The curious case of the missing 54,000 tenants

The Week in Housing 08/09/23

The Week in Housing is back from a summer hiatus. However, this trimester I have much more teaching than usual and so I am not sure I’ll be able to do a weekly Newsletter. For the next while at least, I’ll try have one issue a fortnight. Once I am a month or so into the trimester I’ll have a better idea of how busy I’ll be with teaching. Thanks for bearing with me. On a positive note, the reason I have more teaching this trimester is because I’ll be on sabbatical from January, so I’ll certainly have more time for writing then.

Lots happened over the course of the summer, but probably the most notable new development was the release of some of the CSO data on housing. One of the most notable aspects of that data relates to the number of PRS tenancies in Ireland. The CSO tells us that there are 330,000 households in the PRS. Meanwhile, the RTB registration data records 276,000 tenancies at the end of 2021. So there appears to be a missing 54,000 PRS tenancies.

This is significant for a number of reasons. Firstly, given that the growth of the PRS has been one of the two most significant developments in the Irish housing system since the GFC (the other being the collapse of supply), it would be pretty handy if we actually knew how big the sector is.

Secondly, the size of the PRS, and whether or not it is growing or contracting, is politically charged. The RTB data has been showing a steady decline since 2016 (see table 5 below). This has coincided with data produced by industry bodies (such as real estate firms) and anecdotal evidence that landlords are ‘fleeing the market’. This has become one of the most dominant narratives in the debate on the PRS (I took part in an Oireachtas Committee meeting on this topic last year, see my full statement here), especially evident at the moment in relation to the idea to introduce tax breaks for small-scale landlords (there was a good debate on this on Drivetime recently, with Stephen Kinsella very critical of the idea of tax breaks)

But the CSO shows no such decline. Indeed, it suggests that the number of dwellings rented from a private landlord increased by 7%. This includes increases in all four Dublin Local Authority areas. However, some of the biggest increases where outside the larger cities, such as a 15% increase in Waterford and Carlow. Note that, according to the Census data, the proportion of households renting has not increased, rather the number has increased in line with population growth.

There’s been a bit of a debate over what is going on with this discrepancy. Most comments I've seen (mainly on Twitter) involve the idea that census respondents were making some kind of error in how they categorize the nature of their tenancy, for example AHB residents categorizing themselves as PRS tenants (the number of AHB households seems to be underreported in the census, which supports this analysis).

However, for this to be of sufficient scale to explain the discrepancy between the CSO and RTB figures, it would mean that respondents in 2022 were systematically making errors that those in 2016 were not, and there is no real reason to think this is the case. The Census question on tenure has not changed since 2016, and reads:

Does your household own or rent your accommodation? (Mark one box only)

1. Own with mortgage or loan

2. Own outright

3. Rent

4. Live here rent free

If renting, who is your landlord

1.       Private Landlord

2.       Local Authority

3.       Voluntary/Co-operative housing body

Another issue is licensees, who may record themselves in the Census as private renters but who do not appear on the RTB register (an angle covered by this very good Dublin Inquirer piece).

Finally, there appears to have been an extended period during which it was extremely difficult for landlords to register a tenancy, arising from some kind of error with the RTB’s system. I don’t know too much about it (and this would not explain the trend since 2016), but it might be a factor.

To my mind one (or more) of the following scenarios must be true:

1. It is a data collection/categorisation error (such as those just mentioned),  i.e. the PRS has in fact not grown and the RTB data is more or less correct.

2. Since 2016 there has been a systematic under-registration of tenancies with the RTB, as landlords become more wary of the new regulations and enforcement, i.e. the PRS has in fact grown and the RTB data is misleading.

3. While there have been many landlords selling up and leaving, there has also been an influx of new investors that has offset this. From this point of view, both the 'landlord exodus' and the growth of the sector are compatible. If we consider that the increase in institutional PRS investment took place mainly since 2016, then it should come as no surprise that there has been an influx of new tenancies.

Option 2 above makes a certain amount of intuitive sense. It is well known that tightly regulated PRS markets often lead to large ‘black markets’. Sweden is a good example of a country with a very tightly regulated market in which subletting is very widely used. I believe many German cities have a similar thing going on. In the Irish case, it is noteworthy that the decline in registrations coincides with the wave of regulation in the sector, including RPZs and the new RTB enforcement powers. Moreover, when I was conducting research with Threshold a few years ago, a Threshold adviser told me during an interview that a very large proportion of the clients he deals with have unregistered tenancies.

Sinn Féin’s Eoin O’Broin has been digging into this issue. In a response to his query, the CSO noted that they are looking into the issue and hope to do further analysis. They also noted the following differences between the CSO and RTB data:

The census is filled in by the tenant, RTB registrations are made by the landlord.

The number of AHB properties recorded in the census suggests that some tenants may be recording them as PRS.

There are differences in how student accommodation is treated.

The census may include more informal letting arrangements, e.g. renting from family members.

Turning to another issue, the CSO data indicates that homeownership has declined further (proportion of owner occupied dwellings is now 66%, down from 68% in 2016) . I was somewhat surprised by this, to be honest, because the number of FTB mortgages issued annually has risen very quickly, especially compared with the immediate post-GFC period (2010-2016). So I thought we might see a bit of a move away from declining homeownership. A recovery of homeownership has also been anticipated by some housing scholars, especially those who focus on more demographic factors. In the US, for example, my understanding is that there is quite a strong narrative that declining homeownership was very closely associated with the millennial generation, whose labour market and housing careers were basically stalled by a decade because of the GFC. Once this effect had worked its way out of the system, some commentators expected a return to higher homeownership levels.

Anyway, this has not materialized in Ireland and it seems our current relatively low levels of homeownership (especially for those in their 30s/40s - 36 is now the age at which the majority of households own their own home. If memory serves me right in the 1980s it was around 24) will continue to be at the heart of housing debates in Ireland, especially as we edge closer to the next election.

Events & News

Don’t forget, if you are organizing an event related to housing, be sure to get in touch so I can share it here. The Simon Week conference will take place on September 26th, featuring a talk by Craig Colburn, author of Homelessness is a Housing Problem. This year’s Simon Brooke lecture will take place on September 27th and will look at the issue of youth housing. There’s a post-doc job open with Rob Kitchin (Maynooth Uni) working on housing/planning data.

What I’m reading

There’s been a good bit going in terms of the vacancy issue of late; Kathleen Stokes, Cian O’Callaghan and Maedhbh Nic Lochlainn published a new report (Urban vacancy in Ireland: Addressing recent responses and opportunities), and Cian O Callaghan has a new theoretical piece setting out a research agenda around vacancy and urbanization. Meanwhile, CATU have a new pamphlet, written by Fiadh Tubirdy, looking at the rent strikes of the early 1970s - I believe it is available in Connolly Books (more info here). This new piece of research from the US caught my attention. It finds that public housing landlords evict at similar rates to private landlords in similar neighbourhoods. A salutary reminder that we have to remember that public or social housing provision can come with its own issues, and that public housing will only deliver secure housing if there is a policy, economic and social context which supports. Over the summer I had the opportunity to read some older books I’d always wanted to get too, and both turned out to be brilliant: Anne Haila’s Urban Land Rent and Michael Harloe’s Private Rental Housing in the United State and Europe.