RTB Rent Register

Galway - I'll take a random house in well known Tirellan Heights

From Daft I had to do city center, 6 properties, but it's really 4
- rents are €1150, €1473, €2250, €2500


RTB

1775816245815.webp


- lowest rents are €570, €650 & €911
- highest rents are €1400, €1500 & €1600
- all of those in the RTB registar are in city center, but Tirellan is not, that's the electoral area showing up
 
Last edited:
Galway - let's try a different area, so I'll go with Knocknacarra area. From Dublin side to Salthill side.

1775816810378.webp


- 5 of those are in Barna, like seriously !
- this is funny, the lowest is €550 which is in Taylor's Hill, now that makes zero sense, the next lowest is €700 & €900 (Barna) . Taylor's Hill is the most expensive part of Galway and Barna is a different town!
- Highest is €1950, and 2 at €1800
- percentage wise the Knocknacarragh is 73% at €1500 which makes sense, but the best percentage match is the €700 one in Barna at 86% !

(bored with this now so I'll do Cork a different time)
 
Last edited:
Regarding the the FOI document two points:
  • BER weighting. If your property has a BER rating of C1, they will give a higher similarity score to a comparable property with BER C3 rather than B3. I would have thought the heating value is closer to a B3.
  • Location looks undervalued as a predictor of rent. If your property is in the same electoral division as a comparable property, that comparable will receive a maximum of 12 points. This looks drowned out by other parameters. I would have expected a massive weighting for location.
 
Last edited:
Bluecup said:
Location looks undervalued as a predictor of rent. If your property is in the same electoral division as a comparable property, that comparable will receive a maximum of 12 points. This looks drowned out by other parameters. I would have expected a massive weighting for location.
Click to expand...
It seems to be 17 points, because additionally there are 5 points when the ED is the same and the Bedrooms are a match and the Floorspace are within 10% tolerance. Personally, I'm not sure it would be good to prioritise nearby properties with different number of bedrooms and floor space, over a little bit further away properies (still in same electoral area) with the same number of bedrooms and floor space.

I've started to code the RTB algorithm provided in the RTB document received by @Sr. Tayto, in order to run it over data I scraped from the RTB site. I made a few tests first to make sure I understand the algorithm. First of all, it appears the algoritthm in the document does indeed match an actual test I made with the RTB rent register query tool, so that's good. Secondly, other than the comments by @Bluecup, and the trimming of top/bottom 1%, I don't see any blatant biases in the algorithm, certainly not anything like favouring lower rent levels. It seems to be a reasonably well-thought through algorithm, and while we could all find different ways to do it, it's reasonable enough IMO. I may learn more as I analyse more data....
(Whether it is actually running over a full database of data is another matter, because I have noted suspicions of few registrations from 2021-23 turning up in search results.)

Some have commented on cost-rentals showing up in the 10 matches. This is presumably of the following rule, so in most cases I don't think it would make much difference to the results (I might be wrong... would be interesting later to scrape data for such an area)
Tier 4 (Local Authority Anchored) = 10 points If there aren’t enough strict matches (Tier 1), we add properties from the relevant Local Authority for that property.
Click to expand...

Anyway, I have a piece of basic working code now, but to debug it first of all, and then see if I can replicate the RTB tool results, I need to scrape the data again for a couple of areas. Apart from needing fresh data, I want to make sure I have data from all the Electoral Divisions within an Electoral Area.
Watch this space.
 
Last edited:
@patfert1 thanks again for all this!

I guess the main thrust of the issue is that the register shows the rents of existing tenancies, and not the market value today (which is what the legislation allows to be charged)

I'm curious, have you looked at how your average rent compares to the RTB/ESRI rent index? My average rents from the RTB register spot checks are about 20-25% lower on a reasonably consistent basis for the LEA I looked at. Sample size was not massive though, so curious if you've tried looking.
 
stormy said:
I'm curious, have you looked at how your average rent compares to the RTB/ESRI rent index? My average rents from the RTB register spot checks are about 20-25% lower on a reasonably consistent basis for the LEA I looked at. Sample size was not massive though, so curious if you've tried looking.
Click to expand...
No. Where can I see the rent index you refer to?
 
stormy said:
The index is here. I compared to the "existing tenancies" index.

It breaks it down by LEA.
Click to expand...
Which graph/xls breaks down by LEA? I downloaded a few and didn't find it. I see there it links to a tool on https://data.cso.ie/ which appears to give more granular data.
From what I can see but I may have missed the LEA breakdown, the breakdowns are by county. An average for let's say Leitrim is useful, but I don't think an average for Dublin is.
 
@patfert1 has mentioned the lack of 2021-23 tenancies. But why are these relevant? If I am setting market rent now, I will be doing it based on 2025 and 2026 rental values, not ones from a couple of years ago.

Is this a major flaw in this index, that how recently the rent was set isn't factored in. If I get a 100% match for a house with the same BER, size, location, bedrooms etc. as mine, but its rent was set 4 years ago and has only had RPZ increases since, it really doesn't tell me much at all about what the current market rent for mine is.
 
stormy said:
Figure 5 on the link I put above
"Figure 5: RTB Rent Indices - Standardised Average Rent (€) in New and Existing* Tenancies, by Local Electoral Areas (LEAs)*
Click to expand...
Got it, mea culpa. Thanks!

Greenbook said:
@patfert1 has mentioned the lack of 2021-23 tenancies. But why are these relevant? If I am setting market rent now, I will be doing it based on 2025 and 2026 rental values, not ones from a couple of years ago.
Click to expand...
I agree, you want newer rather than older rent level data, and that should be the case since the RTB algorithm contains this statement:
"Only Private tenancies which are New Registration and Renewed in the last 24 months are included in the base dataset. This includes the most recent rent set amounts."
And one of the sorting criteria (but only the 4th criteria) is: "Tenancy Commencement Date (Newest First) Desc"

But in all the data I downloaded, RTB registration numbers indicating tenancies originally starting 2021-23 are few, whereas there are plenty which are older than that (and newer of course). This suggests the chance that not all tenancies are being queried.

Greenbook said:
Is this a major flaw in this index, that how recently the rent was set isn't factored in. If I get a 100% match for a house with the same BER, size, location, bedrooms etc. as mine, but its rent was set 4 years ago and has only had RPZ increases since, it really doesn't tell me much at all about what the current market rent for mine is.
Click to expand...
See above.
 
patfert1 said:
But in all the data I downloaded, RTB registration numbers indicating tenancies originally starting 2021-23 are few, whereas there are plenty which are older than that (and newer of course). This suggests the chance that not all tenancies are being queried.
Click to expand...
I wonder whether the gap is caused by the collapse of the annual registration system in April 2022. Pure speculation on my part.
 
patfert1 said:
, I don't see any blatant biases in the algorithm, certainly not anything like favouring lower rent levels. It seems to be a reasonably well-thought through algorithm, and while we could all find different ways to do it, it's reasonable enough IMO.
Click to expand...
Fantastic piece of work and always good to put to bed the conspiracy theories that were circulating
 
Banquo said:
Fantastic piece of work and always good to put to bed the conspiracy theories that were circulating
Click to expand...
Just wait until the Landlords Who finish up getting more than Market rent finish up with tenants who realise the are been charged more the market rent and stop paying it will cost them a small fortune to resolve,
Been a Landlord has and still is a very good business when I entered it I looked at several other Businesses that I knew the in and outs of,
Looking back I am glad I choose the rental business a lot less pitfalls than most other business,
 
Last edited:
You must log in or register to reply here.
Back
Top