Joint Committee on Housing, Local Government and Heritage debate -
Tuesday, 15 Oct 2024

We are meeting with the Residential Tenancies Board, RTB, and the Central Statistics Office, CSO. I thank the witnesses for their attendance today, for the written submissions they sent us in advance and for the annual report of the RTB, which we have just been presented with. This follows on from a meeting we had some time ago following the discrepancy in figures about rental accommodation in the country. Both organisations agreed to do further research and come back. We are very grateful to them for coming back to help us get a better understanding of how they are assessing tenancies and the number of them we have.

We are joined today from the CSO by Mr. Richard McMahon, assistant director general. He is accompanied by Mr. Kieran Culhane and Mr. Cormac Halpin, senior statisticians. From the Residential Tenancies Board we are joined by Ms Rosemary Steen, director, Ms Louise Loughlin, deputy director, Ms Emer Morrissey, head of investigations and sanctions, and Ms Sinéad Murphy, head of communications. I thank the witnesses for their attendance. The relevant papers have been circulated to members.

I will read a short note on privilege before we begin. I remind members of the constitutional requirement that they must be physically present within the confines of the place where the Parliament has chosen to sit, namely, Leinster House, to participate in public meetings. Witnesses attending in the committee room are protected by absolute privilege in respect of their contributions to today's meeting. This means they have an absolute defence against any defamation action for anything they say at the meeting. Members and witnesses are expected not to abuse the privilege they enjoy and it is my duty as Chairman to make sure they do not do so. Therefore, if their statements are potentially defamatory in relation to an identifiable person or entity, they will be directed to discontinue their remarks and it is imperative they comply with any such direction. Members and witnesses are reminded of the long-standing parliamentary practice to the effect they should not comment on, criticise or make charges against a person or entity outside the Houses or an official either by name or in such a way as to make him or her identifiable.

After opening statements, we will go to members, who will have eight-minute segments to ask questions and receive answers. I invite the witnesses to make their opening statements, beginning with Mr. McMahon on behalf of the CSO and then going to Ms Steen from the RTB.

I thank the Cathaoirleach and the committee for the invitation to speak here today on this matter. I am assistant director general in the social and demographic statistics directorate of the Central Statistics Office. I am joined by Mr. Cormac Halpin, senior statistician in housing, and Mr. Kieran Culhane, senior statistician in national accounts. Mr Culhane and Mr Halpin represented the CSO at this committee’s hearing on 10 October 2023 when it initially considered this matter. At the meeting, an update was provided on the work commenced in July 2023 with the Residential Tenancies Board following the publication of Census of Population 2022 Profile 2 – Housing in Ireland to help identify the difference between the census and RTB datasets on private landlords. At the committee meeting, the CSO also undertook to provide any additional insights possible on the characteristics of the properties or the rental arrangements.

Developing these insights has been a large project involving considerable data matching as well as in-depth analysis of the characteristics of the identified properties using administrative datasets such as the Revenue Commissioners' local property tax data. In line with our mandate under the Statistics Act 1993, access to such information is allowed for statistical purposes only. In addition, the CSO has no role in the requirement to register with the RTB, and the research paper should not be read as a determination on the requirement of a property to register with the RTB.

The results of this research project were published by the CSO on 29 July 2024 under the title Rented from Private Landlords 2022.

It should be noted that, while the research used statistically sound matching and analysis methodologies, factors related to timing and data quality do impact the results.

The first task of the research project was to identify the gap between the two data sets. On census night, 3 April 2022, 330,632 households indicated that they rented their homes from a private landlord. The RTB’s published figure for the number of private rented tenancies registered with it as of 31 December 2022 was 246,453.

This data discrepancy, or difference between the two sources, amounts to 84,179 tenancies. After the exclusion of certain identified properties, such as those found to be owned by a local authority, 73,002 private rentals were identified which appeared on the census but were not present on the RTB register. We refer to this number as the "difference set" requiring further analysis.

To provide as much insight as possible, the CSO undertook an assessment of administrative data sets on objective characteristics of the dwellings, rent payment data and the reference person of the household or owner of the property for these 73,002 private rentals. This statistical model helped us to identify the characteristics of private rented arrangements based on the characteristics of dwellings registered with the RTB in those administrative data sets. Arising from this, two categories were identified, as follows. The first is dwellings with possible informal letting arrangements, that is, those rental properties which were not registered with the RTB and possessed characteristics which may indicate they may have been subject to informal rental arrangements. The second category is dwellings with possible formal rental arrangements, that is, those rental properties which are not currently registered with the RTB but possess certain characteristics which make them likely to be in the private rental market and therefore required to register with the RTB. The analysis indicated the existence of 47,754, or 65.4%, possible informal rental arrangements and 25,248, or 34.6%, possible formal rental arrangements. As we were limited as to the administrative data we could compare with the RTB register, we could not be definitive on the possible informal arrangements that applied at the time of the census. However, by categorising the data in this way, we were able to use census information to provide additional insights around the characteristics of these potential arrangements as identified utilising administrative data sets. Reviewing the census information under the headings geography, dwelling, householder and rent, we found the following: there was a higher prevalence of possible informal dwellings in predominantly rural local authority areas; that possible informal rented dwellings were more likely to be detached houses, whereas possible formal rented dwellings were more likely to be terraced or semi-detached houses; that households living in possible formal rented dwellings were more likely to contain unrelated people than those living in possible informal rented dwellings; and that those living in possible informal rented dwellings were paying 30% less rent than the households in the possible formal rented dwellings.

I wish to note that, in accordance with the Statistics Act 1993, data published by the CSO is published in aggregate form only, which means no person or household can be identified from the data we publish. On this basis, and in line with fundamental principles for official statistics, no identifiable information related to the data sets. analysed in this report can be shared with the committee, the RTB or any other agency. I thank the RTB and the housing sector stakeholders who engaged with us throughout the project.

We are happy to take questions from the committee to inform its consideration of this matter.

I thank the committee for the invitation to attend this afternoon. I am director of the Residential Tenancies Board and I am accompanied by my colleagues Ms Louise Loughlin, deputy director, and Ms Emer Morrissey, head of compliance and enforcement. We will be happy to answer members’ questions but first I will read a short opening statement and respond to the issues raised in the committee’s letter of invitation.

On the RTB’s role and remit, as members are aware, the RTB was established under the Residential Tenancies Act 2004. We are an independent public body operating under the aegis of the Department of Housing, Local Government and Heritage. Our role covers four key areas that include: maintaining a register of tenancies; providing a dispute resolution service, primarily for landlords and tenants; ensuring compliance with rental law; and providing information, research, data and insights to inform policy.

Maintaining an accurate register of tenancies, and ensuring that every tenancy that should be registered is registered, are key priorities for the RTB. We shared this committee’s concern when census 2022 data reported a higher number of households living in private rented accommodation on 3 April 2022 than we observed on the RTB register of private tenancies on 31 December 2022. We wanted to understand the reasons for this difference and we were pleased to engage with the Central Statistics Office on a research study that would provide answers to this question. Due to its access to a wide range of administrative data sources, the CSO was uniquely positioned to conduct this research.

This July, we welcomed the publication of the CSO frontier series research paper, Rented from Private Landlords 2022, which, for the first time, provided greater insights into the difference between the data sets. As the committee is aware, following an extensive data-matching exercise, the CSO identified 73,002 tenancies that were not registered with the RTB. Of these, 25,248 were classified as possible formal rental arrangements that probably should be registered with the RTB. The remaining 47,754 were classified as possible informal rental arrangements that are likely to be outside the scope of RTB regulations.

Regarding compliance with the requirement to register a tenancy, I, as the director of the RTB, together with my colleagues, take any non-compliance with the requirement to register a tenancy very seriously. Under the Residential Tenancies Act 2004 to 2024, the RTB has two mechanisms to address registration non-compliance. The first is investigation under Part 7A, which can result in a civil sanction of up to €15,000. The second is criminal prosecution under section 144, which can result in a criminal prosecution and-or a fine of up to €4,000. As the RTB’s new director, I have outlined my determination to use the RTB’s full powers to investigate and sanction, or to prosecute, non-compliant landlords as required. We are already taking several steps, as my colleagues will discuss later, to increase our compliance and enforcement activities in this area, including targeted compliance campaigns and building our capacity to identify and pursue unregistered landlords.

Under the Statistics Act, the CSO cannot provide the RTB with a list of potentially non-compliant landlords. It has, however, highlighted several geographic areas where a high percentage of unmatched possible informal and formal rental arrangements were identified. These areas, which include counties Donegal, Cavan, Monaghan, Galway, Clare, Louth and Tipperary, will be targeted as part of a new RTB registration compliance campaign between November 2024 and March 2025.

Last week, I wrote to the chief executives of these local authorities to request their support in planning and delivering a public information campaign to encourage landlords to register tenancies in their area. While our primary concern is the 25,248 possible formal rental arrangements that should be registered with the RTB, these compliance campaigns will also target areas where informal rental arrangements are prevalent. It is important to underline that our primary objective is to reach full compliance in everybody’s best interest. Therefore, public information campaigns play an important role in achieving compliance and avoiding any suggestion of a lack of knowledge or awareness.

The RTB is also working to increase our capacity to identify unregistered tenancies through data shared with other Departments. We currently receive data from the Department of Social Protection on landlords in receipt of the housing assistance payment, HAP. We have also worked with the Department of Housing, Local Government and Heritage on a proposed amendment to the Residential Tenancies Act that will allow for additional data sharing with the Revenue Commissioners to identify potential non-compliance.

In August 2024, we compared HAP data and other referral data with our tenancy registration data, and identified 1,500 records that did not match any RTB registration. Last week, we launched a new campaign as part of which we are writing to those identified to give them the opportunity to register their tenancies. Under the Residential Tenancies Act, we are required to write to the landlord on two occasions to offer them the chance to comply with the requirement to register. Landlords who do not respond to the second notice will be subject to prosecution. Our message is quite simple: we will help and facilitate all those who wish to be compliant to do so. For those who remain non-compliant, we will use all of our significant powers to identify such landlords and enforce the regulations.

I conclude by noting my appreciation for the CSO's support in conducting this research, which has provided us with incredibly valuable information. We greatly value our partnership with the CSO, which has generated insights that are helping to drive our new and energised approach, where research informs and helps to deliver a more targeted approach to the RTB's compliance and enforcement activities. I thank the committee for the opportunity to discuss this issue here today. We look forward to addressing any questions that committee members may have.

I thank both organisations for their presentations and the research. We understand how big the piece of research is because two datasets are being compared that are not recording necessarily the same thing, which creates all sorts of challenges.

In the first round, I want to understand how the datasets have been separated. As mentioned here, characteristics have been identified in the form of geography, dwelling, household and rent. Some assumptions have been about those that may be "possible informal" and "possible formal." When I consider this matter I start with the Residential Tenancies Act, which tells us what kinds of rental properties should and should not be registered. Typically, what should not be registered is if I am not paying rent, if it is related to work, if I am renting off a family member, or if it is a licence arrangement where I am living in somebody else's principal residence and renting out a room. In my own head, I am trying to match those categories with what our guests have called "possible informal" arrangements and how they match those criteria. I am not clear how, for example, a semi-detached rural property is more likely to be "informal" than "formal". I am not saying I disagree but I want to understand. I ask Mr. Culhane to take a couple of minutes to explain in more detail, but in plain English, the methodology or criteria used to separate the 47,000 arrangements from the 25,000 because it is not immediately clear from the report itself, which is a really good piece of work,-----

We tried to add any possible additional information we could by linking to administrative datasets. We looked at a lot of administrative datasets. In the end we came up with four datasets, I think, that were most relevant for the private rented sector and for the types of dwellings we were looking at. They included the local property tax, the HAP scheme, rent supplement and a couple of others. Then we linked the properties on the RTB register to those datasets. Much like the Deputy, who started with the legislation, we could only start with the data. The dataset that we worked off was the RTB register, which we matched to those individual administrative datasets. From that, we could get a profile of what a rented property that is on the register looked like when we compared it with these register datasets. We found four real indicators there. I will not go through everything. For example, when we looked at the local property tax, and we looked at the principal private residence indicator, we found that in the case of the majority of properties on the RTB register it was not their principal private residence, which is what would be expected from a rental property. That, on its own, was not an indicator of a private rented property but when we put it together with the other indicators we had, it built a statistical model for us that helped us to identify what we thought should be on the register. Next, we compared our different set - the 73,000 - to those same four administrative datasets and we looked for the same indicators for the different set. Of the ones that fit our model that look like the RTB-registered properties, we said there were 25,000 that we thought were "formal" rental arrangements, and we called the others "informal" arrangements. It might have been better to call them "not formal" in that it was more by omission that they did not look like the properties on the RTB register. That was how we identified those. Then to get the extra little bit about semi-detached-----

There are datasets that are available to us that have information about private rented properties. If we cannot find them on these datasets, there is not much more we can say about them. That is why we went back and looked at the census again. We were able to categorise our two sets of records in the census file. That showed us some differences; for example, the ones that were in one set were more likely to be rural and to involve detached houses than the ones that were in the "formal" set.

One of the issues we were all thinking about when the different set, as we are now calling it, first emerged was the treatment of licensees. I ask this question for my own clarity: if I own a house and Deputy O'Callaghan rents a room in the house as a licensee, how is that recorded in the census? If I am the homeowner and head of the household, obviously I fill out the census and licensees do not fill out a separate census form because the form is for the property. Is there any visibility of licensees in the census data itself?

The census is conducted on a household basis rather than on a dwelling basis. There is provision, if there are two distinguishable households within a dwelling, for each of them to complete a separate census form. We have defined what is meant by a household. Basically, if separate living arrangements pertain to two separate households within the same dwelling, they can complete separate census forms and thereby complete that first set of questions in which the nature of occupancy is indicated. One person could say that they are an owner-occupier and another person might have a different status within the dwelling.

I guess it depends on whether they consider themselves to be a separate household. When the enumerator was delivering the census form in 2022, he or she would have explained the household concept. If people have separate living arrangements and do not share a meal - in effect, they are living separately within the same household - they have an entitlement. That is subjective and at the discretion of the individual households.

Is there any way for the CSO, perhaps as part of this exercise, to attempt to match or look at where two separate census forms were filled in for a single property to try to establish things? I appreciate that all of this is tricky. I do not mean it as a criticism in any way when I say that 47,000 "possible informal" arrangements seems potentially too large. It is worthy of interrogating more because it is a lot.

If we think of the types of people who do not need to be registered, whether it is licensees, people who rent a room, family members renting from family or people not paying rent, that is quite a large number. Let us assume, for example, that not everybody who is renting a room will have filled in the census form. They may not have been there on the night or whatever. Was there any way of looking at double census forms for a single property and whether the CSO was able to extrapolate the possible number of licensees from that? Let us say I had my census form at my principal private residence and Deputy O'Callaghan had his own census form at the same address, but also ticked the box that he was renting from a private landlord. Is the CSO able to make that kind of analysis of the data?

I do not think it could be done directly. We could identify individual dwellings where more than one census form was completed and analyse the dwelling status that was selected for each of the different census forms responded to as part of it. Therefore, one could be owner-occupier and one could be private landlord, for example. There was not a category for licensee in 2022, so we would not get that precise piece of information. It was not done as part of this exercise. It would be not yet have been possible-----

I am taking the next slot. That is an interesting line of questioning that I wish to pursue. We all know from canvassing houses and knocking on doors that we often do not get an answer. I presume the CSO field workers also have the same experience. The scenario Deputy Ó Broin painted is that somebody renting a room within a property is a slightly different scenario from a lot of housing where the granny flat out the back is accessed by the side gate. How much, if any, of that was captured? People can knock on the front door. There is no requirement to check if there is any other form of dwelling type attached to the back or anything like that. The field officers would not be tasked with doing that, would they? There is nowhere on the main household section of the form to capture whether there is another dwelling out the back. Is that part of the question?

We did put extensive efforts into capturing those types of situations. We used a lot of mapping software and so on to try to identify outdwellings and dwellings that were separate to the main building, and that information would have been provided to numerators. There is obviously a legal obligation on all people who are present in the country on census night to comply, and a very big publicity operation accompanies the census. That example the Chair gave about the granny flat is probably a more likely situation whereby separate households are going to be captured because there might be more separate living arrangements in that situation if people are living in a separate building. The scenario I had in my head is one we see sometimes where a couple might, unfortunately, have separated and say they are living in the same dwelling, but they are living separate lives, if you like, and they can be separate households. That can be a little bit more blurred, but where there are completely separate dwellings then, given the intelligence we provide to our field staff in advance, we might expect to pick up those kind of situations to make sure everybody is covered.

It is a really difficult task to try to identify all of this because the CSO is dependent on somebody voluntarily filling out the CSO form. Many people comply with that and are quite happy to do so. That is one aspect to collecting the data. The other aspect then is to register with the RTB, which is obviously a legal requirement for the landlord. We know some landlords do not do that, however. I do not envy the task the CSO has in this regard. It will never capture in its entirety every single dwelling arrangement in the country. Is there best practice? Does the CSO look at how other countries do registration where they have an RTB equivalent or data collection equivalent to the CSO? Do they say that if you got it down to in or around 10%, you would be doing well? Is there any international comparison on that?

From a census perspective, possibly not, because the way Ireland has traditionally conducted the census is quite different from most European countries and globally. The traditional census is kind of a dying art whereby we go to households and deliver and collect a form. Most continental European countries, for example, would have population registers and housing registers that get linked together to provide information. From our perspective, a huge amount of effort is put into maximising the coverage in some of the areas I spoke about previously in terms of the legal obligation and the publicity. Typically, the response rates we have in Ireland compare very well with other English-speaking countries, which are typically the ones that do the traditional-type census. We have always had great buy-in from the Irish public. It would probably be folly for me to say that absolutely 100% of people respond to the census, but our rates compare very well internationally.

As regards picking up complex living arrangements, the census is a blunt instrument. It has to go into every single household in the country, by definition, with lots of different levels of comprehension literacy and language ability. The questions have to be quite straightforward to pick up the multiplicity of different types of living arrangement. We try as much as we can to reflect the way society is evolving, but it is not always possible to get the degree of nuance that some stakeholders would like.

That is pretty good all right. This is a question for the RTB. If it receives a complaint from a tenant and, in investigating that complaint, realises it is not registered tenancy, what steps or measures are taken? Does it follow that process of two contacts to the landlord that Ms Steen outlined in the opening statement?

We have a couple of processes for people to report non-registration to us. It can come in from a member of public or we do data sharing with other public bodies, such as HAP Shared Services Centre, the Department of Social Protection and local authorities. We also have internal data from our disputes process. If someone is part of a dispute or if a tenant raises a dispute and that property is not registered, we can get that information and pursue it.

The RTB has two mechanisms for pursuing non-registration. As Ms Steen outlined in her statement, we have our criminal process, which is registration enforcement, and then we have our investigations process under Part 7A. Both are slightly different processes. The registration enforcement process is about notifying the landlord that we know he or she is not registered. We give the landlord two opportunities to engage and actually register that property. If the landlord does not comply, he or she can be prosecuted through a criminal prosecution with a fine of up to €4,000.

Our investigations process involves a formal investigation, so it very much looks at the evidence that comes from the person who sent that complaint to us. It investigates what the set-up is and whether it falls under our jurisdiction and if it is not registered. If we get a complaint about something else, for instance, breach of an RPZ, we will always check the registration status. Therefore, even if somebody complains about something that is not to do with non-registration, we will have a look at and address it and add it on to the investigation if we can. We follow that process, and the sanction can be up to €15,000 for non-registration if someone has been found to have breached it. We also give landlords the opportunity to comply throughout that process. They can acknowledge and rectify it and register. They can still be liable for a sanction if they do that as well. It is just that is looked on more favourably depending on a couple of different aspects of that, but there is plenty of opportunity.

I have the statistics for 2024 if the committee is interested. We had 498 referrals from members of the public. We had 477 referrals from the Department of Social Protection with regard to rent supplement allowance. We had 6,097 housing assistance payment referrals. We had 270 referred in from local authorities. From internal information, we have 17 referrals mainly coming from disputes. We also had a particular case recently with a landlord that will be published shortly, which Ms Morrissey might briefly refer to for the committee.

We have a case that involves a landlord who was sanctioned for multiple breaches of non-registration. The landlord was sanctioned more than €11,000 for that. It recently went through court confirmation and will be published soon on our website. That was quite a large case, and the members of the investigation team spent a lot of time looking into that. The tenants really engaged with us to help get that done as well. It was a very successful piece of work by my team.

The details of that will be published shortly on our website. We are hoping it will be later this week.

As director, I would say that we take all of these complaints very seriously in the work that is done and we follow up on them very actively. As part of the campaign that we intend to run, I have written to the chief executives of the counties identified in the CSO work to stress the cost to landlords of being fined for potential non-compliance with registration. Those communications campaigns are running through the autumn to clarify that if someone is in a formal arrangement, it needs to be registered. We hope to provide information to narrow the gap further.

It is important that we have material that is as accurate as possible for a whole range of reasons concerning compliance. As director, I am looking at other potential aspects of fines under the new legislation that could assist us in our work in this area, and we might come back to that later.

I am thinking of a particular scenario. Tenants have come to me in the past about issues with the property they are renting and they would not know the address of the landlord and might only have a phone number for the landlord. It sounds like a very informal type of arrangement. What is available to the RTB to track the owner of the property on the assumption that the owner is likely to be the landlord? If the RTB writes to the landlord at the address, it is possible that no contact is made.

Part of this has to do with the relationships we are developing with other Government bodies in order to learn from best practice, in particular with the Revenue Commissioners. Revenue has a lot of data that is of value to us in this matter, particularly where it also involves companies, which is another form that can be used to avoid registration. Ms Morrissey will discuss what is going to happen in her division.

I will deal with the question of what happens if an individual tenant contacts us. At the moment, we have a certain amount of data sharing with other bodies so we are able to search for a landlord address, and we can also look at our own data to see if the person had previously been party to a dispute or an enforcement order. There is a certain amount we can look at. The investigations team have powers under the Act and can look further and request information from agents, tenants or the landlords themselves. Obviously, there are some cases where there is a difficulty in finding people but we have powers to do that, including personal service in individual cases.

As Ms Steen said, we have some data sharing. Since January, my team have been reviewing our current referrals process and looking at our data sharing agreements and where we can expand on that. We have been working with HAP shared services and Revenue around that, and Ms Steen will speak about that data sharing agreement. We are currently working with Revenue on a joint project to try to specifically address non-registration. While there are some limitations in the data sharing, those changes will help us.

Yes, under the legislation. One of the things we have been discussing is whether a potential amendment is needed to support us with regard to the inspection of a property. That is something we are currently considering as part of putting forward amendments to the legislation. The current legislation only permits us to inspect a property if there is an application in for registration or after an investigation has been commenced. It would be potentially important for the committee to consider this matter. I certainly feel it is important that where there is a very serious matter that needs to be investigated, we have the powers to inspect a property. One of the things we have been working to understand, in the work that Ms Morrissey has been doing with colleagues in Revenue, is how they use those powers and how that operates in practice. That is the next stage in terms of enforcement. We are very clear that it is important for the RTB to have powers to investigate in these very serious situations.

I thank the witnesses for their work in this area. I am trying to understand the 47,754 in the unknown category, as the witnesses put it. The opening statement suggests that those living in possible informal rented dwellings or this unknown category were paying 30% less rent than households in formal rented dwellings. Is that just 30% less in overall terms, or is it comparing like with like and comparing similar types of properties and areas? Is it geographical? Is it just 30% in general?

I am asking because the statement also points out that what the CSO considers to be unknown or informal dwellings are also in rural areas. If it was comparing like with like and was 30% less, that would suggest a number of these may well be informal rental arrangements or discounted rents between family members. However, if it is overall, that does not suggest any family discounts in that area because one would expect rents to be significantly less in rural areas. That is why I was asking. We cannot conclude anything from that but it certainly does not point towards discounted rents.

The statement suggested that informal rented dwellings are more likely to be detached houses. Our general concept of a licensee in the rented sector, for the most part, is someone in the rent-a-room situation and, obviously, there are licensees offering student accommodation, digs and so on. We have nothing to suggest there are a substantial number of licensees in detached dwellings.

It says they are detached houses in rural areas. I am trying to understand why we think they are informal settings. There is nothing in what Mr. Culhane is saying to suggest they are licensees and nothing from the rental database to suggest they are at a particular discount that might indicate family arrangements. Of course, not all family arrangements are going to get a discount but some are going to get a discount, one would think.

As I explained earlier, the definition of “formal” that we use is purely a statistical exercise and is not related to the RTB legislation in any way. It is just using the RTB register and the properties on the RTB register as a basis. We are assuming they are all formal arrangements because they are fully registered. If a property has a similar profile to them, it is formal. If properties do not have a similar profile to them, they are the ones we categorise as informal. We are using the RTB register as our basis of comparison.

Therefore, a possible explanation for those in the informal or unknown category is that they are informal letting arrangements, but the information that Mr. Culhane has given us suggests ruling out licensees, for the most part, because licensees are generally in rent-a-room situations. That is generally ruled out by the description that Mr. Culhane has given me. This is just my summation. If Mr. Culhane feels I am being particularly inaccurate, he can let me know.

Rather than these being informal letting arrangements, it could be just as likely that there is a pattern or description that matches the lack of registration. This could simply be the characteristics of rental properties that are not registered as often. That could be the case. I am just trying to tease it out. It could be either but we do not know.

That is what I am trying to understand.

Just because they are categorised by informant does not mean they are not required by law to register with the RTB. Mr. Culhane is saying we just do not know. It could be that they-----

That is helpful. When this information came out initially, some people were making assumptions that these 47,000 property owners were perhaps required to register. The case is we do not know. It could be that a good number of them are required to register and a good number are not. I thank Mr. Culhane. It is useful to have that clarification.

Will the RTB witnesses provide more information on these 47,000 unknown rental arrangements which we do not know how many are or are not required by law to be registered? How will the RTB go about finding that out, addressing it and ensuring that, within that figure of 47,000, the ones who are required by law to register do so? The witnesses talked about their general approach to it, but will they take a particular approach to those 47,000?

As Ms Steen said, it is important to think about the responsibility of the RTB as regards tenancies. We are responsible for maintaining a register of private residential tenancies, tenancies of approved housing bodies and student-specific accommodation tenancies. When we talk about what we are responsible for, it is just as important to consider what we are not responsible for: tenancies in local authority housing or under shared ownership lease arrangements; holiday letting arrangements; the rent-a-room scheme, particularly where the landlord and tenant share the same self-contained property; one that can often be missed, where a tenant lives with a spouse, civil partner, parent or child of the landlord and there is no written letting agreement in place; short-term lets and lets on licences - typically, most licence arrangements do not come under the remit of the RTB, and while the naming of a licence as such does not necessarily mean it is not regarded as operating as a tenancy for the purposes of the RTB, a licence is usually said to exist where a person is staying in a hotel, guesthouse or hostel; sharing with the owner in rent-a-room or digs accommodation; or staying in rented accommodation at the invitation of the tenant. That is what is excluded from our remit.

I thank Ms Loughlin for her answer. However, we have established that, out of the 47,000 we are discussing, it is not likely that many of those will be licensees, who are generally in the rent-a-room category. The properties in question are predominantly detached dwellings in rural areas. We have also established that there is a good likelihood that a significant portion of the 47,000 - certainly not all of them but a significant number - would be required by law to be registered with the RTB. Ms Loughlin has given me the list of people who are not required by law to register with the RTB. I am asking about the number within the 47,000 who are required by law to register with the RTB and what will be done to get registration up among whatever that number is within the 47,000. I accept that we do not know what the number is, but is there going to be a specific approach regarding the subset of the 47,000 that are required by law to register?

In the director's opening statement, she described the campaign we have commenced with regard to contacting the local authorities. The RTB was not dependent on the CSO data in considering our compliance activities. As my colleague mentioned, we have been planning since the start of this year a range of compliance activities, especially in regard to non-registration. We have referred to the 1,500 dwelling owners we are contacting on foot of information from shared services with HAP as well as a range of other services. We have commenced the work of contacting those landlords. In addition, we are starting with the county campaigns we referred to, which relate to the counties that appear from the CSO data to have the highest level of informal arrangements. There are two campaigns running in parallel in regard to non-registration.

I have one final question on the methodology just so that I am clear. Is it the case that the CSO looked at the RTB registration data - the total registration - and at the characteristics of properties that were registered, such as whether they were more likely to be urban than rural and more likely to be smaller or terraced than semi-detached, bearing in mind that the latter, obviously, would have higher rents? Did the CSO look at the overall characteristics of properties that were registered and that assessment allowed it to divide the different sets? Have I got that right in terms of the methodology?

No, it was not those overall characteristics. It was specifically the characteristics of the RTB dataset when we linked the data on the dwellings to those other administrative datasets. We considered, for example, whether the principal private residence was marked as "No" in the LPT dataset. Had it ever been in the HAP scheme? Had the person living there ever been in the HAP scheme? Had somebody claimed rental tax credit on the property? It was those indicators that suggested the property, either because of the person living in it or the property itself, had previously linked with another administrative dataset as a private rented property.

One of the useful things that could come out of this entire exercise, because it is important, not so much in the sense that the CSO had to do all of this hard work and then come in here to answer our questions but rather that something arises from it, is that it shows there is an argument to take a look at the question in future censuses on private rental arrangements so that, rather than just asking whether people are renting in the private rental sector, they would be given some options to say, for example, whether they are renting a whole property, sharing with other renters or renting a room in the owner's property. The reason this is important from a housing policy point of view is that, for us to understand what is going on in the private rental sector, we need the best data we have. We have had big policy debates and there has been Government policy and legislative decisions on foot of an understanding of quite a significant shrinkage of the private rental sector, looking at RTB registrations, even though we know those registrations are not the definitive data. If we were able to find a way of getting more granular census data in future years, it could be really interesting. In a sense, this is the best effort that can be made with the datasets the CSO has. Our questions are not critical but what they show is that it is very difficult for the CSO, with the information it has, to definitively read these things. It is useful to know that because it means we can move on it. It would be very interesting to look at how the private rental sector question is asked.

With respect to the RTB, I really welcome the indication in the director's opening statement that there will be an equal focus on informal tenancies. While there is clearly a chunk of properties that are possibly informal that would not have to register with the RTB, it is reasonable to say there are some in that category that may have to. We do not know the size of that and that is crucial. In the RTB's annual report, on pages 79 and 80, there is a list of referrals for potential non-compliance, in table 3, and, separately, in table 4, information on the number of issued registration enforcement notice letters. I am interested to know something, if not today then at any time, because the data may not be collated. If we look at the top row of the table on enforcement actions for registration, we see there were approximately 4,500 and 6,000 in 2018 and 2019, respectively. The numbers go down in 2020 and 2021, during the Covid period, but start to move again in 2022.

Obviously, there are the first and second notices, a warning letter and a solicitor's letter. I would be really interested to know how many of the people in the original category who got the original letter ended up being registered. We could look at that table and say it is a lot of activity, but the real test is what number of those then registered. If that data was available, that would be really interesting. If not, it might be worthwhile putting it into the reporting because it would then show if the RTB was doing a good things and that enforcement is leading to greater registration. That is a positive. I would be interested to hear the witnesses' thoughts on that.

We have to confirm that we have it, in terms of whether we were collecting it at the time. The Deputy's point that we should possibly collect that going forward is well made, however. We just need to confirm if it was collected at the time from that bunch of cases referred to. Certainly, however, that is something we do need to consider going forward.

We are also considering the issue of how we potentially deal with people who have not registered. If somebody who is in a dispute comes to us and registers during that process, we have the ability to potentially increase the registration fee at that point. We have not done so yet, but it is one of the measures we are considering with regard to ensuring there is an incentive for landlords to register before they need us and that this is clearly done at an early stage in the relationship with the RTB. I will definitely take away that query and we will look into it. We might write back to the Deputy separately about those-----

To labour the point - I apologise; I am a politician so that is what I do for a living - if we take the 2022 data, for example, there were 3,000 first notices. Obviously, the figure for second notices was 1,600. It would be great to know whether that means that almost half of the people who got the first notice registered and, therefore, a second notice was not needed. Likewise, for example, only 175 warning letters followed on from the 1,600 second notices. Again, if that means the difference all registered, that is something I would recommend the RTB talk about because it is good. Clearly, however, if that is not the case, we have a problem. I am not particularly interested in it historically, but from this year on it would have to be possible because the RTB keeps a record of whom it writes to.

I am not at all putting words in anyone's mouth because, again, I think this is a really brilliant exercise. Is it fair to say that while the RTB is categorising arrangements as possible informal and possible formal because of characteristics that may indicate such, it is not saying that one group requires registration and another group does not require registration, because it does not have visibility on that, but that on the balance of probability, the 25,000 more probably will require registration, whereas the 47,000 are less likely to require registration, or is it not commenting on that? I am just interested to know what the witnesses think that says. Again, I am not trying to pin anyone down.

Well, they look like the ones that are registered on the register, so from that we think they should be as well because they are interacting with the rest of the State like a registered property. The other ones are not. Really, what we are saying is that they do not look like the ones that are registered.

That is a reasonable proposition. That is all the more reason for the commitment that the RTB is not going to exclude those. The only reason I am saying that is because, prior to Ms Steen taking office, I had a conversation with somebody else in the RTB who said that 25,000 is actually only 10% of total rental properties, so that is good. There was a working assumption that the 47,000 could be bracketed off. Although they are much looser in there and there are significant numbers that would not require registration, they are still part of the picture.

I have a couple of questions to follow up on as well. To follow the line of questioning from Deputy Ó Broin, the RTB might engage with a landlord where a report is made that there should be a registered tenancy, and if the landlord does not respond, it sends a second notice. If we look at the figures, the option available to the landlord is to register and become compliant with the law, which is the outcome we would all be looking for. Is there also the option then for landlords to say they are ending the arrangement and that they are no longer a landlord because there is no longer a tenancy, and they are putting the house up for sale? Does the RTB collect that data? In order to rectify or finish off an investigation, is that something a landlord can tell the RTB, which it can then accept and say that is no longer a tenancy?

That is one of the options for the landlords. When the registration enforcements end, when we send out that initial letter asking the person to register because we have information that there is a tenancy, that person either has to register or let us know if he or she is not registered. We do not currently have data on historic cases in which that would have been done, but we will be collecting that going forward. When it goes to investigation, there is an assigned investigator who engages with the landlord throughout the whole process. If a landlord comes forward and says it is no longer a rental property, or that there was a rental property or it is a licence agreement because the tenant is living with him or her, the investigation is closed off because there is actually no improper conduct to investigate. We do have that option. The aim for us is to support landlords who should be registered to become registered and to help them along that process. For those people who are deliberately trying to avoid registration, it is really then we use our full powers, which include enforcement or that sanction process to make sure they feel that the RTB is using its powers.

As we mentioned, Ms Morrissey and the team have embarked on this new campaign. We actually have two campaigns running at the moment. One is related to RPZs, which is a matter we might discuss with the committee on another day, and the other is compliance in terms of registrations. That does point very much to communications and making sure we have a much more active voice in making it clear that people must be registered. We are hoping that taking that regionally into the counties the CSO has very helpfully identified will drive registrations. That will give us very important data. In terms of the resourcing, we might need to follow up on a county-by-county basis. It will also perhaps inform further work the CSO may want to look at in terms of the next census. We are happy to share the results of that with both the committee and the CSO. There are obviously some unique arrangements in place, which may be to do with family members in rural Ireland, that we need to understand as part of the housing demographic. It is important also to note that from the figures we saw from the CSO, Dublin was largely compliant in that sector. That is important to be aware of as well.

I referenced knocking on doors for years, and we come across all sorts of different living arrangements and various set-ups. I have often gone to what looked like one house at point and found there were two doors and two letterboxes. We do not know whether that is just a family member and they have segregated that off. I do not underestimate how difficult it is.

That is an important aspect. Irish communities are very supportive of family relationships, particularly in rural Ireland. Co-locating on sites is something that would have happened way back in terms of the way families operate. We need to be respectful of those arrangements because they are not tenancies under the Act currently. That may be a valid part of the story around the Irish housing sector. We have to respect that this is the way those arrangements operate and that is the way it is.

I appreciate that the RTB has had a considerable workload over the past couple of years. I know it is growing the organisation and the IT systems it has, etc. I would be interested in the results in terms of the number of landlords who choose to leave - I cannot call it the market because they are not really in the market; they should have been registered in the first place to be considered part of the market. I am interested to know how many of them will consider leaving because the RTB is shining a light on them.

We do have the overall figures regarding landlords leaving the market. As we reported previously, there were 230,006 landlords registered with the RTB in quarter 1 of 2024. This was an increase of 7.9% on the 213,777 registered in quarter 2 of 2023, when the data set that the CSO examined began. Therefore, we are not seeing a high level of exits at a macro level as a result of the investigation work under way, but I would like to take the time with colleagues to look into this in more detail if we have the data.

I am interested in knowing how many landlords who were registered state, on investigation, that they have exited.

When a rental property is vacant for two years, it can be reregistered then. Where somebody was evicted from a property, for whatever reason, and the property comes back into the system again, is it flagged with the RTB?

We just began the data collection in 2022. At present, notice has to be served to the RTB, as I understand it, so we will be aware notice was served to the tenant. The question is whether we will be able to look back and see whether a property from which a tenant was evicted was re-rented immediately at the end of the two-year period or whether the landlord exited and sold the property. That is the issue that arises. I do not know whether Ms Loughlin wants to add anything to my answer.

This is connected to the work we did with the ESRI. I do not have all the material with me today but the matter is connected with the property-level analysis and the profile of the register. The property-level analysis will be published in full later this year by the ESRI. We published some of the information with the ESRI in September and it is on our website. The analysis compared like-for-like properties. It is only by virtue of our having annual registration in place since 2022 – this is only the second full year of having the data – that it was possible to compare matched sets of data. On the Cathaoirleach’s question on deregistered properties that are reregistered, we will be able to obtain data over time. We have started the work with the ESRI. When the analysis is published in full later in the year, we will be able to see some of the data in much more detail.

It is like any management system in that one must seek to improve it continuously. I take Ms Loughlin’s point on it.

My final question is for the CSO representatives. With regard to the question to be answered on the CSO form, do they envisage any way of changing the question, adding more questions or seeking to have an explainer for people that might give us better clarity at the time of the next census? It is some time away yet, but-----

Yes. The next census will be in 2027. We have recently gone through a consultation process that we go through before every census, whereby we invite submissions from members of the public and interested stakeholders on what they would like to see on the next census form. I refer to new questions, revisions to existing questions, and questions to be dropped. As part of this, the questions related to housing would have been considered by a group called the census advisory group, which comprises key stakeholders and key data users, primarily from across government but also from other sectors that use data. The group considered the question we are talking about, on the nature of occupancy, and recommended that the current version of it be retained in 2027. Therefore, it will not change for the next census.

The use of administrative data as part of the census is increasing. It was introduced in 2022 to improve the quality of the census, reduce the rate of non-response where it happens and validate the quality of census data. It may be possible to use administrative data in 2027 to explore the possibility of producing more of the type of information in question. We are still two and a half years out from the next census, so I do not want to pre-empt what will happen in that regard; however, we can say with some certainty that administrative data are here to stay. That is good from a data-quality perspective because it brings richness and surety to the census data. We have taken note of what the committee has reflected on regarding the degree of nuance. We will see what is possible as regards the use of administrative data but the census is two and a half years away and a lot of work has to happen between now and then. We are very conscious of the data required not just by the committee but also by other data stakeholders in this space.

It would not be technically correct to say it has been taken, but it effectively has because a census pilot recently took place and that is the vehicle we use to test new questions for potential inclusion. The question tested was the same one. The key reason for the retention of the existing question concerned the time series. The question has been used in several censuses. When an aspect of a question is changed, even by one word, it compromises comparability over time. That consideration was important to the census advisory group. A housing group that was convened examined this, in particular. This is not to preclude the possibility of what has been suggested being done. The direction of travel is such that more and more administrative data will be used as part of the census. It is certainly something we are conscious of, and we will explore it as part of the preparations for 2027.

No member of this committee should be interfering with the methodology for the census. The CSO staff are the professionals and we do not deal with the matter. However, the CSO’s exercise has thrown up some really important questions regarding our understanding of the private rental sector. If there were some way of doing what I suggest – it could be through administrative data rather than changing the census question – it could be beneficial. It would be a terrible shame if all the hard work and the current report did not benefit the 2027 census. However, the CSO has its processes and I will not interfere.

A fine line has to be treaded in that if a form is made too complicated, people will fill it out incorrectly. If I understand Mr. Halpin correctly, changing a question too much affects the ability to compare trends in all the data collected over the years.

My next question is for the representatives of the RTB and was alluded to by Deputy Ó Broin earlier. Let me refer to where there is no requirement to register with the RTB, as with a student, licensee or family member. Are there rental arrangements aside from those listed in the Act that probably should have been included in it to tighten the net a little? I am very conscious that, in the case of many rentals that are not registered, the landlord and tenant might have a perfectly harmonious relationship, maybe for the longer term, notwithstanding the fact that neither party benefits from the protection of registration. I do not foresee circumstances in which every single rental will be registered. What rental arrangements has the RTB come across that are not listed in the Act as requiring to be registered? While there will always be outliers, are any of the rental arrangements in question prevalent?

We see potential issues when people come in search of the mediation services. It is normally when a tenant or somebody in one of the relationships mentioned seeks support. I would like to take the question away with our team and determine the number of people who apply for mediation who are not registered and whose support we cannot move forward as a consequence.

I would like to take that back and have a look at our statistics to see if we capture that currently. That might be the most accurate way of understanding how widespread an issue it is. Obviously, this is an issue for us on which we must stick to our knitting to some degree in terms of ensuring the compliance we are mandated to do under the legislation is done to the highest possible standard. As regards the point Deputy Ó Broin and the Cathaoirleach both made about the fact there is a wider community potentially in need of support and advice around issues in these different types of living arrangements, I will take that point back to see how we might address it. Certainly, through the communications campaign, we will be looking to ensure people understand what formal arrangements in the context of the RTB legislation mean and how they can get in contact with us. There was some feedback that this needed to be slightly clearer on our website. If people want some form of advice, there is then the question of whether we could give greater helpline-type assistance. That is something for the broader sector that needs to be considered. At the moment, there is nothing jumping out at me as the new director coming and looking at the statistics. It is more about this issue of the different types of living arrangements regionally. That is the new outcome or learning we saw from the study. I hope that is helpful.

It is helpful. Ms Steen spoke earlier about a potential amendment around the inspection of properties. Obviously, it is within our remit to amend legislation like that. I advise Ms Steen that if she sees something in the Act, or on the last matter we spoke about, although I understand she has to take time to-----

It is a question we very much wish to understand. In situations that have been brought to our attention, either through a good citizen report or through the various agencies with which we are engaging in terms of data sharing, it is about having a mechanism to knock on a door. We have been discussing this quite a bit in the office. I am keen to potentially undertake a form of pilot study in Dublin to understand how we can visit to confirm data. It is a complex area and we are not currently resourced to do that. We would need support from various stakeholders to bring that forward. As part of these investigations, situations arise in which we would really like to understand more. It may be the case that everything is fine, but, occasionally, we want to have the opportunity to investigate. Ms Emer Morrissey might reference this more. This was at the core of the conversation we had around the partnership or arrangements we were putting in place with the Revenue Commissioners.

At the moment, our power is kind of restrictive. We can only inspect a property when an investigation is under way, which means we have to be aware of a situation. Someone has to report it to us. We would like to be able, for example, to use CSO data to determine a geographical area in which we suspect, maybe through good citizen reports, as Ms Steen said, there are some non-registered or non-compliant landlords and then be able to investigate and decide whether it is something that falls under our remit and can be pursued further through investigations, enforcement and partnerships with other organisations such as Revenue and the pilot we are hoping to do with it. We will be able to join up and do that piece of work as a pilot study in certain areas to see how it would work. Obviously, having stronger legislative powers around that would really help.

We welcome the committee's support for the residential tenancies Bill, which includes this data sharing aspect with Revenue. It is incredibly important to us as an organisation that this aspect progresses as quickly as possible and that the matter of giving us the power to inspect be considered as part of the amendments which might be brought forward. We want to ensure we do that through engagement with all of the relevant policy and stakeholder bodies. Communications campaigns can bring us to a certain point. If the expectation is for us to act in terms of full compliance, however, we need the tools to bring that about. I am very determined to do all that I can with the powers I have. The committee has heard me speak to that earlier. It is my position, as director, that everyone who should be registered is registered. We welcome the issue of giving some additional powers in the context of being able to manage potential queries or outstanding reports. There are exceptional situations, in our view, that require an inspection.

Obviously, we all hope to be back here in the future to assist the RTB with that work. That is another day's story, however. I thank the witnesses from the RTB for their attendance. The RTB has appeared willingly before the committee many times and assisted the committee in its work, as has the CSO as well. The work it does is incredible for everyone. I remember as a student spending a lot of time on the CSO website going down a rabbit hole finding out a lot of stuff I had not intended to find in the first place. I thank it for the work it has done. I will now adjourn the committee until Tuesday, 22 October at 11 a.m.