is 2016/CNECT+/002.

The initiative to ban unjustified geo-blocking forms part of the Digital Single Market Strategy, adopted in May 2015, as well as part of the Single Market Strategy, adopted in October 2015.

2. Inter-Service Steering Group

An Inter-Service Steering Group was set up in 2015. In total, six meetings were organised: on 17 July 2015, on 14 December 2015, on 11 February 2016, on 1 March 2016, on 1 April and on 14 April 2016. The following directorates and services were consulted: CNECT, COMP, GROW, JUST, TAXUD, MOVE, SJ, SG, JRC, ENV, REGIO, AGRI, EMPL, EAC, FISMA, NEAR, TRADE, HOME and DIGIT. The feedback received from these directorates and services has been taken into account in the report.

The ISSG approved the Inception Impact Assessment that was published on 18 December 2015.

The minutes of the meeting of 1 March 2016 were submitted to the Regulatory Scrutiny Board.

3. Consultation of the Regulatory Scrutiny Board

The Impact Assessment Report was examined by the Regulatory Scrutiny Board on 6 April 2016. The RSB issued its opinion on 8 April, requesting a set of revisions. The revised impact assessment was submitted on 15 April 2016. On 21 April 2016, the Board gave an overall positive opinion and provided further suggestions for improvement, which were implemented. The present impact assessment includes these further improvements (see appendix).

4. Studies to support the Impact Assessment

Two studies were launched specifically for this impact assessment.

Firstly, a Mystery Shopping Survey on approximatively 10 500 websites assessed cross border within the EU was carried out between October and December 2015. For details, see annex 6.

The full study can be found at: http://ec.europa.eu/justice/newsroom/consumer-marketing/news/160318_en.htm

Secondly, in February 2016 a Flash Eurobarometer was carried out to determine the extent of geo-blocking which business as end-users face. For details, see annex 10.

The full study can be found at forthcoming

In addition to these two specific studies, there exist a number of other recent studies relevant for the extent of geo-blocking and discrimination based on residence or nationality within the European Union. An overview of the evidence can be found in annex 5.

Finally, for the economic model described in annex 4, the European Commission acquired a data set with nearly 2 million observations of monthly online and offline sales volumes and prices for more than 100 000 models of 4 types of electronic goods in 10 EU Member States 2012-2015 from a data provider.

Appendix - RSB requests for improvements and revisions introduced by services

Subject area	RSB comments	Revisions introduced
(1) Further clarify the scope of the initiative.	The rationale for the extension of the scope to the transport sector should be further elaborated in the report, clarifying what is the underlying problem and explaining why this cannot be addressed by the sectoral transport legislation currently in place that already prohibits discrimination (except for railways).	Sections 1, 2.6, 3.1, 5.0, 5.3, 6.4, 7.2 The extension to the transport sector has been abandoned
(2) Enhance the definition of the problem	The problems related to the ineffectiveness of Article 20(2) of the Services Directive should be further explained in the main report, rather than only in Annex 8.	Section 2.3.2.2 The text explaining the ineffectiveness of Article 20 of the Services Directive has been further developed under "Ineffective enforcement of the existing legal framework".
	The studies used for illustrating the existence of the problem should be quoted with caution and the results should be qualified. For instance, the apparent high demand for arranging cross-border part of the delivery of products by the online buyers (with 32% of consumers ready to do so) is misleading as only 12% of the surveyed online buyers in these countries used freight forwarding due to the reason that the seller did not ship to their country (which may have been true for non-EU countries as well).	Sections 2.2.1 and 6.4 The references to the studies have been adjusted to better express inherent uncertainties. For the example, multiple motivations have been acknowledged.
(3) Better define the options and analyse their impacts	The report should clarify whether or not microenterprises are excluded from the initiative	Sections 5.3, 6.4 and 7.1 The square brackets have been removed. Micro-companies are included in the initiative, but transparency requirements have been significantly reduced.
	The executive summary and conclusions from the impact analysis should be more nuanced and qualified in order to reflect the uncertainties surrounding: the estimations of the magnitude of unjustified geo-blocking; the fact that the baseline will be affected by the effectiveness of related regulatory initiatives that would reduce the main incentives for geo-blocking; and the possibility for the identified negative impacts to be larger than expected.	Sections 6.4, 7.1, 7.2 and executive summary The relation between unjustified geoblocking and total geoblocking has been reinforced, and the importance of other policy initiatives has been strengthened.
	For example, VAT arbitrage is explicitly referred to, but not seen as opening the door to VAT circumvention practices (p. 44).	Section 6.4 The possibility of increased VAT arbitrage has been expressed more clearly.
	The possible negative impacts in some Member States (such as employment effects related to the reduction in off-line shopping in the option of "shopping like a local") should be mentioned	Section 6.4 The possible negative impact has been acknowledged.
(4) Legal certainty and coherence with EU policy instruments	Interactions with other EU regulatory instruments and initiatives will need to be specified. For instance, passive sales restrictions in vertical agreements are mentioned as possible cases of unjustified geo-blocking, but are not clearly part of the scope of the impact assessment (p. 23). It should be clarified whether the envisaged Regulation will include or not such restrictions in its scope.	Sections 1 and 5.3 The interaction with EU competition law, especially regarding to vertical restraints, has been clarified.
	Option 3 on "shopping like a local" but might raise compliance issues and the applicability of Rome I is made clear for services but not for tangible goods (p. 39).	Sections 5.3 and 6.4 The causation between a trader directing its activities to a Member State and the applicable law has been clarified.
(5) Monitoring and evaluation.	The indicators chosen refer to geo-blocking in general without trying to better measure the scope and the trends in “unjustified” geo-blocking.	Section 8 The indicators have been refined to better capture unjustified geoblocking.
(6) Procedure and presentation	The report should consistently describe the inclusion of Article 20(2) of the Services Directive in the revised Annex of the Consumer Protection Cooperation Regulation as part of the package of the DSM proposals and, thus, part of the baseline scenario.	Section 2.3.2.2. A description of the inclusion of Article 20 in the Annex of the CPC Regulation has been added. It is already explained in the baseline scenario under section 5.0.
	Annex 1 should briefly explain how the Board's recommendations have led to changes compared to the earlier draft	Annex 1 - section 3 The information has been added
	Annex 3 should be updated in line with the changes in the main report, in particular regarding the practical implications of the preferred option on businesses	Annex 3 – all sections The implications of selling locally to a resident in another Member states have been clarified.
	The report (including the executive summary) should be proofread for the use of correct terminology, spelling mistakes, redundant words and inconsistencies.	on-going
	The operational objective should rather measure the decrease and not the increase of "the share of websites blocking access to users from other EU Member States"	Section 8 The mistake has been corrected.
	options that are not the preferred option are not being "discarded", but are simply less effective or efficient	Section 5.6 The wording has been changed.

Annex 2: Stakeholder consultation

This annex presents the main findings of the (1) public consultation on geo-blocking and other geographically-based restrictions when shopping and accessing information in the EU and (2) other stakeholder activities.

1. SYNOPSIS REPORT

Summary of Responses to the European Commission's 2015 Public Consultation on

'GEO-BLOCKING AND OTHER GEOGRAPHICALLY-BASED RESTRICTIONS WHEN SHOPPING AND ACCESSING INFORMATION IN THE EU'

EXECUTIVE SUMMARY

A wide majority of the respondents support the problem definition of geo-blocking and other forms of geographically-based discrimination as presented in the questionnaire. At the same time, a number of respondents, especially from business associations, take issue with the wide definition of geo-blocking.

Overall, a majority of respondents from both the consumer and company perspective agree that consumers should be able to purchase and access services from everywhere in the EU.

According to the respondents the goods and services most affected by geo-blocking are clothing, footwear and accessories, media (books), computer hardware and electronics, airplane tickets, car rental, digital content such as streaming services, computer games and software, e-books, MP3s.

A majority of respondents from both the consumer and company perspective agree that traders should inform customers about sales restrictions. The majority of respondents also agree that there are no reasons for website blocking, apart from compliance with explicit legal requirements.

Consumers strongly agree or agree to the suggested policy options, including to require traders to accept cross border transactions without an obligation to deliver and in particular to allow the download of digital content, when certain conditions are met.

Businesses and public authorities urge the Commission to define what is justified or unjustified geo-blocking. The majority of businesses oppose an obligation to sell and deliver throughout the EU, highlighting the need to respect their contractual freedom.

A large majority of all respondent groups agree that enforcement of rules should be improved, improving information requirements and ensuring non-discrimination.

The majority of consumer respondents expect a positive impact of most of the suggested measures to tackle unjustified geo-blocking and related discrimination, while companies are divided as regards expected impacts.

II. INTRODUCTION

The Commission launched a public consultation in the context of its initiatives to tackle unjustified geo-blocking under the Digital Single Market Strategy and to fight discrimination on the grounds of residence or nationality under the Single Market Strategy.

The consultation ran from 24 September to 28 December 2015.

The questionnaire was published in 24 languages. The consultation was publicised on the Commission's websites, social media channels as well as in stakeholder meetings. Responses have been published except where respondents asked for confidentiality.

The questionnaire was subdivided into the following three parts (1) "consumers", (2) "trader"' and (3) "consumer and company perspective", depending on the perspective selected by the respondent.

While most questions were common to all three parts of the questionnaire, each part also contained questions specifically targeted to the above respondent groups. For a more detailed analysis of respondent perspectives, the respondent groups are broken down into further categories (as can be seen in the following section III of the synopsis report). 433 responses were received.

The College of Europe, Bruges, has provided some assistance to Commission services in the analysis of parts of the replies from a selection of respondents groups. A report prepared by the College of Europe will be published separately from this report in the EU bookshop.

The numbers and percentages used to describe the distribution of the responses to the public consultation derive from the answers provided under the EU-Survey tool. Other submissions of stakeholders to the public consultation, such as position papers and contributions by email, have been taken into account when describing and analysing the views of stakeholders, but without being considered for the statistical representation.

Replies to all questions in the EU-survey were optional. Respondents often chose not to answer all questions. Therefore, where percentages are reported below, these derive from respondents who replied to that question. Those who did not reply to a particular question are not accounted for in the percentages displayed. This is to ensure clarity with regards to the interpretation of the data.

III. OVERVIEW OF RESPONDENTS

The following overview of respondents details the Commission's classification of all 433 responses to the consultation.

	Replies
Consumers	251
Consumer Organisations	27
Member States authorities	13
Consumer authorities	6
Companies	58
Business Associations	78
Total	433

Note: Place of Residence as declared by respondents. Total respondents in graph as displayed in the graph above exceed total number of replies (433). This is to account for respondents who stated multiple places of residence.

The consultation sought the views of interested parties, as such the responses cannot be considered to be statistically representative.

Views were sought from consumers, consumer associations, businesses and business associations as well as Member States, their institutions and national authorities.

Based on the responses to the public consultation it is evident that respondents made very different interpretations on which stakeholders that should reply from a “consumer and company” perspective. Instead of the intended target groups of firms with a retail activity, the respondents included many different stakeholders, e.g. a small European gaming company, a multinational electronics corporation, a European association in the tourism sector, and a Member State’s Ministry. Given the low number of respondents in this category (25 total submissions) and the perceived ambiguity of the definition of this stakeholder category, certain caution must be observed when drawing any conclusion from these responses.

When analysing the submission to the consultation the replies have been sorted into three wide categories: consumers and consumer organisations; companies and business associations; Member States and national authorities.

Consumers and consumer organisations

Individual consumers from numerous Member States responded. The largest number of responses from individual consumers was received from Germany, the United Kingdom, Austria and Italy. Two submissions from outside the EU have been received.

Consumer organisations responded, representing both national and European level.

Companies and business associations

Companies from numerous Member States responded. The largest number of responses received are from Germany, Austria, France and the United Kingdom. Five submissions were received from companies headquartered outside of the EU.

The majority of the companies who submitted their views indicate that they either predominantly sell on-line or both on-line and offline. Only one responding company exclusively sells in brick-and-mortar shops.

While some responses were received from large companies, many companies described themselves as small or micro enterprises. Companies indicated they are active in many different sectors including retail, software, IT, telecoms and transport.

Numerous business associations responded to the consultation, mostly from national and European level. We also received responses from associations of businesses not headquartered in the EU.

Some associations represent businesses from all sectors. Most associations responding represent companies from specific sectors such as retail/e-commerce, media, industry, fashion, tourism, gambling or legal services.

Some respondents criticise the questionnaire for being biased.

The questionnaire stated that copyright related aspects are left aside. However, several respondents refer to practices based on copyright, e.g. relating to digital content. The relationship of this initiative and other initiatives under the Digital Single Market Strategy such as portability seem not to be known to all respondents.

In terms of views of stakeholders from specific sectors several lottery operators underline that the regulatory framework on gambling including any geographically-based restrictions is up to Member States due to the specific nature of gambling and is in the public interest. From the music sector, associations are concerned about unintended ripple effect on the delivery of digital goods that incorporate copyright content if the scope of the geo-blocking initiative is not clearly defined. Many associations from the audio-visual and media industry highlight that copyright and licensing practices are not part of the scope of this initiative and stress the importance of territorial licensing, the freedom to conduct a business and the right to property of the EU Charter of Fundamental Rights.

Several associations of the publishing sector point out problems associated with collecting VAT in the country of consumption and that many digital products include copyright-protected content (for example political news). The presence on a given market outside national borders generates costs and obligations that a retailer needs to evaluate before agreeing to sell to customers abroad (VAT, fiscal reporting, service as well as contractual obligations), which is particularly challenging for Small and medium-sized enterprises (SMEs).

Member States and national authorities

Several Member States as well as authorities from Member States responded. National authorities mainly include those in charge of consumer affairs or competition issues. In total, public entities from 16 Member States answered to the public questionnaire or provided their views in separate written submissions.

IV. FORMS OF GEO-BLOCKING

Across the different stakeholder groups responding to the consultation there is strong support for the general principle that all consumers and business should be allowed to purchase and access services from anywhere in the EU.

A number of stakeholders, especially from concerned business associations, have critically commented on the wide definition of geo-blocking in the public consultation. The majority of these respondents see geo-blocking as blocking on-line access to an offer based on the customers' location, which is separate for purchasing and delivery restrictions.

a. Significance of geo-blocking or other geographically-based restrictions

On the question on what forms of geo-blocking and other geographically-based restrictions that are considered to create significant obstacles in the single market, respondents generally agree that practices that either prevent access to offers or access to the actual good/service make up significant barriers. The views expressed regarding the extent to which geo-blocking practices constitute barriers in the single market are also reflected by the importance attributed to the different geo-blocking practices (Fig. 1).

Figure 1. Significance of different forms of geo-blocking 103

Consumers and consumer organisations

Four out of five respondents identified as consumers or consumer organisations agree that the forms of geoblocking listed above create significant obstacles to the single market. Barriers stemming from restricted access to websites are seen as the most significant while many individual consumers have highlighted that geo-blocking related to digital content is a specific concern (e.g. access to TV broadcasts or platforms for apps or audio-visual content)..

On qualifying the relevance of these barriers, consumers considered all of the listed types of geo-blocking practices to be highly significant (with 85-90% of the respondent classifying them as "important"/"very important"). Most significant are barriers blocking access to offers (>90%), while restrictions for digital goods/services (87%) are considered slightly more significant than non-digital goods/services (82%).

As for the reasons given by traders for refusing to sell or charging a higher price because of residence, about half of respondents (46%) did not receive any explanation. Only about one in ten respondents (11%) stated that they had asked for an explanation and in most cases received one from the trader for the discrimination based on residence. In about a quarter of the cases (27%), the consumer asked but did not receive any explanation from the seller as to why they refused to sell or charged them a higher price. In 16% of the reported cases an explanation was provided upfront by the trader.

Companies and business associations

Respondents from companies and business associations take a different view from consumers. Even though a clear majority (75%) of respondents believe that blocking access to a website creates barriers, more than half of the respondents believe that restrictions on the basis of delivery or payment issues create significant obstacles to trade.

A number of stakeholders from business associations and individual companies highlight the need to clarify what the implication would be for copyright/licensing issues in future initiatives.

When ranking the importance of the listed forms of geo-blocking, companies and business associations tend to see them as less important compared to consumers. However, two out of three respondents from companies consider that rerouting or refusal of access to a website to be "important" or "very important" barriers. Even if several company respondents argue that geo-blocking is positive for consumers as it forwards them to websites adapted to the customers' home market. About half of the company responses stated that disproportionate shipping costs or using a country specific format (e.g. addresses, postal codes, phone, etc.) are important barriers.

Businesses selling and buying goods and services received an upfront explanation from the seller in about one of every eight cases (14%), the same frequency by which an explanation was received as to why a sale was denied or a higher price was charged based on the place of residence/establishment. In a little more than one third of all cases (36%) the seller did not provide customers with any explanation, while just over a third of respondents indicated that they have never asked the seller for an explanation.

According to the seller's point of view, less than one in three customers (29%) are provided with an explanation for the use of geo-blocking or other geographically-based restrictions. Around a third of businesses (34%) clarify that they provide a full explanation upfront on their website, in publicly available material or in the shop. Less than a third (31%) of the respondents state that they have not been asked to provide any explanation, while traders providing an explanation upon the request of a consumer association seldom occurred (in just 6% of cases). It has to be noted, however, that only a very small number of replies to this questions were received.

b. Experience of geo-blocking or other geographically-based restrictions

i) Being the subject of geo-blocking or other geographically-based restrictions

Consumers and consumer organisations

When asked about their experience of geo-blocking or other geographically-based restrictions when shopping cross-border, the vast majority of consumers and consumer organisations (89.4%) confirm that they had been subject of geo-blocking.

The most common restrictions which consumers face, regardless of the economic sector or type of purchased goods and services are refusal to sell (including automatic rerouting), refusal to deliver, price difference when shopping cross-border and the refusal of a discount.

Consumers report that they are geo-blocked in the retail sector when attempting to purchase clothing, footwear and accessories, media (books), computer hardware and electronics. In the digital sector, another sector in which consumers face difficulties, the issues are mainly related to the inability to purchase or access streaming services, computer games and software, e-books and MP3s. Further, the field of accommodation and leisure services also poses challenges. Consumers also experience geo-blocking when attempting to purchase airplane tickets, rent a car or when they wanted to access cultural and entertainment services. Some of the respondents specify that they have been geo-blocked when renting a car in another country. Moreover, customers also expressed dissatisfaction at being unable to access on-line gambling websites in other Member States.

Companies and business associations

Companies and businesses associations selling and buying goods and services experience geo-blocking to a lesser extent than consumers, with six out of ten businesses (63%) replying that they are subject to geo-blocking. In contrast, more than a third of the respondents (37%) submit that they have not experienced business-to-business geo-blocking or geographically- based restrictions in the course of their business.

Member States and national authorities

Most national public authorities agree with the forms of geo-blocking identified in the public consultation. Two authorities stress the need to define the scope of the initiative more precisely and make a clear case for the most problematic practices. One Member State gives similar examples of geo-blocking as those identified by the Commission.

Another Member State agrees with the categories identified by the Commission but emphasised the legitimate “supply side” reasons that prevent companies from delivering cross border (see under "V. Justifications for geo-blocking"). It points to the lack of clarity as to whether the initiative would address copyright related geo-blocking. Another Member State presents a strong opinion that copyright aspects should stay outside the scope of the initiative on geo-blocking. In addition, another Member State reports on cases of refused delivery cross-border concerning technical products, toys and homewares. A region of a Member States highlights the importance of portability of digital goods.

ii) Applying geo-blocking or other geographically-based restrictions

Companies and business associations

When asked about their experience of applying geo-blocking or other geographically-based restrictions slightly less than half of businesses (45%) submit that they have applied geo-blocking or other geographically based restrictions in the course of their business (in either a business-to-business or business-to-consumer context). However, only a limited number of responses from companies were received to this question. As the main reasons for the application of geo-blocking practices, most of the traders point out VAT rules, divergent national regulations, consumer protection laws and delivery costs. Others did so in order to comply with copyright and licensing restrictions, the requirements for fiscal reporting and auditing, due to their contractual obligations or business models. Others point towards differences in technical requirements within different Member States, avoidance of bureaucracy and legal disputes, and the aim to provide a better and tailored service to consumers. A few of them also include as a reason the compliance with laws on pricing and misleading advertising, recycling fees to sales transactions and fraud prevention. In all of their responses traders consider the mentioned reasons as justification for the application of geo-blocking.

From the position papers received, it is evident that companies consider some forms of geo-blocking to be good for consumers (e.g. rerouting to websites where goods are adapted to consumers home market). Most of the relevant responses consider geo-differentiation required under local legal compliance requirements (e.g. in terms of different technical, regulatory, legal and fiscal rules in different Member States).

V. JUSTIFICATIONS FOR GEO-BLOCKING

a) Situations when geo-blocking is not justified

The responses from stakeholders show that the majority of respondents believe that the listed practices cannot be a justification for geo-blocking or differential treatment based on residence, even if the views of consumers and those of companies show a larger discrepancy than under most other questions.

Consumers and consumer organisations

Both individual consumers and consumer organisations are opposed to the listed situations being used as possible justifications, especially in situations where delivery is easily accessible where the consumer is willing to pay additional costs (above 80%), when there is no additional cost or administrative burden (above 90%), or when paying with means of payment that are internationally valid and accepted (87%).

Companies and business associations

Among company respondents the replies are more evenly distributed between "justified" and "not justified". With regards to justifications relating to delivery, about half of the respondents consider differential treatment justified, while about a fifth of respondent take the opposite view. Especially in situations when a trader advertises the services/product in that country, or targets the country of the customer with a website, respondents from companies tend to be more accepting of not allowing any justifications. As a general point, many companies underline that each provider of goods/services should be free to decide the geographical scope of their operations and nothing should force them to extend their offer to markets where they have previously not been active. In addition to the public consultation, a number of companies and company associations provide examples on situations where they think that geo-blocking is justified. Differences on warranties, VAT, consumer rights, language requirements are repeatedly listed by companies and business associations as legitimate objective reasons for differential treatment.

Member States and national authorities

Public authorities tend to take the position that the listed practices should not be justified. Among the views presented in position papers it is argued that in some cases geographical price discrimination can give rise to overall economic/consumer advantages. But even when either supply-side or demand-based factors might justify price discrimination, respondents argue that it should not be justified to make consumers pay higher prices based on their willingness to pay, but only on the grounds of nationality or residence.

b) Objective factors justifying geo-blocking (beyond legal constraints)

The responses to the question, "what objective factors – beyond legal constraints – that could justify geo-blocking", reveal a disparity between stakeholders' perspectives. Consumers tend to take the line that no other objective reasons exist, while companies and business associations list numerous grounds for different treatment.

Consumers and consumer organisations

The overwhelming majority of responses from consumers take a general position against all forms of geo-blocking. However, a few individual replies identify some situations where differential treatment could be justified, such as unavailability for after-sale services or very high shipping costs in hard-to-reach areas.

Companies and business associations

Respondents from companies and business associations provide a number of reasons for differential treatment due to e.g. national difference in VAT, health and product safety rules, shipping cost, etc. The views expressed indicate that all current reasons normally invoked (see below) for geo-blocking are considered as legitimate. A number of respondents also take a more general line that companies should be free to set prices within different markets (sometimes referred to as "geo-tailoring"). Others mention the risk of hacking or cybersecurity concerns as a ground to deny access to certain countries. Some companies also refer to the gambling sector, where legal requirements require restrictions in terms of territoriality (e.g. denial of access to licensed gambling websites from consumers from other Member States; IP blocking of access to unlicensed gambling websites or websites based in other Member States).

Member States and national authorities

When asked about possible justifications for geo-blocking, most public authorities agree that nationality or residence alone cannot be considered as legitimate reasons for refusing to sell cross-border. However, several responses also suggest legitimate grounds for such a refusal that may occur in cross-border transactions, including: dispute resolution and after sales services; cost of delivery, security and fraud related reasons; differences in tax rates; different level of purchasing power in the Member State of the consumer and trader, and; different safety standards for products or services.

VI. POLICY RESPONSES

a) Elements of a policy response

Consumers and consumer organisations

When asked about elements of a policy response, the vast majority of individual consumers and consumer organisations strongly agree or agree to more transparency and rules banning certain business practices. Virtually all suggested elements of a policy response are supported by a large majority of respondents.

A large majority (94%) of the relevant respondents favour a ban on discriminatory blocking of access to websites. Nine out of ten (93%) respondents favour a prohibition of refusal to download digital products (such as software or video games). More than eight out of ten (between 84 and 87%) of respondents from this group are in favour of companies having the obligation to explain - either before the transaction or upon request - the reasons for the difference in treatment of customers based on residence/nationality. A similar ratio (between 71 and 87%) of respondents support a list of reasons that may never justify different treatment of domestic and foreign users or a list of reasons that may well justify different treatment. Almost nine in ten (87%) of the relevant respondents agree to rules banning ways and means of discriminatory geo-blocking and other restrictions. Furthermore, a vast majority favours policies requiring traders to obtain consent prior to automatic rerouting. Likewise, a clear majority (70 to 80%) of respondents favour rules requiring traders to accept cross-border transactions from users throughout the EU, either under conditions reflecting additional costs or in cases where users arrange the delivery themselves.

Consumer associations support the suggested EU policy responses, particularly a list of reasons which traders can or cannot use as ground for different treatment.

Companies and business associations

Across companies, opinions are rather divided. More than a half of the relevant respondents strongly disagree or disagree to rules requiring traders to accept cross-border transactions from users throughout the EU under the same conditions as those applied to users of the home country of the provider. Likewise, the majority opposes rules requiring traders to accept those transactions under conditions reflecting additional administrative costs. Rules requiring traders to accept those transactions if the users are willing to organise the delivery themselves are also opposed by the majority.

A relative majority of respondents agree on most policy proposals concerning increased transparency. Just above one in three respondents (35%) believe that "companies should have the obligation to make clear before the transaction the detailed, objective and verifiable reasons why they treat customers differently based on their residence or nationality", five in ten replies (49%) submit that "companies should have the obligation to explain upon request the detailed, objective and verifiable reasons why they treat customers differently based on their residence or nationality". Around a third of the replies (30%) agree to "rules including a list of reasons that may never justify treating domestic and foreign users differently", while around a quarter of the replies (25%) wants policy proposals covering "rules prohibiting traders to refuse the cross-border download of digital products (such as software or video games)".

Around half of the respondents (50%) agree to a policy banning the discriminatory blocking of access to websites. There is no clear majority for or against the other options.

Business associations are generally supportive of the aim to prohibit unjustified geo-blocking while at the same time highlighting the importance of contractual freedom. Many of them call for a clear definition of geo-blocking and what is regarded as unjustified. Some associations do not see a need for new legislation and point to the importance of enforcement of existing rules (Services Directive, Consumer Rights Directive, competition law).

Companies and associations from the gambling sector highlight the legitimate need to block access to websites of illegal operators (consumer protection and public policy reasons).

Some business associations support policy options where a consumer is prepared to organise delivery of a physical product himself, to oblige the trader to non-discriminatory treatment of customers from abroad. That would be under the assumption that this would be within the usual delivery area of the trader and on the same basis as domestic customers or those in the usual delivery area. Similarly, some companies acknowledge that the situation of unsolicited requests and if the trader's law applies is very different situation for a trader compared to active sales.

There is no clear majority across all groups of respondents as to whether respondents favour specific rules for online transactions or for physical purchases and orders.

Member States and national authorities

While most public authorities point to the need of effective enforcement of existing legislation (Services Directive, consumer acquis) several also support targeted instrument to address unjustified geo-blocking.

Most public authorities agree with the need to ensure greater transparency, in particular by allowing consumers to see differences in prices (e.g. by banning automatic rerouting). One Member States expresses strong views against obliging companies to disclose the reasons for geo-blocking upfront and considers that such a measure would place a disproportionate burden on businesses.

Public authorities vary as to the preferred method for addressing geo-blocking practices. Some support the concept of lists of justified or non-justified practices. Others prefer an open approach by which only examples of problematic practices would be provided in the legislation. One Member State points out to the risk that exhaustive lists quickly become outdated. Technologically neutral and principle based legislation is preferred. One public authority supports the option to delegate potential (additional) costs to the consumer whereby the trader has to justify the additional costs.

Some replies support more effective enforcement of the new rules by a combination of enhanced powers of the Commission and more effective cooperation between national competent authorities. Some replies invoke the need to strengthen the Consumer Protection Cooperation (CPC) network.

b) Exclusion of SMEs

When asked whether SMEs, in and particular micro companies, should be exempted, a third of all respondents favour an exemption, while two thirds would not exempt SMEs. Both the consumer and company respondents mirror these exact fractions (one third vs. two thirds). When asked under which circumstances SMEs should be exempted, respondents made several suggestions. Some respondents would accept an exemption only if their small nature practically impedes the offer of goods or services Union-wide, others suggested the use of a threshold/volume criterion based on e.g. the number of employees, company turnover, total revenue or the production volume. Others argue that an exemption would be needed only if the administrative impact would be too large. One business association expressly welcomes exemptions for small firms, however, not by means of a blanket exemption but on a case-by-case basis.

c) Best policy instrument

Consumers and consumer associations

When asked about the best policy instrument, consumers suggest very different approaches. Examples are the introduction of a genuinely free movement of digital and non-digital products within the EU, a removal of all restrictions to the free circulation of goods and an EU legal framework plus national implementation. Better enforcement and the possibility of fines are also mentioned. Overall, many respondents favour harmonised European rules on this topic. Others, however, would not favour a legislative response, suggesting that European Treaties would not allow an intervention to force sellers to be active in specific territories.

Companies and business associations

Companies showed equally divergent views as consumers. Some said there should not be any policy response. Others pointed out the importance of competition. Common rules facilitating cross-border trade are favoured by other respondents. Finally, non-legislative action or guidelines are mentioned as preferred options by several respondents.

Some business associations oppose the enactment of a list of justifications as it would not reflect the complexity of business models would never be comprehensive and soon be outdated.

Member States and national authorities

Only a limited number of replies to this open question were received, but several replies from Member States consider monitoring and enforcement by national authorities in combination with cooperation of national authorities as important. Divergent views are expressed as to whether the existing regime based on Art. 20 of the Services Directive is sufficient.

d) Effective implementation

Overall, the respondents favour almost all suggested ways to improve effective implementation. More than seven out of ten (72%) of the respondents support a monitoring and enforcement role for the European Commission as well as for national authorities. Similar numbers of respondents (70%) support a more effective cross-border cooperation mechanism between national authorities. More than a half of the respondents want alternative dispute resolution bodies to ensure implementation. The support is larger among consumers, with more than seven in ten (respectively 78%, 74% and 77%) to the first three enforcement options. Companies are less supportive of the suggested implementation mechanisms. However, half of the relevant company respondents support monitoring and enforcement by national authorities. As regards the other enforcement options, no majority can be identified among company respondents. When asked to specify, some suggest that the European Commission was best placed to take care of the enforcement due to the pan-European dimension of the problem. Others highlight reservations vis-à-vis the effectiveness of their national enforcement authorities. Additional respondents mention the role of courts. When asked about best practices, including self-regulation of companies, respondents argue either in favour of self-regulation or against it, pointing out that it would not achieve a real solution for consumers.

Member States and national authorities

Most responses from Member States and their authorities support improved enforcement by national authorities. More effective cooperation between national authorities is also welcomed by several Member States.

VII. MARKET IMPACT

a) Overall impact

Consumers and consumer organisations

A large majority (94%) of consumers believe the impact would be very positive or positive. Only a small fraction (4%) believes that the impact would be negative or somewhat negative. Consumers also highlight the opportunities and improved market access. Some argue that, due to increased offers, greater choice and competitive prices, they would buy more. Finally, respondents mention the beneficial impact on migrants and minorities as regards better cross-border access.

The opinions of consumer associations are more divided. All but one of the consumer associations see positively on measures to increase transparency, while a relative majority of respondents (44%) answer that “rules including a list of reasons that may never justify treating domestic and foreign users differently” and “rules banning ways and means of discriminatory geo-blocking and other restrictions” would have a very negative effect on market opportunities.

Companies and business associations

Among companies six out of ten respondents (64%) expect a positive impact on cross border e-Commerce, while similar numbers (59%, 58% and 55%) expect such impact on cross border trade generally, the economy and their business sector respectively. A negative impact on the economy, cross-border e-Commerce, etc. is expected by less than a quarter of the respondents representing individual companies (13% to 25%). Some associations submit that a prohibition of geo-blocking would reduce competition. Because of the de facto obligation to sell, only large traders would be able to comply with such an obligation and economically survive. Smaller traders would not be able to respond to these sales requests and if not competitive enough, this would lead to a concentration in the market. Only big traders would then survive, which would lead to less choice and offers to consumers.

Member States and national authorities

Most responses from Member States and their authorities invoke the principle of freedom to contract which should not be undermined by the new instrument on geo-blocking. They also point out that the legitimate considerations to address unjustified geo-blocking have to be weighed against the risk of putting additional burdens on businesses.

b) Impact of individual measures

Overall, a clear majority of all respondents see a very positive or somewhat positive impact on their activities. An overwhelming majority of respondents (79%) expect a positive impact in case rules are introduced banning ways and means of discriminatory geo-blocking and other restrictions. A similar number of (78%) of respondents expect a positive impact if rules are enacted that prohibit traders to refuse cross border download of digital products. More than seven in ten (73%) respondents expect a positive impact in case of enactment of rules banning the discriminatory blocking of access to websites. There is a clear majority of respondents that see a positive impact also as regards other possible measures such as more transparency i.e. information to the customer, rules requiring traders to accept cross border transactions under certain circumstances and rules including a list of reasons that may or may never justify different treatment.

No clear majority, however, was found as regards the expected impact of rules either specifically addressed at online or offline transactions respectively.

Consumers and consumer organisations

The degree of expected positive impact is even higher if one only looks at the consumer responses. An overwhelming majority of consumers (95%) said that they positive impact in case rules are in force banning discriminatory geoblocking and other restrictions. 88% of consumers see a positive impact if rules are enacted regarding cross border download of digital products. 92% expect a positive impact of banning the blocking of access to websites. Likewise, a clear majority of consumers expects a positive impact of measures on transparency, of requiring traders to accept certain cross border transactions and of a list of reasons clarifying different treatment respectively (81%, 64% and 80%).

Companies and business associations

Based on replies by companies, a majority (56%) expect a negative impact of rules requiring traders to accept cross border transactions under the same conditions as those applied to users of the home country of the trader. Likewise, more than a half (55%) state such negative impacts would have to be expected once rules were enacted that impose on traders to obtain consent of users prior to automatic rerouting. Half of the respondents forecast a negative impact if rules are created including a closed list of reasons that may justify different treatment. However, up to a quarter of companies provide – across possible measures – they are neutral as regards expected impact.

When companies were asked whether such impact would be mostly on the economy, cross border e-commerce, cross border trade generally or their business sector, a clear majority (62%) of companies did not to answer the question. The replies of those who answered are split among the four choices (31% economy, 29% cross border e-commerce, 23% my business sector, 17% cross border trade generally).

When respondents from the group of companies were asked about where an additional burden would be expected to be on, the replies are divided among the four options: respondents chose additional administrative compliance costs (30%), additional costs in the area of marketing or web design (22%), additional personnel costs (24%) and additional delivery costs (25%).

Member States and consumer authorities

Among individual measures listed in the public consultation all respondents from public authorities support rules including a list of reasons that may never justify different treatment of domestic and foreign customers (50% of replies indicate somewhat positive result, 50% very positive).

Three quarters of respondents are supportive of rules including a closed list of reasons that may justify different treatment of domestic and foreign customers, while one forth is neutral.

Respondents are in general supportive of other individual measures listed, with the exception to rules requiring traders to accept cross-border transactions from customers from throughout the EU under the same conditions as those applied to customers of the "home" country of the provider, where a half indicate somewhat negative result. This is mirrored in the replies concerning the rules requiring traders to accept cross border transitions from customers from throughout the EU and to provide delivery cross-border if the customer is willing to organise themselves the delivery and cover the additional shipping costs. Half of the respondents would be against such rules.

VIII. CONCLUSION

The responses to the consultation show that there is overwhelming support for the general principle that consumers and business should be able to purchase goods anywhere in the EU. Across the board, regardless of their geographical location, the responses from consumers and consumer organisations expressed a general dissatisfaction with the current state of affairs. Companies, across various sectors, agree with the description of existing barriers in the single market while urging caution in introducing sweeping new measures seeking to address the situation. Member States also consider geo-blocking to be of high importance.

Regarding the different forms of geo-blocking and other geographically-based restrictions currently being practised a clear majority of respondents agreed to their general significance. Very few additional restrictions were brought up in the responses. Both from the perspective of consumers and companies the replies show a general discontent with the current state of fragmentation of the single market. However, on more specific issues the views of consumer and consumer organisations, on the one hand, and companies and business associations on the other, are somewhat divergent. Although concerned about existing barriers, companies and business associations were more accepting of the current state of affairs, attributing it to divergent legal regimes within different Member States. Consumers and consumer organisations expressed a view of principle that the prevalence of existing practices is an important obstacle to the single market.

The difference in the views expressed by consumer and companies becomes even more pronounced when considering the possible justification for allowing geo-blocking and geographically-based restrictions. Consumer and consumer organisation tend to qualify more of the listed practises as "unjustified", while companies, although more divided in their positions, tend to be more accepting of the number of objective factors that would justify geo-blocking. The replies also give voice to a concern, mainly stemming from companies but also public authorities, that the consultation's separation of "justified" and "unjustified" practices is vague and difficult to apply in practice. If future policy measures address these practises, many of the respondents from companies and business associations underline the need to clearly define - in a legally certain way - what is considered to be a "justified" or "unjustified practice". Similar concerns have been voiced by respondents from Member States which argue for a cautious approach, being wary that attempts to define justifications might in practice lead to unintended consequences or unacceptable infringements of the freedom to contract.

In terms of favoured policy responses to address the current situation, a vast majority of consumers and consumer organisations show support for increased transparency by preventing the blocking of access to websites. An equally large proportion also favours a prohibition to refuse downloading of digital products. Consumers also support a list of reasons which can never justify different treatment, which is also the policy response consumer associations deem as appropriate. From the perspective of companies, a common thread is the emphasis on respect for contractual freedom. Even if the majority of companies support measures increasing transparency, many replies highlight the risk of introducing overly burdensome information obligations. At the same time a majority of company respondents oppose measures that would force them to accept cross-border transactions from users throughout the EU under the same conditions as those applied to users from the provider's home country. A majority of respondents from companies and business associations also reject potential policy responses that would require traders to accept transactions which come with additional administrative costs. They are divided, however, as regards transactions where the customers organise the delivery themselves.

With regard to ensuring effective implementation, there is strong support from consumers for all envisaged mechanisms, be it increasing the role of the Commission, national authorities or providing a better cross-border cooperation mechanism. Companies take a different view with none of the suggested mechanism getting support from a majority of the respondents.

Concerning the expected impact of measures to address geo-blocking, companies take a more reserved view, with a majority expressing concerns about the effect of having to accept cross-border transactions or requiring consumer consent before rerouting. Some business associations also warn that proposed measures would amount to a de facto obligation to sell, which could reduce competition, as small actors would not be able to respond to sales requests. Consumer organisations are supportive of measures on transparency and clarification of non-discrimination. They expect a positive impact of most of the suggested measures. Likewise, the vast majority of consumers expect a positive market impact, with some respondents expecting increased offers, greater choice and competitive prices leading to increased overall consumption.

Further consultation activities

The Commission exchanged views with stakeholders in numerous meetings, in particular:

Meetings with Member States in January 2016.

Meetings with business and consumer organisations in 2015/2016.

STAKEHOLDER CONSULTATION WORKSHOP

On 18 February 2016, the Commission held a workshop on geo-blocking and geo-discrimination in Amsterdam, with a good balance of representation of stakeholders. 49% of participants represented business interests, 30% consumer interests (including European Consumer Centres) and the remaining was composed of various organisations, including enforcement authorities.

Participants discussed and commented on possible policy options as set out in the Inception Impact Assessment. Some options were welcomed, others were criticised by stakeholders.

Convergence of views occurred on the following aspects: Stakeholders would favour enhanced transparency for consumers, if that would not become overly burdensome for businesses. Stakeholders asked the Commission to address the drivers of geo-blocking and discrimination first rather treating the symptom. The policy option of a "black list" was seen risking to be too prescriptive, burdensome and not future-proof. The policy option "shop like a local" raised concerns by some stakeholders, particularly traders who pointed to the risk of being made forced to apply foreign law.

The relevant remaining barriers to cross-border trade were discussed (e.g. applicable law, taxes, labelling, right of return, warranties).

SME Representatives considered there is no need for new legislation on transparency, which would involve high compliance costs (translation of websites). The view was that more coordination was needed. Consumers considered however that transparency would increase market opportunities for companies and could increase their market share, although it could be more complicated for SMEs. Some enforcement authorities questioned the need for transparency if there was no eventual access to the product or the service.

The possibility to pick-up goods at the location of the trader with no obligation to deliver cross-border was discussed. Businesses questioned whether it would require a pure online player to set up physical pick-up points. Most considered that this should be based on a voluntary approach rather than an obligation. Questions were raised as to the compatibility with the logistical chain of several traders. Possible effect on prices was discussed. Consumer representatives stressed the complexity of the proposal. In their view, it can help in neighbouring countries or for very expensive goods (buying a car cross-border).

On purely digital services, businesses focused on how to address the various drivers spurring companies to restrict cross-border transaction, especially the issue of payments systems not supporting cross-border sales. Questions were also raised on the scope of this option

For the policy option "same location, same deal", the scope was discussed and clarified

Finally, with respect to the reasons that should or should not be accepted as being based on objective criteria, discussions pointed at the complexity of this issue an. Generally, traders would favour a stable and clear list of non-acceptable justifications, which should however remain sufficiently broad so as to take into account technological and economic developments. They stressed that the issue of the applicable consumer legislation generates a uncertainty for traders. They also stressed, with regard to refusals of means of payment, that such refusals are often due to the payment service provider and not the trader himself. Consumers generally were on a similar line, favouring more "black list" approach but stressing that devising it would practically be very complex.

Annex 3: Who is affected by the initiative and how

Stakeholders who would be affected by the initiative as contained in the preferred policy option 3 (equal treatment in specific situations "shop like a local"):

Consumers searching and shopping across borders for tangible goods and those buying digital services or services provided at the premises of the trader would benefit from the initiative:

oTangible goods

Consumers could compare offers and prices cross-border without facing discriminatory practices including geo-blocking. They would have access to goods and services like domestic consumers in the country of the trader, i.e. at the same conditions. However, as in order to take advantage of this possibility, they would need to organise the transport of the good across the border themselves. They could either pick up the good themselves, which would often be the case in border regions. It would also help consumers who want to send gifts to a person in another Member State or consumers who travel, study or work abroad. Alternatively, consumers could contact a parcel delivery intermediary, which consumers in some Member States already do. It should be noted that in order to find the intermediary, they will typically not be able to rely on the seller, as such an intervention by the seller in the delivery would trigger VAT registration requirements which many sellers will want to avoid.

As a result, consumers will be able to benefit from lower prices in other Member States and buy a larger variety or quality of goods available in the EU. Both online and offline prices can be expected to decrease slightly, which increases consumers' purchasing power and real expenditures. In the example of the electronic goods sector, price decreases are estimated at -0.5% offline and -0.6% online, on average across the EU. Based on the analysis of the electronic sector, the total EU28 consumer surplus increase from online purchases of tangible goods could amount to 0.8% on average for the EU28, with variations across Member States..

Consumers would be treated as domestic customers and, hence, would have to accept the language of the trader and certain legal rules of that country.

oDigital Services

The abolition of geoblocking will allow foreign consumers access to services in all Member States, at the same prices as domestic online shoppers in these countries. The major difference compared to tangible goods is that in this situation there are no transport costs increasing potential price arbitrage. Unrestricted access to all offers could in theory lead to progressive convergence of prices in excess of transport costs. However, consumers also face information costs resulting from language differences and the fact that most price comparison sites are national.

As regards non-audiovisual online content services, although at this stage non-discrimination principle is not applicable to them, from the outset consumers would not be blocked or rerouted without their consent when accessing the online interfaces from other Member States and should not be discriminated as regards payment means..

oServices consumed at the location of the traders (such as hotel rooms or rental cars)

Consumers could not be denied access to a services or made to pay different prices based on their place of resident or their nationality, in situation where the services is used by the consumer at the place of establishment of the trader or the location where the trader provide the services. The situation will mainly affect situations where the consumer is temporarily visiting another Member States be it on a tourist, family or business trip, where consumers reside in a Member States where they do not hold the nationality or those residing in cross-border regions. The transport costs inherent in this situation are likely to keep price arbitrage to a minimum. In fact, empirical evidence suggests that price arbitrage in practice might not lead to price convergence, even where transport costs are minimal.

Businesses

oCompanies that currently automatically reroute website visitors would have either to abandon this rerouting practice or install an approval procedure for website visitors to agree. This concerns a small number of mostly larger companies – 2% of website operator practice automatic rerouting, and in order for this to be possible, the company must operate at least two distinct websites.

oCompanies selling tangible goods, digital services or services consumed physically (hotel rooms, rental cars etc.) that currently treat customers differently on the basis of nationality or place of residence would have to accept foreign buyers at the same conditions as national customers. For traders operating online, this may either require modifications to the interface in order to ensure that details of foreign buyers can be entered, or the installation of an alternative possibility (e.g. e-mail ordering). It is important to note that under this option traders would not be obliged to deliver tangible goods across borders. In order to avoid VAT registration requirements in other Member States, they would also have to abstain from putting customer in contact with parcel delivery intermediaries.

In so far as companies still treat customers differently, they would have to set up a mechanism to respond to requests for justification (e.g. via e-mail) or to provide an explanation on their website.

If the company had previously several distinct websites/points of sale with different conditions, it is possible that there will be a limited shift of sales channels, with some customers moving to the more favourable sales channel.

oCompanies that source cross-border goods or services for which they are end-users (i.e. supplies which are neither sold nor transformed) could take advantage of the same increase in variety and better prices as consumers.

Member States

Member States' authorities responsible for consumer protection will need to familiarise themselves with the new rules.

Annex 4: The economic impact of lifting geo-blocking restrictions

1 Introduction

This annex summarises the method used to assess the economic impact of removing geo-blocking (GB) practices in the EU and the results obtained. A more detailed explanation of both issues can be found in Duch-Brown and Martens (2016).

The 2015 Mystery Shopping Survey (GfK, 2016) provided evidence and measurements on the prevalence of GB at different stages in the cross-border shopping process by product category. The objective of this annex is to quantify the economic impact of the (partial) elimination of GB restrictions. This is done in several steps. First, we estimate the impact of lifting all GB restrictions on cross-border e-commerce for a limited set of online electronics goods in 10 Member States that represent about 85% of the EU electronics market. The Mystery Shopping Survey showed that electronics goods are subject to widespread GB restrictions; lifting the restrictions could have a strong impact in that sector. Moreover, we have detailed data for electronics goods sales that enable us to estimate the impact of lifting trade restrictions on buyers (consumers) and sellers (producers), both in the online and offline distribution channels. Without GB restrictions, buyers can price-arbitrage between country markets and find new product varieties that are not available in their domestic markets. Sellers would adjust prices in response to these new market conditions. Second, starting from these overall results for a scenario whereby all GB restrictions are lifted, we estimate the economic impact of intermediate policy scenarios that provide for a partial lifting of GB restrictions only. Some cross-border online trade restrictions remain in place and result in relatively higher trade costs and barriers, compared to the full lifting of all GB restrictions. We simulate the impact of these relatively higher trade costs and the corresponding decline in economic benefits, compared to the full GB scenario. Third, we expand the estimates obtained from this limited electronics sales dataset to all other online goods sales and to all EU28. We apply the Mystery Shopping Survey results on the prevalence of GB restrictions and cross-border price difference in other online products.

We find that lifting GB restrictions has a positive impact on consumer and producer surplus in all EU MS. Overall, consumer surplus increases by 1.2% in the EU28 and producer surplus (profits) by 1.4%. The relative impact is especially strong in some smaller MS where consumer choice is currently relatively limited and price competition is lower because of limited market size. Consumer gains are stronger than producer gains in these countries. However, for the overall EU28, producers/sellers benefit slightly more – in relative terms – than consumers because of economies of scale and cost reductions when online sales increase. Absolute figures, however, are higher for consumer surplus than for firms' profits. The overall result is largely driven by the largest EU economies. Partial lifting of GB restrictions reduces the magnitude of the benefits but still generates a positive economic impact.

The remainder of this annex is structured as follows. Section 2 explains the methodology employed. Section 3 discusses the data sources. Section 4 presents the results, for each stage in the simulation process.

2 Modelling methodology

The ideal methodology would have implied the estimation of a general equilibrium trade model linking the observed extent of GB for different products groups to bilateral cross-border trade volumes by product category. However, there are no detailed online cross-border trade data by country pair and product category for the EU. In the absence of such data the estimation of the trade response to lifting GB related trade restrictions is not possible. In addition, there is no computable general equilibrium model at the EU level that explicitly incorporates e-commerce. Solving these severe modelling restrictions required alternative solutions in order to estimate the economic impact of removing GB.

The quantitative methodology adopted can be conceptualised as a partial-equilibrium structural bottom-up approach. The model is grounded in a partial-equilibrium framework since it uses very detailed online and offline price and quantity data to identify with high precision consumers' substitution patterns in only four product categories and ten EU Member States. It is structural in the sense that economic theory is used to develop statements about how a set of observable endogenous variables are related to another set of observable explanatory variables, and sometimes also to a set of unobservable variables. However, economic theory alone cannot provide enough information for the estimation of the model. For this reason, there is a need to add statistical assumptions about its observed and unobserved variables. A key reason to use economic theory, beyond the specification of the relationship between the variables, is to clarify how institutional and economic conditions affect these relationships. This specificity is essential to make causal statements about the estimated relationships, or use them to perform counterfactuals, i.e., scenarios that have not been implemented but that can represent the likely outcomes of policy interventions. Finally, it is bottom-up because from the detailed results obtained for the four product categories, we extrapolate to the rest of the online economy and the other MS –using proportions – to produce a figure for the whole EU-28.

2.1 The core model

The methodological approach is framed in the tradition of structural estimation in empirical industrial organisation in the economics profession. This approach uses discrete choice models for the estimation of demand and adds a simulated supply side to compute the industry equilibrium given by the observed data. Adding a simulated supply side to account for firms' pricing behaviour, the observed market equilibrium can be found. Moreover, by changing supply or demand conditions, the framework allows for the design of counterfactuals that simulate policy changes.

The model used here is a modified version of Duch-Brown et al. (2015), and was developed by researchers from the JRC/IPTS. The model is a partial equilibrium approach using detailed pricing data for both online and offline sales of specific products, allowing estimating with a high level of accuracy demand substitutability and market equilibrium.

From a market analysis perspective, there are three potential competitive constraints: demand substitution, supply substitutability, and potential competition. Demand substitution constitutes the most immediate and effective disciplinary force on suppliers, and in particular to their pricing decisions. Supply substitutability and potential competition are relevant in the medium to long terms, since they imply the need of adjustments through tangible or intangible assets, additional investments or strategic decisions, all of which would imply significant changes in the markets under consideration 104 . Hence, a precise estimation of demand substitutability is essential to the analysis of the effects of changes in the institutional setting of a given sector, and this is the basis for the approach taken here.

We consider the demand for four consumer electronic products: smartphones, tablets, desktops and laptops. Consumers can choose among a large variety of products that are differentiated in quality. Furthermore, consumers can either purchase these electronic products in a traditional brick-and-mortar shop (offline) or they can purchase the products through an online distribution channel. Finally, consumers can also decide not to buy an electronic product at all, in which case they can spend their money on other goods. To model the substitution patterns, a two-level nested logit model is used which allows for market segmentation according to two discrete dimensions: i) quality, which can be either high or low; and ii) the distribution channel, which is either offline or online (as in Duch-Brown et al., 2015). This model is useful since the nesting parameters enable one to assess to which extent consumers view products in the same distribution channel and/or quality category as closer substitutes.

Assuming that consumers choose the product with the highest utility, one can obtain the choice probabilities for every product in every country, including the probability of purchasing the outside good (McFadden, 1978). At the aggregate level, these choice probabilities can be equated to the market shares, relative to a hypothesised potential market, defined here as representing 40% of country population. 105 The demand model can be used to compute consumer surplus (McFadden, 1978 or Anderson et al. 1992), and to compute the own- and cross-price elasticities of demand (see Duch-Brown and Martens, 2016 for details). If the model conforms to the basic principles of consumer theory, the model translates preference correlations into aggregate substitution patterns. Products in the same subgroup will have higher cross-price elasticity than products in a different subgroup. Hence, one can therefore assess to which extent the online distribution channel substitutes for the traditional brick-and-mortar channel, or to what extent it provides a new source of differentiation that raises total sales for consumer electronics rather than displaces existing sales.

An oligopolistic supply side is added to the model to infer marginal costs and current economic profits; as well as to define the observed market equilibrium. The model assumes that firms maximize profits, and that they compete in prices in a differentiated products setting (Bertrand competition). As shown by Berry (1994) and Berry, Levinsohn and Pakes (1995), the profit maximising conditions can be used to compute the current marginal costs. Furthermore, this system can be used to perform policy counterfactuals, and in particular the effects of removing the GB restrictions to cross-border e-commerce. The model also calculates consumer welfare (consumer surplus) and producer welfare (profits) changes, by computing the welfare measures in the different counterfactuals and in the observed market equilibrium.

2.2 Policy options

The policy counterfactuals implemented assume that, from an observed equilibrium, lifting GB restrictions will allow consumers to buy online from the cheapest provider, thus partially arbitraging away price differentials. With the help of the model, the system of profit maximising conditions can be used to compute the counterfactual equilibrium prices if online products have become accessible across the border. It is possible, in turn, to compute the counterfactual sales, profits, and consumer surplus. The model includes both transport costs for every country pairs as well as VAT rates, which would also have an effect on price differences.

A further step in the model is to incorporate potential reactions from retailers from lifting GB restrictions. One way to do it is to assume a two-stage game in which the first stage is described by the consumers' reaction to an integrated EU online market. In this stage, we assume that consumers are able to buy online from the cheapest provider, independently of their respective locations. This triggers online cross-border purchases. Both offline and online retailers in those markets more affected by sales diversion will react, undercutting prices to adapt to the new competitive scenario, basically effective competition from cheaper prices available online abroad. At this stage, firms recognise that consumers can now buy from anywhere and adjust their prices accordingly. This will bring prices further down, both offline and online. The new equilibrium will also change consumer and producer surplus. Consumers will face even lower prices. However, it is a-priori uncertain how consumers will react when both offline and online domestic retailers cut prices. It may or may not be cheaper to purchase across the border. Due to the existence of transaction costs, a simple price reduction could make the domestically available products cheaper for domestic consumers. Domestic firms will sell more, but with a lower mark-up, so the net change in profits is an empirical question that cannot be settled a-priori.

This scenario describes a situation where there is "full lifting of all GB restrictions". We translate that scenario in costless price arbitrage and search for new varieties in online cross-border shopping, subject to parcel delivery costs plus a trade cost margin of 25% on top of the parcel delivery costs. A less ambitious policy scenario is also simulated that only partially lifts GB restrictions. The "shop-like-a-local" option implies delivery of the order in the country of the trader from where it can be picked up by the buyer or by a transport intermediary who delivers the good to the buyer in another country. We assume that these intermediary services add a fixed trade of 20€ per purchase, irrespective of the country of residence of the buyer and seller. There are already a number of online firms that provide this type of intermediary shipment services to help consumers get around GB restrictions. They have different fee structures, either flat rates (around 5-10€ per parcel) or a surcharge on shipping rates, or a combination of both. We combine all these options in a fixed rate of 20€ on top of parcel delivery costs. In summary, the less ambitious policy scenario results in model simulation with relatively higher trade costs than the "full" scenario that lifts all GB restrictions.

We do not simulate results for the "improving transparency" and "banning blocking of access" policy options because their quantitative impact on cross-border online trade is likely to be marginal. Under the transparency option, web shops should indicate clearly at the beginning of the ordering process whether any delivery restrictions apply and inform consumers upon request about the justifications for different treatment. We assume that this makes no difference in the number of cross-border purchases. The "Banning blocking of access" option would prohibit denial of access to a website. Currently, about 98% of all websites can be accessed and 49% deliver the product order (See Annex with results from the Mystery Shopping Survey). An increase to 100% access implies a 2% increase in the 49% that are already successful at the moment, or 50% successful orders in total. The difference compared to the current situation is only marginal and falls within the error margins of the estimation results.

2.3 Extension to all online sales

In order to extend the results from electronics products to all online goods sales, we use a simple linear extrapolation method that proceeds as follows:

First, we extend the results from the 4 electronics products to all electronics sales. To do that, we use the share of these products in the aggregated consumer electronics industry. On average, these four products represent about 60% of total consumer electronics retail sales, with differences across countries. We linearly extrapolate our estimated measures of welfare and sales by country to cover the entire electronics sector using the corresponding shares for each of the 10 countries for which we have detailed data, assuming that these 4 products are representative for the entire electronics products sector, and that the share would remain constant in the absence of GB. Second, we extend geographically from EU10 to EU28. The 10 countries in the sample represent 85% of the EU-28 consumer electronics industry. We have information of the share of each of the countries not included in the detailed database, so we can impute a value that corresponds to the observed share in 2014. For this purpose, using the share of electronics in total retail sales by country, and assuming proportionality, we linearly extrapolate to get the corresponding values for each one of the missing countries. We can then compute also the EU28 aggregate for consumer electronics. Third, we extend from electronics to all online goods sales. In a similar procedure, we use the share of electronics in total e-commerce by country, considering that this proportion does not change across countries and linearly extrapolate the results. It is again possible to compute the EU-28 aggregated figure.

Finally, we introduce corrections for two indicators that change with respect to our benchmark sector (consumer electronics): price differences between country markets and the extent of GB. First, we take the ratio of average cross-border price differences between the benchmark product (electronics, as observed in the detailed GfK dataset) and other product categories (as observed in the more aggregated Mystery Shopping Survey). We weight the figures obtained before by this measure in order to take into consideration that more price arbitrage will occur in sectors with higher online price differences between countries. Second, we take the ratio of the probability that GB occurs in our benchmark product (electronics) and in the target product (as measured by the Mystery Shopping Survey) to correct for the potential trade expansion in each product category after lifting GB restrictions. In so doing, the procedure considers also the maximum potential trade creation opportunities derived from online price differences and GB intensities. Detailed results in Duch-Brown and Martens (2016) based on the Mystery Shopping Survey indicate that cross-country price differences in clothing are 4% higher than in consumer electronics. Hence, there is more price variation in clothing than in consumer electronics. Moreover, attempts to buy clothing online cross-border are only 82% as likely to get geo-blocked as attempts in consumer electronics. In other words, clothing has more cross-border price variation but less GB, compared to consumer electronics.

2.4 Model assumptions, limitations and simplifications

There are two dimensions were assumptions and limitations apply: i) the core model; and ii) the extrapolation method. Explanations in terms of assumptions, limitations and simplifications are given first for the core model, and then for the extrapolation exercise.

The core model is based on discrete choice models of product differentiation. These models have gained considerable importance in empirical work in economics in the past two decades. Because they treat products as bundles of characteristics, they offer the possibility to uncover rich substitution patterns with a limited number of parameters. Berry (1994) developed a framework to estimate a class of discrete choice models with unobserved consumer heterogeneity based on aggregate sales data. His framework includes the random coefficient logit model of Berry, Levinsohn and Pakes (1995), the nested logit model (with special random coefficients on discrete product characteristics) and the logit model (without consumer heterogeneity). This is the state of the art in the literature. Alternatives based on considering homogeneous products 106 , representative consumers, preferences over goods instead of characteristics, are inferior in both theoretical and empirical terms.

The logit and nested logit models have been popular because of their computational simplicity. At the same time, they have long been criticized because they yield too restrictive substitution patterns. In particular, the nested logit model allows products of the same group to be closer substitutes than products of different groups, but the aggregate substitution patterns remain restrictive: cross-price elasticities within the same group are symmetric, and substitution outside a group is symmetric to all other groups. In contrast, the random coefficients logit model incorporates random coefficients for continuously measured product characteristics. This creates potentially more flexible substitution patterns, where products tend to be closer substitutes as they have more similar continuous characteristics. However, the random coefficients model is computationally more demanding, and several recent papers have studied a variety of problems relating to its numerical performance (Knittel and Metaxoglou, 2008; Dubé, Fox and Su, 2012; and Judd and Skrainka, 2011).

In this respect, Grigolon and Verboven (2014) suggest that the choice between the more tractable nested logit model and the computationally more complex random coefficients model depend on the application. In the present analysis of removing GB restrictions, consumer heterogeneity regarding the offline/online origin is particularly relevant, and the nested logit model should capture this reasonably well. Needless to say, the model used would gain from a more flexible specification of substitution patterns, or at least from a combination of the two.

On the supply side, the most relevant assumption is that of price competition. However, this is also a standard assumption in the literature, reflecting the fact that prices tend to be considered more relevant from a differentiated product perspective than competition in quantities or different strategic variables. Additional assumptions are necessarily imposed by the lack of appropriate data. For instance, the data does not include separate information about manufacturers and retailers. Hence, we assume vertical integrated firms. The database, however, includes information about brands. Although there is no complete correspondence between brands and manufacturers, the best option is to define each brand as a multi-product firm. This assumption introduces some degree of competition in each country, since the number of brands sold by country is different, but normally large.

The main counterfactual completely removes all GB restrictions and allows consumers to buy online from the cheapest provider, thus partially arbitraging away price differentials, subject to trade costs. In addition, consumers are able to access a wider choice set of new products that were not accessible because of GB. This implicitly assumes that consumers can incur in costless search and purchasing activities. In order to take this into consideration, trade costs are inflated to accommodate other "costs". The alternative “shop like a local” scenario allows online sellers to avoid additional cross-border trade costs by delivering the goods to a domestic address in the country of establishment of the shop. Hence, the buyer has to arrange for pick-up, either in person or through an intermediary. Such commercial intermediaries already exist as a way to overcome GB restrictions and they charge an additional cost for this task. In the counterfactual simulation, we add a fixed fee of 20€ per purchase, over and above physical transport costs, to capture the cost of these intermediary delivery services.

An important caveat is that the model does not take into account that sellers may start making losses on some products because of increased price competition and exit the market nor the fact that, in order to export, firms may incur in additional costs. Some smaller sellers may also merge to gain more from economies of scale and become more competitive. Additionally, the simulations assume that the change in total welfare is the sum of total consumer surplus and producer surplus. Although we use VAT rates, we do not compute tax revenues. Moreover, the exogenous variables in the model are assumed not to change after removing GB. In particular, the exogenous part of utility 9consumer side) and marginal costs (supply side) remain unchanged. This means that the focus is thus entirely on the quantification of the allocative effects of removing GB. A more complete analysis would also incorporate the efficiency effects, which may enter through changes in utility, marginal costs or fixed costs.

The extrapolation is a simple exercise that tries to give an overall picture of the effects obtained by the core model. In this case, the limitations come mostly from the lack of appropriate data to translate the estimations of the four product categories and ten countries to the rest of the online EU economy. Please refer to Duch-Brown and Martens (2016) for a more detailed discussion.

2.5 Appropriateness, validation and robustness

These types of models are common in the industrial organisation literature and in competition policy analysis in particular. As a matter of fact, these models are used regularly by competition authorities around the world to simulate the effects of mergers, and other issues related to market structure, trade, vertical restraints and competition. The model used here has been developed by the JRC/IPTS exclusively for the impact assessment study presented. A similar model was produced by JRC/IPTS researchers in collaboration with researchers from the University of Leuven and Telecom ParisTech. The original purpose of that model was to study the economic impact that the appearance of the online distribution channel has had on the European Economy (Duch-Brown et al. 2015). A similar model was developed by the JRC/IPTS to analyse the welfare effects of trade in digital music (Aguiar and Waldfogel, 2014).

This class of econometric models is a useful tool to evaluate the effects of public policies in oligopolistic industries, to understand business strategies, or to identify collusive or anti-competitive behaviour. The original model has been presented in several workshops and conferences and discussed broadly with scientists, academics and e-commerce stake-holders. It relies on similar but older data than the one used in this report. The data comes from a well know marketing company that indicates that their coverage of sales of the products and countries they supplied is well above 90%. Hence, product market coverage is extensive supporting the validity of the results. Scientific experts in the fields of modelling and data management have been consulted and both the modelling approach and the richness of the data have been confirmed.

Several counterfactual scenarios have been estimated to corroborate the results of the model. In addition, several parametric modifications have also been introduced in order to test the model consistency. All these tests are reported in the main text and they all suggest that the model results are consistent. Efforts have been put in place to assure that the maximum market coverage is achieved and that the modelling tool is effective and up-to-date, in order to address both the potential uncertainty in the model results and also the robustness of the model results to changes in its structure.

3 Data

We use data from three different sources: detailed online and offline sales data for electronics goods for 10 EU MS from the marketing data company GfK; data on price differences and the probability of GB for other product categories from the 2015 Mystery Shopping Survey; and general e-commerce market data for the EU from Euromonitor.

We use data on online and offline sales of electronic goods for the simulation of the impact of lifting GB restrictions. We selected this dataset because it was the only dataset available in the EU market with sufficiently detailed information on monthly online and offline sales volumes in several Member States, prices and detailed characteristics of the goods over a sufficiently long period of time. Moreover, electronics products are an important online product, representing 13% of all online sales. We selected four electronics products (smartphones, laptops, tablets and desktops) that represent about 60% of all online electronics sales. The 10 countries for which we have data represent about 85% of the total online market in the EU. Last but not least the Mystery Shopping Survey demonstrated that electronic goods are most likely to be subject to GB (GfK, 2016). An earlier 2009 Mystery Shopping survey had already come to similar conclusions (Cardona and Martens, 2014). This enables us to use electronic goods as a benchmark for the GB impact simulations. Geo-blocking will be less severe in other product categories.

We acquired data on online and offline sales of 4 types of electronic goods in 10 EU MS for the period Jan 2012 – Jun 2015 107 . For each country, the data include the brand name and model label of each product, the technical characteristics of the products, the volume of sales and the average price in each distribution channel and country. We also relied on Euromonitor International's Passport Database which includes detailed information on retail sales and particularly online retail by sector, at the country level. These data are required to produce an estimation of the impact for all online sales and for all EU-28.

Table 1 gives an impression of the extent of geographical market segmentation in the EU for these 4 electronic products. Few models are sold in all countries but the models that are widely sold account for a large share of total sales, especially for tablets and smartphones. While average price differences between online and offline prices are rather low, the average difference between the highest and lowest priced product across countries and models can be very substantial. These data give an indication of the large potential for price arbitrage and searching for product varieties (models) when GB restrictions would be lifted.

Table 1: Indicators of market segmentation in the EU

		Models sold in
		All countries		Only 1 country		Price differences (%)
Product category	Models	% models	% sales	% models	% sales	Offline vs. Online	Offline	Online
Desktops	33253	0.6	18.9	75.2	20.8	2	42	30
Tablets	6398	3.0	64.3	60.6	11.7	3	31	38
Smartphones	6136	6.8	77.4	47.9	2.8	2	47	57
Laptops	66473	0.4	16.4	75.2	34.3	9	31	26

Notes: Average price difference between the highest and the lowest price for the same product.

Source: JRC/IPTS calculations with data from GfK.

Online sales overall have grown over the period 2012-2015, from 13.9% to 17.5% of total retail sales (Fig. 1) in these four product categories. Figure 2 shows how the proportion of online and offline sales varies significantly across the 10 EU MS covered by the dataset, from only 4.1% online sales in Italy to 31.4% in Slovakia. On average, there are also significant price differences between countries (Figure 3). Price differences are much wider at individual product level. Table 2 shows the share of e-commerce in total retail by country. It also shows the distribution of e-commerce in 2014 by MS. In the first case, Denmark and the UK have the highest proportions of e-commerce over retail. In the second case, the UK, Germany and France represent almost 70% of total e-commerce in the EU in 2014.

Figure 1: Evolution of e-commerce as a share of total retail

Note: Product-level data covering 4 product categories and ten countries Jan 2012 to Mar 2015. The aggregate figure for the EU in 2014 was 7% (Duch-Brown and Martens, 2015).

Source: JRC/IPTS calculations with data from GfK

Figure 2: Online sales as a proportion of total sales

Note: Product-level data covering 4 product categories and 10 countries Jan 2012 to Mar 2015.

Source: JRC/IPTS calculations with data from GfK.

Figure 3: Potential for price arbitrage across the EU

Note: EU-10=100 in each channel.

Source: JRC/IPTS calculations, data from GfK.

Table 2: Market shares of e-commerce

	Share of e-commerce in
Country	Domestic retail	EU e-commerce
Austria	5.2	1.8
Belgium	5.7	2.4
Bulgaria	1.3	0.1
Croatia	1.5	0.1
Cyprus	3.5	0.1
Czech Republic	6.9	1.1
Denmark	11.0	2.5
Estonia	5.5	0.1
Finland	9.0	1.9
France	6.2	15.1
Germany	7.2	18.9
Greece	3.3	0.7
Hungary	3.2	0.5
Ireland	8.1	1.4
Italy	2.3	3.8
Latvia	2.7	0.1
Lithuania	4.0	0.2
Luxembourg	4.3	0.2
Malta	5.9	0.0
Netherlands	8.1	4.5
Poland	5.8	2.8
Portugal	2.6	0.6
Romania	2.4	0.4
Slovakia	4.6	0.4
Slovenia	2.6	0.1
Spain	3.1	3.4
Sweden	7.1	2.7
United Kingdom	12.2	34.3
EU-28	6.8	100.0

Source: Euromonitor.

4 Results

4.1 Policy Scenario 1: full lifting of all GB restrictions

The results from all these calculations are brought together in a few tables. Table 3 reports additional cross-border trade in electronics products after the lifting of all GB restrictions 108 . Trade increases by nearly 630 Mln €. This can be decomposed in a trade diversion and a trade expansion effect. Trade diversion occurs because consumers shift from buying offline and online in their home market to importing online because foreign markets are cheaper and offer new varieties of goods. Trade expansion occurs because electronics products become cheaper on average and consumers buy more as a result. Table 4 shows these substitution and net increase effects. The UK and Poland gain most in terms of trade. Both are very competitive markets. The decline in domestic sales hides two opposing forces: domestic consumers who shift to cross-border purchases and foreign consumers who start buying in your country market.

Table 3: Additional online cross-border trade in the lifting of all GB restrictions scenario (Mln €)

	BE	DK	FR	DE	GB	IT	NL	PL	SK	ES	Imports
BE		0.1	0.4	1.6	18.8	0.0	0.2	23.1	0.1	0.1	44.4
DK	0.0		2.2	5.5	21.2	0.1	1.8	100.6	1.7	1.9	134.8
FR	0.1	-0.2		1.0	45.3	0.2	1.3	43.7	0.3	0.6	92.2
DE	0.9	10.1	7.9		49.4	2.5	6.6	75.5	1.5	1.2	155.6
GB	0.2	0.0	0.7	1.5		0.0	2.2	7.6	0.1	0.1	12.5
IT	18.8	0.4	4.7	2.2	25.4		0.7	46.4	0.3	3.3	102.3
NL	0.0	0.4	1.0	1.0	23.9	0.1		45.1	0.2	0.4	72.1
PL	0.0	0.8	0.6	0.2	2.0	0.0	1.2		0.1	0.2	5.1
SK	0.0	0.0	0.2	0.3	0.6	0.0	0.1	4.2		0.4	5.8
ES	0.0	0.0	0.2	0.0	-3.8	0.0	0.1	7.1	0.0		3.8
Exports	19.9	11.7	17.8	13.3	182.9	3.0	14.1	353.4	4.4	8.2	628.6

Source: Duch-Brown and Martens (2016).

Table 4: Changes in sales volumes between online and offline channels, net trade effect (in Mln € and %)

		Change in volume				Percentage change
		Offline	Online		Net effect		Total	Offline	Imports	Exports
	Market size		Domestic	Cross-border
BE	1,569	-23.4	-0.3	19.9	-3.8		-0.2	-1.5	2.8	1.3
DK	1,759	-89.6	-0.7	11.7	-78.6		-4.5	-5.1	7.7	0.7
FR	6,448	-101.1	-7.6	17.8	-90.8		-1.4	-1.6	1.4	0.3
DE	14,033	-213.8	67.2	13.3	-133.3		-0.9	-1.5	1.1	0.1
GB	6,816	-15.9	-4.0	182.9	163.0		2.4	-0.2	0.2	2.7
IT	5,663	-78.9	4.1	3.0	-71.8		-1.3	-1.4	1.8	0.1
NL	3,007	-74.2	7.1	14.1	-53.0		-1.8	-2.5	2.4	0.5
PL	1,679	-1.4	-0.8	353.4	351.1		20.9	-0.1	0.3	21.1
SK	232	-3.4	-0.1	4.4	0.8		0.4	-1.5	2.5	1.9
ES	3,649	-5.6	8.4	8.2	10.9		0.3	-0.2	0.1	0.2
Total	44,854	-607.5	73.4	628.6	94.5		0.2	-1.4	1.4	1.4

Source: Duch-Brown and Martens (2016).

Table 5 shows the dynamic impact of lifting all GB restrictions on electronics products prices in each country (by policy scenario, see below). Prices decrease across the board in all countries, both online (-1% on average) and offline (-0.5% on average). The decrease is stronger is more highly priced country markets as competition increases. Offline prices also decrease because there is more competition from online sellers.

The main question is not so much the increase in cross-border trade but how it affects the welfare of consumers and producers (or sellers). Trade is not an economic policy objective in its own right; it is a means to increase the well-being of citizens. For the 4 electronics products we estimate that consumers gain about 500 Mln € or 0.7% in consumer surplus thanks to lower prices and increased variety of products available. Producers gain 283 Mln € or 1.3% in profits from these new trade opportunities – all this compared to a total market size of about 45 Bln €.

Table 5: Impact on prices (% change)

	Full removal		Shop like a local
	Offline	Online	Offline	Online
BE	-0.7	-1.2	-0.7	-0.7
DK	-0.5	-2.2	-0.5	-1.4
FR	-0.6	-0.6	-0.6	-0.3
DE	-0.5	-0.8	-0.5	-0.3
UK	-0.2	-0.2	-0.2	-0.1
IT	-0.7	-1.1	-0.7	-0.6
NL	-0.5	-0.8	-0.5	-0.4
PL	-0.5	-0.6	-0.6	-0.3
SK	-0.5	-2.9	-0.5	-1.9
ES	-0.8	-0.5	-0.8	-0.2

Source: Duch-Brown and Martens (2016).

We the above estimates we can now estimate the EU-wide impact of a hypothetical policy scenario that would remove all GB restrictions for all goods in all MS. In Figure 4 we extend the results from the above analysis to all online sales in all EU28. We find that lifting GB restrictions has a positive impact on consumer and producer surplus in all EU MS. Overall, consumer surplus increases by 1.2% in the EU28 and producer surplus (profits) by 1.4%. The impact is especially strong in some smaller MS where consumer choice is currently relatively limited and price competition is lower because of limited market size. Consumer gains are stronger than producer gains in these countries. However, for the overall EU28, producers/sellers benefit slightly more than consumers because of economies of scale and cost reductions when online sales increase. The overall result is largely driven by the largest EU economies. Partial lifting of GB restrictions reduces the magnitude of the benefits but still generates a positive economic impact.

Figure 4a: Impact of the full GB removal scenario (EU28, all online products)

CS = consumer surplus; PS = producer surplus.

Source: JRC/IPTS calculations.

Figure 4b: Impact of the full GB removal scenario (EU28, all online products), Bln€

CS = consumer surplus; PS = producer surplus.

Source: JRC/IPTS calculations.

4.2 Policy scenario 2: "shop like a local"

In the "shop like a local" scenario the shop deliver to a domestic address in its country of establishment only; the buyer has to arrange for pick-up, either in person or through an intermediary. This has been modelled in the simulations by means of an increase in cross-border trade costs compared to the baseline scenario. In the "shop like a local" scenario we add 20€ per purchase for intermediary delivery services. Such intermediary delivery service providers already exist and this scenario could be a reality without a policy initiative. We assume however that the policy initiative would give more visibility to this option and would bring more service providers to this market.

Because of these higher trade costs, the trade effect of the alternative option is lower than in the baseline scenario. The dynamic price effect is also somewhat lower, -0.5% offline and -0.6% online on average across EU28 (table 5). The growth in consumer and producer surplus diminishes a bit compared to the baseline. EU28 consumer surplus increases by 0.8%, producer surplus by 1.0%. Figure 6 (panels a and b, respectively) below show the results per MS. The ranking of impact across MS does not change substantially.

Figure 5a: Impact of the "shop like a local" scenario (EU28, all online products)

CS = consumer surplus; PS = producer surplus.

Source: JRC/IPTS calculations.

Figure 5b: Impact of the "shop like a local" scenario (EU28, all online products), Bln€

CS = consumer surplus; PS = producer surplus.

Source: JRC/IPTS calculations.

References

Aguiar, L. and J. Waldfogel (2014). Digitization, Copyright, and the Welfare Effects of Music Trade. JRC/IPTS Digital Economy Working paper 2014-05.

Anderson, S. P., A. de Palma, and J. F. Thisse (1992). Discrete Choice Theory of Product Differentiation. Cambridge: MIT Press.

Berry, S., J. Levinsohn, and A. Pakes (1995). Automobile Prices in Market Equilibrium, Econometrica, 63, pp.841-890.

Berry, S.T. (1994). Estimating Discrete-Choice Models of Product Differentiation, RAND Journal of Economics, 25(2), pp.242-262.

Dubé, J-P., J. T. Fox, and C.-L. Su (2012). Improving the Numerical Performance of Static and Dynamic Aggregate Discrete Choice Random Coefficients Demand Estimation. Econometrica, 80(5), 2231 .2267.

Duch-Brown, N. and Martens, B. (2015). The European Digital Single Market. Its Role in Economic Activity in the EU. JRC/IPTS Digital Economy Working Paper 2015-17.

Duch-Brown, N. and B. Martens (2016). The economic impact of removing GB restrictions in the EU Digital Single Market. JRC/IPTS Digital Economy Working Paper 2016-02.

Duch-Brown, N., L. Grzybowski and F. Verboven (2015). The Impact of Online Sales on Consumers and Firms: Evidence from Household Appliances. JRC/IPTS Digital Economy Working Paper 2015-15.

GfK (2016), Provision of a Mystery Shopping survey on Territorial Restrictions and GB in the European Digital Single market.

Grigolon, L. and F. Verboven (2014) Nested logit or random coefficients logit? A comparison of alternative discrete choice models of product differentiation. The Review of Economics and Statistics 96(5), 916-935.

Judd, K. L. and B. Skrainka (2011). High performance quadrature rules: how numerical integration affects a popular model of product differentiation. CeMMAP working papers CWP03/11.

Knittel, C. and K. Metaxoglou, (2008). Estimation of Random Coefficient Demand Models: Challenges, Difficulties and Warnings, Working Paper 14080, National Bureau of Economic Research.

McFadden, D. (1978). Modelling the Choice of Residential Location, in Spatial Interaction Theory and Planning Models. Karlquist, A., L. Lundquist, F. Snickers, and J. Weibull (eds.). Amsterdam: North Holland.

Annex 5: Evidence overview

The evidence showing the extent of restrictions based on nationality or place of residence including geo-blocking

Discrimination based on nationality or place of residence including geo-blocking is not a rare occurrence: there is ample evidence of this phenomenon. Indeed, there are seven recent quantitative sources independent of each other all of which point to the existence of widespread restrictions based on nationality or place of residence:

1.Mystery Shopping Survey 2015

2.eCommerce competition sector inquiry 2015

3.Inquiries into car rental and amusement park industries in 2014 and 2015

4.DSM Consumer survey 2015 (reported in the Consumer Conditions Scoreboard 2015)

5.Consumer Flash Eurobarometer 397 (2015)

6.Business Flash Eurobarometer 434 (2016)

7.Community survey on ICT usage 2015

8.Complaints database

9.Studies specific to non-audiovisual digital content

The mystery shopping survey and the eCommerce competition inquiry as well as the sectoral inquiries look at the supply side, the two consumer surveys and the Community Survey look at the demand side. In addition, the complaints database provides evidence on consumers motivated enough to take action.

1. Mystery Shopping Survey 2015

At the end of 2015, a large mystery shopping survey for goods and services (but not digital content) was carried out looking at approximately 10 500 websites that were visited first by mystery shoppers as domestic users and then as cross border users from another Member State. It covered goods/services from the eight sectors most commonly purchased online in the European Union according to the DSM Consumer Survey published in 2015 109 .

The survey found that while only 2% of all websites completely block access to their website to visitors from another EU country (either by denial of access or automatic re-routing), 26% do not allow such visitors to register (in order to initiate a purchase). Of the remaining 72% of websites which did not block access and allowed visitors to register, approximately one third did not provide delivery options to the country of the visitor. Finally, a quarter of those websites which did allow registration and delivery did not accept payment from the users. So, in the end, geo-blocking practices were identified in approximately 63% of all assessed websites. The chart below displays the success rate in terms of not encountering various forms of geo-blocking, as a proportion of all websites assessed. It can be seen that only 37% of websites actually allowed foreign visitors to reach the stage of successfully entering their payment card details, i.e. the final step before pushing the "order/buy" button,

Chart 1: Mystery shopping success rates (% of all shopping attempts)

Source: JRC-IPTS based on Mystery Shopping Survey 2015 (GfK Mystery Shopping)

Significant differences of the prevalence of geo-blocking restrictions were found depending on the sector. Among the selected sectors, electronic household appliance had the highest share, with only 14% of cross-border shopping attempts succeeding. However, even the sector least prone to geo-blocking i.e. travel services, still produced a failure in one third of the cases. Most of the remaining tangible goods sectors have fairly similar total success rates (between 21% and 40%).

In short, the survey found that geo-blocking restrictions are a significant and widespread problem when attempting to buy cross-border online.

Table 1: Mystery shoppers success rates by product category
	sent to same website	find same product	can register successfully	can have product delivered	possible to pay	can enter payment details successfully
Clothing, shoes and accessories	97%	98%	75%	45%	37%	35%
Electronics & computer hardware	98%	97%	64%	32%	25%	21%
Travel services (hotels, transport)	99%	94%	86%	80%	74%	67%
Cosmetics and healthcare products	96%	98%	74%	44%	40%	37%
Books	99%	98%	75%	66%	52%	40%
Computer games and software	98%	98%	62%	42%	32%	27%
Electrical household appliances	98%	99%	59%	25%	19%	14%
Online reservations of offline leisure	98%	96%	83%	71%	66%	60%

These results confirm a 2009 Mystery Shopping Survey concerning online shopping of tangible goods which found that only 40% of websites allowed the completion of a cross-border purchase, with 19% of websites not allowing registration, 52% not offering shipment and 47% not accepting foreign credit cards. 110

See annex 6 for more detail; see full report at:

http://ec.europa.eu/consumers/consumer_evidence/market_studies/geo-blocking/index_en.htm

2. e-Commerce Competition sector inquiry 2015

In 2015 the European Commission initiated an inquiry into the eCommerce sector for physical goods and digital content within the framework of EU competition law. For consumer goods, it found that 36% of online retailers (out of more than 1000 online retailers who participated in the inquiry) do not sell cross border in at least one product category in which they are active. The practice is widespread across all the sectors covered by the inquiry and can take place at different stages of the purchasing process.

Chart 2: Respondents who gather location information for each geo-blocking purpose – EU 28

Interestingly, it appears that there is a positive relationship between the total turnover and the proportion of respondents that gather location information for geo-blocking purposes. With the exception of the lowest turnover bracket, the higher the turnover, the higher the proportion of retailers that gathers information for geo-blocking purposes.

Chart 3: Respondents who gather location information for geo-blocking purposes for each 2014 turnover category

Finally, the proportion of respondents who reported that they face a contractual limitation to sell cross-border in at least one product category at EU level is 12%. Therefore, most of the geo-blocking practices seem to be based on unilateral decisions by the traders. .

As regards digital content, the vast majority of the 262 retail providers of digital content that were contacted sell audio-visual content, including music videos. A total of 58 digital content providers were active in selling music content. Music products overall are among the least geo-blocked product categories; however 44% of contracts licensing music require digital content providers to geo-block. The reasons behind these restrictions will be analysed subsequently.

Proportion of agreements requiring providers to geo-block by category – Average for all respondents – EU 28

This percentage is higher in large groups (companies active in 4 or more MS). 70% of contracts between rights holders and these providers require them to geo-block.

Proportion of agreements requiring providers to geo-block by category – Large groups

See full report at: http://ec.europa.eu/competition/antitrust/sector_inquiries_e_commerce.html

3. Inquiries into car rental and amusement park industries in 2014 and 2015

Inquiries into the car rental and amusement park industries in 2014 and 2015, as well as the analysis of complaints have shown that companies applied different conditions on the basis of residence even when the service they provide takes place at an identical location and on the same time for all customers concerned; i.e. in situations where the trader did not have to deliver the product or to apply different consumer law or taxation rules. In one particularly egregious case, a customer booking a rental car from Germany was found to be charged twice as much as a comparable customer based in the UK 111 , while the amusement park charged visitors up to 80% more if they came from a different member state 112 . Although some of the specific practices identified have been brought to an end following the Commission's involvement 113 , these examples indicate that price discrimination can reach very large dimensions.

4. DSM Consumer survey 2015 (reported in Consumer Conditions Scoreboard)

On the side of the online consumers, a 2015 survey found that among consumers whose most recent problem concerned (an attempted) purchase of tangible goods/offline services purchased online from another EU Member State, 7% reported that they could not access the foreign seller’s website (or only limited content was displayed to them), 6% indicated that foreign sellers refused to sell to them because of their country of residence, 5% were automatically redirected to a website of their country, 6% reported that the foreign seller charged them a higher price than was available in the seller’s country and 4% that their means of payment was refused by the foreign seller.

See full report at: http://ec.europa.eu/consumers/consumer_evidence/consumer_scoreboards/11_edition/docs/ccs2015scoreboard_en.pdf

5. Consumer Flash Eurobarometer 397 (2015)

In spring 2014 a Flash Eurobarometer survey was carried out on "Consumer attitudes towards cross-border trade and consumer protection", interviewing approximately 26 600 consumers from all Member States. It found that among consumers shopping goods and services online cross-border, 19% experienced geo-blocking restrictions in total, with 10% not being offered any delivery, 8% being automatically redirected and 5% having their credit card refused 114 .

Chart 4: Problems encountered when shopping cross-border online for goods or services

http://ec.europa.eu/COMMFrontOffice/PublicOpinion/index.cfm/Survey/getSurveyDetail/instruments/FLASH/surveyKy/2031

6. Geoblocking Flash Eurobarometer 434 (2016)

In February 2016, a Flash Eurobarometer survey was carried out context in order to collect evidence on the prevalence of geo-blocking and other discriminations faced by businesses based on their place of residence or establishment. It covered purchases of goods and services, whether online, offline or by distance sales (phone/fax), subject to publicly available general sales conditions, excluding purchases of goods/services for resale. The study surveyed 4.200 companies from 15 Member States, finding that around 10% of those having engaged in cross-border sourcing in the last 12 months were not able to conclude the purchase without encountering any problems.

Chart 5: Problems encountered by companies when buying goods and services from another EU country 115

The respondents' feedback shows that problems were encountered across various goods and services sectors.

Chart 6: Types of goods and services for which companies faced problems when purchasing within the EU 116

See full report at: forthcoming

7. Eurostat survey on ICT usage

As part of the 2015 Community survey on ICT usage in households and by individuals, which surveys more than 200 000 individuals in all EU Member States, an option "Foreign retailer did not sell to my country" was included in the questionnaire. This option was chosen by 3% of all online shoppers. Given that 30% of online shoppers purchase cross-border, this means that 10% of cross-border online shoppers experienced restrictions as a result of their place of residence in the last twelve months.

http://ec.europa.eu/eurostat/documents/2995521/7103356/4-11122015-AP-EN.pdf/276b6a7c-69a6-45ce-b6bf-488e975a8f5d

8. Complaints database

Between 2008 and 2015, the European Commission has received through various channels 117 more than 1500 complaints by citizens who felt they were discriminated. These complaints concern restrictions based on nationality or place of residence taking place online and over the counter. More than a third of the consumer complaints concern traders who automatically re-route their cross-border customers. In about a third of the handled cases the consumers suffered from a price difference as a direct effect of being faced different terms and conditions based on their nationality or where they live. A significant amount of consumers also felt discriminated against because traders refused to deliver to their member state.

Chart 5: Complaints received by the European Commission

See annex 7 for details

9. Studies specific to digital content

A recent study 118 looking at the cross-border availability of music for download on iTunes found that whilst cross-border purchases were not possible, availability stood at 78% for the top 300 songs, although with price differences across countries. 119

Another recent study 120 looked at the cross-border availability of e-books 121 based on Amazon sites. This study (based only on the 100 top books) showed that availability is high due to universal access to the generic amazon.com site, but there are severe restrictions on cross-border access to the existing country extension in the EU 122 . Similarly, the Commission undertook recently a scrapping of Amazon sites. Preliminary results based on crawling 1148 top selling e-books on all 6 Amazon domains in Europe show that for a given Amazon domain, Amazon does not price discriminate with respect to where a consumer lives (to the limited extent cross-border access is possible). For instance on amazon.de prices seem always the same, regardless of whether a customer resides in DE or AT or LU. However, there are price differences across Amazon domains..

Finally, the public consultation on geo-blocking and other geographically–based restrictions 123 contained a disclaimer 124 upfront excluding geo-blocking or other restrictions related to copyright and licensing practices. Nonetheless, the consultation contained a number of questions on barriers to downloading digital content (see also annex 2 for details) which is very often protected by copyright and neighbouring rights. 125 Therefore there is some uncertainty whether respondents took into account the above disclaimer in their replies. The end result is that

86% of consumer respondents regarded restricting the access to/download of a digital product by traders as an important barrier; 90% of consumers felt that geo-blocking was particularly unjustified if there is no additional cost of delivery/administrative burden (e.g. download of digital content) and 93% of consumer favoured a prohibition of refusal to download digital products (such as software or video games);

25% of the businesses who responded wanted policy proposals covering rules prohibiting traders to refuse the cross-border download of digital products (such as software or video games).

Annex 6: Summary of Mystery Shopping Survey 2015

a) Methodology

At the end of 2015, a large mystery shopping survey was carried out by GfK Belgium PS looking at approximately 10 500 websites that were visited first by mystery shoppers as domestic users and then as cross border users from another Member State.

A total of 8 sectors (6 tangible goods sectors and 2 services sectors) were assessed in order to provide sufficient breadth in terms of sectors covered whilst keeping a balance in delivering sufficient mystery shopping surveys per sector to enable post-fieldwork analysis. The two services sectors, in particular, were relatively challenging in terms of being able to compare and sample similar services in multiple countries, but their inclusion was key in order to ensure that the mystery shopping survey focuses on services as well as on tangible products.

The sector selection was based on those goods and services sectors that are most commonly purchased online 126 in the EU according to the 2015 DSM Consumer Survey 127 .

Sector share of users purchasing online

Clothing, shoes and accessories 76%

Travel services (hotel and transport) 68%

Electronics and computer hardware 66%

Books 64%

Online reservation of offline leisure

(e.g. event tickets, restaurants) 63%

Electrical household appliances 61%

Cosmetics and healthcare products 60%

Computer games and software 50%

The survey targeted 143 country pairs where cross-border eCommerce is more intense. For each of the country pairs and each of the 8 sectors, 9 websites were visited, resulting in a target of 10296 websites (eventually, a total of 10537 websites were assessed). Each website was accessed twice, once from the country of the website (online retailer) and a second time from the country of the mystery shopper (cross border).

b) Results

Overall, the survey found that:

2% of all websites, accounting for 7.5% of cross-border traffic, engaged in automatic re-routing or directly blocked mystery shoppers

During website registration, 27% of online retailers blocked cross-border mystery shoppers by preventing them from successfully registering on their website

Even after allowing mystery shoppers to successfully register on their website, 32% of online retailers refused to deliver a product/provide a service to the shoppers’ country

At the payment stage, in 26% of websites shoppers couldn’t pay because their means of payment was not accepted/offered or because they were unable to enter their card details

In 37% of all websites, no information on delivery restrictions was clearly displayed on the starting page or during the ordering process (at 44% among those websites where geo-blocking practices were identified)

Only in 37% of all websites were shoppers successful in reaching the final stage of successfully entering their payment card details in order to verify the purchase

The Figure below shows the success rate in terms of not encountering various forms of geo-blocking as a proportion of all websites assessed (not within each stage of the shopping process).

For example, out of all 10537 websites assessed, on 49.2% i.e. 5184 websites no delivery restrictions were imposed (i.e. mystery shoppers reported that it was possible to have the product delivered to the country of the shopper). Or, to take another example, it was possible for mystery shoppers to register successfully on 72.4% of the 10537 websites assessed, i.e. on 7629 websites.

Figure 1: Mystery shopping success rates (% of all websites N=10537)

In terms of the eight sectors covered, large differences in geo-blocking restrictions were observed. In the Electrical household appliances sector, 86% of the mystery shopping assessments ended prematurely because of geo-blocking restrictions compared to only 60% within the Books sector. With services, geo-blocking practices were not as common, resulting in geo-blocking restrictions in 33% and 40% of cases for Travel services and Online reservations of offline leisure. Service providers were more likely to place restrictions at the access stage of the shopping process (between 7% - 9%) compared to tangible goods (3%-6%). At the access stage 128 , the highest proportion of geo-blocking practices took place when shopping cross-border for Flight Bookings (17%), Car rentals (13%), followed by amusement/theme or adventure park tickets (11%).

Figure 2: Geo-blocking practices prevalence at different stages of the online shopping process by sector (expressed as a proportion within each stage of the shopping process)

Source: Q11, Q13, Q15 - Access, N = 10537; Q19 - Registration, N = 10382; Q20 – Delivery, N = 7628; Q24, Q25 – Payment, N = 5180; weighted data

It should be noted that notable differences can be observed with products within the same sector, not only between different sectors. In the Travel services sector, for example, overall geo-blocking practices stand at 56% for Car rentals but only at 25% for Accommodation bookings.

Regarding the size of the seller, medium-sized retailers were the least likely to engage in geo-blocking practices (55%) compared to small (66%) and large retailers, (69%) while marketplaces were the most likely to engage in overall geo-blocking practices with as many as 70% of mystery shopping assessments facing geo-blocking restrictions.

Figure 3: Geo-blocking practices prevalence at different stages of the online shopping process by retailer size

Source: Q11, Q13, Q15 - Access, N = 10537; Q19 - Registration, N = 10382; Q20 – Delivery, N = 7628; Q24, Q25 – Payment, N = 5180; weighted data

For most websites assessed, registration was required in order to be able to complete a purchase. Only for 8% of the websites purchase was possible without having to register on the cross border website. For one in five of the websites assessed during the registration stage, problems with registration were reported but it was still possible to complete the registration, whilst registration could not be completed for 27% of those websites where shoppers tried to register. Registration failed most often for websites selling electrical household appliances (40%), computer games and software (37%) or electronics and computer hardware (35%).

Figure 4: Prevalence of encountered problems during registration by market

Source: Q19 - Did you encounter problems during registration? N = 10366, weighted data

During the delivery stage where shoppers need to provide their delivery/shipping address related to a specific online purchase, for almost a third of online retailers, mystery shoppers found that delivery is possible to all EU28 countries. This suggests that two-thirds of online retailers apply in some degree geo-blocking practices when it comes to delivery within the EU. According to mystery shoppers’ assessments and the way information was provided on websites, most commonly, retailers either only deliver to the country where they are based (26.4%) or where the consumer is based (in addition to the country of the retailer, but not to other countries) (31.5%).

Figure 5: Overall delivery restrictions – countries where retailers deliver to

Q20: To which EU countries would it be possible to deliver the product, in addition to country of the retailer? N = 7628, weighted data

During the payment stage, it was found that issues with blocking payment are more prevalent by small retailers, while there is only little difference between medium and large sellers. Market places are the least restrictive regarding means of payment.

Figure 6: Prevalence of payment errors by size of retailer

Source: Q24 - Does it appear possible for you to pay for this product on the website? N = 5135, weighted data; Q25 - Now, please try to enter the payment card details provided without confirming the order. Are you able to do it and if so? N = 4381, weighted data

Annex 7: Factual analysis of discrimination-related complaints collected by the Commission

1.Introduction and methodology

The purpose of this Annex is to present the complaints-based databank, which is a factual analysis of complaints, rather than an evaluative one. It is to serve as a broader overview of the challenges which consumers and businesses face within the Single Market, when providing or purchasing cross-border services. In the beginning of this exercise over 4000 complaints were made available to the Commission from various sources. Out of those, only one thousand five-hundred and fifty complaints from 2008 until 2015 were assessed to be relevant to the databank. They provide an ample number of real life examples of restrictions based on nationality or place of residence.

In comparison to the public consultation on the topic, the database was created through additional efforts by the Commission to utilise available internal channels containing information pertinent to our current efforts. It is compiled through analysing complaints received via ECCRS 129 , Your Europe Advice (YEA), European Consumer Centres (ECC), complaints addressed directly to the Commission, questions posed by MEPs and Parliamentary petitions. The possible explanations of the divergent nature of the complaint extraction and the post-analysis numbers have been discussed. The Commission has learned that in some databases where the consumers themselves have to enter the sectors concerned or the relevant law which they allege is infringed, mistakes occur.

The database is not an exhaustive catalogue of consumer complaints and due to the lack of a centralized system is only a representation of the success of the Commission's internal efforts. The factual individual assessment of the complaints, unlike in the above mentioned databases, has enabled the creation of a more detailed point of reference with regards to difference in treatment based on nationality or residence under Article 20 of the Services Directive.

Instead of having a pre-set questionnaire, each complaint has been individually assessed and recorded. The database is complimentary to the public consultation. It also goes a step beyond the information gathered through EU-barometer surveys, since the type of encountered problems is more comprehensively recorded.

2.Main results

Fig.1

Sources and complaints

The graph above is a visual representation of the breakdown of sources of complaints year-by-year. At a first glance the red trend line shows that complaints received through ECCs are on a much stronger rise that complaints received through various other channels. There is a slight distortion in the YEA trend line which is due to the fact that the process of accessing some over 8000 entries with complaints from 2013-15 which might 130 be relevant to our database is still pending. Once the process is complete the YEA complaints count are expected to surpass the data provided by the ECCs. In March 2016, the Commission is also expecting to receive additional data of complaints handled by various Member States, some of their enforcement authorities, small and local consumer associations, which will also feed into the databank of the Commission.

Fig.2

The graph above (Fig.2) is a representation of the total number of consumer complaints (year by year) which have been logged in the database. It shows the composite count, regardless of the channel from which they were obtained. The Commission is expecting the bars of 2013-15 to grow with a similar measure as in the years 2009-2012, once it has fully finalised the analysis. It is, however, evident that complaints concerning possible discrimination based on nationality or residence are on the rise.

Transparency and access to information

It is evident from the data that there is not enough transparency within the Single Market. A significant amount of the handled complaints indicate that consumers are often not provided with a specific justification as to why the service was refused.

Fig.3

The 'Undetermined' figure of 85% shows that in those cases, based on purely factual analysis, it was difficult to distinguish whether or what objective justification was invoked by the trader. In most sources, this data is not mandatorily recorded and although in the majority of those cases it was signalled that the consumers have asked for an explanation by the trader, there was only follow-up data in about 15% of them.

Most of the times when the service is refused the consumers are left with a general justification (if any is indeed provided) which is not specific to their situation. In the predominant lot of cases consumers are not informed about the possibility of accessing offers on country-specific websites or there is no prior information on delivery restrictions.

Fig.4

Out of the 15% of the cases where it is known what, if any, was the justification given by the trader, 44% of the consumers who actually enquired as to why they were refused the service or good did not receive an explanation. Regardless of the stage at which the explanation was given, whether it is a full explanation upfront on the website or after the consumer specifically requested it, in the remaining 56% of cases consumers felt that they were given a generic explanation. In 21% of the complaints where we know the justification of the traders, the objective criteria stated was contractual obligations. 11% of justifications relate to the lack of required intellectual property rights. Similarly, around 10% are connected to different tax regimes in Member States. Only 7% of justifications relate to additional costs which will be incurred due to the transport and delivery related price fluctuations, if delivered cross-border. About 4% of the cases state that discrimination was applied in order to avoid fraudulent practices. And in 3% of the cases the difference in treatment resulted from the commercial practices of the trader.

From the factual analysis of each individual complaint in the database, one can conclude that in the majority of the complaints with known justifications, the complainants are of the opinion that traders should inform customers about sales restrictions (e.g. outright on the website) or at least provide them with the case-specific justification of why their transaction with the trader was rejected, rather than giving them a generic reason.

Type of restrictions

Difference in treatment is frustrating consumers attempting to utilize the Single Market.

According to the analysed data, about a half of the consumers have managed to transact, while the rest were denied the transaction at all (total refusal to sell or deliver).

Fig.5

In all of the cases in the databank, whether the transaction was concluded or not, there was a sale but with an accompanying difference in treatment based on residence or nationality. Fig. 5 is showing the outcome of restrictive practices, not taking an account of whether the different treatment was justified or not.

A denial of access to a website (blocking the access or automatically re-routing the consumers) and refusal to sell are among the most common outcomes for consumers when they feel that they have been discriminated against. More than a third of the consumer complaints handled indicate that traders often automatically re-route their customers when they try to access the website of the trader in another Member State, rather than the one provided within the country of origin of the consumer's IP address. In about a third of the handled cases the consumers suffered from a price difference as a direct effect of being geo-blocked. Fig. 5 above is not differentiating between cases with or without objective justification but it is rather showing the outcome of all the complaints logged in the database.

A significant amount of consumers also felt discriminated against because traders refused to deliver to their Member State. In some cases it is not just the product or the service having to cross borders but also customers. EU consumers legitimately expect the same treatment when receiving services or products in the same location regardless of where and how the transaction took place (e.g. car rental pick-up in a certain location on a particular dare for a particular duration with the same car and options, amusement park entrance on a particular date and for a particular duration). The complaints by consumers show a trend that some companies within the Single Market refuse to sell or differentiate prices between customers based on the place of residence or nationality, regardless of the fact that they are receiving the service or product at the same location.

Type of difference of treatment

There are still cases where consumers are treated differently based on their nationality within the Single Market.

Based on the obtained records of different complaints, almost one-fifth of the analysed cases relate to difference in treatment of consumers based on their nationality. Fig. 6 is showing the basis for different treatment, not taking an account of whether the different treatment was justified or not.

Fig.6

Although different treatment of consumers based on residence is far more common, a significant portion of consumers are still feeling discriminated against based on the Member State from which they come. The most often encountered methods of differentiation which have been registered are the automatic geo-location of consumers' IP addresses; differentiations based on billing- and delivery addresses; credit- and debit card or bank account identification; identification through national identification documents (including, but not limited to, passport, ID card, driver's license, disability- and student cards, license plates, phone numbers); proof of residence; language and country of establishment.

Are differences of treatment based on residence or nationality purely an on-line issue?

It is not only in the digital context that consumers feel discriminated against. The complaints handled show that 'off-line' differentiation of consumers is also quite significant.

In about a third of all handled cases it is evident that the method of sale is an off-line one. From the data gathered, it is apparent that differences of treatment based on residence or nationality are not just purely an on-line issue, although being mainly such. Consumers are frustrated a significant amount by difficulties and discrimination in the off-line world when purchasing services or goods. In about 10 % of the cases the method of sale was not entirely clear. Sometimes, it was not recorded, other times it could not be certain as to whether it was an on-line or an off-line transaction. In another 8% of the cases, consumers were affected by this difference in treatment when trying to access the same service in both the on-line and off-line markets.

Fig.7

Fig.7 is representative of the percentage of complaints across all sectors and for all of the years for which the Commission has managed to gather information (see Fig. 1and 2). It is showing the method of sale described in the complaints, not taking an account of whether the different treatment in the individual complaint was justified or not.

Sectors mainly affected by geo-blocking practices

Restrictive practices are affecting consumers in a wide array of sectors. Some of them fall within the scope of the Services Directive, but there are also other significant ones, which remain outside of its reach.

Fig.8 shows the main sectors in which different treatment was applied, not taking an account of whether the different treatment was justified or not. It merely purports to visualize the distribution, in percent, of complaints within different sectors of the Single Market. As mentioned earlier, the database is limited to the success of our internal efforts to obtain the maximum number of entries from various Commission sources.

Although the Commission specifically searched for cases which are only within the scope of the Services Directive, the different data sources yielded varying results, including sectors which are not covered by the Services Directive but were logged in the different systems as pertinent cases.

About two-fifths of the analysed complaints relate to the retail sector (which includes tangible consumer goods). In comparison, the latest data has about 10% of cases relating to purely digital issues (excluding tangible goods).

In the retail sector, the complaints which have been collected are mainly concerned with the cross-border (attempted) purchase of clothing, footwear and accessories (including tailor-made goods) and electronics and computer hardware. Each of the two sub-sectors represents one-fifth of all complaints related to retail.

In the digital sector, a predominant number of complaints, with 71%, relate to e-books, streaming services and MP3 downloads. In most of those cases, consumers have complained that they do not have access to such paid services from another member state. Still, in a considerable one-fifth of all the complaints which we have concerning the digital sector, consumers claim to have been treated differently when attempting to purchase computer games and software.

Fig.8

In the transport services sector, we were mainly interested in finding cases where there were issues with the car rental sub-sector. Out of the 16% of the complaints related to transport services, 4% relate to cases where consumers had issues when renting, leasing or sharing a car.

The remaining 12 % of cases refer to other transport services. About a third of those relate to local public transport. Maritime-, air-, and rail transport related issues amount for about 15% each. Road transport and transport infrastructure services related cases amount to 10% each.

The accommodation and leisure services cases are 16 % of all the cases in our databank. Some of them explicitly relate to accommodation and food service related issues within the Single Market, but the majority of cases communicate issues which consumers had when attempting either to purchase or to use various package-travel options from different member states. Complaints including cultural and entertainment services also represent a significant amount. The minority of cases which the Commission assessed concerns services related to sports and hobbies, especially in the cross-border context.

Out of the 8% network industries cases, the most problematic area, based on the assessment of complaints, are the telecom services. More than two-thirds of all issues are related to the latter sub-sector with a fifth attributed to gas, water and electricity related services and about one-tenth representing issues in the Radio and TV sub-sector.

As stated earlier, the presence of the sectors in others, namely education, financial services and gambling is not representative of the actual number of complaints addressed at those sectors. The Commission has not specifically searched for cases which relate to financial services or gambling.

The extraction made from the IT Tools available from the Commission's sources (YEA, ECCRS) yielded results, which included the complaints in those sectors. They were registered as complaints under the Services Directive, pertaining to difference in treatment expressed as discrimination based on nationality or residence as per Art.20.

3.Concluding remarks

When internally comparing the conclusions from the data, whether it is method of sale year-by-year, basis for different treatment year-by-year or main Member States involved, certain consistency stemming from the results has been established. The number of complaints received through the various channels is growing. The bigger the economy the more complaints related to that particular Member State there are. The proportion of issues related to a particular sub-sector from within the main sectors remains in the large part similar.

As already mentioned, further assessment work is still undergoing in an attempt to compile an even more comprehensive database. However, the Commission is not expecting results which are too divergent from the ones presented in this document. It believes that the current pool of examples is sufficiently telling for general trends to be recognized and observed within the Single Market.

Annex 8: Assessment of the functioning of Article 20(2) of the Services Directive

1. Introduction

The purpose of this Annex is to assess the effectiveness, efficiency, coherence, relevance and EU added value of the current provision addressing discrimination by service providers based on nationality or place of residence, namely, Article 20(2) of Directive 2006/123/EC on services in the internal market ("Services Directive") 131 . Due to the limited scope of the exercise (only one article of the Directive), this is not a fully-fledged evaluation in the spirit of the Better Regulation Guidelines. The findings will, however, provide further evidence as to the problems described in the Impact Assessment.

The public consultation on geo-blocking and geographically-based discrimination was carried out in the framework of the preparation of this Impact Assessment (see annex 2 for more details). In February 2015, the Commission organised a public consultation in the form of a conference dedicated to Article 20 of the Services Directive; "Buying Services everywhere in the Single Market". 132 In addition, other sources of data i.e. the complaints database, the results of the surveys were also used to inform the findings of this assessment (see annex 7 and annex 5 for more details).

As described in the problem definition of the Impact Assessment, discrimination based on place of residence or nationality is an obstacle to a fully functioning Single Market. Customers wishing to access better deals across the borders are faced with outright refusals to sell or charged higher prices based on where they live; undermining customers’ trust in the Single Market. Differences in treatment based on nationality or place of residence when shopping abroad is mainly applied by private service providers, not Member States (i.e. public service providers). The complaints received by the Commission show that 4% of them concern alleged discrimination by Member States (e.g. price discrimination in public museums). 133 The assessment of the functioning of Article 20 of the Services Directive is therefore limited to Article 20(2) covering discrimination by private service providers.

Article 21 of the Services Directive obliges Member States to designate assistance bodies to ensure that service recipients can obtain information and assistance in case of dispute between a provider and a recipient. Article 21 is partly covered by the assessment since it has an impact on the functioning of the non-discrimination principle in Article 20(2).

2. Background

The Single Market has delivered advantages to European businesses. The implementation of the Services Directive has contributed to the removal of barriers for companies across the EU. A 2015 Commission analysis of the effects of the Services Directive showed that an even more ambitious implementation of the Directive could generate up to 1.7% of EU GDP in addition to the 0.9% that will be realised by reforms undertaken since the adoption of the Services Directive. 134 However, facilitating the provision of services across national borders is not enough for the establishment of a genuine Single Market. It is equally important to ensure that service recipients, and in particular, consumers can make full use of the opportunities of the Single Market.

2.1 Services Directive

The objective of the Services Directive is to release the full potential of services markets in Europe by removing legal and administrative barriers to trade. The Directive was adopted in 2006 and transposed by all EU countries in 2009.

To enhance the rights of service recipients and strengthen their confidence in the Single Market, the Services Directive obliges Member States to remove obstacles for service recipients wanting to purchase goods or services across the borders, such as obligations to obtain a specific authorisation to receive services from another Member State. It also introduced an obligation on Member States to ensure that discriminatory requirements based on service recipients' nationality or place of residence were not applied against them.

Article 20 of the Services Directive prohibits discrimination against service recipients, by both public and private service providers, on the basis of their nationality or place of residence. The purpose of the Article is to help service recipients, especially consumers, access offers available in other Member States and make the most of the Single Market.

Article 20(1) is targeting public service providers such as public authorities. It obliges Member States to ensure that the recipients are not made subject to discriminatory requirements based on nationality or place of residence under national, regional or local regulations. It includes situations where public authorities treat service recipients differently.

Article 20(2) cover situations where private service providers, i.e. private companies, treat service recipients differently based on their nationality or place of residence when trying to access offers across the borders. It obliges Member States to ensure that the general conditions of access to a service, which are made available to the public at large by the provider, do not contain discriminatory provisions relating to the nationality or place of residence of the recipient, but without precluding the possibility of providing for differences in the conditions of access where those differences are directly justified by objective criteria. According to Recital 95 of the Directive, objective criteria could be additional costs incurred because of the distance involved or the technical characteristics of the provision of the service, or different market conditions, such as higher or lower demand influenced by seasonality, different vacation periods in Member States and pricing by different competitors, or extra risks linked to rules differing from those of the Member State of establishment. The lack of the required intellectual property rights in a particular territory could also constitute objective criteria according to the Recital.

2.2. National implementation of the non-discrimination principle

Member States were obliged to implement Article 20 into their national legislation. Article 20(2) is implemented through national provisions that make the prohibition of discrimination on grounds of nationality or residence binding on service providers. Most Member States have introduced in the horizontal laws transposing the Services Directive the provision on non-discrimination. Many simply reproduced literally the wording of Article 20(2). Others have relied on already existing legislation that achieves the same purpose. It is for the relevant national competent authorities to ensure compliance with the relevant national provisions implementing Article 20(2) into their national law. 135

Apart from enforcement authorities, Member States were also obliged to designate assistance bodies according to Article 21 of the Services Directive. The objective was to enhance confidence of recipients of services by giving them the means to make informed choices and comparisons when engaging in cross-border transactions. It introduces the right to recipients to obtain, in their home state, general information and assistance on the legal requirements, in particular on consumer protection rules, and on redress procedures applicable in other Member States. The "Article 21 bodies" may assists service recipients who have experienced discrimination based on residence or nationality in violation of Article 20(2).

2.3. Commission guidance on the application of Article 20(2) of the Services Directive and the brochure "Buying services everywhere in the EU – A practical guide to consumers"

In 2012, the Commission issued guidance on the application of Article 20(2) of the Services Directive. The purpose was to improve the application and enforcement by national authorities and courts. In light of the situation where national authorities received numerous complaints but there were hardly any enforcement actions, the guidance was an attempt to give full effect to Article 20(2) by further clarifying the interpretation of the non-discrimination obligation. 136

In 2014, the Commission published a brochure "Buying services everywhere in the EU – A practical guide for consumers" explaining the non-discrimination principle by giving real life examples of differences in treatment and advice on how to act if experiencing alleged discrimination including a list with contact details to the "Article 21 bodies".

3. Effectiveness

- To what extent have the objectives been achieved?

- What have been the (quantitative and qualitative) effects of the intervention?

- What factors influenced the (lack) of achievements?

The objective of the non-discrimination principle in Article 20 of the Services Directive is to enhance the rights of service recipients and strengthen their confidence in the Internal Market by ensuring that service recipients are not subject to discriminatory practices based on their place of residence or nationality when shopping across the EU. The Commission has collected and factually analysed over 1500 complaints between 2009 and 2015 from service recipients across Europe on alleged discrimination. 137 The complaints show that customer experience restrictions/differences in treatment based on their nationality or place of residence which in turn might constitute discrimination. However, in order to establish discrimination causality link, a case-by-case assessment has to be conducted, which falls under responsibility of the national enforcement authorities . 85% of the complaints received by the Commission did not contain clear information on service providers' justification for treating a customer differently, or enough information about the situation to fully assess whether the difference in treatment was justified or unjustified. Therefore, it is not possible to draw any conclusions as regards the percentage of discrimination cases among the complaints. However, the number of complaints show that restrictions based on nationality or place of residence are a major concern for consumers. Despite the number of complaints the Commission is not aware of a single company that has been subject to sanctioned enforcement actions by national enforcement authorities since 2009.

In 2012, the Commission published guidance on the application of Article 20 (2) of the Services Directive where it was stated that the application of Article 20(2) has to be made on a case-by-case basis. The Commission also identified a number of situations of differences in treatment; such as online transactions (geo-blocking) where customers are treated differently based on various proxies (e.g. IP address, the country where their credit card has been issued, delivery address, etc.), and the situation where the consumer travels to receive the service at the premises of the trader (e.g. hotels, amusement parks, museums, tourist attractions, etc.). The guidance addresses the same situations covered by option 3 of the present Impact Assessment. The publication of the guidance, including the advice given on thee specific situations, did neither result in an increased number of sanctions, nor did Member States amend their national regulations in order to ensure a more effective transposition of Article 20(2) of the Services Directive.

In 2014/2015, the lack of enforcement actions was discussed by the “Article 21 Network” which is a sub-group to the Expert Group on the implementation of the Services Directive. The “Article 21 Network” consists of the designated assistance bodies referred to in Article 21 of the Services Directive. In general, the assistance bodies do not have any enforcement powers. Their role is to help consumers solving Article 20 disputes with service providers. If the bodies do not manage to solve the dispute they may transfer the case to the relevant enforcement authority. As part of the discussions, the Article 21 Network reported problems with the interpretation of Article 20(2) and the list of possible justifications in Recital 95, the identification of the relevant enforcement authority and continuous difficulties to get local enforcement authorities to act.

In addition, the conference on Article 20 organised by the Commission in February 2015, brought together all relevant stakeholders (in total 122 participants representing consumers, businesses and competent authorities) to discuss the non-discrimination principle from a legal and economic point of view. The objective was to discuss (1) How the non-discrimination principle is working in practice for consumers, (2) Challenges for companies as regards Article 20(2) and (3) The way forward – what should be the next steps. Moreover, particular attention was brought to the question on how to improve cooperation between various players active in the enforcement of Article 20. The conclusions of the conference by all stakeholders were that the framework set in, as well as the implementation of, Article 20(2) is nowhere working in practice. Competent authorities highlighted that the enforcement has to be intensified, and that unsettled legal questions regarding the application of Article 20(2) and what constitutes objective criteria justifying differences in treatment has to be clarified. Consumers supported that they are not in a position to claim concrete rights under Article 20 (2) of the Services Directive. Business representatives stressed the need to respect that there are objective criteria which may justify differences in treatment; however, there is a legal uncertainty as regards what is justified and what is not. 138

The following issues have been identified as the main reasons to the lack of effectiveness.

3.1. Legal uncertainty regarding objective criteria justifying a difference in treatment

Not every difference in treatment based on service recipients' place of residence or nationality constitute discrimination. According to Article 20 (2) of the Services Directive, differences in the conditions of access are legitimate if directly justified by objective criteria. To determine whether a difference in treatment adds up to discrimination or not, it has to be assessed if the reasons invoked by the trader indeed constitute objective criteria justifying the difference in treatment. At present, this must always be assessed on a case by case basis, thus creating uncertainty for traders, consumers and national enforcement authorities.

The provision itself does not provide any further guidance on what constitutes such objective criteria. Recital 95 of the Directive gives numerous examples of objective criteria that may be invoked by service providers to treat customers differently; additional costs incurred because of the distance involved or the technical characteristics of the provision of the service, or different market conditions, such as higher or lower demand influenced by seasonality, different vacation periods in the Member States and pricing by different competitors, or extra risks linked to rules differing from those of the Member State of establishment. The lack of the required intellectual property rights in a particular territory would not constitute an unlawful different treatment. The 2012 Guidance of the Commission on how to apply this part of the Services is not delivering the expected outcomes, as confirmed at the February 2015 conference on Article 20(2) of the Services Directive "Buying Services everywhere in the Single Market" and by the respondants to the public consultation who perceived there to be no improvement after the publication of the guidance.

At present, companies use a wide range of justifications for treating customers differently, as observed in a 2009 Commission study on business practices. 139 For example, the following reasons have been given by companies to justify price differences based on the consumers place of residence: regulatory environment (e.g. compliance costs, fragmentation of consumer or environmental legislation, regulatory uncertainty, etc.), different marketing costs in different Member States (e.g. online ads and costs for translating brochures); corporate structure (e.g. franchisees having their own pricing policy); exchange-rate fluctuations; need for a bank account in the country where the company is established, taxation and credit card processing fees; operational drivers (e.g. costs such as re-registration of cars or verification of foreign driving licenses in the car rental sector); competition; market growth, risks related to stricter consumer protection laws; seasonality (e.g. different holiday periods) and simply the fact that the cost for different services varies between the Member States. These justifications have been given by companies operating in different fields including car rental, leisure parks and rental of summer accommodations and were confirmed in recent interviews with the European Consumer Centres and Your Europe Advice. 140

In a study titled "Discrimination of Consumers in the Digital Single Market" from 2013, it has been concluded that Article 20 of the Services Directive has had hardly any effect on the Single Market. It particularly addresses the issue of justifications, and the question what constitutes objective criteria. According to the study, the list of justifications in Recital 95 is so long that in the vast majority of cases it would be easy for businesses to find (or feign) a reason for any refusal to sell or for offering different conditions. Furthermore, even for services which are purely digital (e.g. cloud computing services) different prices can easily be justified by "different market conditions" or the "pricing by different competitors". The European Parliament draws the conclusion that the lack of clarity regarding objective criteria is the reason why there is currently no discernible enforcement activity throughout Europe. 141

This is also confirmed by the results of the public consultation (see annex 2) which seek views from stakeholders concerning justifications for allowing geo-blocking and geographically-based restrictions. Consumers and consumer organisation tend to qualify more of the listed practises as "unjustified", while companies, although more divided in their positions, tend to be more accepting of the number of objective factors that would justify geo-blocking. These replies also give voice to a concern, mainly stemming from companies but also public authorities, that the separation of "justified" and "unjustified" practices is vague and difficult to apply in practice. In the same manner, the conclusions of the conference on Article 20 of the Services Directive also indicated difficulties with the application of Article 20(2) due to the legal uncertainty regarding acceptable and non-acceptable justifications. 142

The legal uncertainty created by the broad list of possible justifications for differences in treatment seem to be the main problem as regards effectiveness. The publication of guidance has not been perceived as improving the situation.

3.2. Lack of enforcement actions by national authorities

To this date, the Commission is not aware of any company that has been sanctioned for violating Article 20 (2) of the Services Directive. In the 2012 Commission guidance on Article 20, a list of the competent enforcement authorities in all Member States was introduced as an Annex. The purpose was to clarify which authority that is responsible to enforce Article 20 (2) in each Member State. However, in several countries consumers would have to go to court in order to proceed with a case of non-compliance with Article 20 of the Services Directive and in others it is very difficult to identify the competent enforcement body. 143

The European Parliament made a call, already in 2010, for an effective implementation of Article 20(2), as well as the proper enforcement by national authorities and courts of the national provisions implementing the non-discrimination principle in the legal systems of Member States. 144

The European Consumer Centre Network ("ECC Network") has analysed complaints received by the network related to Article 20 of the Services Directive. Between 2010 and 2012 it received 222 Article 20-related complaints, but believes that many complaints pertaining to situations whereby consumers could not fully benefit from the Single Market went unreported. This may be attributed to the lack of awareness of the protection consumers enjoy under the Services Directive and their inability to recognise which business practices may constitute a breach of the non-discrimination principle. In more than 32% of the 222 complaints, active intervention from the ECC on behalf of the consumer was necessary with a successful outcome reached in nearly 50% of those cases. 8 service providers changed their business practice following the intervention by the ECC network. Furthermore, it reported 12 cases to the relevant enforcement authorities, but only one of all these referrals resulted in a decision made by an enforcement authority. The fact that very few cases result in enforcement actions at national level, may be due to the authorities' inability to handle individual complaints or their failure to interpret the existing rules correctly. The ECC Network has called for the urgent need to make the Services Directive work in practice. 145

Regulation (EC) No 2006/2004 on consumer protection cooperation (CPC Regulation) lays down the general conditions and framework for cooperation between national enforcement authorities in the EU. It covers situations when the collective interest of consumers are at stake and allows authorities to stop breaches of consumer rules when the trader and the consumer are established/residing in different Member States. The cooperation is applicable to consumer rules in various areas, such as unfair commercial practices, distance selling and passenger rights which are listed in its Annex. However, Article 20 (2) of the Services Directive is not included in the Annex. It means that there is currently no cooperation in terms of coordinated enforcement of Article 20 (2) by national authorities. The situations covered by Article 20 (2) are cross-border situations which require Member States to cooperate in order to ensure an effective enforcement. As evidenced by the evaluation underpinning the CPC regulation review, the lack of such cooperation constitutes an obstacle to an effective enforcement of Article 20(2). As part of the review of the CPC Regulation, the Commission has proposed to include Article 20 in the Annex of the CPC Regulation.

The assistance bodies referred to in Article 21 of the Services Directive play an important role in the enforcement of Article 20(2) even though they, in general, do not have any enforcement powers. Consumers may turn to the assistance bodies first, which in turn may submit the complaints to the competent enforcement authorities. The ECC network has taken the initiative to organise themselves regarding the application of Article 20(2) (see report from 2012 in footnote 15), and the Commission meets them twice a year as observers. However, only half of the Member States have appointed the national ECC as their Article 21 assistance body, and the Commission is therefore organising separate meetings with the Article 21 assistance bodies in addition to the meetings of the ECC network. Overall, the resources are scarce among the assistance bodies, in particular, in some of the eastern European countries. Furthermore, consumers usually turn to the assistance body in his/hers home Member States whereas the trader normally is established in another Member State. It means that the assistance body has to contact the enforcement authority of the country where the trader is established, the complaint may never reach the competent enforcement authority if the assistance body does not transfer the complaint due to e.g. the lack of resources.

The need for more effective enforcement was also clearly voiced in the public consultation (see annex 2). More than seven out of ten (72%) of the respondents support a monitoring and enforcement role for the European Commission as well as for national authorities. Similar numbers of respondents (70%) support a more effective cross-border cooperation mechanism between national authorities.

The lack of enforcement by national enforcement authorities has been identified as one of the reasons for the lack of effectiveness of the non-discrimination principles in Article 20(2) of the Services Directive. However, only ensuring that all Member States have designated enforcement authorities that cooperate with each other, and that these authorities have the resources to properly enforce Article 20(2) would not be sufficient, as confirmed in the Impact Assessment on the review of the Consumer Protection Cooperation Regulation. Improved enforcement in terms of coordinated enforcement across the EU would not be sufficient on its own to resolve discrimination based on nationality or place of residence. However, coordinated enforcement across the EU is a necessary element to address discrimination covered by Article 20(2). To summarise the conclusions of section 3.1 and 3.2, the objective of Article 20(2) of the Services Directive has not been fully achieved. The legal uncertainty regarding acceptable and non-acceptable justifications for differences in treatment makes it difficult for Member States to effectively enforce the non-discrimination provision.

4. Efficiency

- To what extent are the costs involved justified, given the changes/effects which have been achieved?
-To what extent has the intervention been cost effective?
- If there are significant differences in costs (or benefits) between Member States, what is causing them?

The transposition of Article 20-21 of the Services Directive required Member States to designate an assistance body, and to ensure proper enforcement of Article 20 (2). The role of the assistance body is to provide general information and assistance on e.g. redress procedures applicable in other Member States. It includes assisting consumers facing discrimination when buying services in the EU. The costs and the administrative burden for Member States to fulfil the obligation of an assistance body may vary across Member States due to the organisational structure in the Member States at the time of the transposition of the Directive, and the quality of the services provided by the assistance bodies. However, there is no available data on the costs in each Member State. No feedback from stakeholders concerning the functioning of the assistance bodies was provided during the public consultation on geo-blocking and other forms of geographically-based discrimination. During the public conference held in February 2015, stakeholders, including the assistance bodies confirmed that the current system is not working. In addition, the ECC network cannot be fully used because Member States did not always appoint their national ECC as interlocutors under Article 21 of the Services Directive.

However, the existence of assistance bodies is justified to fully achieve the objectives pursued, namely, enhance the rights of service recipients and strengthen their confidence in the Internal Market. Merely enforcing the provision by sanctioned decisions against businesses violating the provision is not sufficient, since it does not provide a solution to the dispute between the trader and the customer. The assistance body, which in many Member States is the ECC, helps the consumer by contacting the trader to try to solve the dispute. The number of contacts and complaints dealt with by European Consumer Centres has been steadily increasing, thus demonstrating a growing awareness of their role and usefulness. 146 As regards designated enforcement authorities, a majority of the Member States have designated the national consumer and/or competition authority, i.e. make use of already existing enforcement authorities.

The costs and the administrative burden for Member States to fulfil the obligation of an assistance body varies across Member States due to the organisational structure in the Member States at the time of the transposition of the Directive, and the quality of the services provided by the assistance bodies. However, there is no available data on the costs in each Member State to assess cost effectiveness. Furthermore,it has been established in section 3 that Article 20(2) is not effective; the objectives have not been fully reached and despite the publication of guidance there is no progress in terms of an increased number of sanctioned decisions. Therefore, no matter how Member States have chosen to organise their enforcement and assistance bodies, and the level of quality of the services provided, Article 20(2) cannot be considered as an efficient provision.

5. Relevance

-To what extent is the intervention foreseen under Article 20(2) intervention still relevant?
-How well do the original objectives still correspond to the needs within the EU?
-How relevant is the intervention for EU-citizens?

The non-discrimination principle was introduced in the Services Directive to enhance the rights of service recipients and strengthen their confidence in the Single Market, and to contribute to the overall objective of the Services Directive – a fully functioning Single Market. Since the adoption of the Services Directive cross-border trade has increased. The Digital Agenda Scoreboard showed an increase of online cross-border purchases of goods and services between 2008 and 2014; from 6% as an EU average in 2008 to 15% in 2014. Today, customers travel on a frequent basis across the borders to purchase goods and services (35% of EEA citizens travel at least once a year abroad for different reasons 147 ), and the Internet revolution has made it a lot easier to search for the best possible offers in the EU. For example, in the travel and tourism sector – one of the activities most affected by the Internet revolution – online sales represent 40% of total retail sales. In 2014, online sales in this sector reached a volume close to EUR 70 billion. 148 As a consequence of customers travelling more frequently and the digital development facilitating e-Commerce, consumers expect to be able to fully enjoy the benefits of the Single Market.

However, the number of complaints received by the Commission on alleged discrimination since the implementation of the Services Directive in 2009 shows that consumers are frequently faced with restrictions based on their nationality or place of residence when trying to make use of the Single Market. In 2014, the Commission initiated a dialogue with the car rental industry based on numerous complaints from consumers on discriminatory practices. Consumers were charged different prices based on their place of residence, and when trying to complete a booking from another country-specific website (e.g. a German consumer tries book from the country-specific website targeting French consumers) they were sometimes automatically re-routed to the website targeting their own country. The Commission and the car rental companies reached an agreement which has improved the situation. 149 In 2015, a web scraping study showed that a major leisure park in Europe charged different entry prices for the access to the park depending on the consumer’s place of residence, and prevented them from accessing better deals available at other country-specific webpages of the same company. Consumers in France and Belgium had access to specials deals on their country-specific webpages which were not accessible for consumers residing in other Member States. 150 The car rental initiative and the amusement park exercise attracted media attention in nearly all Member States, showing that EU consumers care about these issues and the importance of addressing discriminatory practices across Europe. 151

In 2011, a Commission survey showed that the main reason why consumers did not buy products from an online seller based in another country was related to conscious choices by consumers (i.e. the choice of products was enough in their home countries). 152 However, the survey did not offer online discrimination based on residence or nationality (geo-blocking) as an option apart from non-acceptance of foreign credit cards. The recent 2015 Digital Single Market (DSM) survey of online consumers projects geo-blocking as one of the main challenges that online customers face. Furthermore, the results of the public consultation on geo-blocking and other forms of geographically-based discrimination showed that 80% of consumer respondents indicated that they have experienced geo-blocking (i.e. discrimination based on residence or nationality in an online context), and a majority strongly agreed to the statement that geo-blocking and other forms of geographically-based discrimination create significant obstacles to the Single Market. 153 Most recently, a large mystery shopping survey looking at 10 500 websites first as a domestic user and then as a cross-border user found that only 37% of these websites actually allowed shoppers from another EU Member State to complete an online purchase. 154

It's clear that refusals to sell and the application of higher prices purely based on the customer's nationality or place of residence is a major cause of frustration for those who wants to make full use of the Single Market.

The need to enhance the rights of EU-citizens by addressing discrimination based on nationality or place of residence remains highly relevant, and might be even more important today than in 2009. Effectively addressing discrimination based on nationality or residence could increase consumers’ trust in the Single Market, and contribute to the overall objective of a fully functioning Single Market.

-How well adapted is the intervention to subsequent technological or scientific advances?

Article 20 (2) of the Services Directive only introduced a general non-discrimination principle, and the guidance document on the application of the provision gives no further guidance on how to address the use of technological measures to apply discriminatory practices.

Digital technology has advanced since the implementation of the Services Directive in 2009. The development of new technologies has made it easier for traders in an online context to identify the location of the customer. It is well-known today that traders use different technological measures (e.g. tracking of IP addresses or the use of credit card details) to identify the location of the customer to refuse to sell, apply different conditions or, in an online context, even prevent access to the actual website or application.

The non-discrimination principle in Article 20(2) is technological neutral. However, new technological developments have given traders more possibilities to identify the location of the customer and apply discriminatory practices (e.g. unjustified charge a higher price).

6. Coherence

- To what extent in the intervention coherent with other interventions with similar objectives (non-discrimination)?

Treaty on the Functioning of the European Union (TFEU): In primary law Article 18 of the TFEU establishes the general prohibition of discrimination between persons on the grounds of nationality. Articles 34 and 56 of the TFEU provide for the free circulation of goods and the freedom to provide services. The Court of justice further clarified the implications of Article 56. It said that Article 56 TFEU gives rights not only on the provider of services but also on the recipient. 155 It also added that the abolition of State barriers to freedom to provide cross-border services would be compromised if it could be neutralised by obstacles resulting from publicly announced decisions of private operators 156 (for instance publically available terms and conditions) or private associations. In those situations, decisions of such private entities might have effects equivalent to those of public authorities to which Article 56 of the TFEU is primarily addressed. The Court judged that equality of treatment between nationals and non-nationals prohibit not only overt discrimination based on nationality but also discriminations linked to the place of residence. 157 It went as far as saying that even a measure putting at an advantage only residents of a specific region or area may be considered discriminatory; it does not have disadvantage only nationals of other Member States.

Transport: The non-discrimination principle in the transport sector prohibits traders to discriminate passengers in terms of price based on their place of residence or nationality. 158 Traders could only deviate from this principle if it concern social tariffs, e.g. preferential rates for citizens living in remote areas. It means that passengers of different modes of transport (air, maritime, bus and coach) can, in general, purchase tickets without any discrimination based on the customer's nationality or place of residence or on the place of establishment of carriers or ticket vendors.

Consumer protection cooperation: The review of the CPC Regulation concluded that an update of the Annex of the CPC Regulation would be necessary to ensure consistency between the sectorial and horizontal legislation which is already listed in the Annex. The Commission proposes to introduce Article 20 as one of the new provisions, Directives and Regulations that could be added to the Annex. In addition, Regulation (EC) No 1008/2008 on common rules for the operation of air services which also contain a provision on discrimination will be added.

Article 20 (2) is based on Article 18 and 56 of the TFEU. The non-discrimination provisions in the transport sector are similar to Article 20(2), but much stricter since traders, in general, may not invoke objective criteria justifying price differences based on residence or nationality. The consequence of the transport legislation and Article 20(2) is that customers have the right to equal treatment when buying tickets for transportation, but might face price discrimination when they arrive at the destination (e.g. in shops, museums, etc.).

-To what extent is the intervention coherent internally?

As stated above, the objective of the Services Directive is to realise the full potential of the Single Market for services by removing legal and administrative barriers to trade, which include removing barriers for service recipients. However, the Directive mainly focuses on addressing national requirements affecting the service itself and the service provider, not the service recipient. Furthermore, Article 20(2) is one of very few Articles in the Services Directive which indirectly creates an obligation for companies. The other Articles only put obligations on the Member States.

Article 20(2) contributes to the overall objective of the Services Directive, and even though it may be considered as different compared to the other Articles, it constitutes the backbone of service recipients' rights under the Directive and is therefore considered to be coherent internally

-To what extent is the intervention coherent with wider EU policy?

Discrimination based on nationality or place of residence including geo-blocking (i.e. online discrimination) is addressed in both The Digital Single Market Strategy (DSM) 159 and the Single Market Strategy for goods and services (SMS) 160 .

The DSM was presented by the Commission in May 2015. It is built on three pillars: (1) Better access for consumers and businesses to online goods and services across Europe; (2) Creating right conditions for digital networks and services to flourish; and (3) Maximising growth potential of our European Digital Economy. One of the initiatives under the first pillar is to prevent unjustified geo-blocking (i.e. online discrimination based on nationality or place of residence). It constitutes one of five initiatives (the other initiatives concern improved cross-border parcel delivery, harmonised EU contractual rights for online purchases, reducing VAT related burdens for traders, etc.) which together aims to fulfil the objective of the first pillar, namely, give better access for consumers and businesses to online goods and services across Europe.

In October 2015, the Commission presented the SMS. It aims to improve the functioning of the Single Market for products and services and to guarantee appropriate protection for EU citizen by targeted actions in three key areas: (1) Creating opportunities for consumers, professionals and businesses; (2) Encouraging and enabling the modernisation and innovation that Europe needs; and (3) Ensuring practical delivery that benefits consumers and businesses in their daily lives. The first key area is covering an action where the Commission has committed to prevent discrimination of consumers and entrepreneurs. The objective is to comprehensively fight discrimination based on nationality or place of residence, regardless if the transaction takes place online or offline.

The DSM and SMS have identified discrimination based on nationality or place of residence as an obstacle to a fully functioning (Digital) Single Market. Both strategies have announced action to address these issues. The objective of Article 20(2) of the Services Directive is coherent with the objectives of the recently launched initiatives.

7. EU added value

- What is the additional value resulting from the EU intervention, compared to what could be achieved by Member States at national level?
-To what extent do the issues addressed by the intervention require action at EU level?

As concluded in the assessment of the provision's effectiveness, Article 20(2) of the Services Directive has not brought substantive added value to service recipients or businesses trying to make full use of the Single Market. The lack of enforcement actions allow companies not complying with Article 20(2) to continue to do so without any repercussions from national enforcement authorities. The ECCs, which a majority of the Member States have designated as their "Article 21 bodies", has intervened against companies on behalf of consumers regarding violations of Article 20(2). However, the number of successful interventions where traders have changed their business practices remains very low.

Situations covered by Article 20(2) have a strong cross-border nature which is challenging for national enforcement authorities. It requires coordination at EU level in order to achieve an effective enforcement and functioning of the Article. The coordination between the different "Article 21 bodies" assisting consumers when facing discriminatory practices in violation of Article 20(2) could not have been achieved without the introduction of the Services Directive. Member States would have been confined to their national boarders without proper cooperation and coordination tools, and the Single Market would be undermined.

The need for the EU action has been also confirmed in the public consultation (see annex 2). Even though the views of stakeholders differ as to the elements of the policy response, there is a wide consensus that the problem of discrimination by service providers based on nationality or place of residence can be only tackled by the EU action.

No substantive EU added value has been achieved due to the lack of effectiveness of Article 20(2). However, in principle, it would not be possible to achieve the objectives of the provision without an EU intervention.

8. CONCLUSIONS

The assessment has shown that Article 20(2) of the Services Directive is not an effective EU intervention to achieve the pursued objectives. Several factors contribute to this: legal uncertainty regarding acceptable and non-acceptble justifications which companies can use to treat customers differently, and consequently, lack of enforcement by national authorities. Due to the lack of data, the efficiency of the provision could not be assessed. However, the objectives of the non-discrimination principle remain highly relevant, and coherent with other EU policy actions, such as the Digital Single Market Strategy and the Single Market Strategy for goods and services. The introduction of Article 20(2) has not brought any substantive added value due to its lack of effectiveness, but the analysis shows that EU intervention to address the issues concerned is necessary to reach the objectives.

Annex 9: Glossary

For the purposes of the Impact Assessment on geo-blocking and other discriminations based on nationality or place of residence the following definitions shall apply:

'Service' means any self-employed economic activity, normally provided for remuneration, as referred to in Article 57 of the Treaty. (It covers for example retail of goods and services, leisure services, services provided by architects and engineers, construction services such as installations and maintenance of equipment, food services such as hotels and restaurants, and leasing services.)

'Customer' means a consumer who or an undertaking which intends to purchase or purchase goods or services which are not for resale to another customer;

'Trader' means any natural person or legal person, irrespective of whether privately or publicly owned, who is acting, including through any other person acting in her name or on her behalf, for purposes related to her trade, business, craft or profession.

'Contract' means any agreement under which the trader transfers or undertakes to transfer the ownership of goods to the recipient, supplies or undertakes to supply a service, and the consumer pays or undertakes to pay the price thereof.

‘Distance contract’ means any contract concluded between the trader and consumer under an organised distance sales or service-provision scheme without the simultaneous physical presence of the trader and the consumer, with the exclusive use of one or more means of distance communication up to and including the time at which the contract is concluded;

'General conditions of access' means all terms, conditions and other information, including sale prices, regulating the access of customers to goods or services offered for sale by a trader, which are set, applied and made available to the public at large by or on behalf of the trader and which apply in the absence of an individually negotiated agreement between the trader and the customer.

'Discrimination based on nationality or place of residence' means unjustified differences in the general conditions of access to goods or services made available to the public at large, i.e. outright refusals to sell or the application of different conditions, based on customers' nationality or place of residence when intending to purchase or purchase goods or services which are not for resale to another customer. Differences in treatment do not amount to discrimination if the differences can be justified by objective criteria.

'Nationality or place of residence' cover also proxy factors which may end up being tantamount to nationality or residence, such as the country of the driving licence, the country of credit card issuance, the place of delivery, the country of origin of specifically dedicated IP addresses, the lack of credit history in a particular Member State, the lack of registration in the population registry, etc.

'Unjustified geo-blocking' means online discrimination based on nationality or place of residence.

'Re-routing' means any practice by information society service providers the purpose of which is to redirect the customer to an alternative version of the online interface to an information society service, on the basis of the customer's nationality or place of residence;

'Online interface' means any software the purpose of which is to allow customers to conclude a commercial transaction by means of information society services, such as a website;

'IP address' stands for Internet Protocol address and is a numerical label assigned to each device participating in a computer network that uses the Internet Protocol for communication. An IP address serves two principal functions: host or network interface identification and location addressing. IP addresses are often used for deducing location of an Internet user through geolocation services

'Geolocation services' means a software service used to deduce the geographical location (i.e. geolocation) of a device connected to the Internet. The most common way of obtaining a geolocation is through user's IP address. IP-based geolocation services use databases mapping IP addresses to Internet providers, cities and countries that allow them to locate a device based on its IP address with a high accuracy. Using this method, determining the country in which the device is located is 95%-99% percent accurate, while accuracy rates on naming the city from an IP address vary between 50%-80%. Geolocation information can be obtained in a number of other ways, including MAC address, RFID, Wi-Fi connection location, or GPS coordinates.

'Micro enterprise' means 161 micro those businesses having less than 10 employees and a turnover of €2m or less.

'Small enterprise' means those businesses having less than 50 employees and a turnover of €10m or less.

'SME Small and medium enterprises' are those businesses having less than 250 employees and a turnover of €50m or less.

Annex 10: Flash Eurobarometer 434 on geoblocking in B-to-B

a) Methodology

In February 2016 a Flash Eurobarometer 162 survey was conducted by TNS Political & Social with the view of collecting additional evidence on the extent of geoblocking in cross-border business-to-business (b2b) purchases. These purchases cover goods and services such as travel and accommodation (car rental, train tickets, hotels), office equipment (computers, stationery, software, coffee machine), uniforms and safety equipment, etc. Furthermore, these purchases must be subject to publicly available general sales conditions, i.e. not subject to specific conditions agreed through negotiations between the contracting parties. Excluded are also purchases of goods and services for resale.

The survey targeted businesses of any size that had acquired experience purchasing cross-border in 15 Member States (Austria, Belgium, Denmark, Estonia, France, Germany, Greece, Hungary, Italy, Ireland, Netherlands, Poland, Slovakia, Spain and United Kingdom). The survey was carried out by phone during the second half of February 2016, addressing a total of 4200 companies active in the following sectors: Manufacturing; electricity, gas, steam and air conditioning supply; water supply, sewerage, waste management and remediation activities; construction; wholesale and retail trade, repair of motor vehicles and motorcycles; transporting and storage; accommodation and food service activities and finally information/communication services. 163

b) Results

Overall, the survey found that:

Out of the companies surveyed two thirds had not tried to purchase a good or service in another Member States in the last 12 months. The main reasons for these firms not sourcing cross-border are

Out of the companies with experience of cross-border purchases, 10% faced problems.

The likelihood of experiencing problems is partly linked to the size of the company, e.g. larger companies with a turnover above €2 million are more likely not to have experienced any problem (65%) compared to smaller companies with a turnover below €100,000 (56%).

The companies reporting problems mainly experienced problems with “very high shipping prices” (17%), an inability to "get an invoice due to different VAT regulations” (10%).

Companies report more problems when trying to purchase goods compared to services.

For goods, the most problematic product categories include “electronic or ICT goods” (16%), “raw materials and construction material” (15%) and “digital goods” (11%).

For services, most expressed problems concerned after-sale services, such as “maintenance and repair services” (11%) and “sales and customer support services” (11%)

I. Purchasing goods and services cross-border

The introductory question on the respondents' experience of cross-border purchases of goods and services served as a filter question, as it was not possible to address the questions to a population only made up by businesses with experience in cross-border activity. Based on the responses to this question, more than half of the consulted businesses (68%) had not tried to purchase any good or service cross-border in the last 12 months prior to the survey.

Smaller businesses are less likely than larger ones to have engaged in cross-border purchases. For business with a yearly turnover less than €100,000, only 21% of the surveyed companies tried to purchase cross-border while the corresponding number for businesses with a turnover above €2 million amounts to 52%. The Member States with the highest proportion of companies that have tried to purchase outside of their own country in the last 12 months are Greece (61%), Austria (60%) and Ireland (60%). The Member States with the lowest proportion of companies having tried to purchase goods and services cross-border are Italy (19%), Poland (19%) and Spain (30%).

Graph 1: Companies having tried to purchase goods or services cross-border in the past 12 months (per Member States)

Base: All companies (N=4,200)

II. Reasons for not purchasing goods and services from other Member States

Companies that indicated that they had not tried to purchase any good or services cross-border were subsequently asked about their reasons for not doing so. A majority (77%) indicated that they had not tried because “they do not need anything abroad”. The second most mentioned reason, although significantly smaller by comparison, is the “worry about after-sales service not working from abroad” (13%). Slightly more than one in ten companies mentioned “standards or certifications for goods and services are different abroad” (12%). A similar proportion of companies that have not tried to purchase in other EU countries, explained that “shipping or delivery costs are too high” (11%).

Micro companies tend to be less likely to have purchased abroad compared to small companies (14% micro versus 10% small enterprises). They have more concerns about after sale-services than small companies. Microenterprises are also more likely to raise different standards or certifications for goods and services as a reason for not buying a product or service from abroad than small size companies (12% micro versus 8% small companies).

Table 1: Reasons for not trying to purchase goods and services cross-border

Base: Companies that have not tried to purchase any goods or services from another EU country (N=2,845)

III. Experience with purchasing from other Member States

A clear majority of companies that tried to buy cross-border did so without experiencing any problems (91%). A small minority of companies mentioned the product or service was not delivered successfully (4%) while only 2% of companies could not carry out the purchase because the seller gave a generic refusal to sell given their location abroad.

Table 2: The result when trying to buy goods or services cross-border

Base: Companies that have tried purchasing any goods or services from another EU country (N=1,355)

IV. Problems encountered when purchasing from other Member States

When asked about the problems encountered when purchasing from other Member States, almost two thirds (63%) of companies having purchased cross-border replied that they had not experienced any of the listed problems. 164 The most commonly cited problem (17% of companies) was "very high shipping prices”, while one in ten companies mentioned difficulties of getting an invoice due to different VAT regulations. Less common issues were different prices (excl. shipping costs) mentioned by 9% and the unavailability of after-sale services 8%. A small proportion of firms reported the unavailability of special discounts (4%), refusal to sell outside the country (3%) and difficulties accessing offer due to website being blocked (2%).

Table 3: Problems encountered when trying to purchase goods and services cross-border

Base: Companies that have tried purchasing any goods or services from another EU country (N=1,355)

V. Problems experiences during the purchasing process

Examining when firms encounter problems, most reported issues concern the delivery phase either due to goods not being delivered in time or not in the condition they expected (28%). More than one in ten companies indicated problems when “indicating the place of delivery” (14%) or “when informing that your company was located in another EU country” (12%). The least amount of reported problems concerned "downloading the purchased good or service” (7%).

Sample size limitation prevent a deeper analysis broken down by company characteristics level, but broadly it can be noted that companies selling goods are less likely to have experienced any of the listed products compares to companies selling goods (25% vs. 15%).

Table 4: The stage of the purchasing process when problems were faced

Base: Companies that have experienced a problem when buying from another country in the last 12 months (N=312)

VI. Type of products and services for which companies experiences problems

Little less than half of the respondents (47%) had not experienced any problems purchasing goods from another Member State, while about six of every ten companies (63%) had not experienced problems with purchasing services.

The type of goods which posed the most problems to purchase cross-border were electronic or ICT goods (16%), raw materials and construction material (15%) and digital goods (11%). Fewer than one in ten companies experienced problems with travel transport (8%), accommodation or hospitality (7%), and office supplies and equipment (8%). Just less than one in ten respondents reported problems in "other" products, and only a minor number had issues purchasing with uniforms, safety equipment (2%).

Most problems were encountered when seeking to purchase after-sale services such as “maintenance and repair services” (11%) and “sales and customer support services” (11%). Problems with other listed services were less common with fewer than one in ten companies have experienced issues with postal and communication services (8%), financial services (8%), IT services (7%) and less than one in twenty businesses experienced issues with professional services (3%) or utility services (2%).

Annex 11: barriers leading to geoblocking

9.Onerous VAT rules

The complications of having to deal with many different national systems represent a real obstacle for companies trying to trade cross-border both on and offline. The complexity of VAT for making intra-EU B2C supplies is cited by business as one of the top three barriers to cross-border e-commerce. 165

Since 1st January 2015, with the entry into force of new "place of supply" rules, VAT on all electronic services such as online gaming or cloud computing services is levied where the customer is based, rather than where the supplier is located. In parallel, an electronic registration and payment system (the Mini One Stop Shop - MOSS), has been implemented to reduce the costs and administrative burdens for businesses concerned. A first analysis of the implementation of the MOSS indicates that in 2015 EUR 3 billion VAT was collected through the system which is utilised by approximately 12 000 EU business and which has resulted in administrative burden savings of EUR 500 million as opposed to the alternative of having to register and account for the tax due in the Member State where the customer is located (the scope of the new VAT rules is however wider than the present initiative). This represents a substantial simplification and in such circumstances business should no longer be able to legitimately justify differences in treatment.

However, this possibility does not currently exist for businessess supplying tangible goods to consumers (retail). This means that if a business wishes to make supplies of goods to customers in other Member States they are still required to register and account for VAT in each of those Member States. The compliance costs have been estimated at EUR 8 000 annually per Member State 166 . This clearly represents a significant obstacle. While there is an existing simplification measure in place whereby businesses do not have to register up to the distance sales thresholds for intra-EU Business to Consumer (B2C) supplies of goods (EUR 35 000 or EUR 100 000 to be determined by the Member State of destination) 167 , these thresholds have caused uncertainty and complexity for business when the threshold is exceeded.

Under the DSM Strategy, the Commission has made a commitment to bring forward a proposal before the end of 2016 to extend the Single Electronic Mechanism for VAT registration, declaration and payment to all ecommerce supplies, both within and outside the EU.

The situation of B2B cross-border transactions is simpler, because in many cases, the seller has no obligation to charge VAT and it is the customer who accounts for VAT in his own country. It is particularly so when goods are moved from one Member State to another and for most services when the customer is a businesswith a VAT registration number and the supplier is not established in the Member State of the customer.

10.Costs of adapting to the consumer law of other Member States

Differences in national consumer laws may be a reason for refusing cross-border sales in a B2C context (this would not be relevant in a B2B relationship).

The Consumer Rights Directive has fully harmonised certain aspects of consumer and contract law applicable to distance sales to consumers (including online sales), such as pre-contractual information the consumer should receive and the right of withdrawal from the contract. However, in other areas (e.g. guarantee regarding the conformity of goods) there are only minimum EU rules, which Member States may supplement with stricter national requirements for the benefit of consumers. Following the 2008 Rome 1 Regulation, 168 traders may in principle choose the contract law of their own Member State in B2C transactions. However, in the event of a company directing its activities to another Member State, the possibly higher mandatory protections afforded to that consumer by the law of their country of residence apply.

The legal situation for distance sales means that traders of goods and services who wish to serve a pan-European market may potentially need to know about, and comply with certain specific mandatory consumer protection rules from 28 Member States. Finding out what such rules are case may be difficult and costly especially for SMEs 169 . Adding up only costs related to consumer contract law differences, the costs suffered by businesses when they sell to consumers in other EU Member States have been estimated at between EUR 4 and 8 billion 170 . These set-up costs are of course more difficult to amortize for smaller businesses interested in online cross-border sales.

This complex legal situation is being addressed in the context of the Digital Single Market Strategy by the Commission's proposal for a Directive on certain aspects concerning contracts for the online and other distance sales of goods and the proposal for a Directive on certain aspects concerning contracts for the supply of digital content. A key objective of these proposals is to ensure full harmonisation of key consumer rights in relation to distance sales of goods as well as digital content.

Differences between consumer laws, however, cannot be a reason for discrimination when the trader does not direct its activities to the country where the consumer has his habitual residence. In such a case a trader can choose the law of his country which will be the only applicable law. .

In case of conflict, the consumer may go to the courts of the country of their residence when the trader targeted activities to his/her Member State. Although cross-border court litigation is rare in B2C case, it may oblige suppliers to invest in legal advice in a foreign jurisdiction.

11.Different product-related rules

Technical specifications or rules on labelling and selling arrangements may also differ depending on where in the EU the customer is located. They may require the trader to adapt their products and packaging accordingly, in particular concerning linguistic versioning. The recent public consultation on product-related national legislations 171 (in particular labelling) showed that business operators find that these national requirements (including national application of rules harmonised at EU level) can hinder intra-EU trade. As shown notably in the replies to this public consultation, businesses perceive that the information and compliance costs can be prohibitive.

In addition, the B2B Eurobarometer shows that around 12% of the companies that have not purchased any goods or services from suppliers in other Member States refrained from doing so on account of standards and certifications for goods (and services) being different from their home Member State.

The Single Market Strategy announced an EU-Wide Action Plan to improve a better understanding of the principle of mutual recognition in the single market for goods.

12.Organisation of after-sales services in cross-border situations

Traders generally strive to provide good after-sales services, not only to enhance long-term relationships and customer loyalty, but also to meet the expectations from the customer and encourage them to come back and buy more. Every customer who is unhappy with the availability or quality of after-sales services will negatively affect the reputation of the trader. Customers do not expect products to be perfect, but they expect manufacturers to fix things quickly. 172 The lack of possibilities to arrange for good after-sales services could therefore be a reason for traders to deny cross-border sales.

Furthermore, it is also known that extra costs from after-sales services is one of retailers' main concerns about selling online to other EU countries; 32% of the respondents of a survey considered it to be an important factor for not selling abroad. 173 However, the survey did not distinguish between companies present in one country, and companies present, through subsidiaries or branches, in several countries.

Among the companies surveyed as part of the B2B Eurobarometer, those having experienced problems also indicated that in 8% of cases it was not possible to obtain after-sales services for goods purchased cross-border. Furthermore, 13% of those firms not sourcing cross-border stated that their concern about after-sales service not being available in their Member State was the reason for not buying cross-border.

13.Limited cross-border delivery options

EU law requires Member States to provide for cross-border postal parcels delivery up to 20 kilograms. 174 However, a lack of affordable, high-quality delivery services is consistently cited amongst the top reasons mentioned by both e-retailers and consumers 175 for not engaging in cross-border e-commerce. Stakeholders complain about a lack of transparency, the high prices of delivery for small shipments, the lack of interoperability between the different operators typically involved in a delivery of a cross-border shipment and the lack of convenience for the final consumer. Over 60% of manufacturing and retail (including wholesale) companies selling online across borders declared the high prices of cross-border delivery to be a problem for them 176 . In fact, several research studies estimate that list tariffs for cross-border parcel delivery (for non-account customers) charged by national postal operators are often two to five times higher than comparable domestic prices 177 .

As observed in the responses to the public consultation, as the quality of delivery varies across Europe and postal operators, traders who want to ensure a high quality service often invoke the lack of quality of existing delivery options to refuse to deliver to a specific country. Furthermore the responses to the B2B Eurobarometer shows that high prices is a frequent problem for companies buying goods cross-border, with 17% of the those having experienced problems attributing it to very high shipping costs. The high cost of delivery was also mentioned by 11% of respondent as the reason for why they have not purchased any goods or services cross-border.

Some of the challenges of parcel delivery will be addressed, as announced in the Digital Single Market Strategy, by the proposed Regulation on cross-border parcel delivery. 178 It will aim at improving transparency of costs for cross-border parcel delivery and the regulatory oversight of the cross-border parcel market.

14.Difficulties in cross-border payments

As consistently mentioned in studies 179 and public consultations, some operators prefer not to sell to other Member States because their fraud prevention system is based on access to informational resources such as public registries which may not be available in other countries. As a consequence, they may find that they are suffering higher fraud rates on cross-border sales than on national sales, which may result in the unprofitability of cross-border operations. However,71% of card fraud is linked to domestic transactions and transactions outside the Single Euro Payment Area (SEPA), while cross-border fraud within SEPA only accounts for 29% of frauds. Although only 2% of all transactions were acquired from outside SEPA, they accounted for 22% of all fraud. 180 According to the database of complaints collected by the European Commission, only 1% of the justifications used by traders for difference in treatment referred to fraud avoidance.

Besides, several EU law provisions have taken away the relevance of this justification, such as the European order for payment procedure 181 and the European Small Claims Procedure 182 . As a consequence, difficulties to recover debts below EUR 2,000 (respectively in future EUR 5000) in cross-border cases have been substantially lowered. However, the small claims procedure only helps when the merchant is able to identify the target for the claim.

With modernisation of payments technology and after the introduction of EU legislation there are fewer reasons for internet retailers to refuse transactions for payment-related concerns. With the SEPA Regulation 183 , as of October 2016, all Member States will be using the same schemes for credit transfers and direct debits in euro and it will no more be possible to discriminate on the basis of the country where the payment service provider is located where payments in euro are concerned.

The Regulation on cross border payments 184 in euros also provides that charges levied by banks on retailers in respect of cross-border payments shall be the same as the charges for corresponding national payments 185 . This means that retailers have no justification for refusing to receive transactions by credit transfer or direct debit for payments in euro from any Member State.

Regarding card payments, the Regulation on interchange fees for card-based transactions was adopted in 2015 to address the problem of widely varying collectively-agreed inter-bank fees regarding card and card based transactions 186 . This Regulation introduces EU wide ceilings for inter-bank fees regarding transactions with consumer debit and credit cards. It also addresses rules limiting retailers' possibilities to steer consumers to using cards with lower fees. The Regulation limits the application of Honour all Cards Rules when this would oblige retailers who accept cards with regulated fees also to accept cards with unregulated fees ('Honour all Products'). However, card schemes can impose Honour all Cards Rules when they oblige retailers to accept the same sort of card regardless of the country in which the card was issued or the bank which has issued the card ('Honour all Issuers'). This means that retailers cannot refuse cards issued in other Member States with fees capped by the Regulation. Nevertheless, retailers often decide not to accept foreign cards.

Finally, to adress concerns of security and liability, the revised Payment Services Directive (PSD2) adopted in November 2015 187 will introduce common security requirements to authenticate the user of a payment service; secure customer authentication 188 will be required for all remote electronic payment transactions 189 unless a specific exemption has been agreed at EU level (eg for low value payments). Where the retailer or his bank fails to adopt strong customer authentication, they shall be liable for any fraud. Once PSD2 will be of full application in 2018, those retailers whose bank has adopted strong customer authentication for payments will not run any greater risk of fraud if a payment is issued from another Member State.

15.Legal barriers

Companies may refuse to sell to residents of other Member States because of legislation forcing them to do so. This could for example concern services which may be legal in some Member States but not in others, such as the sale of "snus" in Sweden which traders are prohibited to sell in all other Member States. 190 Similarly, firearms or certain alcoholic products may be legally available in one Member State but not in another. Insofar as these restrictions derive from national law which is compatible with European law, geoblocking would be justified.

(1) COM(2015) 192 final

(2) COM(2015) 550 final

(3) COM(2015) 634 final

(4) COM(2015) 635 final

(5) Inception impact assessment on "Modernising VAT for cross-border e-commerce", http://ec.europa.eu/smart-regulation/roadmaps/docs/2016_taxud_002_iia_vat_en.pdf

(6) COM(2016) 148 final

(7) Adoption by the Commission scheduled for 18 May 2016

(8) COM(2015) 627 final

(9) Regulation (EC) No 2006/2004 of the European Parliament and of the Council of 27 October 2004 on cooperation between national authorities responsible for the enforcement of consumer protection laws, OJ L 364, 9.12.2004, p. 1.

(10) SWD(2016) 70 final, Geo-blocking practices in e-commerce, Issues paper presenting initial findings of the e-commerce sector inquiry conducted by the Directorate-General for Competition, preliminary report scheduled for mid-2016, final report scheduled for early 2017.

(11) Directive 2006/123/EC of the European Parliament and of the Council of 12 December 2006 on services in the internal market, OJ L 376, 27.12.2006.

(12) Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal aspects of information society services, in particular electronic commerce, in the Internal Market ('Directive on electronic commerce') , OJ L 178, 17.7.2000.

(13) The following services are excluded from the scope of either the Services Directive and/or the e-Commerce Directive: non-economic services of general interest, financial services, electronic communication and network services, transport services, services of temporary agents, healthcare services, audiovisual services, gambling services, services related to official authorities, social services, private security services and services provided by notaries and bailiffs. See Art 2(2) of the Services Directive and Art 1(5) of the e-Commerce Directive

(14) This means that Information society services should be supervised at the source of the activity, (recital 22 of the e-commerce directive)

(15) See Annex 7.

(16) Fixed book prices (by law or business agreement) exist in Austria, France, Germany, Greece, Slovenia and Spain. Belgium notified the Commission about its plans to introduce fixed prices for e-books in Flanders.

(17) Notably Article 101 TFEU; Commission Regulation (EU) no 330/2010 of 20 April 2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices, OJ L 102/1, 23.4.2010, ("Vertical Block Exemption Regulation") and the Guidelines on Vertical Restraints, OJ C 130/1,19.5.2010

(18) See for instance Case C-388/01, Commission vs Italy, ECLI:EU:C:2003:30, points 13-14

(19) Recital 95 specifies that such objective criteria is for example additional costs because of the distance, different market conditions, extra risks linked to different rules in different Member States and lack of intellectual property rights .

(20) However, the e-commerce Directive does not affect the law applicable to contractual obligations relating to consumer contracts.

(21) See for instance Case C-291/13, Sotiris Papasavvas v O Fileleftheros Dimosia Etaireia Ltd and Others, ECLI:EU:C:2014:2209, points 26-30.

(22) Directive 2011/83/EU

(23) Commission Regulation (EU) No 330/2010 of 20 April 2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices, OJ L 102/1, 23.4.2010, Articles 4 (b) and (c).

(24) See Vertical Guidelines, paragraph 61.

(25) Those travelling spend on average 11.6 days per year abroad. BEREC International Roaming Analysis of the impacts of “Roam Like at Home” (RLAH), 17 December 2014, http://berec.europa.eu/eng/document_register/subject_matter/berec/opinions/4826-international-roaming-analysis-of-the-impacts-of-8220roam-like-at-home8221-rlah

(26) The total consumption made by visitors for and during tourism trips was estimated to 187 billion in 2013

(27) 40% of the EU population lives in border regions

(28) Gomez-Herrera, E., Martens, B., and Turlea, G. 'The drivers and impediments to cross-border e-commerce in the EU', JRC/IPTS Digital Economy Working Paper 2013-02, 2013

(29) Nevertheless, sometimes this does happen, but typically only when identification of the customer is required (e.g. registration for a discount card).

(30) European Commission, Flash Eurobarometer 397, 'Consumer attitudes towards cross-border trade and consumer protection', 2015

(31) Eurostat News Release 218/2015 - 11 December 2015 , http://ec.europa.eu/eurostat/documents/2995521/7103356/4-11122015-AP-EN.pdf/276b6a7c-69a6-45ce-b6bf-488e975a8f5d

(32) ECCRS, Your Europe Advice, European Consumer Centres, complaints addressed directly to the Commission, questions posed by MEPs and petitions to the European Parliament.

(33) European Commission, Flash Eurobarometer 434, 'B2B Geoblocking', 2016

(34) See Annex 10. This percentage varies between Member States, from only 19% in Italy and Poland to 60% in Austria, also depending on the sectorial composition of the firms surveyed in respective Member State.

(35) Defined in the study as DE, FR, IT, ES , UK, NL, SW, IR, AT and CH

(36) Ipsos MORI 2015 for PayPal

(37) See also Consumer Conditions Scoreboard 2015, p. 82, column "online service"

(38) https://ideas.repec.org/p/ipt/decwpa/2015-04.html

(39) http://ec.europa.eu/competition/antitrust/sector_inquiries_e_commerce.html

(40) ; https://ideas.repec.org/p/ipt/decwpa/2015-13.html ;

(41) Martens and Milkians (2013), 'Large-scale Web-scraping Evidence on EU On-Line Price Differences Driven by the Country of Purchase'.
Study of March 2016, covering 19 national branches (Austria, Belgium, United Kingdom, Czech Republic, Denmark, Spain, Finland, France, Croatia, Hungary, Ireland, Italy, Lithuania, Netherlands, Poland, Portugal, Sweden, Slovenia, Luxembourg) and the non-country specific EU website of one of the largest European car rental companies. The car rental options were checked for 25 locations (airports in capital cities of 25 EU Member States) and for 9 different date ranges, covering holidays and workdays in order to account for seasonality. On each website, the default search options were chosen, which do not include add-ons like insurance, etc. Each website and location/date combination was queried from computers with IP addresses in all 28 Member States. Thus, the price discrimination was detected based on users' IP addresses. For each user, around 6000 quotes for rental cars were scraped from each website. The web scraping study was performed for the Commission in March 2016 by Prof Christo Wilson, Northeastern University, USA.

(42) Discrimination of consumers has not been detected in terms of availability of offers. Consumers from all Member States can in addition to the non-country specific website access and purchase on any of the country-specific websites of the same car rental company, where no price discrimination has been detected. Prices of the same offers (i.e. same rental location, dates, car types and other parameters) may differ among websites (incl. all country-specific and the non-country specific website), but none of the country-specific websites modifies the price based on the IP address of the consumer.

(43) For the reasons explained in the assessment of options (see option 3), even if there is directing of activities to other Member States, in some specific cases, difference of treatment should be prohibited .

(44) see Press releases of 11/08/2014 ( http://europa.eu/rapid/press-release_IP-14-917_en.htm ) and 28/10/2014 (http://europa.eu/rapid/press-release_IP-14-1209_en.htm)

(45) The figure does not show problems occurring after the order has been placed (delivery) or when an experienced problem did fit into the answer options ("others"). For more information see Annex 10.

(46) The responses are filtered based on those companies having engaged, and experienced problems with cross-border purchasing process in the last 12 months. Multiple problems could be indicated by each respondent and thus the percentages are not cumulative.

(47) Ecommerce Europe, Analysis of the survey “Barriers to Growth”, 2015, http://www.ecommerce-europe.eu/stream/survey-barriers-to-growth-ecommerce-europe-2015.pdf

(48) Regulation (EC) No 593/2008 of the European Parliament and of the Council of 17 June 2008 on the law applicable to contractual obligations (Rome I)

(49) Around a quarter of online consumers have concerns about high delivery costs (27 %), high return shipping costs (24 %), and long delivery times (23 %) when it comes to purchasing online from another EU Member State. GfK for the European Commission, Consumer survey identifying the main cross-border obstacles to the DSM and where they matter most, 2015

(50) Directive (EU) 2015/2366 of the European Parliament and of the Council of 25 November 2015 on payment services in the internal market

(51) Marc Michel Josemans vs. Burgemeester van Maastricht, Case C137/09 of 16 December 2010, ECLI:EU:C:2010:774

(52) Guidelines on Vertical Restraints, OJ C130/1 of 19.5.2010

(53) A ban of internet sales by retailers is considered as a passive sales restriction which has the object of restricting competition. It qualifies as a hardcore restriction under Article 4 (b) and (c) of the Vertical Block Exemption Regulation and runs afoul of Article 101(1) TFEU unless justified by clear efficiencies under Article 101(3) TFEU.

(54) SWD(2016) 70 final, final report scheduled for early 2017

(55) DG COMP report, scheduled for early 2017

(56) http://ec.europa.eu/internal_market/services/docs/services-dir/studies/20091210_article20_2_en.pdf

(57) See Annex 8

(58) See Commission Staff Working Document with a view of establishing guidance on the application of Article 20(2) of Directive 2006/123/EC on services in the internal market ("the Services Directive") (SWD(2012) 146 final) and its annex providing an overview of the competent authorities. In some Member States, only national courts are mentioned as such authorities.

(59) Impact Assessment

(60) Commission Staff Working Document with a view of establishing guidance on the application of Article 20(2) of Directive 2006/123/EC on services in the internal market ("the Services Directive") SWD(2012) 146 final

(61) Regulation (EU) No 260/2012 of the European Parliament and of the Council of 14 March 2012 establishing technical and business requirements for credit transfers and direct debits in euro

(62) GfK Mystery Shopping Survey and JRC/IPTS calculations (forthcoming)

(63) Ibid.

(64) See section 2.1.2.3

(65) Directive on certain aspects concerning contracts for the supply of digital content (COM(2015)634 final) and the Directive on certain aspects concerning contracts for the online and other distance sales of goods (COM(2015) 625 final)

(66) http://ec.europa.eu/internal_market/services/docs/services-dir/implementation/report/SWD_2012_146_en.pdf

(67) Buying services everywhere in the EU - A practical guide for consumers, Publications Office of the European Union, 2014

(68) see Press releases of 11/08/2014 ( http://europa.eu/rapid/press-release_IP-14-917_en.htm ) and 28/10/2014 (http://europa.eu/rapid/press-release_IP-14-1209_en.htm)

(69) Case C-233/09, Dijkman

(70) Case C-281/98, Angonese, paragraph 32

(71) Case C-103/08, Gottwald, paragraphs 27 and 28

(72) Case C-388/01, Commission v. Italy, paragraph 14

(73) COM(2015) 634final

(74) COM(2015) 635 final

(75) For B2B purchases, only those purchases subject to general selling terms and conditions made available to the public at large would be covered by this option.

(76) Services covered by the VAT Mini One-Stop-Shop (Council implementing regulation EU No282/2011)

(77) Directive (EU) 2015/2366 of the European Parliament and of the Council of 25 November 2015 on payment services in the internal market, amending Directives 2002/65/EC, 2009/110/EC and 2013/36/EU and Regulation (EU) No 1093/2010, and repealing Directive 2007/64/EC, OJ L 337, 23.12.2015,

(78) in accordance with Regulation (EU) 2015/751

(79) See Vertical Guidelines, paragraph 61.

(80) Commission Regulation (EU) No 330/2010 of 20 April 2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices, OJ L 102/1, 23.4.2010.

(81) For B2B purchases, only those purchases subject to general selling terms and conditions made available to the public at large would be covered by this option.

(82) For B2B purchases, only those purchases subject to general selling terms and conditions made available to the public at large would be covered by this option.

(83) Price (P) equals tariff * time; Tariff: The wage cost of the employee assessing the request and replying. The wage costs will differ between Member States, but the EU-28 is used as a proxy (€24.6/hour). Time: The time taken up by the reply to each request for justifications should be minimal considering that the trader should be well aware the reason used for applying different condition (as it requires a concerted effort to do so) and as the reply can be provided in the trader's own language. Assessing the request and providing a reply should not take more than 5 minutes (or 1/12 hour). Using these general estimations for price and quantity, the total cost for each justification would be 24.6*(1/12) = €2.05.

(84) On the basis of an EU average of €24.6/hour; Eurostat, Labour cost levels, Industry, construction, services (2014)

(85) ITIF, The Economic Costs of the European Union’s Cookie Notification Policy BY DANIEL CASTRO AND ALAN MCQUINN | NOVEMBER 2014,

(86) Council Directive 2006/112/EC — the EU’s common system of value added tax (VAT)

(87) Unless stated otherwise the evidence in the following sections refers to specific situation a) distance sale of tangible goods when the customer arranges for cross-border delivery. The analysis of the electronics sector is based on four product categories in ten Member States accounting for more than half of EU online sales of electronic goods. For the economic model used in the assessment of this option and its limitations, please see annex 4; Full report:. Duch-Brown, N. and B. Martens (2016) , The Welfare effects of lifting geoblocking restrictions in the EU Digital single Market. JRC/IPTS Digital Economy Working Paper (forthcoming).

(88) Defined in the study as DE, FR, IT, ES , UK, NL, SW, IR, AT and CH

(89) Ipsos MORI 2015 for PayPal

(90) 2015 Study by lesfrontaliers.lu ( http://www.lesfrontaliers.lu/finances/faut-il-faire-ses-courses-au-luxembourg-en-france-ou-en-belgique ). An older similar study comparing the prices of seven product categories between Belgium, the Netherlands, France, Luxembourg and Germany had demonstrated a price differential of 9% ( http://www.test-achats.be/action/espace-presse/communiques-de-presse/2011/comparaison-de-prix-produits-electroniques-la-belgique-9-plus-chere ). A 2011 study found large differences in the prices of 74 consumer products sold in Austria and Germany, with for example a difference of 74% in the price of the same face cream sold by the same drugstore (http://www.konsument.at/markt-dienstleistung/preisvergleich-oesterreich-deutschland-318879174916?pn=1). A study conducted in 2014 found that the price differences of drugstore products sold in Austria in Germany had persisted ( https://tirol.arbeiterkammer.at/beratung/konsumentenschutz/TestsundPreisvergleiche/_Oesterreich-Aufschlag__bei_Drogerieartikeln.html ). Similarly, according to a 2015 test, a basket of 10 food products sold by the same supermarket cost 47 € in Finland and 39 € in Estonia ( http://www.mtv.fi/uutiset/kotimaa/artikkeli/mtv-vertasi-ruokakorin-hinta-viron-ja-suomen-prismoissa/4712440 ).

(91) http://www.europe-consommateurs.eu/fileadmin/user_upload/eu-consommateurs/PDFs/publications/etudes_et_rapports/4.4.8_synthese_comparaison_des_prix_ikea-cuisine_france-allemagne.pdf

(92) Cardona, Duch-Brown, Francois, Martens and Yang (2015) "The macro-economic impact of cross-border e-commerce in the EU Digital Single Market", JRC/IPTS Digital Economy Working Paper nr 2015-09.

(93) Ibid.

(94) CapGemini: "VAT Aspects of cross-border e-Commerce - Options for modernisation", (forthcoming)

(95) The vast majority of Member States apply the EUR 35 000 threshold, notably France who recently reduced the threshold due to concerns on distortions of competition.

(96) For the economic models used in the assessment of this option, please see annex 4; Full report:. Duch-Brown, N. and B. Martens (2016), The Welfare effects of lifting geoblocking restrictions in the EU Digital single Market. JRC/IPTS Digital Economy Working Paper (forthcoming).

(97) Cardona, Duch-Brown, Francois, Martens and Yang (2015) "The macro-economic impact of cross-border e-commerce in the EU Digital Single Market", JRC/IPTS Digital Economy Working Paper nr 2015-09.

(98) Ibid.

(99) The extent to which different options would achieve the objectives

(100) The relations of costs to benefits

(101) The coherence of each option with the overarching objectives of EU policies

(102) For presentation purposes the answer options "strongly agree"/"agree" and "strongly disagree"/"disagree" are presented

(103) Alternatively, one can see this as the difference between (comparative) static and dynamic approaches.

(104) Alternative definitions of the market size give similar results. See Duch-Brown and Martens (2016) for further details.

(105) Product differentiation is an important source of market power in many industries. One of the main motivations to estimate demand/supply of differentiated products is to understand the role of product differentiation in consumer choice, changes in market structure and the intensity of competition.

(106) The data were purchased by DG JRC from GfK, a German market intelligence company.

(107) Note that we have no data on trade flows prior to lifting GB restrictions. The trade flows reported here are estimated increases in trade, whatever the level of trade before lifting restrictions.

(108)

(109) "Supply-side Barriers to Cross-border eCommerce in the EU Digital Single Market", Melisande Cardona and Bertin Martens, JRC-IPTS Digital Economy Working Paper 2014-13; JRC92294. See also: YouGov Psychonomics (2009), Mystery Shopping Evaluation of Cross-Border E-Commerce in the EU

(110) http://europa.eu/rapid/press-release_IP-14-917_en.htm

(111) http://www.euractiv.com/sections/competition/eu-probes-disneyland-paris-alleged-price-discrimination-316675

(112) http://europa.eu/rapid/press-release_IP-14-1209_en.htm

(113) Figures do not add up due to overlaps

(114) The responses are filtered based on those companies having engaged in cross-border purchases in the last 12 months. Multiple problems could be indicated by each respondent and thus the percentages are not cumulative.

(115) The responses are filtered based on those companies having engaged, and experienced problems with cross-border purchases in the last 12 months. Multiple goods and services could be indicated by each respondent and thus the percentages are not cumulative.

(116) Your Europe Advice, European Consumer Centres, complaints addressed directly to the Commission, questions posed by MEPs and Parliamentary petitions

(117) http://ipts.jrc.ec.europa.eu/publications/pub.cfm?id=6084

(118) iTunes accounted for 63% of worldwide digital media downloads in 2013.

(119) https://ec.europa.eu/jrc/en/publication/eur-scientific-and-technical-research-reports/geographic-fragmentation-eu-market-e-books-case-amazon

(120) In the sector inquiry e-books are included in a category of consumer goods together with tangible books, CDs and DVDs and BluRay.

(121) Most Kindle readers may be bought from any European Amazon website, except for amazon.nl, which serves only Dutch consumers. From the US, only consumers without a dedicated EU website can buy a Kindle reader (except for NL). Certain Kindle readers (Fire tablets with 4G connectivity and "Special Offers devices") are available only for customers whose country of residence is FR, IT, ES, UK, DE.

(122)

It is to note that the results of the public consultation cannot be considered representative as it was the choice of respondents to take part in the consultation and not pre-selected sample of respondents. In total 251 consumers, 58 businesses and 78 business associations responded to the questionnaire.

(123) "This questionnaire refers only to geo-blocking or other restrictions in so far as they do NOT those related to copyright and licensing practices (such as sport events), which will be addressed by the Commission in separate initiatives. They do cover, however, geo-blocking of non-copyrighted content, including political news reporting."

(124)

Questions 2, 3 7, 8 and 13.

(125) The only sector in the top 8 that was excluded was that for Non-electrical household products & interior design. This sector was replaced by Computer games and software in order to investigate a more diversified sample of markets.

(126) http://ec.europa.eu/consumers/consumer_evidence/market_studies/obstacles_dsm/docs/21.09_dsm_final_report.pdf

(127) Geoblocking during ‘access’ is either due to rerouting, blocking access to a website or not offering the exact same product that was found when the website was assessed by a domestic shopper

(128) European Consumer Complaints Registration System

(129) See additional explanations about the sources of complaints in point 2.6 and 3.

(130) http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32006L0123&from=EN

(131) http://ec.europa.eu/growth/tools-databases/newsroom/cf/itemdetail.cfm?item_id=8076

(132) See Annex 7

(133) http://ec.europa.eu/growth/single-market/services/economic-analysis/index_en.htm European Commission analysis available under

(134) With a view to establishing guidance on the application of Article 20(2) of Directive 2006/123/EC on services in the internal market, SWD(2012) 146 final, see Annex I

(135) With a view to establishing guidance on the application of Article 20(2) of Directive 2006/123/EC on services in the internal market, SWD(2012) 146 final

(136) Between 2009 and 2015, the European Commission has collected 1500 complaints on alleged discrimination direct or indirect by utilising various internal channels, namely ECCRS, the European Consumer Centre Network and Your Europe Advice, see Annex 7

(137) http://ec.europa.eu/growth/tools-databases/newsroom/cf/itemdetail.cfm?item_id=8076

(138) European Commission, 'Study on business practices applying different condition of access based on the nationality or the place of residence of service recipients – Implementation of Directive 2006/123/EC on Services in the Internal Market', 2009. The study covered car rental, digital downloads, online sales of electronic goods and tourism.

(139) A Single Market Strategy for Europe – Analysis and Evidence (SWD(2015) 202 final) p. 52-53

(140) Study "Discrimination of Consumers in the Digital Single Market" (2013), European Parliament Directorate-General for internal policies, Policy department A: Economic and scientific policy (IP/A/IMCO/ST/2013-03)

(141) http://ec.europa.eu/growth/tools-databases/newsroom/cf/itemdetail.cfm?item_id=8076

(142) For example; Finland, Latvia, Luxembourg and Poland refer to national courts and Germany refer to 7000 trade and business authorities, chambers of auditors, lawyers and tax consultants (Annex II of SWD(2012) 146 final)

(143)

European Parliament resolution of 21 September 2010 on Completing the Internal Market for E-Commerce

(2010/2012(INI)); points 31 and 32; Conclusions of Competitiveness Council on Digital Single Market and

Governance of the Single Market of 30-31 May 2012

(144) Enhanced Consumer Protection – the Services Directive 2006/123/EC, Analysis of Article 20.2 and Article 21 related consumer complaints reported to ECC-Net between 2010 and 2012.

(145) 2015 ECC Anniversary report, p.8; http://ec.europa.eu/consumers/solving_consumer_disputes/non-judicial_redress/ecc-net/docs/ecc_net_-_anniversary_report_2015_en.pdf

(146) http://berec.europa.eu/eng/document_register/subject_matter/berec/opinions/4826-international-roaming-analysis-of-the-impacts-of-8220roam-like-at-home8221-rlah Those travelling spend on average 11.6 days per year abroad. BEREC International Roaming Analysis of the impacts of “Roam Like at Home” (RLAH), 17 December 2014,

(147) Duch-Brown, N. and Martens, B., 'The European Digital Single Market', JRC/IPTS Digital Economy Working Paper [DSM Strategy SWD]

(148) http://europa.eu/rapid/press-release_IP-14-1209_en.htm

(149) Commission web scraping study performed by College of Computer and Information Science, North-eastern University, Boston 2015.

(150) Articles on the initiative on the amusement park were published in UK, FR, DE, BE, NL, ES, PL, DK, LV, FI, IT, SE, IE, CZ, AU, RO, SI, CY, HU, LT and HR, and in the US and Brazil.

(151) Consumer market study on the functioning of e-commerce and Internet marketing and selling techniques in the retail of goods, Civic Consulting (2011), final Report for DG SANCO.

(152) Annex 2 of the Impact assessment on the proposal for a Regulation of the European Parliament and of the Council on addressing geo-blocking and other discriminations based on residence or nationality.

(153) Annex 7 of the Impact assessment on the proposal for a Regulation of the European Parliament and of the Council on addressing geo-blocking and other discriminations based on residence or nationality.

(154) Case C-233/09, Dijkman

(155) Case C-281/98, Angonese, paragraph 32

(156) Case C-103/08, Gottwald, paragraphs 27 and 28

(157) Article 1(a) and Article 4(2) of both Regulation (EU) No 181/2011 concerning the rights of passengers in bus and coach transport and Regulation (EU) No 1127/2010 concerning the rights of passengers when by sea and inland waterway regulate the principle of non-discrimination. Article 23(2) and 16(1 of Regulation (EC) No 1008/2008 on common rules for the operation of air services refer to non-discrimination in air transport.

(158) http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52015DC0192&from=EN

(159) http://ec.europa.eu/DocsRoom/documents/14007/attachments/1/translations/en/renditions/native

(160) http://ec.europa.eu/enterprise/policies/sme/facts-figures-analysis/sme-definition/index_en.htm .

(161) http://ec.europa.eu/public_opinion/index_en.htm For additional information on the methodology and results of this Eurobarometer please consult:

(162) The sample size in each Member States is a follows: N=400 in France, Germany, Italy, Poland, Spain and United Kingdom. N=200 in Austria, Belgium, Denmark, Estonia, Greece, Hungary, Ireland, Netherlands and Slovakia.

(163) Available reply options: Difficulty to access the offers online (access to website blocked); Refusal to sell outside the suppliers country; Refusal to grant a special discount due to the fact that your establishment is located in another EU country; Could not get an invoice due to different VAT regulations; Very high shipping process; Different prices (excluding shipping costs) because your establishment is located in another EU country; Impossible to get after sales services back home; None of the above; Other; Don’t Know

(164) Ecommerce Europe, Analysis of the survey “Barriers to Growth”, 2015, http://www.ecommerce-europe.eu/stream/survey-barriers-to-growth-ecommerce-europe-2015.pdf

(165) Ongoing study by Deloitte “Options for the modernisation of VAT for cross-order e-commerce”. These calculations are based on the standard cost model.

(166) The vast majority of Member States apply the EUR 35 000 threshold, notably France who recently reduced the threshold due to concerns on losses.

(167) Regulation (EC) No 593/2008 (OJ L 17716). The Regulation does not apply to Denmark

(168) The impact assesment accompanying the proposed Directive on certain aspects concerning contracts for the supply of digital content (COM (2015)634) and the Directive on certain aspects concerning contracts for the online and other distance sales of goods (COM(2015) 635) estimated the costs stemming from differences in contract law suffered by businesses when they sell to consumers in other EU Member States amount to between EUR 4 and 8 billion. These set-up costs are of course more difficult to amortize for smaller businesses interested in online cross-border sales

(169) See Impact assessment to the Common European Sales Law, SEC (2011) 1165 final, and the Impact assessment on Digital contracts rights, SWD(2015) 274 final/2

(170) http://wcmcom-ec-europa-eu wip.wcm3vue.cec.eu.int:8080/justice/newsroom/contract/opinion/150609_en.htm

(171) "Winning in the Aftermarket", by Morris A. Cohen, Narendra Agrawal, and Vipul Agrawal, Harvard Business Review, May 2006

(172) Consumer Conditions Scoreboard, 2015 edition

(173)

In line with article 3 (5) the amended Directive 97/67/EC of the European Parliament and of the Council of 15 December 1997 on common rules for the development of the Internal market of Community postal services and the improvement of quality of service, Member States shall ensure that postal parcels received from other Member States and weighing up to 20 kilograms are delivered within their territory. Where Member States have increased the weight limit of the universal service in accordance with Article 3(5) of Directive 97/67/EC, there is no lack of standard delivery of parcels up to 10 kg, or up to 20 kg. If, however, the parcel is heavier than 10 or 20 kg, such a lack of delivery option can be invoked. The trader may also in any case legitimately invoke the lack of quality of existing delivery options to refuse to deliver to a specific country.

(174) Around a quarter of online consumers have concerns about high delivery costs (27 %), high return shipping costs (24 %), and long delivery times (23 %) when it comes to purchasing online from another EU Member State. GfK for the European Commission, Consumer survey identifying the main cross-border obstacles to the DSM and where they matter most, 2015

(175) European Commission, Flash Eurobarometer 413, 2015

(176) Meschi, M. et al., Intra-Community cross-border parcel delivery, FTI Consulting for the European Commission, 2011; Okholm, H. B. et al., e-Commerce and delivery - A study of the state of play of EU parcel markets with particular emphasis on e-commerce, Copenhagen Economics for the European Commission, 2013; Civic Consulting for the European Commission, Consumer market study on the functioning of e-commerce and Internet marketing and selling techniques in the retail of goods, 2011; YouGov Psychonomics for the European Commission, Mystery Shopping Evaluation of Cross-Border e-commerce in the EU, 2009

(177) COM(2016)XXX final upcoming

(178)

See for instance “Study on business practices applying different condition of access based on the nationality or the place of residence of service recipients - Implementation of Directive 2006/123/EC on Services in the Internal Market , November 2009, http://ec.europa.eu/internal_market/services/docs/services-dir/studies/20091210_article20_2_en.pdf”

(179) European Central Bank, 4th Report on Card Fraud, July 2015, https://www.ecb.europa.eu/pub/pdf/other/4th_card_fraud_report.en.pdf

(180) Regulation (EC) No 1896/2006 of the European Parliament and of the Council of 12 December 2006

(181) Regulation (EC) No 861/2007 of the European Parliament and of the Council of 11 July 2007 establishing a European Small Claims Procedure

(182) Regulation 260/2012 on SEPA credit transfers and direct debits in euro ,OJ L 94/22, 30.3.2012,.

(183) Regulation 924/2009 on cross border payments, OJ L 266/11, 9.10.2009

(184) For payments up to EUR 50 000, of the same value and in the same currency.

(185) Regulation 2015/751 on interchange fees for card-based transactions, OJ L 123/1, 19.5.2015.

(186) DIRECTIVE 2015/2366 of 25 November 2015 on payment services in the internal market, OJ L 337/35, 23.12.2015.

(187) 'Strong customer authentication’ means an authentication based on the use of two or more elements categorised as knowledge (something only the user knows), possession (something only the user possesses) and inherence (something the user is) that are independent, in that the breach of one does not compromise the reliability of the others, and is designed in such a way as to protect the confidentiality of the authentication data.

(188) See Article 97.

(189) Tobacco products for oral use, except those intended to be inhaled or chewed, are prohibited in the EU according to Directive 2014/40/EU. Swedish snus is tobacco which is neither smoked nor chewed and is therefore prohibited to sell within the EU. Sweden has been granted a permanent exemption from the sales ban on snus.