EUROPEAN COMMISSION

Brussels, 7.11.2022

SWD(2022) 350 final

COMMISSION STAFF WORKING DOCUMENT

IMPACT ASSESSMENT REPORT

Accompanying the document

Proposal for a Regulation of the European Parliament and of the Council

on data collection and sharing relating to short-term accommodation rental services and amending Regulation (EU) 2018/1724

{COM(2022) 571 final} - {SEC(2022) 393 final} - {SWD(2022) 348 final} - {SWD(2022) 349 final}

Contents

1.Introduction: Economic, Political and legal context

2.Problem definition

3.Why should the EU act?

4.Objectives: What is to be achieved?

5.What are the available policy options?

6.What are the impacts of the policy options?

7.How do the options compare?

8.Preferred option

9.How will actual impacts be monitored and evaluated?

Annex 1: Procedural information

Annex 2: Stakeholder consultation

Annex 2a: Results of the additional targeted survey to Public authorities

Annex 2b: Results of the additional Targeted Survey to Platforms

Annex 3: Who is affected and how?

Annex 4: Analytical methods & SME Test

Annex 5: Description of the sector

Annex 6: List of existing and planned instruments

Annex 7: Examples of Litigation

Annex 8: Overview of Registration Procedures and Requirements on Platforms concerning STR in EU Member States

Annex 9: Literature review

Glossary

1.Introduction: Economic, Political and legal context

1.1Economic and societal context

Short-term accommodation rentals (STRs) are accommodations used for short-stays by guests. STRs may take the form of a room in a dwelling, an entire apartment or house where very minimal complementary services (e.g. cleaning services, breakfast), if any, are provided on location. 1  STR services consist of the repeated letting (on a daily, weekly or monthly basis), for remuneration, whether on a professional or non-professional basis, of furnished accommodation.

STRs have existed for many years as an alternative to more traditional accommodation services such as hotels. The number of STRs is however increasing significantly across the EU, a trend, which has been confirmed during the COVID crisis 2 . For example, the number of listings on Airbnb grew from 2 500 listings in 2007, to over 5.6 million listings in 2021. 3 The growth of STRs is mainly driven by the emergence of online platforms 4 , which allow hosts to reach out directly to guests from all over the world and facilitate bookings by the latter. ‘Platformisation’ has greatly contributed to the diversification of the STR offer, both in terms of property rented out (room in or entire apartment or primary or secondary residences) and locations. In 2019, 512 million guest nights were booked in the EU via the main online platforms (Airbnb, Booking.com, Expedia Group and TripAdvisor) 5 , representing an increase of 15.8% compared with 2018. 6  Together, the four online platforms referred to above represent approximately 83% of the STR segment in terms of number of guest nights booked, with Airbnb leading with an average traffic share of 55% 7 . There are also many smaller STR online platforms, many of which have developed in the EU. 8  Several of them offer more ‘niche’ services, focusing on specific types of STR accommodation (e.g. Fairbnb 9 , promoting fair and sustainable hosting) or STRs located in specific geographical areas. Some smaller players also combine intermediation services with management services. 10  It is estimated that there are a total of 710 online platforms in the EU, many of which operate cross-border.

New profiles of hosts have also developed. Whilst STRs mainly started as an offer from one individual to another individual (peer-to-peer), the share of ‘professional’ hosts, letting several properties on a more commercial basis 11 , has more than doubled between 2014 and 2020, with professional hosts accounting today for more than 45% of the offers and representing the 13% of all hosts. 12  A small minority of them have STR listings in several EU Member States: between 2014 and 2019, 2.5% of EU hosts have offered STR services in more than one Member State. 13 It should be noted that 16.3% of the hosts that responded to the public consultation stated that they offer STR services in an EU country that is not their country of residence/place of establishment. 14

STRs create opportunities for guests, hosts and the entire tourism ecosystem. Guests appreciate STRs because they increase choice and, often, provide more competitive prices, better facilities and better locations. 15  For hosts and other service providers active in the tourism sector, STRs are a source of income and employment. Some public authorities also help promote STRs, e.g., as a means to attract tourism to more rural or undervalued areas (in France, for example, rural areas have developed partnerships to create 15 000 new STRs 16 ), as a means to complement the hotel infrastructure 17  and/or to pursue other objectives, such as the renovation of buildings. The development of the STR segment has also boosted the development of a multitude of auxiliary service providers, such as companies offering management, cleaning and/or reception services, some of which operate cross-border. 18

However, the growth of STRs is also triggering concerns and challenges. STRs have deeply impacted certain areas, in particular touristic urban and coastal areas in the EU. 19  Public authorities and citizens in these areas have expressed concerns, ranging from increase in nuisances by tourists, such as noise, congestion and waste, to negative impacts on price and availability of long-term housing. 20  Studies point at a correlation between the increasing number of STRs in certain neighbourhoods and the affordability of housing 21 and/or the quality of life in those neighbourhoods, although findings are mixed. 22 STRs also raise concerns related to health and safety and tax compliance. 23  For representatives of hotels, STRs create allegedly unfair competition in certain areas, as they are not subject to similar requirements as hotels. 24  Associations (e.g. of local residents) often denounce the negative impacts of STRs on their neighbourhoods. 25  

1.2Political context and legal context

As a result of the abovementioned challenges, STRs are increasingly regulated, at national, regional and/or local level. Public authorities have taken actions to enhance the transparency of the STR segment, by imposing, for instance, registration requirements and/or by requesting online platforms to share data. In some cities, steps have been taken to limit the offer of STRs by imposing a variety of market access restrictions (e.g. zoning restriction, limit on the number of nightly stays). In some areas STRs have been subject to outright bans. 26  STR regulations are often being challenged in court however, often on grounds of incompatibility with EU law (notably the Services Directive and the e-Commerce Directive). 27  

This situation has triggered increasing calls for action at EU level. In 2016 and 2018, the Commission offered guidance on how to design proportionate market access rules and obligations on online platforms, via its Communication on the collaborative economy 28  and the development of policy principles. 29  A number of stakeholders did not consider this sufficient, however, as exemplified by calls for more legal certainty and transparency coming from the Council 30 and the Parliament. 31  

Recently, the Court of Justice of the EU (CJEU) in its Cali Apartments decision has confirmed that under the TFEU and the Services Directive public authorities can impose market access restrictions on STRs in order to protect public interest objectives, such as combating the long-term rental housing shortage, as long as the rules adopted are proportionate. In this context, the CJEU emphasized the importance of available data and analysis for proportionate policy making. 32  

However, public authorities struggle to have reliable data. A first step to obtain specific data on STRs at EU level was achieved in 2020, when the Commission signed a landmark agreement with Airbnb, Booking, Expedia Group and Tripadvisor, to publish data on STRs offered via these online platforms across the EU. 33 These online platforms now provide regular statistical/aggregated data on the number of nights booked and the number of guests, per municipality in the EU. The Digital Services Act 34 and the taxation transparency requirements under DAC7 35 will also help to bring more transparency for public authorities on STRs, without however fully addressing the specific problems and drivers of this initiative. 36  

In its SME strategy for a sustainable and digital Europe 37 , the Commission announced that it would look into developing a framework for STRs. In its IIA published in September 2020, the Commission identified problems related to a lack of transparency and data sharing, and problems related to burdensome market access restrictions for hosts. Stakeholder consultations following the IIA clarified a need for EU action to bring more transparency and clearer rules on data access and sharing, whilst pointing at the local dimension of market access requirements for hosts. 38  In the context of the Transition Pathway for Tourism 39  and the Urban Agenda 40 , stakeholders also called for the adoption of an EU framework bringing more transparency to the STR segment.

It has therefore been decided to focus, as a matter of priority, on problems relating to transparency via a dedicated initiative. As will be explained, it is expected that the initiative will help reduce unjustified market access restrictions for hosts as well. The Commission will continue to monitor market access restrictions and might take steps (including possible guidance) at a later stage.

Any action under this initiative will seek to address gaps and overcome remaining uncertainties in a targeted manner, to facilitate a balanced development of STRs. It will be in line with, and contribute to, the fulfilment of the Sustainable Development Goals (‘SDG’) 41 , in particular SDG 11 (sustainable cities and communities), by giving public authorities the tools and the data to regulate the STR sector in a proportionate and sustainable way. It will also support the proposed measure from the Conference on the Future of Europe to “invest in an economy based on tourism and culture, including the many small destinations in Europe” 42 . It will complement and build on all existing legal instruments be in line with EU competition law, international trade commitments 43 and the proposed European Declaration on Digital Rights and Principles. 44

2.Problem definition 

2.1        What are the problems? 

This impact assessment focuses on two interrelated problems, which both concern transparency.

Public authorities that are concerned by STRs need specific and granular data about hosts and their offers to be able to monitor, regulate and control the fast-moving STR segment. Consultation results show that a large majority of public authorities across the EU need reliable data on STRs.  Needs for data have been expressed by public authorities from 20 Member States 45 , reflecting the importance of STR bookings in these areas 46 . In response to a targeted survey, 72 out of 80 authorities said they needed reliable data on STRs. Needs are expressed by all types of public authorities, mainly cities but also regional authorities, and to a lesser extent by national authorities.

Among the authorities that require data, the type of data needed is very similar across the EU. They need identification data about hosts (i.e., who is renting out) and about the accommodation rented out (in particular address, type of real estate unit and accommodation capacity) 47 to understand who is renting out what and where, and verify compliance with local rules (e.g., local health and safety requirements). Public authorities wanting to monitor and control STRs also need activity data, that is the number of nights each accommodation is rented out for 48 and the number of guests welcomed. Activity data aim to trace and measure the activity of STR service providers, helping public authorities to monitor tourism flows, design appropriate policy responses and enforce the rules. Survey results indicate that public authorities need activity data on a very regular basis, at least once a month (29 out of 80 authorities) or once per year (14), with a significant minority (16) asking for such data in real time and 3 authorities requesting data on a daily basis.

Public authorities often use a variety of means to obtain data, with diverse efficiency results. The most used and also most successful tool for public authorities to obtain identification data is the use of a registration system, whereby hosts offer details about who they are and the accommodation they offer. Activity data, however, cannot be obtained via registration schemes. Public authorities tend to turn to platforms to obtain activity data. Only 16 out of 80 authorities in the targeted survey indicated that they oblige hosts to report their activity ex-post. 49  This can be explained by the fact that the data received from online platforms is considered more reliable, as hosts may not always declare properly their activity, and/or easier to manage one source of data instead of data from individual hosts). In cases where data is received from hosts, public authorities may need to cross-check it with data from online platforms.

At present, there are two main problems relating to data generation and data sharing on STRs. First, online platforms are increasingly confronted with numerous and divergent data sharing requests and or type of transparency requirements (such as an obligation to display of registration numbers). Second, public authorities struggle to get the data they need about STRs in an efficient manner. These problems have a series of negative consequences, as will be explained below.

Problem 1: Online platforms are confronted with numerous and divergent requests for transparency and data sharing on STRs creating market fragmentation and barriers

Online platforms, in particular those operating in more than one city or Member State and having more data, are increasingly confronted with a variety of national and local rules concerning data sharing requirements and other transparency obligations (such as obligations to display registration numbers). A growing number of public authorities are putting in place regulations to collect specific STR data from online platforms 50  and in some places, authorities have entered into voluntary agreements with platforms to obtain data. Also outside these frameworks, authorities increasingly reach out to platforms with requests for data. This is confirmed by the survey targeted for public authorities (out of 80 public authorities replying, 33 state that their national law has rules allowing them to obtain data from platforms, nine state to have voluntary agreements with platforms). 

This is an EU-wide phenomenon which appears to affect the large majority of platforms: requests come from almost all over the EU 51 and from all levels of public authority (national 52 , regional or local level 53 ). For example, based on the 15 replies received to a targeted survey sent to online platforms 54 , 12 online platforms declared that they have received at least one data request from a public authority. 55  Not only larger platforms are requested to share data, also small platforms are. In terms of numbers of data requests received, online platforms indicate that, in 2017, they have received a total of 13 data requests from public authorities. This number has increased over the years, with 43 requests received in 2018, 616 in 2019, 1 732 for 2020 and 869 so far for 2021 56 .

While the core type of data requested is similar, platforms declare that the data requests otherwise diverge in terms of frequency, origin of the data request (the type of authority), additional information or evidence requested (in addition to the core data set) and procedural requirements/technical means used to request and transfer the data. 57  Data requests from public authorities for public policy purposes often have a regional or local focus, concern specific information and frequencies and thus are different by their very nature. Increasing and divergent requests, combined with regulatory complexity and a lack of legal certainty, make it difficult and costly, in particular for smaller online platforms who would like to expand their services in other Member States, to enter markets and/or scale up. 58 In their response to the targeted survey, 7 online platforms indicated they have already denied a data request from a public authority, mainly for legal and cost-related reasons. 59  Platforms are also confronted with diverging requirements to display information on their web interface, such as e.g., registration numbers, although such requirements are not always complied with (see below, under the drivers). 

This market fragmentation affects small, medium and larger platforms which operate in several cities or have a cross-border activity, or have the intention to do so. 60   Indeed, this is not a problem that affects only larger operators. Evidence points to the fact that also smaller platforms active in this sector operate cross-border. 61

62 63 64 In addition, fragmentation also limits the capacity of local platforms to expand their operations to other markets. Neither network effects nor economies of scale can develop in fragmented markets, where the potential numbers of users and transactions are limited by the reduced dimension of local or national markets and by the existing barriers to market integration.

Problem 2: Public authorities struggle to get the data they need about STRs

Consultation results show that public authorities have similar needs with regards to a core set of data (identification of the host, the accommodation and the activity), while certain public authorities can require additional information or evidence based on the local needs. Public authorities report that they use different sources to obtain this data: registration schemes for hosts, data from platforms, reporting from hosts, or independent data sources.

This consultation also indicates that public authorities that need data have difficulties in obtaining them, and often end up with insufficient, incomplete or inaccurate data on STR 65 . This is not the case for other forms of accommodation such as hotels 66 . Out of 80 replies from public authorities to the targeted consultation, 77 declared they have difficulties in getting data. 67  Problems occur at all levels of government, national, regional and local, and problems relate to both identification and activity data. These findings are also corroborated by a study carried out in 12 large touristic cities (from 12 Member States). 68 Whilst the situation may vary across and even within Member States, public authorities report similar difficulties across the EU, including delays, oppositions or rejections of their data requests. 69  In spite of national, regional or local initiatives to obtain data, public authorities are often confronted with several data gaps. Public authorities put two main reasons forward to explain why they are not able to obtain the data they need: insufficient level of cooperation of online platforms, and the difficulties in enforcing registration schemes, in particular without the cooperation of online platforms. These causes are further explained below.

2.2    What are the drivers?

Several root causes lead to the two main problems identified.

Driver 1: Registration schemes differ widely and are often not complied with

As mentioned above, registration schemes have proven to be an effective way for public authorities to obtain identification data. Registration numbers, granted per accommodation, are a key tool to enforce registration schemes. 70  However, there are a few challenges connected to registration schemes.

First, there is a significant level of divergence of registration schemes across (and sometimes within) Member States. An increasing number of national or local authorities have put in place registration schemes. Registration schemes now exist in 22 Member States and at various levels (national, regional and/or locally) and other Member States are considering introducing them too. Some Member States have registration schemes at local level (e.g. the Netherlands, with currently nine local registration schemes 71 , France with over 50 local registration schemes 72 ), or both at regional and local levels (e.g., Spain has regional registration schemes in place in 19 regions and at least 6 local registration schemes 73 ). Registration requirements differ however significantly both in terms of procedural and substantive requirements. Some authorities impose particularly burdensome requirements (e.g., requirements to submit a variety of documents) or procedures 74 . This, in turn, can have a chilling effect on the number of hosts, and hence on the business opportunities for online platforms. 75  

In addition, public authorities also impose diverging requirements on platforms to co-operate in the enforcement of registration schemes (e.g., diverging requirements relating to the control on and/or display of registration numbers). This is particularly burdensome for online platforms.

Second, according to public authorities who answered the public consultation, hosts do not systematically register, even where an obligatory registration scheme is in place. This may be because they are not aware of the obligation to register, or, if they are, they chose not to comply and are not technically prevented from offering their accommodation on online platforms without registering (or in other words, online platforms allow hosts without a registration number to publish their accommodation on the platform). Research indicated that 87% of Berlin listings on the AirBnB platform did not include a registration number 76 , and it appears that this trend is widespread across the EU 77 . Non-compliance with registration requirements is not only a problem for public authorities, but may negatively affect platforms as well. 

In certain jurisdictions, hosts experience challenges in understanding and completing the registration procedures. 78 Public consultation results also show that platforms are not always aware of registration requirements, in particular where schemes are local and are changing.  In addition, registration requirements are often not being taken seriously by hosts. 79  Registration avoidance most often occurs where online platforms do not cooperate with public authorities to share data or do not require (and then display) registration numbers when a host is listing its accommodation on their websites, even when they are required to do so 80 . Public authorities cannot identify hosts who do not register and need therefore data from online platforms intermediating STR services. As a result, the ability of public authorities to enforce registration obligations is weakened, and they are likely to rely even more on data from online platforms to carry out checks on the legality of the offers. The cooperation of online platforms is therefore key to reduce the burden resulting from data sharing requests. 81  

When online platforms disregard their obligations, some public authorities have decided to initiate legal proceedings leading in some cases to fines under national law. 82  

Driver 2: Lack of effective and enforceable legal frameworks, standards and tools for sharing data on STRs

Even where effective registration systems are in place, public authorities struggle to have complete sets of data on STRs, offering them insights on STR activity on their territories, for different reasons.

Data requests from public authorities are often rejected by online platforms, notably because of an alleged absence of a (clear and undisputed) legal basis for data sharing (48% of the replies) 83 . Online platforms also have pointed out they sometimes do not have the requested data (14% of the answers), or have technical (e.g. mismatch between data sets and data collection systems) and practical problems. Platforms also point at the heavy resources they needed to respond to data-sharing requests (costs, manpower and time).

In fact, diverging national data sharing requirements create fragmentation and additional complexity for online platforms operating cross-border. Whilst the number of data requests is starkly increasing, the type, the format and the frequency of the data requested by public authorities differs across the EU. 84 This creates a lack of traceability and interoperability of the data. In particular, a lack of standardisation and legal framework including an agreement for a shared identifier (e.g. the registration number, the host’s name and address, the cadastral number of the STR) used by both hosts and online platforms makes it difficult to have compatible data and easy to process. 85 This implies that the same data produced by different stakeholders might not be reconcilable because data is not generated and stored in the same format (or, more generally, with the same IT standards).

In addition, electronic tools to share data might not be in place or may be inadequate. For example, large amount of data might be shared by online platforms through inefficient and inappropriate tools – e.g. excel sheets. The absence of a standard for data sharing might result in corrupted data. Finally, using non-automated data sharing systems increase the probability of data corruption. This creates extra costs, possibly mistakes and translates into further delay of the availability of data about STR activity. All this leads to additional barriers for online platforms to operate in the single market.

Driver 3 - Newly introduced data sharing requirements are not sufficiently timely and granular and data is not reaching local authorities

A number of initiatives have been taken recently at the EU level to address the calls for greater transparency on STRs, and on the collaboration of online platforms in this regard. However, they still do not fully address the needs of public authorities. 86

In March 2020, ESTAT entered into a voluntary data sharing agreement with the four main online platforms. 87 The agreement allows ESTAT to obtain key data from these online platforms and publish experimental statistics on STRs. Public authorities thus have access to data about STRs offered via these online platforms. This offers an interesting but yet not fully complete picture of the sector, because the data that ESTAT receives and publishes is aggregate (non-personal) data and therefore not granular, detailed and timely enough, in particular for local authorities, for policy making and law enforcement purposes. Public authorities therefore impose further obligations on online platforms to share data with them, which can, in turn, create barriers for these platforms to operate in the single market. 

Some online platforms have also started making available aggregated data 88  through dedicated portals. Public authorities consider however that this data is often incomplete or difficult to process 89 and not sufficient for policy-making purposes, in particular at local level, where much targeted measures, which may differ depending on the districts or streets concerned, are often needed to regulate STRs. Aggregated data may for example provide a number of STRs but not the type of STRs (room or apartment…) and do not provide any information on the owner (natural or legal persons), which are often essential for policy-making purposes 90 .

In the area of taxation, under DAC7, new rules create an obligation for online platforms to report the income earned by sellers pursuant to, among other activities, the rental of immovable property on a short-term basis 91 . A significant part of the information to be reported by online platforms is relevant for the enforcement of STR rules, too. However, this data is not immediately reusable. A legal basis would have to be introduced in each national legal system in order to allow the transfer of data from the tax authorities to the relevant public authorities enforcing STR rules. Data under DAC7 may also not be sufficiently granular, as it does not provide information on the number of nights and the number of guests hosted. In addition, the reporting frequency under DAC7 is annual, after closure of the fiscal year, and that is too late for STR purposes (as explained above). 92  

Finally, whilst the DSA imposes certain obligations on platforms to respond to specific requests for information, it does not require online platforms to share data with public authorities on a systematic basis. In addition, the DSA requires online platforms to enable the display of registration numbers only for hosts that qualify as ‘traders’ (a term which does not necessarily cover all hosts active in the short-term rental sector) and this may lead to enforcement gaps in a sector, like the STR sector, where hosts are often private individuals offering assets or services on an occasional peer-to-peer basis. 93

2.3        What are the consequences?

As a consequence of the insufficient and inefficient gathering of STR data by public authorities, and the fact that online platforms are confronted with numerous and divergent requests for data sharing and obligations to display registration numbers, both public authorities and online platforms are faced with burdensome efforts (in terms of costs and time) to get and share data. Online platforms have stressed the increasing costs related to data requests. 94  Current costs for online platforms to manage each request for data submitted by a public authority concerning STRs on its geographical territory are estimated to amount on average to EUR 13 549 and take from one to three months to be processed and dispatched 95 . These costs result from the diversity of the requests in terms of data requested, format, frequency and technical means used to transmit the data. Resources spent by online platforms to handle data requests in 2019 is estimated to have amounted to around EUR 30 million 96 . Over recent years and due to the increasing number of data requests, the number of requests that online platforms have answered and the associated costs have also increased significantly. The fragmentation of data sharing requests and obligations to display registration numbers increases the costs for online platforms, which in turn, may affect their ability to expand cross-border and offer their services in the internal market.

Public authorities, in turn, indicate that the data they receive from online platforms often needs to be double checked, as there can be incomplete or contain incorrect values even for identification data, and that no clear improvement in quality has been observed over the years. 97 As a result, they have to continue collecting data with techniques such as web scraping, which are time consuming and costly (for example, a big EU city signed a yearly contract in 2022 of EUR 7 500 to collect this kind of data from online platforms). 98  

The insufficient and inefficient gathering of STR data also creates difficulties for public authorities to design proportionate STR rules, even if other factors (e.g. political factors) can affect the design of such rules. For example, the city of Barcelona explained that before they started a collaboration with online platforms to obtain data on STRs, there were around 6 000 illegal listings in the city; the exact location of such listings was impossible to determine without data from online platforms; the city had to rely on citizens to flag the illegal activity; and the absence of data possibly made it difficult to design new tourism and housing policies. 99  There is a vast body of literature confirming that better regulation and sound policies must be based on evidence and data, in any sector. 100 The CJEU has also stressed the importance of data for proportionate policy making. 101  There are also a number of initiatives aiming at improving policy making, e.g. urban planning processes, with data. 102  Ill-designed STR rules can have a negative impact on the STR / tourism ecosystem.

Finally, these problems can contribute to poor enforcement of local STR rules, where such rules exist. As mentioned earlier, data is needed for enforcement purposes. For instance, where no registration exists, authorities may simply not be aware of STR offers on their territories; where registration exists but online platforms fail to display registration numbers, identifying illegal STR listings (i.e. without valid registration number) may be challenging for authorities. In addition, where online platforms do not share activity data, authorities are unable to assess whether possible restrictions (in terms of numbers of nights a property can be rented out, or how many guests can be accommodated) are being respected 103 . The problems identified can therefore create problems of illegal STRs in local markets and possibly problems of unfair competition (e.g. with hotels).

2.4        How will the problem evolve?

STRs are expected to stay and some studies project a possible significant growth of STRs in the EU 104 . Public authorities will need to monitor the development of STRs and new trends like the greater use for teleworking or increased professionalization of hosts. As STRs may develop in new locations, more authorities may need to carry out monitoring activities and adopt rules (either limiting their presence to combat negative societal externalities or to promote STRs in order to attract more tourists). They will have to ensure compliance with their rules and be able to adapt them swiftly once they are in place. For these reasons, it is likely that public authorities will need more, better quality and sufficient detailed data, on a more frequent basis in order to pursue their policy objectives in a justified and proportionate manner. The City of Porto for example indicated that they needed more data on the compliance of the accommodation with health and safety regulations. 

Online platforms are likely to receive an increasing number of diverse data requests from local and national public authorities and could be subject to new obligations to share data. 105  This will increase burdens on platforms, in particular smaller platforms. However, it is likely that online platforms would not be able to address all the requests they receive, or address them swiftly, particularly should it remain unclear on what legal basis they can share data to comply with the request.

As public authorities would not get the data they request, litigation would likely increase as public authorities could try to turn to courts to force data sharing and to ensure that requests to delist illegal listings are acted on 106 . In order to avoid such litigation, online platforms may agree to share data on a case-by-case basis to avoid the risks if they do not reply to public authorities’ requests. Only case-by-case solutions would therefore be provided that may be detrimental especially for smaller / less organised urban areas that are not able to negotiate with big online platforms or do not have in place a legal framework to request data.

In the absence of change, the size of the data gap for public authorities could continue to increase. The data needs of public authorities may never be fully addressed, and more and more authorities will experience the same problems (costs, time) when they try to get the data from online platforms. This may limit their ability to enforce existing rules and deprive them of relevant information e.g., on trends or limit the possibility to compare their local situations with others. Possibly, future STRs rules will not be based on evidence. Public authorities would still be required to combine several methods to collect data, including web scraping, and several types of data sets which may not be interoperable. Inefficiencies in data collection and processing are likely to remain or increase.

Online platforms however could also try to improve their cooperation with public authorities. Airbnb for example has created a city portal open to communities, which provide certain data and tools to their partners.

3.Why should the EU act? 

3.1 Legal basis

If the EU intervention takes the form of a legislative proposal, Article 114 TFEU would be the appropriate legal basis. Article 114 TFEU allows for the adoption of measures that are considered necessary for the approximation of the provisions laid down by law, regulation or administrative action in Member States that have as their object the establishment and functioning of the internal market. It is, therefore, the appropriate legal basis for a legislative intervention covering online service providers in the internal market and addressing divergences between Member States’ regulations and requirements, which affect the functioning of the internal market for online platforms.

This initiative intends to ensure the proper functioning of the single market, in particular in relation to the provision of intermediating services. The majority of online platforms intermediating the offer of STR services operate across the EU and are confronted with several and diverging requirements and requests concerning information sharing on hosts and their activity and the display of information on STR services. Differences in national laws exist and are likely to increase, as national laws will continue to develop, given that some Member States have legislated or intend to legislate on the intermediation of STR services. Online platforms will also continue to be confronted with more and diverging data requests as a result of the growing need for transparency in the STR segment. Moreover, as a matter of Union law, information society services are in principle subject to the law of the Member State in which the service provider is established, and the adoption of national measures that restrict the cross-border provision of such services is subject to strict substantive and procedural conditions 107 .

In line with this objective, this intervention aims to establish a harmonised European framework for data generation and data sharing on STR services. The approximation of rules applicable to intermediary services is necessary to avoid the diversity of rules and requirements in the single market and facilitate data-sharing with public authorities and the cross-border provision of STR services. The data sharing framework proposed is expected to have a positive effect on market access, as it will provide the authorities with the data they need, supporting Member States in discharging their obligation to enact and maintain proportionate rules. This initiative will complement other horizontal acts/proposals (e.g. the DSA) which are also based on Article 114 TFEU. 108  The use of Article 114 is also justified by the fact that the vast majority of data will be provided by online platforms operating cross-border (giving their share in the traffic). Covering, as a complement, online platforms with no or limited cross-border activity will contribute to giving a comprehensive set of data to public authorities and thus make the proposed EU framework fully efficient.

If the intervention takes the form of recommendations, then Article 292 TFEU would be the appropriate legal basis.

3.2 Subsidiarity: necessity and added value of EU action

According to the subsidiarity principle 109 , the EU should only act where the objectives of the proposed action cannot be achieved sufficiently by Member States alone and can, therefore, by reason of the scale or effects of the proposed action, be better achieved by the EU.

Several Member States have established mechanisms that require hosts to register their activity; without prejudice to the compliance of such rules with EU law, in some cases Member States have introduced an obligation for online platforms to display these registration numbers on STR services intermediated by them. Several Member States are also imposing data reporting requirements on platforms. However, without the cooperation of online platforms in enabling that only listings with registration numbers are displayed, public authorities struggle with the enforcement of the obligation on hosts to register.

In addition, as explained in more details in chapter 2, divergent and burdensome frameworks concerning data sharing are affecting the ability of online platforms to comply with such requirements and therefore to operate cross-border. The fragmentation of data sharing requirements therefore currently creates barriers for online platforms to operate in the single market. In addition, even where data sharing rules are in place at local, regional or national level, online platforms do not always share the data with public authorities for legal, technical or cost-related reasons. For all of these reasons, the current interventions at national, regional and local level are often not effective, and authorities are complaining that they have difficulties in enforcing existing data generation and sharing requirements and therefore in obtaining data from platforms and hosts.

This proposal primarily aims at streamlining data requests across the EU to facilitate compliance with such data requests for online STR platforms, hence reducing costs and removing barriers to operate in the single market for these platforms. An EU legal basis for data sharing by online STR platforms with public authorities, as well as an EU framework for such data sharing to take place, will increase legal certainty, ensure that the obligation to share data is compliant with EU law (GDPR, DSA…) and consistent with other reporting requirements (e.g. DAC7). It will also ensure that the data exchanged is standardised and interoperable. The EU framework will be applicable across the EU and will in particular simplify/streamline the landscape of reporting obligations on platforms. This harmonised framework will also prevent differentiated implementation of the obligation on platforms between Member States as the conditions for data sharing are the same for all (no selective avoidance).

The combination of an EU obligation for authorities to issue registration numbers and for online platforms to enable hosts to display only listings with such registration numbers, where a registration system is in place, will ensure that such requirements on hosts are easily enforceable. It will also increase the traceability and facilitate data exchanges based on this registration number. The common EU framework will hence offer national and local authorities the level of transparency they need to enforce the rules and adopt informed policy responses in line with existing EU law.

The principle of subsidiarity is also respected by giving local, regional or national authorities flexibility with regards to the implementation of registration schemes (whether to introduce them or not, and at which level), but also with regards to the information and evidence that each authority can request from hosts, to ensure that local circumstances and needs are respected and taken into account. The measures proposed also strike a balance between the interests of all stakeholders involved, by requiring concrete action from public authorities, hosts and online platforms, and offering benefits to these actors, ensuring the proportionality of the foreseen intervention.

4.Objectives: What is to be achieved?

4.1 Objectives tree

4.2    General objective

The main policy objective of this initiative is to enhance transparency in the STR segment as part of efforts to promote a balanced tourism ecosystem. The initiative will improve and harmonise the framework for data generation and sharing on STR across the EU, remove related barriers to the provision of online services by online platforms and ensure that public authorities have sufficient and reliable data to develop sound policies in line with public policy objectives. This will ensure that online platforms do not face unjustified and disproportionate barriers to offer their services in the internal market. It will also help ensure that illegal STR services would no longer be offered, providing more certainty on enforcement of registration.

4.3    Specific objectives

The specific objectives pursued with this initiative are to:

1.Reduce the burden on online platforms of STR data collection and sharing, and avoid fragmentation of these transparency requirements across the EU;

2.Ensure that public authorities have the tools they need to gather data for policy making and enforcement of STR rules. 

In order to reduce the burden on online platforms and avoid fragmentation of these transparency requirements across the EU, the initiative aims to put in place a common approach to data generation and data sharing, whilst clarifying the responsibilities of online platforms, hosts and public authorities.

With respect to the second objective, the intervention aims at ensuring that authorities have access to the information that is needed for policy making and enforcement of STR rules (e.g. to ensure a sufficient supply of affordable long-term rental housing), in accordance with the applicable national rules.

5.What are the available policy options? 

5.1 What is the baseline from which options are assessed?

The baseline scenario entails a ‘no policy change’ approach.

The EU would not specifically act to solve the transparency issues of the STR segment. It is expected that an increasing number of Member States and/or public authorities would continue to develop STR rules and mechanisms to obtain data from hosts and online platforms at national, regional and local level. The STR framework would therefore remain fragmented, uneven regulatory burdens would remain (and might even increase, as online platforms are confronted with increasing data sharing requests), and fragmented enforcement issues are likely to persist. Litigation is, in particular, likely to continue and increase, with STR rules being challenged in national courts and the CJEU. 110 Public authorities and online platforms could also continue developing voluntary cooperation solutions, which take time to negotiate and only bring solutions for parties to the agreement. In their general contract conditions with hosts, certain online platforms provide that hosts are responsible for complying with local STR laws, or require hosts to acknowledge that they have complied with all the existing local rules before proceeding with the listing of the STR property. 111 These conditions may also foresee the possibility for these platforms to take appropriate action in response to a host’s breach of applicable rules that they may become aware of, in particular following a notification of an illegal listing by public authorities.

With no specific policy action, existing voluntary agreements for data sharing, as well as the recent DAC7 and DSA will bring some improvements to the data sharing and enforcement landscape in the EU without, however, comprehensively solving the problems identified (see above, section 2 ‘problem definition’). The DSA introduces due diligence requirements for online platforms to make sure that they behave more responsibly and transparently. Among these, a specific provision (applicable to online platforms that constitute marketplaces) will, in particular, require online marketplaces to ensure ‘compliance by design’ through an obligation to enable the display of registration numbers (where a registration procedure is in place), but only with respect to listings offered by hosts who qualify as ‘traders’ 112 , a term which does not necessarily cover all hosts active in the short-term rental sector. In addition, the DSA will give public authorities a tool to facilitate the enforcement of orders to provide specific information about individual users, and thus to investigate in more detail individual providers of STR services that authorities suspect act in a fraudulent manner. However, its provisions to do not extend to an obligation to collect and share data that applies to all hosts and their activities from platforms. 113 These limitations may lead to enforcement gaps with respect to hosts’ compliance of STR rules, including compliance with registration obligations. 

A common European data space for tourism is currently being developed. 114 This data space aims to bring together various data sets on tourism for use by tourism businesses and public authorities alike (e.g., allowing them to monitor tourism trends). The creation of a tourism data space will not, however, solve the problems identified above (inter alia because it will not as such lead to generation of the data that is required by public authorities). Public authorities would still require specific STR type of data, notably specific identification and activity data regarding host offers with a certain frequency to design precise policy responses and to enforce STR rules. 

Finally, CJEU jurisprudence will provide binding interpretation, the Commission’s 2016 Communication on the collaborative economy and the 2018 policy principles and good practices will continue to serve as a guidance 115 , notably to ensure that measures taken by public authorities in relation to STR are justified in light of the TFEU, the Services Directive and the e-Commerce Directive in particular. The Commission will continue monitor and might update its guidance documents.

5.2 Description of the policy options

Three policy options have been identified. All aim to facilitate data generation and data sharing, by clarifying the respective duties of public authorities and online platforms and ensuring that effective tools are in place. Whilst the three options contain comparable measures, there differ considerably in terms of intensity of the intervention. The below intervention logic shows how the measures considered relate to the drivers, the problems and the specific objectives.

Policy option 1 – Soft measures to promote data generation and data sharing

Under this option, the EU would issue a Recommendation 

·Encouraging public authorities to put in place registration schemes for hosts. Under such schemes, hosts would have to register with public authorities to offer STR services. As part of the registration process, they would be asked to provide certain data to the relevant public authorities about themselves and the property they put out for rent. The EU could encourage Member States to ensure that registration schemes are user-friendly and online, and lead to the automatic grant of a unique registration number per accommodation. It could also recommend to put in place a digital single entry point, to facilitate registration and data sharing (see further below, under Option 2).

·Requiring online platforms to design and organise their online interface in a way that enables all hosts to display the registration numbers on their websites for each listing. 116 Online platforms would have to require a registration number for each accommodation from hosts who offer their accommodation on the online platform. This measure would ensure that hosts register their activities with public authorities (as without registration, they would not be able to offer STR services on online platforms), which would enable public authorities to better control compliance of STRs with local rules. 

·Requiring online platforms to share pre-agreed activity data on each listing with public authorities. Information on such data sharing would be clearly communicated to the hosts at the time of the registration. This will enable public authorities to have a detailed knowledge of the activity of each STR accommodation. Activity data is relevant for areas where a limit on the number of nightly stays is in place, but also in order to design and implement proportionate rules and ensure compliance with them. 

Such Recommendation would be based on good practice examples, whilst taking into account applicable Union law.

The Recommendation could be complemented by an EU Code of Conduct between public authorities and online platforms, and national legislation aiming to facilitate data sharing in practice in accordance with the GDPR. 117  The specific STR Code of Conduct could build on and complement other voluntary agreements to data sharing on STRs, such as the Eurostat agreement or the future Code of Conduct for data sharing in tourism. 118 It would in particular seek to:

·Specify which type of data online platforms could share with public authorities and with what frequency;

·Include agreements to use certain technical solutions to facilitate data sharing (e.g. the use of APIs 119 or simpler alternatives such as ‘drag and drop’ for SMEs which might not have a sufficiently developed technological infrastructure);

·Include IT specifications to ensure, to a limited extent, interoperability of the data shared between public authorities and online platforms and security safeguards.

The Code of Conduct could be initially developed between a limited group of online platforms and public authorities, giving other online platforms and public authorities the possibility to enter similar commitments voluntarily once it is adopted.

The data collected voluntarily through the STR initiative could, in an aggregated format, feed into the Tourism Data Space.

Policy option 2 – Legislative initiative containing a common approach to data generation and data sharing

Under this option, a legislative initiative would be adopted, covering the following key measures:

1.Obligation for public authorities to maintain a registration system for hosts and their accommodations if they want to obtain STR data from online STR platforms (opt-in)

Public authorities (national, regional or local) who would like to obtain data on STRs for policy making and enforcement purposes (e.g. to ensure a sufficient supply of affordable long-term rental housing) from online STR platforms would be required, as a first step, to establish or maintain a mandatory registration scheme for hosts that announce their offers via online platforms. 120

Hosts would be required to provide information relating to the host/owner (name, address, contact details) and the accommodation (type of accommodation, capacity, address, cadastral number). Public authorities would also be allowed to ask for additional information or evidence, such as health and safety certificates, fire safety certificates, insurance, landowner permission, etc…), on the condition that these additional requirements are necessary and proportionate to achieve a public interest objective (as required by the Services Directive and TFEU).

The registration procedure would lead to the automatic grant of a unique registration number per accommodation. This would ensure full traceability of who offers what STRs on their territories. Verification of the information and supporting documents provided by the host, as well as any inspections, if necessary, would be carried out by public authorities ex post.

The requirement to maintain a registration scheme would apply only to public authorities (at national, regional or local level) interested in obtaining data on STRs for policy making and enforcement purposes (e.g., to ensure a sufficient supply of affordable long-term housing) from online platforms. This may result in a situation wherein some local authorities maintain registration systems, and others not. Only the authorities that have put in place registration systems would be able to benefit from the data reporting obligations on platforms under this initiative (see below). This ‘opt in system’ aims to keep administrative burdens for authorities limited and to ensure that data is only exchanged where it effectively serves a public purpose.

To reduce administrative burdens for online platforms and hosts, and in line with the Single Digital Gateway Regulation, 121  Member States would be required to ensure that the information regarding the existence of (and links to) registration schemes at local, regional or national level is accessible via a national single digital entry point(the single digital entry point would also facilitate data sharing, see below). Member States should also ensure that hosts with multiple listings can re-use data they have already submitted during the first registration (notably data relating to the hosts itself), in accordance with the once-only principle. 122  

An EU Member State in the territory of which public authorities are not, at any level (national, regional or local), interested in obtaining data from online platforms would not be required to establish or maintain a registration system, nor a related single digital entry point (opt-in system, as explained above). Where registration schemes are already in place at national, regional or local level, Member States could maintain them, but would need to ensure that their existing registration schemes are in line with EU requirements. The obligation on a Member State to set up the single digital entry point is triggered by the wish of a local, regional or national authority to participate in the registration and data sharing systems as described under this policy option.

2.Obligation on online platforms to enable the hosts to display their registration numbers

Online platforms would be obliged to create a field for registration numbers, which is already envisaged, in case of listings published by traders, under the “compliance by design” obligation under the DSA, where registration systems are in place. This obligation would therefore complement the DSA, covering all hosts, and not only hosts that would act as traders.

In line with what the DSA currently foresees with respect to listings published by traders, online platforms would be required to make best efforts to assess, prior to the allowing of the offering of the service, that such registration information is provided. Similarly, online platforms would be required to make reasonable efforts to randomly check that the information provided by the host is reliable through the use of any freely accessible official online database or online interface made available by Member States (assuming such database would exist at national, regional or local level). 123

Where an online platform would obtain sufficient indications or would have reasons to believe that the information provided by the host is inaccurate, incomplete or not up to date, this would, pursuant to the DSA, trigger an obligation for online platforms to suspend the provision of their services to those hosts until they provided all the required information.

The obligation for platforms to enable all hosts to display their registration numbers would ensure that hosts effectively register their activities, which facilitates compliance by local authorities.

3.Obligation on online platforms to share activity data

Online platforms would be required to report a closed list of specific data on the activity of hosts to public authorities, notably data on the number of nights an accommodation is rented out, and the number of guests to which it rented out. They would be required to share data only with public authorities that have put in place a registration scheme, and the reporting would be linked to the unique registration number issued for each rental unit for each host.

This will allow public authorities to combine the activity data shared by the online platforms (i.e., registration number, number of night and number of guests) with the data provided by hosts through the registration scheme (i.e., identification of the host and the property), to obtain the complete information on STR services provided by each host. Complete information should enable authorities to order (based on applicable national or EU law) the online platforms to remove, under the rules of the DSA and in full compliance thereof, any specific illegal listing that these authorities have identified (e.g. listing with invalid registration number, or breach of any other requirement in place). The DSA rules will apply with respect to orders for removal of specific illegal listings, orders for obtaining specific information about individual users, insofar as those orders fulfil the conditions of the DSA. The limitations of liability of online platforms for the illegal content they store at the request of their recipients will be governed by the DSA.

Platforms would have to share data via an API (or similar tool) linked to a single digital interface made available at the national level, and linked to the single digital entry point referred to above. This interface would in practice serve as a data hub, collecting all relevant data from all relevant online platforms, and sending on these data to relevant public authorities within the national territory (with each public authority receiving data concerning its respective jurisdiction only). National authorities would therefore need to develop an IT infrastructure (or adapt any existing one already in place) to allow online platforms to share data through it.

Online platforms would be requested to report activity data at least on a weekly basis. They are expected to share data through automated means – e.g. with API – through the single digital entry point. More lenient reporting obligations would apply to small and micro online platforms (within the meaning of Recommendation 2003/361/EC), 124 unless these online platforms reach a certain threshold of active users (hosts) in the EU. Small and micro online platforms that do not reach such threshold are not expected to share data through automated means and can report data to public authorities every three months. This solution should ensure that authorities get data from all platforms, whilst minimising burdens on the smallest platforms with a limited impact on the market. 

Member States would be responsible for the implementation of the measures foreseen under this option, and would also have to ensure that the necessary safeguards are in place to comply with the GDPR. The responsibility for the enforcement of these measures would lie in principle with the Member States and consistently with the DSA framework.

The data collected by public authorities through the STR initiative could also, in an aggregated format, feed into the Tourism Data Space.

Policy option 3 – Legislative initiative containing a common approach to data generation and data sharing, with mandatory national registration schemes for all hosts

A legislative initiative would be adopted, covering the three key measures set out above, including more lenient obligations for small and micro online platforms, but with the following difference: all Member States would maintain a single registration system at national level for all hosts offering STR listings via online platforms.

This option would require all Member States to adapt their current practices in terms of registration schemes. Member States where registration schemes currently do not exist, would have to introduce one, and where registration schemes have been created at local or regional level, they would need to be centralised at the national level.

Whilst this system can be more burdensome for Member States on the short-term, it would increase efficiency for online platforms, hosts and public authorities on the long-term, and is likely to be more future proof (as, with the rise of STRs all across the EU, it is likely that overtime all Member States would want to have data on STRs). Online platforms, in particular, will benefit in terms of efficiency by setting up a single IT system applicable across all EU at the same time.

As under option 2, Member States would be responsible for the implementation of the measures foreseen under this option, and would also have to ensure that the necessary safeguards are in place to comply with the GDPR. The responsibility for the enforcement of these measures would lie in principle with the Member States and consistently with the DSA framework.

The data collected by public authorities through the STR initiative could also, in an aggregated format, feed into the Tourism Data Space.

5.3Options discarded at an early stage

Five measures have been discarded upfront, for the reasons set out below.

1.Mandatory EU registration scheme

The EU would set up a unique registration system for all hosts across the EU. All hosts would receive a unique registration number. The EU would collect data from online platforms, linked to the registration number, through a single digital entry point set up at EU level. Local and national public authorities would be able to extract the necessary data from this database (manually or with automated tools such as APIs) as well as EU institutions and bodies 125 (aggregated data). While bringing some simplification and clarity to hosts and online platforms in the EU with regards to their obligations to register and share data, this option does not sufficiently take into account the principle of subsidiarity.

2.A legislative approach providing for a legal basis for national tax authorities sharing data they obtain under DAC7 with local authorities (G2G dimension)

The DAC7 126 will apply as from 1 January 2023. It creates an obligation for online platforms to report annually some data on STR activity (including the income earned) and for the Member State where reporting takes place, to exchange this information automatically with the competent tax authorities of the other relevant Member State(s). Online platforms must report, inter alia, the following data to the relevant Member State: details of the platform; identification data on the host and the accommodation 127 ; the revenue received during each quarter and the number of relevant activities provided with respect to each property listing; any fees, commissions or taxes withheld or charged by the platform during each quarter; where available, the number of days each property listing was rented and the type of each property listing.

Under this measure, a legal basis would be introduced at the European level to allow the transfer of this data from the competent tax authorities of Member States to their public authorities enforcing STR rules (regional or local authorities). However, this option was discarded as the frequency of data sharing (annually) and timing (year+1) would prevent public authorities from obtaining up-to-date information on hosts that would allow them to undertake effective and timely law-enforcement action. The type of data collected is also incomplete, as it does not cover for instance the number of guests.

3.A legislative approach imposing obligations on online platforms only; no obligations on Member States

This option would foresee the obligations on online platforms to share data as presented under options 2 and 3, without however considering the obligation to set up a registration scheme or to issue a registration number. While with this option, public authorities could get some data on STRs from online platforms directly, it would not allow public authorities to cross-check the data received from online platforms with data received directly from hosts. This could lead to poor data quality and traceability. The burden on the online platforms would also be increased, as in the absence of registration schemes and registration numbers, online platforms might have to share also identification data. As this option would put all the burdens on online platforms, without creating any efficiency gains for them (i.e. harmonised data set to be shared with public authorities, automated data sharing mechanisms, information available on a single digital entry point), it has been discarded upfront.

4.Exemption for certain online platforms from data-sharing obligations

The option of a full exemption for certain small and micro online platforms from reporting obligations, based on the volume of STR activity intermediated (number of active hosts) was assessed but also discarded at an early stage. For the purpose of this assessment, it has been identified that 680 companies 128 could fall into the reporting exemption. There would be a direct positive effect for online platforms which would be able to save “costs” deriving from the data sharing obligations: the implementation of this variant will translate into around EUR 1.6 million savings per year for all EU companies falling into this reporting derogation. However, other considerations (below) need to be taken into account.

Whilst such exemption might entail benefits for smaller online platforms (because they would be able to save costs related to data sharing), excluding certain online platforms from the scope of the initiative could create a loophole in the stream of information public authorities will be able to receive. This will lower the completeness and reliability of information at the disposal of public authorities to tackle positive and negative externalities linked with STR activity. Consequently, the enforceability of STR rules in place will be jeopardised. In addition, an exemption from reporting requirements could incentivise circumvention of the rules, by directing traffic towards online platforms that would be exempted from the reporting obligations. Hosts who do not want to see their activities reported partially or fully could easily resort to these platforms. This would clearly defeat the objective of the reporting requirements by hampering the ability of public authorities to monitor compliance and enforce the rules. At the same time any initial benefit for these smaller online platforms will not represent a consistent incentive prolonged in time (because as soon as these smaller online platforms will reach the minimum volume of transactions they would need to report data). It should be noted that also the DAC 7 does not entail an exemption from data sharing for smaller platforms.

5.Proportionality assessment of market access conditions for hosts

Under this option, the EU would have developed a dedicated proportionality assessment to ensure that any additional market access requirements, going beyond registration and reporting requirements (for instance quantitative restrictions such as night caps, or zoning restrictions), do not go beyond what is strictly justified and proportionate. This assessment would include both procedural and substantive requirements that public authorities must respect when introducing market access restrictions on hosts. Public authorities would be required to evaluate market access requirements to demonstrate they are necessary and proportionate to meet a specific public interest objective. They would also have to differentiate between types of hosts and types of STR offers (e.g., differentiate between ‘peers’ and professional hosts). A list of un-recommended market access requirements, that are not considered justified and proportionate, such as the imposition of undifferentiated bans on STRs, would be imposed. However, this option was discarded for a number of reasons. Stakeholders, and in particular public authorities, did not express support for such intervention. The Cali apartments judgment already clarified the requirements under the Services Directive, and notably the requirement to conduct a proportionality test based on objective facts, already apply.

6.What are the impacts of the policy options? 

Whenever possible, economic, social and environmental impacts were assessed quantitatively. Data needed for the analysis has been obtained from multiple sources 129 and triangulation between sources was used to confirm the cost estimation. For indirect impacts on STR players, a more qualitative analysis has been conducted in order not to rely on weak estimations, which could be misleading.

Direct impacts of the initiative are assessed on three categories of stakeholders. Online platforms and public authorities are directly targeted by the initiative (as a result of new recommendations/obligations for them). Hosts are directly affected but in different degrees depending on the policy option analysed. We refer to ‘other players’ as the set of economic agents operating in the tourism ecosystem impacted by the STR segment; these are impacted mildly and indirectly.

For public authorities, the maximum cost in the estimated range was considered. It is assumed that Member States shall assess what existing IT infrastructure they can adapt to comply with recommendations or provisions set up by the initiative, being able to lower costs by re-using existing solutions.

With regards to social and environmental impacts, quantification of impacts were requested to public authorities, but the feedback provided did not allow to quantify substantially these possible impacts of the measures. These are therefore described in a more qualitative manner.

An implementation (transition) period of two years will provide public authorities and online platforms sufficient time to adapt and implement the new requirements.

6.1Policy option 1 - Soft measures to promote data generation & data sharing

Economic impacts:

In light of the voluntary nature of this option and the absence of information on whether Member States shall implement all or only some of the recommendations, the estimation of the economic impacts at EU level in quantitative terms depends on how many Member States are likely to implement them, the exact nature of commitments made under a Code of Conduct and the amount of participants, as well as the diversity of existing registration schemes. Some elements to assess qualitatively the impact on the different players are also presented below, in order to allow for a comparison between the options both quantitatively and quantitatively.

Impacts on online platforms:

1.Benefits

The cost for an online platform to process a single request for data submitted by a public authority amounts to EUR 13 549 130 . As explained in chapter 2, the dedicated survey for online platforms has shown an exponential growth in data requests in the past years. Online platforms reported to have received more than 1700 requests for data sharing in 2020 and 800 in 2021 (the number lowered due to the impact that the COVID pandemic had on tourism flows and therefore on the need of local public authorities to check STR activity). Taking these numbers of data requests as a baseline, we can estimate a cumulative administrative cost for online platforms in the next 5 years between EUR 54 million (based on 800 requests per year) and EUR 115 million (based on 1700 requests per year) at least. This estimate is based on a conservative assumption of only 60% of the biggest platforms (18) are confronted with data requests so any requests directed to smaller platforms are not included in these estimates (see table 7 annex 4). At the same time is assumed that each online platform will receive requests only from 40% of the total 150 local public authorities identified A leaner and more predictable way to share data will reduce the time and full time equivalents (FTEs) needed to deal with each data request. This option foresees soft tools to streamline the way data is generated and shared by online platforms. If the recommendations are put in place consistently by public authorities and online platforms, uncoordinated requests for data would be replaced by a more streamlined approach based on pre-defined data sets and data sharing mechanisms. This will benefit all platforms, especially SMEs.

The benefits for online platforms will however strongly vary depending on the exact principles agreed in the Code of Conduct and the number of public authorities voluntarily joining the Code of Conduct and implement the recommendations. It is therefore not possible to quantify the benefits resulting from a reduction in the number of uncoordinated requests. In addition, it should be noted that some public authorities will continue to request data via means other than those set forth in the recommendation and/or Code of Conduct. This means that online platforms could continue to be exposed to uncoordinated requests by public authorities.

2.Costs

Online platforms would bear costs to adapt their IT infrastructure to ensure automated sharing of data with public authorities. The one-off costs per platform that would decide to share data via an API could be estimated at EUR 30 000 with yearly maintenance costs of around EUR 36 000 131 . However, as these costs would depend on the characteristics of the available technical solutions (automated and less-automated ones) to facilitate data sharing and would only apply to online platforms who decide to voluntary comply with the framework, it is not possible to properly model the costs estimations. 

In Member States where registration schemes are put in place and registration numbers are issued, online platforms may also be required to display such registration numbers. For online platforms to enable the display of registration numbers, minor technical adjustments would be needed, the cost of which is expected to negligible.

Impact on public authorities:

1.Benefits

Positive impacts would be only ensured where and to the extent public authorities and online platforms implement the soft tool set forth under this option. Benefits for public authorities would in that case include a reduction of costs in obtaining the data they need. Authorities would receive better quality and more complete data sets, in a much more streamlined and cost-effective manner, with less litigation.

The benefits are difficult to quantify under this option, as they would depend on the willingness of online platforms to comply with the soft tools. Better quality data will also reduce public authorities’ costs in monitoring and securing compliance with local rules, as authorities would be able to identify and act against illegal listings in a more effective manner.

2.Costs

Authorities deciding to implement the soft tools would face a set of administrative costs, related to the creation or adjustment of registration schemes, and the creation of an IT infrastructure facilitating data sharing with online platforms.

The one-off cost per national authority for setting up a new IT infrastructure is estimated to be around EUR 137 000 132 but assuming that every national authority will exploit (at least partially) existing IT infrastructure, 30% of cost savings could be made. Yearly costs for hosting and maintenance have been identified respectively to around EUR 60 000 and EUR 36 000, for a total of roughly EUR 96 000. Assuming that 16 Member States 133  will implement the recommendations under option 1, total costs for national authorities under this option will amount to around EUR 1.55 million for setting up the system and EUR 1.55 million for each subsequent year for hosting and maintenance.

In addition, regional or local authorities may also bear costs related to the connection of their systems with the IT infrastructure created by national authorities, which are estimated at EUR 30 000 134  with yearly maintenance costs at EUR 36 000. In the conservative assumption that around 100 cities/areas 135 in the 16 Member States mentioned above will want to implement the provisions and will adapt their existing systems gaining additional 80% savings in the above mentioned cost to connect their systems, this could entail a cumulated adjustment cost of EUR 600 000 and a recursive yearly maintenance cost of about EUR 720 000.

There may also be indirect costs associated with upskilling public authorities in relation to data analysis and business intelligence, depending on the existing level of expertise. It is, however, difficult to quantify such costs as it is impossible to know in advance which public authorities will need to develop from ex-novo or are already in possession of such skills.

Finally, it should be noted that under option 1, Member States would still have to adopt national legislation facilitating data sharing, to be compliant with the GDPR, since there would be no formal legal basis created under EU law. This would reduce the (cost) effectiveness of this option.

Impacts on hosts:

1.Benefits

According to the study made by Oxford Research 136 , hosts suffer from burdensome registration requirement and have, on average, to wait 15 days before their registration is complete (e.g., because of ex ante checks carried out by public authorities). Where the soft tools are implemented, new hosts entering the market in the assumed 16 Member States implementing the recommendations under this option, would benefit from more streamlined registration procedures and from a significantly reduced waiting time, translating into a total saving of about EUR 438 million in the first 5 years after implementation (based on baseline number of 2019 and growth rate estimations 137 ). This estimation represents the monetisation of the time hosts are expected to dedicate to this activity and it is based on the average hourly wage and the assumption that the waiting time for registration will be reduced from 15 days to between 1.4 and 1.7 day, which is in line with time currently needed to complete a registration procedure in Portugal and Greece. These two have been identified as benchmark because they are in line with the proposed soft tools. Based on the assumption that 87% of the hosts are peers 138 and 13% professional hosts, the cumulative cost savings for citizens over five years are estimated at around EUR 381 million and for professional hosts at around EUR 57 million.

2.Costs

Hosts in the 100 cities/areas 139  in the 16 Member States mentioned above will possibly be requested by competent local authorities to adapt their registration information. Using the assumption that data already used in the active systems will be re-used, it is possible to assume that any adaptation will not cost more than ~30% of a new registration for each host (the cost associated with a new registration under the characteristics required in this initiative is at around EUR 320 140 ). A rough estimation brings a total one-off administrative cost for hosts of around EUR 15.8 million. 141  Based on the assumption that 87% of the hosts are peers and 13% professional hosts, the one off administrative costs for citizens are estimated at around EUR 13.7 million and for professional hosts at around EUR 2.1 million. 

Impacts on Tourism Ecosystem and other accommodation service providers:

If the soft tools referred to above are implemented, the tourism sector as a whole is likely to benefit, in an indirect manner. Better data about STR would give authorities the tools to assess and address any problems related to unfair competition between STR providers and providers of other accommodation services, such as hotels. This is likely to have a positive economic impact on the latter.

Further, public authorities might decide to make data collected on STRs available on an aggregate basis to the various stakeholders active in the tourism sector. This is the aim of the envisaged tourism data space. Better quality data on STRs would allow a variety of actors in the tourism sector (e.g., restaurants, tourist guides, cleaning services, management companies – often operating cross-border) to monitor and anticipate trends, and to adjust their service offers to these trends.

Social impacts:

The potential impacts are presented below in a qualitative manner, as the social impact of the measures under this option will depend on the implementation of the measures by public authorities and online platforms, and also on what public authorities will do exactly with the better quality data they will receive. Expected impacts include:

·Increased trust of consumers/guests

In the areas where the recommendations foreseen in option 1 will be implemented, guests’ trust is expected to increase thanks to the use of the registration number. In addition, public authorities would be able to better monitor and enforce health and safety requirements, thanks to the data received from hosts and online platforms.

·Less illegal listings

Since traceability of hosts and their activity, including with the obligation to enable all hosts to display the registration numbers, will increase substantially the ability of public authorities to enforce local rules on STR activity, it is expected that fewer unlawful STR accommodations will be offered. This would benefit the final consumers – i.e. the guests – and/or the local communities, as authorities would be better able to ensure the quality of offer to tourists and the quality of life of residents.

·Better management of tourism flows and increased ability of public authorities to assess and mitigate negative externalities of STRs

Thanks to better data-sharing between online platforms and public authorities, the latter would have better information and tools to predict and manage touristic flows. Completeness, reliability and extensiveness of information on the STR segment, if properly used, will help public authorities in their public policies duties. They will be able to design evidence based policies, producing better targeted policies to address specific externalities of the STR activity – positive or negative – to the benefit of local communities. This could, for instance, have a positive impact on the availability of affordable housing in high pressure areas. In these areas, public authorities would be able to better regulate STRs to ensure a balance between the STR offer and the availability of affordable housing. Better data could also positively affect the capacity of public authorities to deal with health and safety, as well as possible security issues related to STRs.

·Impacts on fundamental rights

The recommended provisions under policy option 1 are expected to affect the fundamental right of the protection of personal data 146 , since it would increase data collection and sharing. However, such processing would be necessary and proportionate to achieve the objectives as described above. With respect to the impact on the fundamental right of protection of personal data, the Code of Conduct on data–sharing between online platforms and public authorities could specify the application of the GDPR to the processing of STR data and provide for data protection safeguards to mitigate the effects of data sharing by setting standards on data, their format and the tools needed to perform such operation which are in line with the principles of GDPR (data minimisation and purpose limitation, integrity and confidentiality, etc.). If implemented by players, the agreement will streamline the operations, reducing the heterogeneity of formats data are shared at the moment. The expected result is the development of a data-sharing process that better guarantees the privacy and data protection of individuals, i.e. the hosts. If widely (voluntarily) adopted, the Code of Conduct should assist with the compliance with GDPR rules. In order to ensure the lawfulness of the processing of those data, a national legislative act to establish the legal basis for the processing would still be required.

Environmental impacts:

The potential impacts are presented below in a qualitative manner as the actual quantitative impacts of the measures under this option will (i) depend on the level of implementation of the measures by public authorities and online platforms, and (ii) on what public authorities would decide to do with the data they receive. Nevertheless, it is possible to forecast certain types of indirect positive impacts as presented below, if public authorities get more data, and if the data collected is used by public authorities to develop green policies:

·Increased ability of public authorities to assess and mitigate the ecological footprint of STR activity

More data on tourism flows via STRs could give public authorities the possibility to size energy consumption of such accommodation and their contribution to the ecological footprint. Public authorities would also possibly be able to better assess the needs for waste management and coordinate resources at local level accordingly, reducing the urban impact of STR activity.

·Impact on rural areas

Better data should allow public authorities to attract STRs in rural areas where STRs have a positive overall impact, for example where STRs can allow to invest in the renovation of houses, including for the greening of buildings. It should be noted that all positive impacts mentioned before would materialise only if and to the extent the soft tools under this option are complied with.

Stakeholders’ views on the policy option 1 148 : All stakeholders groups supported in some manner the measures presented under this option during the stakeholder and the public consultations. However, public authorities and online platforms – as well as local associations - notably have stressed the limits of soft approaches in this regard. In particular, they stress that soft approaches have already been put in place since the Communication of 2016 and the policy principles of 2018, but that these have not led to major improvements in terms of transparency in the STR sector. 149  Barcelona explained that, while the collaboration they have put in place with online platforms to obtain periodic data on STRs and flag illegal listings brought some improvements in policy making and enforcement, they still observe major quality problems with regards to the data received, which has forced the authorities to continue to collect data through web scraping. 150 This example therefore shows the limits of voluntary cooperation.

Stakeholders, including local associations, therefore, express a preference for a more structured framework for data sharing, taking the view that only a binding common EU framework, by preference of a legislative nature, could achieve it. This is confirmed in the Memorandum of the Urban Agenda Culture and Cultural Heritage 2021 151 (which concludes that without mandatory collaboration of online platforms, the regulatory powers of public authorities remains ineffective).

6.2Policy option 2 - Legislative initiative containing a common approach to data generation and data sharing

Economic impacts:

The main difference with respect to option 1 is that option 2 makes the system compulsory through legislation where there is a need for STR data. Impacts are similar in nature, but different in probability of realisation and magnitude. While in option 1, the implementation of the framework (and therefore the costs) is voluntary, and solely rely on the good will of public authorities and online platforms, in option 2, the implementation of the framework is mandatory for all public authorities (local or national) who would like to collect data on STRs, and for all online platforms. Repetitions of the descriptive parts of the analysis are omitted and analysis below will rather focus on the differences.

Impacts on online platforms:

Under option 2, all online platforms in the EU would have to put in place the data sharing framework. This will result in much bigger (and much more certain) benefits for online platforms, whilst also increasing compliance costs.

1.Benefits

The main benefit for online platforms results from the replacement of uncoordinated data requests by more streamlined and proportionate requests (in particular with regards to the type of data to be shared, the frequency and the technical means), which can be translated into cost savings on the long term.

Contrary to option 1, online platforms will have the same data sharing obligations and obligations to allow the display of registration numbers (when data are requested) within the entire single market. Under this option, the data set that online platforms have to share with selected public authorities (those that have put in place the data framework) will be pre-defined and closed. This will therefore create economic benefits for online platforms, and in particular for smaller platforms, as they will always have to share the same data set with the same frequency with selected public authorities.

A streamlined data-sharing framework, which will be automated, will help concretely online platforms to better plan and allocate resources due to the predictability of the data requests as well as the predictability of the data to be shared, the tools to be used and the data standards to be respected. This will imply certainty for online platforms on the type and timing of data-requests coming from public authorities since they will only receive data requests under the framework set up by the initiative, reducing the burdensomeness of requests.

Savings for online platforms are expected to be substantial. A streamlined data-sharing framework will enable online platforms to eliminate the costs for legal assessment and put in place an automated way to process incoming data requests from public authorities, thus making the cost to process a single request very limited and negligible (therefore reducing the cost to process single requests from EUR 13 549 to EUR 0). On this basis, it is estimated that savings for online platforms over a period of 5 years will amount to between EUR 54 million (based on 800 requests per year) and EUR 115 million (based on 1700 requests per year) and net savings to around EUR 35.8 million and EUR 96.8 million (baseline costs related to data requests from public authorities minus one-off costs for online platforms 152  of around EUR 8.2 million as presented below and cumulative maintenance costs of EUR 10 million 153 .

Small and micro-enterprise online platforms with average monthly active hosts equal or lower to 1 000 will benefit greatly from the possibility to use adapted reporting obligations (every three months, and manually, rather than weekly and automated for bigger online platforms). Moreover, the targeted survey for online platforms has shown that online platforms that would qualify for more lenient reporting obligations are already recipients of requests for data submitted by public authorities and this is likely to continue considering that there are national frameworks (e.g. France) that do not foresee any exemption to the collection of STR data from online platforms, including small and micro-enterprises. The data available doesn’t enable to quantify the exact number of data requests received by small and micro-platforms and the related cost savings for small and micro-platforms. However, a single data request might currently cost EUR 13 549 to process, which is significantly higher than the estimated cost of the adapted reporting obligations (around EUR 600 for each reporting activity per small or micro-platform or EUR 2400 per year) 154 .

Additional legal certainty will be given to online platforms by the legal basis that provisions in this policy option will provide – differently from option 1. Due to the diversity of requests, online platforms often fear that data requests from public authorities could breach privacy laws and other legal instruments at the level of the Member States’ domestic law or at the level of Union law (e.g. GDPR). Without a specific basis in Union law, online platforms are therefore obliged to analyse every single data request in order to make sure that these requests do not breach any existing legal instrument, incurring in extra costs and delaying the provision of data to public authorities. The new ad hoc legal basis set by this policy option will increase legal certainty for all online platforms and therefore reduce costs for legal assessment and processing time linked to data requests, while still complying with applicable Union legislation in the field of data protection, such as the GDPR.

2.Costs

Under this option, online platform will mainly incur administrative costs linked with the adaptation of their IT infrastructure and the connection to the single digital entry point.

Larger companies will face the one-off cost of adapting their IT infrastructure to share data by automated means (APIs) through each single digital entry point of Member States in which they operate and in which local, regional or national authorities request data. Small and micro-enterprise online platforms with 1 000 average monthly active hosts will benefit greatly from the possibility to use adapted reporting obligations (every three months, and manually, rather than weekly and automated for bigger online platforms).

The cost of creating one connection between an online platform and the single digital entry point is estimated at around EUR 30 000. In the computation of the costs per online platform, the complexity linked with the number of connections to establish has been taken into consideration. An intermediary that operates in more than one country will need to connect to each national single digital entry point but once the first connection has been developed, a 20% extra cost should be considered for each additional connection. Additionally, in the breakdown of the costs, costs for maintenance have been identified with one third of FTE for each year. Considering an hourly IT specialised salary of 75 EUR, the computation for yearly maintenance is estimated at EUR around 36 000 per year 155 .

Under these hypotheses, the one-off administrative costs for online platforms 156 are estimated at around EUR 8.2 million, while yearly maintenance (administrative) costs are expected to be around EUR 2 million that over a period of five years it amounts to a total of around EUR 10 million (EUR 18.2 million including one-off administrative costs).

As stated before, small and micro-enterprise online platforms with average monthly active hosts equal or lower to 1 000 will benefit greatly from the possibility to use adapted reporting obligations. Therefore, they will not face costs to adapt their IT infrastructure but they will be able to share data by uploading them every three months with an estimated cost to be around EUR 600 for each reporting activity per platform (or EUR 2 400 per online platform per year) 157  which is significantly lower than the estimated cost for processing one data request (EUR 13 549). Based on the estimation that 680 online platforms would qualify for the more lenient reporting requirements 158 , there is only a yearly total reporting (administrative) cost of around EUR 1.6 million. Adjusted reporting obligation will ensure the completeness of data collection – not creating loopholes in the system – and a very limited cost impact for small and micro-enterprises, while making sure that collection and data sharing are embedded in the business process since early stages of company life. This will also help to avoid the creation of artificial obstacles for this type of online platforms to their business development and growth. 159

Online platforms would also incur the (expected negligible) cost related to the creation of the field to request the registration number to hosts and to enable its display in the relevant listing.

Impacts on public authorities:

Under option 2, all public authorities (whether local, regional or national) who want to collect data on STRs would have to implement the framework, whereas under option 1, the implementation of the data framework is purely voluntary. The number of public authorities affected is therefore expected to be higher than in option 1 (for the purpose of calculating costs and benefits, it is assumed that 25 Member States compared to 16 Member States under option 1 will implement the framework). However, the net benefits at EU level will also be higher.

1.Benefits

·Reduction of costs to obtain data

All public authorities interested in data will be able to have the full picture on the online STR activity intermediated by online platforms. This is due to the fact that all online platforms, SMEs companies included, will be required to share activity data, avoiding the creation of loopholes in the data sharing flow. Additionally, the provisions foreseen by policy option 2 will ensure certainty on data traceability and optimisation of data-sharing processes through the implementation of the framework in areas interested in data. The ability to link data on STR activity to the registration number will ensure the traceability, giving the ability of public authorities to consolidate properly data coming from different online platforms on the same host. On the other hand, the defined data set to be shared by online platforms as well as the standards and the tools to be used for sharing it will ensure consistency in the data shared and optimizing the process, hence reducing overall costs.

The full transparency on STR activity that will be achieved will additionally ensure “transparency level-playing field” between STR activity and activity of other accommodation services (hotels, hostels, etc.) that are already monitored by local authorities. 49 out of 80 (61%) respondents declared to the dedicated survey for public authorities that they collect data from other tourism businesses, most of the time (46%) at least bi-monthly 160 . Public authorities will obtain another important piece of information on the accommodation services in their areas and will be able to properly address public interest objectives as management of tourism flows or affordability of housing. Acquiring the comprehensive and reliable knowledge on activity of STR accommodations will give a complete overview of the dynamics of the whole accommodation sector.

While it is not possible to quantify the exact cost reduction resulting from the implementation of the framework under option 2 161 , it is possible to assume that such benefits would materialize for public authorities implementing the data sharing framework. Indeed, while some costs (see below) are associated with the implementation of the data sharing framework, these are one-off costs (adjustment costs). The current costs for gathering data based on fragmented and non-efficient means are expected to be higher than the costs to gather data in a more automated way. Therefore, while on the short-term, absolute costs for public authorities may be higher due to adjustment costs, the benefits would be higher and outweighing the costs on the long term.

·Reduction of enforcement costs of STR rules

Areas that implement the data-sharing framework 162 would obtain the full benefits of transparency and have detailed, traceable, interoperable and timely data. Therefore, this would enable better policy-making, more targeted and proportionate rules, and greater enforceability of their STR rules, helping public authorities better achieve public policy objectives. Additionally, costs will likely be offset by a consistent reduction of the delay in enforcement that public authorities need to bear due to their inability to (a) collect data on a timely manner and (b) communicate to online platforms which listings are unlawful and ask them to be delisted. For example, as obligations will be clearly set by the provisions, savings are foreseen coming from the decrease of legal challenges between public authorities and online platforms to get the data. Similarly, as the single digital entry point shall enable public authorities to match the activity data shared directly by hosts (e.g. in relation to direct bookings), this shall also facilitate better enforcement and a complete understanding of the STR segment.

2.Costs

·One-off costs to adapt to the new registration system and the single digital entry point

The number of public authorities that will implement the provisions will vary based on their present or future appetite for data on STR activity. We can estimate that, at the moment, there is interest for data in at least one locality (big city) in 25 of the 27 Member States of the European Union. This number is calculated taking into account that in 22 Member States some kind of registration scheme is present – at least in one locality in their territory and other 3 Member States have introduced market access restrictions and they are planning to move towards taking additional measures including the creation of registration schemes.

·Costs per national authority for setting up an IT infrastructure including API connection, as well as yearly costs for hosting and maintenance are similar to those described in option 1. National authorities are expected to have one-off administrative costs estimated at around EUR 3 million to setup the infrastructure for the registration schemes and develop the national single digital entry point. Additionally, a yearly cost for hosting and maintenance is calculated to be around EUR 96 000 for each Member State, for a cumulative cost of EUR 2.4 million per year 164 . 

·Costs for local/regional authorities to connect to the single digital entry point

It is expected that local authorities will connect their systems to the national one in order to retrieve data from online platforms. The assumption is that (at least) local authorities with already a system for a registration scheme in place will develop such connections. In line with Option 1, we can assume that 100 local authorities within the 25 Member States have already a system in place and at least 100 will develop a new one (in total at least 200 165 local authorities within the 25 Member States). The 100 local authorities that have already a system in place will therefore save ~80% of the costs estimated for both the creation and yearly maintenance of a new system and the other 100 local authorities will face the whole cost of setting up the connection with the IT infrastructure created by national authorities, which are estimated at EUR 30 000 166 and yearly maintenance costs at EUR 36 000. It is possible to foresee a cumulative one-off cost for local authorities of around 3.6 million. Additionally, a cumulative yearly cost for hosting and maintenance is estimated to be around EUR 4.3 million.

Impacts on hosts: 

As the sharing of identification and activity data shall become consistent and reliable for all hosts and online platforms in all areas interested in data, the benefits under option 2 will materialise in a more systematic manner in those areas. Considering a scenario based on the status quo, it is expected that all areas where there is already an interest in data will use the framework set by provisions in policy option 2.

1.Benefits

·Savings in the time spent to complete registration procedures

There will be a marginal impact on hosts who are already subject to similar obligations, as it is expected public authorities will only request the data that is missing to have the minimum set of data required in the EU template and reuse the data already provided by hosts. Thanks to the provisions on a minimum, easy registration process, all future hosts in all areas should benefit from a more user-friendly and fast registration process in order to obtain a registration number and start to offer accommodation STR services.

In line with Option 1 and taking as a benchmark the most efficient and fast systems in place in the EU, it can be estimated that the waiting time for the hosts to complete a registration process will be reduced from 15 days to between 1.4 and 1.7 days. When translated into the monetisation of the time, hosts are expected to benefit from the reduction of around EUR 3000 167 to around 320 EUR per new registration (based on the average hourly wage)  168 . Taking into consideration that 25 Member States will be adopting the data sharing framework and for the cost considerations made above, it could be possible to save more than EUR 1480 million (monetisation of time saved in the registration process) for the new hosts that will be starting their activities (hence need to register in those areas) in the first 5 years after implementation (based on baseline number of 2019 and growth rate estimations) 169 . Based on the assumption that 87% of the hosts are peers 170 and 13% professional hosts, the cumulative cost savings for citizens over five years are estimated at around EUR 1287.6 million and for professional hosts at around EUR 192.4 million.

·More proportionate other STR rules

As expressed in the previous section, a medium/long-term positive effect for hosts expected is the higher probability of more proportionate rules on STRs issued by public authorities, with a higher probability of realisation with respect option 1.

2.Costs

As stated before, the estimated monetisation of the time hosts are expected to dedicate to a new registration is around EUR 320. We assume that 50% of the hosts will need to register ex novo and the other 50% will need to (eventually) adjust the info they already provided, which in line with Option 1 will not cost more than ~30% of a new registration. Considering the number of the hosts in the above-mentioned 25 Member States, it is expected one off administrative costs of around EUR 112.4 million for them and cumulative yearly administrative costs over a period of five years of around EUR 171.3 million. 171  Based on the assumption that 87% of the hosts are peers 172 and 13% professional hosts, the cumulative administrative costs for citizens over five years are estimated at around EUR 149 million and for professional hosts at around EUR 22.3 million.

Impacts on Tourism Ecosystem and other accommodation service providers:

The same types of impacts as described under option 1 would apply, with a wider benefit expected overall as the number of public authorities putting in place the framework would be higher than under option 1. More data would be available for the tourism data space which could allow for e.g. for more innovative services.

Social impacts:

Social impacts are expected to be similar to those described in option 1, but are expected to be more probable and more important, given that a consistent data framework would be in place in all areas in the EU where STR is currently an issue. It is however not possible to quantify the impacts, as social impacts would largely depend on what authorities would do with the data. In addition, better rules on data sharing are expected to have a positive impact on respect for privacy (see under option 1).

Environmental impacts:

Environmental impacts are expected to be similar to those described in option 1, but are expected to be more probable and more important, given that a consistent data framework would be in place in all areas in the EU where STR is currently an issue. However, it is not possible to quantify those impacts as explained above, as impacts would only materialize where public authorities make use of the data collected to design green policies.

Stakeholders’ views on policy option 2 173 : all stakeholders support in substance the measures put forward under this option (similarly to option 1 and option 3). The public consultation show that respondents, in particular public authorities, local associations and representatives of the tourism industry, are in favour of requiring online platforms to share data on STRs with public authorities to increase transparency and facilitate public policy and enforcement activities. The majority of respondents also consider that such measures should be put in place and/or facilitated at EU level. 174 Public authorities are strongly in favour of this option, compared to the other options, as it leaves some flexibility to Member States and local authorities to regulate, taking into account local preferences and needs, both in terms of data needs and data sharing requirements. Online platforms, in turn, are usually in favour of a more harmonised approach to data sharing, avoiding fragmentation notably of registration obligations which they need to enforce by enabling the display of registration numbers.

6.3Policy option 3 - Legislative initiative containing a common approach to data generation and data sharing, with mandatory national registration schemes for all hosts

Option 3 foresees the obligation for all hosts in the EU to register in national registration schemes. Public authorities and hosts are the stakeholders that will be directly affected compared to option 2. Online platforms will also be further impacted, but only marginally.

Economic impacts:

Impacts on online platforms:

1.Benefits

On a general level, the types of impacts on online platforms are similar to option 2. Online platforms would slightly gain in terms of efficiency, as diverging local and regional registration systems for hosts would be replaced by national ones, making it also easier for platform to track what systems exist and which requirements their hosts need to comply with. However, while online platforms would have to share the same type of data with public authorities, they would have to share a higher volume of data, namely data regarding all hosts in the EU (compared to data on hosts for selected public authorities under option 2). Similar to option 2, small and micro-enterprise online platforms with average monthly active hosts equal or lower to 1000 will benefit greatly from the possibility to use adapted reporting obligations (every three months, and manually, rather than weekly and automated for bigger online platforms).

On this basis and in line with the calculations under Option 2, it is estimated that net savings for online platforms over a period of 5 years will amount between around EUR 36.3 million and EUR 97.3 million (baseline costs related to data requests from public authorities minus one-off costs of EUR 8.2 million and yearly maintenance (administrative) costs of EUR 2 million 175 ). Similarly to Option 2, for online platforms that would qualify for the more lenient reporting obligations, the data available doesn’t enable to quantify the exact number of data requests they receive and the related cost savings. However, a single data request might currently cost EUR 13 549 to process, which is significantly higher than the estimated cost of the adapted reporting obligations (around EUR 600 for each reporting activity per small and micro-platform or EUR 2400 per year) 176 .

2. Costs

Specifically on the cost structure, it will follow exactly what has been discussed in the previous section of Option 2 177 . One-off costs associated with data reporting for online platforms 178  under this option are estimated around EUR 8.2 million, while annual administrative costs would be of around EUR 2 million. For small and micro-enterprises with average monthly active hosts equal or lower to 1000 and based on the estimation that 680 online platforms would qualify for the more lenient reporting requirements 179 , there is only a yearly cumulative reporting (administrative) cost of around EUR 1.6 million (EUR 2400 per year per online platform that would qualify for the more lenient reporting obligations).

Impacts on public authorities:

1.Benefits

Benefits are expected to be the same as under option 2.

2.Costs

Costs are expected to be of the same nature as under option 2. However, in Member States where registration schemes already exist at local or regional level, national authorities will have to centralise these registration schemes at the national level. Important adjustment costs are therefore foreseen for Member States where registration schemes already exist at local or regional level.

In addition, all 27 Member States will have to create a single digital entry point, and adapt/create ex-novo a centralised national registration scheme as well as the needed IT infrastructure, including the APIs predisposition. We assume that 20 Member States will create a new system (IT infrastructure costs at around EUR 170.000 per Member State) and cost savings (30%) will apply only to 7 Member States that already have in place some form of IT infrastructure at national level 180 . Therefore, for the setting up of the systems in the 27 Member States, a total one -off cost of around EUR 3.6 million is expected to set up the IT infrastructure at national level. Additionally, a yearly administrative cost for hosting and maintenance as under option 2 is calculated at around EUR 96.000 for each Member State 181 .

It is expected that local authorities would develop automated connections to the single digital entry point in order to receive activity data, linked with the (newly created) registration scheme at national level. It can be assumed that at least 220 182 local authorities within the 27 Member States will develop a new system or adapt an old one. In line with option 1 and 2, it is possible to assume that 100 of the urban/local areas opting in in the system have already a system in place (45% of the 220) and therefore will save ~80% of the costs estimated for the creation of a new system. The other 120 (55% of the 220) of the urban/local areas will face the whole cost of setting up (of around 30.000 EUR for APIs development). It is possible to foresee a cumulative one-off cost for local public authorities at around EUR 4.3 million. Additionally, a yearly cost for hosting and maintenance is calculated to be around EUR 5.2 million.

Impacts on hosts:

The same types of impacts as under option 2 apply. However, the impacts will affect (positively and negatively) all hosts in the EU, rather than only hosts in areas putting in place the framework. Hosts that had not been subject to any kind of registration before will pay the full cost of registration estimated.

1.Benefits

In line with options 1 and 2 and taking as a benchmark the most efficient and fast systems in place in the EU, it can be estimated that the waiting time for the hosts to complete a registration process will be between 1.4 and 1.7 days across all 27 Member States (around 320 EUR per new registration based on the monetisation of an average hourly wage)  183 .The cost savings are expected to be similar as in Option 2 184 . Option 3 introduces registration schemes in all 27 Member States, covering all local authorities, including the ones not having any intention to do so (e.g. small communities in Member States that have no intention to register or regulate STRs) so creating additional registration costs for hosts that would probably never need to register.

2.Costs

Costs associated with hosts follows the same reasoning of option 2. Under option 3, however, all hosts will be subject to the same obligation to register. Hosts in Europe that had not be subject to any kind of registration before will incur full cost of registration estimated (i.e. 320 EUR). The total one-off costs for hosts will be at around EUR 376.6 million. This estimation takes into account the total number of hosts in the EU and that the majority of the hosts will need to perform a new registration (i.e. all hosts in 2 MSs and 80% of all hosts in the remaining 25 MSs) as well as the hosts who would need to adapt their actual registration to the new template (20% of hosts in 25 MSs already subject to some form of registration). This means that compared to Option 2, as the registration is obligatory across all 27 Member States, around 825 683 additional hosts would need to register under Option 3, with an estimated one-off cost of EUR 264.2 million. Additional yearly administrative costs for the next 5 years after the implementation of the provisions foreseen by option 3 are estimated to be around EUR 438.4 million for new hosts that will be starting their activities (hence need to register in those areas)  185 . Based on the assumption that 87% of the hosts are peers 186 and 13% professional hosts, the one off administrative costs for citizens (peers) are estimated at around EUR 327.14 million and cumulative yearly administrative costs over a period of five years at around EUR 381.4 million. 187

Impacts on Tourism Ecosystem and other accommodation service providers:

The same types of impacts as under option 2 apply. However, the magnitude of the impacts are much wider as data will be available on all hosts in the EU. If public authorities make the data available at aggregate level, for e.g. through the future tourism data space, players in the tourism ecosystem will have more predictability of the market. Local communities could also potentially enjoy the complete data availability.

Social impacts:

The indirect impacts on society will be more consistent and structural compared to option 1. However, less positive impacts on the society are expected under option 3 as this option reduces the regulatory freedom of local and regional authorities to maintain their own registration requirements and decide on the type of data they would like to request from hosts and online platforms. As for fundamental rights, the same types of impacts as under option 2 apply, but such impacts will apply more widely to all hosts.

Environmental impacts:

With a global picture on the STR activity on their territory, public authorities will be able to better address the negative externalities of STRs impacting the environment. However, as also mentioned above, national registration schemes might reduce the scope for specific local and/or regional actions to promote sustainability (e.g., the scope for requiring the submission of additional certificates etc.)

Stakeholders’ views on policy option 3 188 : this option is mainly supported by online platforms, for whom a full harmonisation of registration requirements, obligations to enable the display of registration numbers and data sharing obligations will substantially decrease compliance costs. While hosts stand to benefit from more proportionate registration schemes and STR rules in general, they are not necessarily in favour of more administrative burden (registration obligations) from the outset. Finally, public authorities have stressed the need for any regulatory intervention to respect subsidiarity, and are therefore not in favour a full harmonisation at EU level, but would rather welcome more flexibility to implement the framework where it is really needed.

7.How do the options compare? 

Table 1 - Policy objectives and policy options

Legend: +- almost no impact; + reduced positive impact; ++ positive impact; +++ significant positive impact; - minor negative impact; -- negative impact; --- significant negative impact; +/- mixed impacts.

The options that allow meeting the policy objectives best are those relying on a legislative approach. In both Options 2 and 3, public authorities will have the possibility to retrieve data and online platforms will gain certainty on how to share data. With the flexibility to opt-in, Option 2 makes sure that public authorities with immediate needs get the data while under Option 3 public authorities will have data on all hosts, even where there is not an immediate need for data.

Table 2 – Cost and benefits of policy options for key stakeholders

Legend: +- almost no impact; + reduced positive impact; ++ positive impact; +++ significant positive impact; - minor negative impact; -- negative impact; --- significant negative impact; +/- mixed impacts.

The cost-benefits ratio as a whole is the best where flexibility is provided to public authorities to opt in. Public authorities will incur costs only where there is a real need for data – hence a potential big benefit in using data and better assess STR related issues. Option 3 will provide data to all public authorities, including those not asking (yet) for data. Only in the medium/long term will the benefits for public administrations exceed the initial costs they will face to comply with Option 3. For hosts, Option 3 will have mixed effects since all hosts in EU will have to register, even where public authorities do not need data, therefore creating additional unnecessary costs for such players. However, hosts will benefit in the medium/long term from preventive assessment of externalities linked with STR activity, avoiding undifferentiated policies for STR activity.

Table 3 - Comparison of policy options

Legend: +- almost no impact; + reduced positive impact; ++ positive impact; +++ significant positive impact; - minor negative impact; -- negative impact; --- significant negative impact; +/- mixed impacts.

The key difference in terms of effectiveness between the 3 options lies in a legislative approach imposing obligations on stakeholders, as foreseen under Options 2 and 3. Options 2 and 3 are the most effective as they reduce the burden on platforms and ensure that public authorities have tools to get the STR data.

When it comes to efficiency, the impacts differ. The most efficient option is the one where the flexibility of the opt-in system is provided to public authorities immediately interested in data. Option 1 seems less efficient due to the voluntary nature of the measures, and the efficiency of Option 3 is reduced at EU level as it imposes an obligation even on public authorities not interested in data.

Finally, with regards to proportionality Option 2 scores the best, as it takes the preferences of local authorities into account, in full respect of the subsidiarity principle. Some liberty is also achieved with Option 1, but the possible lack of collaboration of online platform reduces the overall proportionality of this option. Option 3 seems to be the least proportionate as it imposes obligations on all Member States even though they might not have any needs for STR data.

Table 4 - Impact of policy options

Legend: +- almost no impact; + reduced positive impact; ++ positive impact; +++ significant positive impact; - minor negative impact; -- negative impact; --- significant negative impact; +/- mixed impacts.

Summarising how the three options compare under the economic, social and environmental impacts, Option 2 seems to maximise the general positive impacts on the three levels. The reasoning for each option is developed in chapter 6.

Table 5 - Support of stakeholders to policy options

Legend: --- completely adverse; -- not in favour; - not so in favour; +- neutral; + marginally in favour; ++ in favour; +++ significantly in favour – for a more detailed explanation, see Section 6.

The interactions with stakeholders involved in the STR segment and therefore potentially impacted by this initiative show that all of them welcome an EU initiative that could streamline data sharing between online platforms and public authorities. This is also the case of local association. Stakeholders, however, stressed their need for flexibility (for public authorities, obligations only where there is a need for data) and for certainty (for online platforms, clarity on data set to be shared and tools to be used; for public authorities: clarity of the legal basis for data requests and consistent enforcement of their rules). Hosts favour simplified registration but only in areas where this is considered necessary.

8.Preferred option

8.1Preferred policy option

Option 1 does not ensure that the burden on platforms and access to data will be adequately addressed. It will only lead to improvements in some areas and only between limited online platforms and public authorities. This option, as voluntary, is heavily reliant on cooperation and the good will of those making commitments and abiding by them and lacks any means of enforcement to make the more effective. This option will provide little or no improvement in addressing the problems faced by platforms with burdensome requests and public authorities for data.

Option 2 does meet the objectives in a flexible and proportionate way. It would provide a clear and coherent framework for data access and sharing, which provides detailed, traceable, interoperable and timely data in a simple and balanced way for all stakeholders. This would enable better policy-making, greater enforceability of their STR rules and help public authorities better achieve public policy objectives. The burden on platforms will be significantly reduced.

Option 3 would provide an effective data sharing framework across the EU, but requires high adjustment costs for authorities, whilst reducing possibilities to address specific concerns at local and/or regional level. This option ensures full transparency on the STR segment in the EU, including in areas where data are not of interest for public authorities. As one registration system will be created at national level in each Member State, local and regional authorities would not incur any direct costs, but Member States who are not currently interested in STR data would incur costs regardless of their preferences. The costs on hosts across the EU would also be increased, as all hosts in the EU would be required to register. Additional benefits will be reached in the medium/long-term, since public authorities will be given the right tools to have full transparency and assess in advance negative and positive externalities linked with STR activity.

The preferred option is therefore, Option 2. It addresses the specific objectives in a consistent, proportionate, effective and efficient way across the areas in the EU where main problems are de facto present. It ensures the use or adaptation of existing national and regional registration schemes and IT infrastructures (if any). It optimises costs and preserves the ability for public authorities to set the data-sharing framework in compliance with the principles of subsidiarity and proportionality. The principle of subsidiarity is respected. Member States are not able to address alone the issues faced by online platforms operating cross-border. It is also respected by giving local and regional authorities flexibility with regards to the implementation of registration schemes (whether to introduce them or not, and at which level), but also with regards to the information and evidence that each authority can request from hosts, to ensure that local circumstances and needs are respected and taken into account. The measures proposed also strike a balance between the interests of all stakeholders involved, by requiring action from public authorities, hosts and online platforms, and offering benefits to these actors, ensuring the proportionality of the foreseen intervention.

It also contributes to the fulfilment of the Sustainable Development Goals, in particular SDG 11 (sustainable cities and communities), by giving public authorities the tools and the data to regulate the STR sector in a proportionate and sustainable way.

8.2REFIT (simplification and improved efficiency)

REFIT section is not applicable to this proposal. 

9.How will actual impacts be monitored and evaluated?

The Commission will monitor the implementation, the application and the compliance to the new system a view to assessing its effectiveness. To do so, it will work together with the Member States. For instance, it intends to assess whether the rules effectively lead to better quality data for public authorities, based on registration schemes for hosts and data sharing rules for platforms that are effectively complied with.

After the adoption, a transitional period of 2 years will be given to Member States to map the interested local authorities, create or adapt local/national existing registration systems (connecting local registration schemes, if any) and set up the IT infrastructure at national level to streamline data-sharing with online platforms (through the single digital entry point). The first evaluation should take place not earlier than five years after the full implementation of the Regulation (i.e., five years after the initial first two years which will be the transition period). The general effectiveness will be monitored based (but not solely) on the set of KPIs identified in Table 6 below.

Table 6 – Key Performance Indicators for the evaluation of the implemented measure

Legend: # number, % percentage, ∆% percentage difference

Annex 1: Procedural information

1.Lead DG, Decide Planning/CWP references

Lead DG: DG Internal Market, Industry, Entrepreneurship and SMEs (DG GROW)

Directorate: G - Ecosystems II: Tourism & proximity

Decide number of the underlying initiative: PLAN/2021/11047 (Short-term rental initiative)

2.Organisation and timing

The inception impact assessment was published on 16 September 2021. It was followed by a feedback period that lasted from 16 September 2021 to 14 October 2021. 71 stakeholder feedbacks were received.

The Commission held a public consultation from 27 September until 13 December 2021. This consultation was available on the Have your say portal and open to anyone who wished to reply. The public consultation received 5696 stakeholder feedbacks. The Commission launched two additional targeted EU surveys in April 2022: the first one to public authorities, for which it received 80 replies, and the second one to online platforms, for which it received 15 replies. The results of the targeted surveys are presented in Annex 2a and Annex 2b.

The following DGs (Directorates General) have been invited to contribute to this impact assessment: SG (Secretariat-General), CLIMA (Climate Action), COMP (Competition), CNECT (Communications Networks, Content and Technology), DIGIT (Informatics), ECFIN (Economic and Financial Affairs), EMPL (Employment, Social Affairs & Inclusion), ENV (Environment), ESTAT (Eurostat), JRC (Joint Research Centre), JUST (Justice and Consumers), REGIO (Regional and Urban Policy), MOVE (Mobility and Transport), SJ (Legal Service), TAXUD (Taxation and Customs Union) and TRADE (Trade).

The ISSG met three times in 2021 (15 July, 25 October, 15 December) to give an update on the ongoing work and discuss preliminary versions of the Impact Assessment report, together with all the supporting documents. The final ISSG meeting took place on 14 January 2022. Following the negative opinion of the RSB, an additional ISSG meeting took place on 14 June 2022 to discuss the revised impact assessment.

3.Consultation of the RSB

The RSB was consulted in an upfront meeting on 15 November 2021. The impact assessment report was submitted to the RSB on 19 January 2022. The impact assessment was discussed with the RSB on 16 February 2022, and the RSB issued a negative opinion on 18 February 2022. Based on the RSB recommendations, the impact assessment has been significantly revised in accordance with the following points:

The revised impact assessment was submitted to the RSB on 23 June 2022. The Board issued a positive opinion with reservations on 22 July 2022. The recommendations of the Board were addressed in the impact assessment as follows:

4.Evidence, sources and quality

DG GROW conducted and contracted several studies in support of the impact assessment:

·Oxford Research. “Study on national regulatory approaches to short-term accommodation services” - European Commission, Luxembourg (2021) (upcoming).

·Van Haaren, Jeroen, Susan Vermeulen, Jeroen Klijs, Ko Koens and Jorrit Bijl. “ Short-term accommodation rental in Amsterdam. An empirical investigation of statistical correlations between short-term rental, housing prices and quality of life index .” European Commission, Luxembourg (2021).

·Ipsos European Public Affairs. “ Flash Eurobarometer 495. Short-term rentals in the EU .” European Commission, Luxembourg (2021).

·Spark Legal Network and Valdani Vicari & Associati. “ Study on the Assessment of the Regulatory Aspects Affecting the Collaborative Economy in the Tourism Accommodation Sector in the 28 Member States .” European Commission, Luxembourg (2018).

A series of external analytical papers contracted by the Commission shed light on issues related to market access requirements in the short-term accommodation rental segment:

·Duch-Brown, N. (Coord.), The Short-Term Accommodation Rentals market in the EU, JRC Digital Economy Working Paper 2021-09

·Ratin Legis. “ Impulse Paper on the Business Authorisation/Licensing Requirements Imposed Both on Peer-Providers and Platforms. Barcelona, Berlin and Amsterdam .” (2016)

·Guido Smorto. “ Impulse Paper 2 on the Business Authorisation/Licensing Requirements Imposed Both on Peer-Providers and Platforms in Paris, Rome, Milan and London ” (2016)

·Sofia Ranchordás et al. “ Impulse Paper on Home-Sharing in the Digital Economy: The Cases of Brussels, Stockholm and Budapest .” (2016)

The Commission (DG GROW) signed a landmark agreement with Airbnb, Booking, Expedia Group and Tripadvisor, which began collaboration between these platforms and Eurostat, to publish data on short-term accommodation rentals offered via these platforms across the EU.

DG GROW conducted dedicated workshops with all stakeholders:

· Developing a fair single market for short-term rental services , 10 December 2021.

· Enhancing transparency on short-term rentals in the EU , 22 October 2021.

·SIMFO Workshop 2021 – The holiday and other short-term accommodation sector in Dublin, 10-11 February 2021 189

·  SIMFO Workshop 2021 - The holiday and other short-term accommodation sector in Lisbon , 28-29 January 2021

· SIMFO Workshop 2021 - The holiday and other short-term accommodation sector in Valencia , 3-4 December 2020

DG GROW carried out a literature review, based on the references listed in Annex 10.

Annex 2: Stakeholder consultation

Consultation strategy

The objectives were to collect feedback and data on the identified problems and drivers, as well as on the possible solutions. The activities included a public consultation of all stakeholders, a roadmap, workshops and bilateral meetings with relevant stakeholders. The main communication channel has been the Commission’s Have Your Say portal, a dedicated webpage , and twitter . Consulted stakeholders include hosts and online platforms (and their organisations), public authorities (national, regional and local level), service providers such as management companies, hotels, citizens and local associations.

71 feedbacks were received on the inception impact assessment, during a consultation from 16 September to 14 October 2021. Hosts stressed the need to increase the responsibility of online platforms in ensuring that only legal listings are displayed. They warned against the multiplication of restrictive rules for hosts at local level. Property managers (mainly SMEs) suggested that simple registration procedures should be in place for STRs and that a national database should be available to map STR operators. For online platforms, data sharing requests should be proportionate and in compliance with EU law, notably GDPR. Smaller platforms called for data minimisation and the ‘once only’ principle for data sharing. They stressed that new data sharing obligation should build on existing obligations (e.g. under DAC7 and the DSA) and fill remaining gaps. The hospitality sector (mainly hotels) supports registration schemes for hosts and increasing data sharing by online platforms. It calls for a level playing field between STRs and traditional accommodation providers. The majority of stakeholders supports STR rules that differentiates between different types of hosts (no one-size fits all).

Bilateral meetings with the stakeholders and five workshops took place. 190 A public consultation was open from 27 September to 13 December 2021, available in 23 EU languages. In addition, two targeted surveys were launched to public authorities on the one hand, and to platforms on the other hand, in April 2022. The results are presented in Annex 2a and Annex 2b respectively.

Public consultation

Stakeholders group. Respondents were asked to self-identify themselves under one of the 7 categories of respondents that match the composition of the STR ecosystem: hotels and similar, online platforms and their associations (traditional STR platforms, booking platforms, online travel agencies, and their associations), other service providers (e.g. property management companies, maintenance and cleaning services), individual and professional hosts, public authorities (local, regional and national) and other (e.g. academia, host associations, housing and tenant associations, trade associations). The results are presented based on these specific categories of STR stakeholders, rather than on the standard categories of respondents of EU survey. 191 The affiliation with a certain respondent category is based on the self-declaration of the respondent. Thus, is should be bared in mind that there may be errors with regard to the affiliation chosen by the respondent. The following graph shows the responses per respondent category:

Figure 1. Responses by stakeholder group

Total respondents: 5,479.

Figure 2. Responses by country of respondent

Total respondents: 5,695.

The graphs above shows the number of responses per stakeholder category. The majority of responses were submitted by individual (78.1% - 4,278) and professional (11.1% - 606) hosts. Most of the replies were submitted from France (24.8% - 1,414), Italy (16.8% - 957), Spain (10.6% - 601) and Germany (9.8% - 558).

Methodology and tools to process the data. For all seven groups, we received 5696 responses. One reply was disregarded at the request of the respondent. We identified 216 respondents who wrongly self-identified as online booking platforms. We do not include those respondents in the disaggregate analysis, but we include their answers at aggregate level. Therefore, at disaggregate level we only consider 5.479 respondents. As the survey aims to measure the perception and possible issues related to STR, frequency analysis was carried out. Data was categorised into one of seven stakeholder groups, defined above.

The survey is composed of closed and open questions. Closed questions were assessed based on a proportion analysis, measuring the responses across different stakeholder groups. Proportions are displayed individually in each of the questions. For open questions, a qualitative analysis was performed to measure the responses proportions. Filters by stakeholders and responses were applied in both cases if necessary for the study purposes.

Campaigns. A possible campaign by online platforms to alter some of the answers was considered. The periods from 28 October to 3 November 2021 and 25 to 29 November 2021 were analysed. The analysis consisted of the identification of duplicate answers across stakeholders. Duplicates were identified among responses from individual hosts but with a 1% frequency in the first period and a 0.3% in the second one. The analysis does not suggest any possible campaign and any possible alteration is discarded.

Contributions received outside the formal consultation process. 11 contributions were received by email, for either confidentiality or technical reasons, and have been taken into account in the results below.

Results

1.General market information

Definition of STR. For respondents, STR include the provision of accommodation, typically on a daily or weekly basis principally for short stays by visitors (95.9% - 5,462 out of 5695) of the respondents strongly or somewhat agree); 67.4% (3,838 out of 5695) also strongly or somewhat believe it includes stays on a monthly basis. 94.8% (5,399 out of 5695) of the respondents believe it includes furnished rooms or entire apartments or houses, and 50.1% (2,853 out of 5695) agree that it excludes furnished short-stay accommodation with daily cleaning, bed-making, food and beverage services.

Benefits generated by STRs. For all respondents, STRs generate economic and tourist benefits. Most of the hosts (97.5% - 4,762 out of 4884) agree with that statement. Online platforms, public authorities and hotels usually give greater importance to the economic impacts rather than tourist benefits. Nevertheless, both benefits are significant.

Figure 3. Benefits generated by STRs by Stakeholders group

Total respondents: 5, 479.

Economic benefits group the options: It is an additional source of income for citizens; It creates employment; It attracts visitors; It encourages investment in local services (e.g. housing, tourism, transport).

Benefits for tourists group the options: It caters to unfulfilled tourism needs (e.g. low-income tourists, (large) families); It is an alternative to more traditional accommodation offers.    

Negative impact of STRs. 48.2% (2,355 out of 4884) of the hosts report no drawbacks. However, 47.6% (2326 out of 4884) of them is aware of housing affordability and availability issues that STRs could create. Public authorities reported security issues (73.1% - 19 out of 26). 32% (8 out of 24) of online platforms consider the rental pricing issue relevant and 32% (8 out of 25) report no drawbacks.

Figure 4. Negative impacts generated by STRs by Stakeholders group

Total respondents: 5, 479.

Economic: It has a negative economic impact on traditional accommodation providers.

Rent Price: It increases the price of renting or buying housing in cities; it decreases the availability of housing for rent or sale in cities.

Security: It raises security issues; it increases nuisance for local residents.

No Drawbacks: There are no drawbacks.                                                                

Hosts activity

The next graphs show the type of accommodation rented out by both individual and professional hosts and their location (urban, rural area or both).

Figure 5. Type of accommodation offered by hosts

Total respondents: 4884.

Room: Room(s) in my main residence (apartment or house)

Main Residence: The entirety of my main residence (apartment or house)

Own & Other Apartment: An apartment or house other than my main residence

Several Apartments: Several apartments or houses other than my main residence

Other Apartment: Among the “other” type of accommodation, the following were mostly listed: (a) mostly apartments built next to the main residence; (b) other atypical rental spaces (e.g., boat, bungalow, yurt, tent).

Figure 6. Locations where accommodations are offered by hosts

Total respondents: 4884.

Means used to offer STR services. Hosts mainly use online platforms to advertise their STR services. In addition, 19% (928 out of 4884) of hosts consider word mouth as an important means to offer their services.

Online platforms used to offer STR services. Hosts use mainly large platforms to offer their STR services, while only 8.6% (420 out of 4884) of hosts use small platforms only. The most popular large platforms cited were Airbnb, Booking, Vrbo (Expedia) and TripAdvisor. Airbnb consolidates as the most used one (~85% for individual hosts 192 and ~72% for professional hosts 193 ). Some examples of small platforms used by individual hosts included Leboncoin, Rentalia, Homeaway (8.4% - 359 out of 4278), and those used by professional hosts included Gîtes de France, PAP vacances, Agoda and other local platforms (32.67% - 198). 

Figure 7. Main online platforms used by hosts

Total respondents: 4884.

Cross-border activity

Hosts and online platforms were asked whether they offer their services only in their country of residence (and in how many cities), or whether they also offer their services cross-border. While platforms did not provide an answer to these questions, it can be assumed that almost all online platforms operate cross-border. At the same time, while most hosts provide their services in their country of residence (89.4% - 4,366 out of 4884) some hosts declared to offer their services in more than 1 city (13.6% - 664 out of 4884) and some even cross-border (16.3% - 796 out of 4884).

The main reasons put forward by hosts for not operating cross-border were: not interested in operating cross-border (55.2% - 2,696 out of 4884) too costly to do so (21.8% -1,065 out of 4884), don’t know what rules apply (27.6% - 1,348 out of 4884), too many rules apply (11.7% - 571 out of 4884).

Hosts indicated that significant 194 expenses arise to comply with obligations in different Member States. Respondents report cleaning and maintenance costs and taxes. Hosts point out commission fees from platforms.

Ancillary service providers

39.2% (1913 out of 4884) of hosts declared that they collaborate with providers of other services (ancillary services, e.g. property management, cleaning, maintenance…) or are considering doing so. 60% (15 out of 25) of online platforms declared they help arrange such services. 83.4% (121 out of 145) of ancillary services providers declared that they gain a significant part of their customers/revenue from services requested by hosts and/or by referrals from online platforms.

Impact of COVID

When asked about the impact of COVID on the STR sector, public authorities indicated that the STR sector suffered similar setbacks as other industries related to tourism. However, the STR sector shows greater resilience and recovered volumes faster than other types of accommodation. The lockdown period has – reportedly – heavily affected STRs. However, most respondents point out a current recovery to pre-pandemic levels. Hosts, in turn, described the impact of COVID to be significant. On the first year of the pandemic loses range from 20% to 100%. The lockdown implied a drop in international tourists arrivals, sharply reducing STRs revenues. No respondent who identified as a platform replied to specific questions on this issue.

2.Transparency

Data needed to design STR rules

Public authorities indicated that they need non-personal (73.1% - 19 out of 26) and personal data (26.9% - 7 out of 26) to design STR rules. The following data was mentioned most: number of available STRs, property information (location), number of nights rented, information on the hosts.

Data needed to enforce STR rules

For enforcement purposes, 53.8% (14 out of 26) public authorities indicated they need non-personal data, and 61.5% (16 out 26) public authorities indicated they need personal data. The following data were mentioned most: Number of nights the property is rented, host identification and property information (location).

Access to data by public authorities 

26.9% (7 out of 26) of public authorities indicated that they have access to non-personal, aggregated data on the number of STR and of nights rented out and 11.5% (3 out of 26) personal data on who rents what and how often. Public authorities get access to both personal and non-personal data mostly by using publicly available data or private/commercial data. Among those who declared a lack or limited access to personal (88.46% - 23 out of 26) and non-personal data (73.07% - 19 out of 26), the main reasons behind were a lack of appropriate legal framework for data-sharing (for personal 78.26% (18 out of 23) and non-personal 63.15% (12 out of 19)) and lack of response from online platforms to requests by public authorities (56.5% (13) and 52,63% (10) respectively). Ten public authorities declare to receive data either from online platforms (5 out of 26) or using publicly available and commercial data (5 out of 26). Some of those public authorities declare to receive the data on a yearly basis or even on a timely (daily) basis. Among all public authorities, a low percentage (23.07% - 6 out of 26) require a particular delivery format.

Public authorities also receive data directly from hosts (57.7% - 15 out of 26). Those who receive data from hosts do so via online platforms, register procedures and tax payments. Those who do not receive data from hosts declared the following reasons: lack of legislation, lack of cooperation from hosts, and privacy concerns.

Finally, fifteen public authorities declared to receive data from hotels. They particularly have access to licenses, tourist tax information and official statistics data.

Data sharing by hosts

82.9% (4,049 out of 4884) of hosts declared to share information on STRs with local public authorities. In particular, 79.4% (3,878 out of 4884) of hosts declared that they have to register with local authorities and provide information in that context. 83.2% (4,063 out of 4884) of hosts declared that they share information when complying with tax obligations. Some hosts declared to share other type of information, such as the identity of hosts (Spain, Greece, Portugal, Italy). Other: information on income, duration of stay (Greece), monthly statistics to city council on occupation rate (Germany).

Data sharing by platforms

76% (19 out of 25) of platforms respondents declared to share data with public authorities from at least one Member State. 20% (5) declared to share data only in certain Member States (but not all in which they operate). Platforms that do not share data with public authorities (24% - 6 out of 25) declared the following reasons for not doing so: four have not received requests; one has received requests, but they believe they are not subject to the jurisdiction of the relevant Member State; one cannot legally share the data.

With regards to public authorities that request data, platforms identified: National and local tax authorities, tourism authorities and tourism information offices, policy authorities and statistical institutes. When they share data with public authorities, it is often on a mandatory basis (44% - 11 out of 25), less so on a voluntary basis (12% - 3 out of 25), and sometimes on both/it depends on the data (12% - 3 out of 25). One platform explained that they usually create reports on voluntary basis since they want to be open and transparent.

Platforms that share data with public authorities share non-personal data (60% - 15 out of 25), personal data (32% - 8 out of 25) or sometimes other types of data such as income details or tourist information. They share the data mostly on a yearly basis (28% - 7 out of 25), but sometimes also every six months (20% - 5 out of 25) upon requests of public authorities (44% - 11 out of 25) or on a timely basis through e.g. APIs (12% - 3 out of 25). With regards to the level of diversity of the data requests from public authorities, 12% (3 out of 25) declared that the difference is high, while 8% (2 out of 25) declared that there is low or no difference at all.

Platforms declare that data requests continue to pose legal challenges and costs, as they are often fragmented, inconsistent (often duplicate). Platforms identify costs according to (i) legal costs due to challenges with local public authorities and (ii) operational costs linked to the internal product development required to support local data-sharing compliance schemes (and lack of solutions at scale). Platforms report that data-sharing requirements can be unclear, complex and much more costly to build compliance schemes around.

Cooperation of platforms in enforcing STR rules

72% (18 out of 25) of platforms declared that they are subject to obligations to cooperate with public authorities to enforce existing STR rules (e.g. obligation to display registration numbers, delist listings without a registration number).

Platforms specified that the following obligations could apply (non-exhaustive listing) depending on the market:

·Delisting non-compliant listings when ad hoc notices are received from public authorities,

·Displaying registration numbers,

·Identifying whether a listing needs a registration number or not,

·Enforce nightly stay caps restrictions,

·City tax collection and remittance on behalf of home owners.

In terms of costs to comply with such obligations, platforms indicated that: enforcement actions are staff intensive and very costly, there are high administrative costs, and the lack of transparency favours the non-compliance with the rules. Platforms reports that the time needed vary depending on the underlying regulation, information requested, reporting cadence, and adherence to EU law – it can take one week to several months.

How should public authorities get access to data. The graph below shows the perception by respondents as to whether and how public authorities should get access to data on STR to design and enforce policies:

Figure 8. How should public authorities get access to data?

The tables below show the perception of the respondents with regards to the different means public authorities should use to get access to data (and as to whether they should get access to data at all).

a.They should not get access to such data

Total respondents: 5695.

b.They should carry out checks and enforcement actions directly on the ground

Total respondents: 5695.

c. They should obtain it directly from the (registered) hosts

Total respondents: 5695.

d.They should get it from the online booking platforms, upon request

Total respondents: 5695.

e.They should get it from the online booking platforms, periodically

Total respondents: 5695.

Registration obligation for hosts

56.6% (3,224 out of 5695) of total respondents see merit in a registration obligation for hosts as a tool to increase transparency on STRs. Figure 9 explains at which level the registration scheme should be introduced according to the respondents. The majority of the respondents believe that the registrations scheme should be managed at national or at national/regional/local level.

Figure 9. At which level a host registration scheme for STRs should be organised?

Total respondents: 3225.

As shown by the figures below, there is undoubtedly big support for a more harmonised approach to host registration schemes within the EU. Most respondents see merit in having national/local registration schemes aligned with an EU template.

Figure 10. Do you see merit in a more harmonised approach to host registration schemes within the EU?

·I see merit in having one single registration scheme at EU level, but it should replace existing local/national registration schemes

·I see merit in having national/local registration schemes aligned to an EU standard

·I see merit in having an optional EU standard for national/local registration schemes

·I see no merit in having a harmonised approach at EU level

Total respondents: 5695.

Figure 11 shows hosts’ experience with registration schemes.

Figure 11. Hosts’ experience with registration schemes

Total respondents: 4884.

Data sharing by platforms

51.4% (2,927 out of 5695) of respondents are in favour of requiring online platforms to share data on STRs with public authorities to increase transparency and facilitate public policy and enforcement activities. Respondents indicated: they expect positive impact on legal STRs operation (and fight against illegal STRs), less burden on hosts, a higher transparency and better knowledge of the market, better data quality with a facilitated cross-checking and validation of data, better control including on paying taxes, improved quality standards for customers. 85.6% (4,875 out of 5695) of total respondents also believed that such requirements should be put in place/facilitated at EU level.

To facilitate the sharing of data by online platforms with public authorities, respondents are very much in favour of obliging hosts and booking platforms to display registration numbers when advertising a property, but also to develop technical tools for data sharing at EU level, or a more harmonised format and frequency for data sharing.

Figure 12. What would be the better way to facilitate the sharing of data by online booking platforms with authorities?

Obligation for hosts and booking platforms to display the registration number when advertising a property

Development of technical tools at EU level (such as application programming interfaces, ‘APIs’)

Development of a standard format to share data

Development of harmonised rules for sharing such data (type of data, frequency)

Total respondents: 5695.

22.5% (1,281 out of 5695) of respondents also indicated that if data sharing rules were adopted, they consider that some online platforms should benefit from an exemption (e.g. start-ups, platforms with low revenues, etc.) More specifically, respondents identified the following types of platforms: emerging or small platforms, with low revenues or turn-over, local platforms with little scope, platforms that facilitate individual hosting.

3.Market access requirements

Requirements for STR

The graph bellow shows the perception by respondent as to which rules and requirements could be acceptable and if registration schemes, authorization requirements or limits should be imposed. The results show a great divergence of opinion regarding market access requirements, which could be explained by the local dimension of market access requirements (subsidiarity).

Figure 13. What type of rules and requirements for STR could be acceptable?

Total respondents: 5695.

Standardization of rules and requirements

The graph bellow shows the perception of respondents as to whether “peers” (individual hosts) and professional hosts should be subject to the same rules. The results also show that there is no agreement on this question and opinions diverge among respondent groups, which indicates that there is no EU wide support for a standardization of STR rules and requirements, in particular the distinction between peers and professionals.

Figure 14. Should 'peers' be subject to the same rules as professionals?

Total respondents: 5.479.

Requirements on hotels. To start their activity, hotels must get authorisations (e.g. planning permission/building permits). They must meet requirements related to company registration, fiscal registration, compliance with necessary health and safety requirements (e.g. fire protection and food safety), public liability insurance etc. In order to exercise their activity, hotels must elaborate a number of procedures and risk assessments and deal with customer complaints, luggage handling, processing of payments, dealing with emergencies. They must also comply with public order requirements (e.g. registration of guests), privacy regulations, consumer protection law, copyright law, contract law, social and employment law and many other legal requirements.

Annex 2a: Results of the additional targeted survey to Public authorities

The targeted survey to public authorities complemented the results of the public consultation and gathered additional information on their data needs, the purpose for which such data is needed, how and how often public authorities currently collect data and the level of cooperation they have achieved with online platforms. The targeted survey to public authorities was open from 8 April to 2 May 2022 and it was communicated via a targeted email campaign. 80 responses were received from all levels of public authorities (national, regional, province/municipal). The results of the survey are presented below, following the order of the questions.

1.Identification of respondents

Public authorities were asked to identify themselves according to the territorial level they represent. The majority of the public authorities responding to the targeted survey (61.25% - 49) carry out their activity at a province/municipal level. 18.75% (15) of the public authorities have competences at national level and 16.25% (13) at regional level. 3.75% (3) of the respondents declared to be tourism offices.

Figure 1. Number of respondents: 80

2.Data needs for public authorities

Type of data needed on to short-term accommodation rental activities

Public authorities indicated that they need identification data of the host/owner, of the real estate unit and activity data.

Identification data of the host/owner

The table 1 shows the type of identification data of the host/owner that is needed by public authorities at all level. Most public authorities need identification data of the host, and some need the hosts’ tax identification number.

Table 1. Number of respondents: 80

Identification data of the real estate unit

The table 2 shows the data needs of public authorities with regards to the real estate unit (i.e. the accommodation). The address, the type of accommodation and the capacity are among the most selected data types.

Table 2. Number of respondents: 80

Activity data

Table 3 shows the type of activity data needed by public authorities. Public authorities need mostly the number of nights rented out and the number of guests hosted.

Table 3. Number of respondents: 80

Frequency of identification data needs

The frequency at which public authorities need identification data on the host/owner varies greatly, depending on the level of public authorities, as shown by the figure 2 below.

Figure 2. Number of respondents: 76

With regards to identification data of the real estate unit, the frequency at which the data is needed also varied depending on the level of public authority, as shown by figure 3 below.

Figure 3. Number of respondents: 79

Finally, with regards to activity data, public authorities declare to need it mainly on a daily basis, but with important differences depending on the level of public authority.

Figure 4. Number of respondents: 76

For which purposes is data needed

Most of the public authorities require the information about hosts, owners and activity to enforce rules. On the other hand, less than a half of the respondents indicate they need the data to design and justify rules.

Table 4.

Could you specify for the design and justifying of which rules you would like to collect identification data on the hosts/owner?

26% (9) need identification data on hosts/owner to manage tourism flows and design housing market regulation policies. For the same policies, 33% (12) and 30% (11) of authorities indicate they need identification data of real estate units. Finally, 25% (8) would collect activity data for the same purpose. Real estate unit information would be required by 27% (10) of the authorities in order to design better health and safety policies.

Table 5.

Could you specify for the enforcement of which rules you would like to collect identification data on the hosts/owner?

Authorities need data mainly to enforce rules on local housing market. 44% (27) need data on host/owner and real estate unit information for this purpose. 40% (24) of the authorities consider that host/owner information is also important for taxation enforcement rules. Around 40% (23) of authorities also consider important information of real estate units for security and health rules.

Table 6.

3.Current collection of identification data by public authorities

Current collection of data on hosts and real estate units

A considerable percentage of authorities operating at national level collect data from hosts via a registration process (40% - 6), whereas a lower number of them use other methods as contacting platforms or using alternative tools. 85% (11) of the regional authorities collect the data through a mandatory registration process. They also consider the other methods. 23% (3) require the data directly to the platform, use an ex-post framework or use alternative tools. Province and municipal authorities mostly collect the data through the registration process (39% - 19) or via alternative tools (47% - 23).

Table 7. Number of respondents: 80

Type of identification data currently collected

92% (12) of the regional authorities currently collect identification data on hosts, while only 32% (4) are collected identification data on owners. Some province and municipal authorities collect identification data both on hosts and owners (53% - 26 and 58% - 28 respectively). The address of the real estate unit is also considered as crucial information by regional authorities. The type and capacity of the unit rented would be also asked by all public authorities. On the contrary, social security number would be rarely requested by the authorities at any territorial level.

Table 2 Number of respondents: 80

Frequency of collection of identification data

Most of the authorities collect identification data of the host/owner on a short-time basis. 53.8% (7) of the regional authorities collect this data every year. On the other hand, 27.3 % (3) of national authorities collect it every week and 36.4% (4) every year. A considerable percentage of province/municipal authorities (25.6% - 11) collect this information once.

Figure 5 Number of respondents: 70

Similar findings apply to the collection of identification data of the real estate unit.

Figure 6. Number of respondents: 70

4.Current collection of activity data by public authorities

Type of activity data collected and from whom

Number of nights

Table 9 Number of respondents: 80

Number of guests

Table 10. Number of respondents: 80

Reason of the booking

Table 11. Number of respondents: 80

Revenues generated

Table 12. Number of respondents: 80

Frequency of data collected from hosts

The frequency of activity data collection varies greatly among public authorities. National authorities mostly collect data every day (36.4 – 5) or on a weekly basis (18.2% - 3). Similarly, province/municipal authorities collect their data on a daily (35.7% -17) or weekly (16.7 – 8) basis. Moreover, 19% (9) province/municipal authorities collect this data only once. On the other hand, regional authorities usually access to this data on a daily basis (66.7% - 9).

Figure 7. Number of respondents: 70

Frequency of data collected from platforms

Most of the authorities collect activity data from platforms on a daily basis. However, provincial/municipal authorities also collect this data only once (25% - 12), on a weekly basis (13.6 – 7), or every month (13.6 – 7).

Figure 8. Number of respondents: 70

5.Current level of co-operation with platforms

Existence of a legal basis to request data from platforms

A legal basis to request data from platforms seems to currently exist in half of the cases, as shown by figure 9 below.

Figure 9. Number of respondents: 77

Existence of voluntary agreements with online platforms for data sharing

The vast majority of public authorities do not currently have voluntary agreements with online platforms in place.

Figure 10. Number of respondents: 78

Among the small number of voluntary agreements, public authorities declare to have agreements with Airbnb (38.1% - 8), Expedia (19.05% - 4) and Booking (14.29% - 3) in place.

Figure 11. Number of respondents: 9. Number of total agreements: 21

Experience with data requests to platforms

Even though most public authorities declare not to have requests data to platforms ever, it appear that public authorities at all level have already requests data to public authorities, and different types of data (identification data, on hosts, on the real estate unit, and activity data) showing the great diversity of data requests.

Table 14 Number of respondents: 73

If you have already requested data from (at least) a platform, please specify from which one (multiple replies possible):

Among those that have already requested data from booking platforms, a 31.94% (23) of the requests were done to Airbnb, a 18.06% (13) to Booking and a 16.67% (12) to Expedia.

Figure 12. Number of respondents: 27. Number of total requests: 72

Public authorities declare that their data requests are not always answered by platforms, showing the difficulty of public authorities in accessing the data on STRs. In the next plot we can observe the evolution of data request by public authorities (17) over the years compared to the number of requests addressed by platforms.

Figure 133. Number of respondents: 17

Format of data requests (e.g. based on registration numbers)

Authorities’ requests are mostly based on the name, address and contact details of the host. However, a considerable percentage of regional authorities also base their data requests on the local or national registration number (80% - 4). This shows the great diversity in the format of data requests overall.

Table 15 Number of respondents: 27

Requests rejected by online platforms and reasons

A small number of authorities agree in that platforms always collaborate in the data sharing process. On the other hand, among the respondents, a 60% of the national and regional authorities (3) and a 89% (16) of the municipal authorities report that booking platforms refused to share data, mainly due to legal reasons. This again stresses the difficulties faced by public authorities in accessing STR data.