Mr. Patrick Robinson:

Since I am used to a format where I am asked one question at a time, forgive me if I skip anything. I am sure we will cover everything as we go.

I will start with Deputy Ó Broin's questions. The question of what kinds of property are on the site is a complex one. We do not know the status of a property that someone lists on Airbnb. We can see from patterns of usage whether a property looks like it might be someone's home. The data we have provided the committee show that the majority of properties are booked for such small amounts of time that someone is likely to be living in them. To my knowledge, there is no definition in Irish law of when something is someone's primary home. That definition exists in many other places. In France, for example, someone's primary home is where he or she lives for eight months of the year. In the UK, it is based on a series of circumstances to do with where a person pays tax and votes and typical patterns of occupancy.

One of the issues we are facing is that, once people start considering specific definitions, they need to get more precise. For example, what is a short-term rental? In some countries and cities, a short let is anything less than 30 nights. In other places, it is more than that, for example, anything less than 90 nights. We also need to develop a shared understanding of what we mean by "primary home" or "short let". That will enable us to start examining how we as platforms and the whole industry can begin acting on that evidence.

Regarding the question on multi-unit developments, we do not know what the properties are. Clearly, we have an address. There are approximately 3 million places to stay on Airbnb tonight. There is a great deal of information about them that we do not have, but we know what hosts are declaring when they list. For example, they declare whether a property is an igloo, tree house or apartment. Due to the way in which our platform works - I will address the matter of standards and vetting in a moment - if guests are not getting what they expect, feedback is usually bad. Therefore, the host is either removed or suspended because there has been a mismatch between what the guest has been promised and what has been delivered, or the host gets a bad review score and moves down the search results.

There is much that we do not know and the work of the working group could well drive us towards much better information about who is using these platforms, why they are using them - that is another question that arose - and in which properties they are using them.

The question on availability versus bookings is very interesting. In the traditional hospitality sector, occupancy is easy to measure, whereas in our world it is incredibly difficult to measure. One of the issues we have with some of the sites that look at our website, scrape off the data and publish it, for example,insideairbnb.comand airdna.co, is that they look at a host's calendar. One of the features of the Airbnb site is that hosts keep their calendars up to date and indicate when their places may or may not be available. Two things arise from that. First, when a calendar is blocked it does not mean the host has a booking but probably means the host is at home and the property is, therefore, not available for booking. On the other side, just because a calendar indicates that a property may be available, it does not mean the host will accept a booking in the relevant period. This can be a source of some frustration to guests who try to book places that turn out not to be available. We try to ensure that hosts keep their calendars up to date.

Availability does not necessarily tell us anything about the nature of the property that is being shown to us. I am also unaware of any other jurisdiction, certainly in my patch, where the operating framework refers to how often one makes something available as opposed to how often something is used. Everywhere that we have an experience - night limits or intensity thresholds - they are typically about bookings rather than availability. Some of this is related to the fact that it is far easier to measure bookings than it is to measure availability. I caution against disappearing down a rabbit hole around availability. Bookings are a much more appropriate way of looking at intensity of use.

On the issue of Barcelona and New York, I am much more au fait with the position in the former than in the latter. Barcelona faces a challenge which, on the surface, many places would love to have, namely, it is incredibly popular as a place to visit. Since the Barcelona Olympics in 1992, there has been a humongous boom in tourism in the city. The rise of cheaper flights and new accommodation has meant that Barcelona is a fantastic and very popular place to visit. The regional Government of Catalonia and Barcelona city hall are trying to find policies to manage tourism effectively. One of the decisions they took a couple of years ago was to squeeze demand by cracking down on supply. They have drastically limited the number of licences they offer for professional accommodation in the city and stopped building new hotels. As somebody stated, there is still tonnes of demand and this is met by other forms of accommodation. The conversation Airbnb is having with the Catalonian Government and Barcelona city hall is really about how we distinguish, much as the committee is trying to do, between professional operators which should be licensed and home sharers who are contributing to the solution to the Barcelona tourism issue by housing tourists in existing accommodation, driving them away from the central tourist zones and accommodating them in other areas. The disagreement we have is that the ideas on the table tackle the professional end very vigorously but do nothing to protect casual home sharers. That is why we have a disagreement that continues to this day. I am still extremely optimistic that we will resolve that disagreement before long because Barcelona's reputation as an innovative city is beginning to suffer as a result of its dogmatic attitude towards this topic. I totally understand the issue and I was in Barcelona last week for meetings. It is a unique case and there are not many commonalities, particularly given that the authorities in Barcelona are dealing with tourism law rather than housing law. If members wish me to follow up on the position in New York, I can do so but I am afraid I would probably make up much of my reply.