100% correct.
That the Rent Register is rigged, just like the RTB tenancy and landlord numbers, is becoming more obvious by the day.
RTB Rent Register
- Thread starter Sr. Tayto
- Start date
That the Rent Register is rigged, just like the RTB tenancy and landlord numbers, is becoming more obvious by the day.
Shakespeare
Registered User
- Messages
- 148
Also from the article "Controversial new rental reforms that came into effect last weekend will result in more evictions as landlords falsely claim they are selling properties in order to charge new tenants higher rents, an opposition housing spokesman has claimed."
This really frustrates me that it's not challenged. Even if landlords claim they are selling so they can evict and then decide, for whatever reason, not to sell then it's still a "no-fault eviction" and the landlord couldn't charge new tenants higher rent (theoretically market which is what the scaremongering suggests but we all know there's no acceptance of the real market rent per the RTB register) beyond the 2% cap unless left unlet for 2 years so it would be a pointless exercise for a landlord. If they have good tenants, they run the risk of getting rogue tenants for the sake of getting people out, possibly refurbishing just to relet at 2% rent increase.
This really frustrates me that it's not challenged. Even if landlords claim they are selling so they can evict and then decide, for whatever reason, not to sell then it's still a "no-fault eviction" and the landlord couldn't charge new tenants higher rent (theoretically market which is what the scaremongering suggests but we all know there's no acceptance of the real market rent per the RTB register) beyond the 2% cap unless left unlet for 2 years so it would be a pointless exercise for a landlord. If they have good tenants, they run the risk of getting rogue tenants for the sake of getting people out, possibly refurbishing just to relet at 2% rent increase.
Some LLs might already have done that in the past few years, evicting on the pretence of selling and then renting again on a higher price. It was against the old rules, it's also against the new rules. There's not really any reason to think it will happen more now than before.
Rory Hearne is unfortunately brainless. Despite writing two books on the subject he was sure until recently that a landlord could grant a one year lease, not renew it and re-let at a higher price to a new tenant. I heard him say that in a podcast over the summer. He thought the new rules were useless because of this.
This is the standard of 'Housing Expert' that you are dealing with, so no wonder the PRS is so messed up
It did happen, of course. There are always some who break the rules in any walk of life.
The vast majority of landlords who were evicting to sell were genuine though
Its supposed to be similar properties, and the tool seems to be fairly random, so if you are letting a house with 2 bedrooms, a C3 rating and a floor size of say 74m2, you compare your property to other C3 2-bed houses of that size, not a new build A3 apartment with 84m2 floorspace.
Obviously that's more challenging for niche properties or rural areas where there might be fewer or more unique properties.
The other role being we get to see what rents are actually being charged and not the narrative either side of the rental sector wants to hear.
Of course and that is very important. No more than the property price register, I don't think anyone has an issue with that aspect of the register. Transparency is beneficial to landlords, tenants and society as whole.
What is amiss though is that RPZ rents are being presented as market rents which they are not. That is the whole purpose of RPZs - keep rents below what the market would allow.
Also, cost rentals are included. The whole purpose of cost rentals is to charge rent at 75% of market rent or lower. More importantly, they are not flagged as cost rentals.
No, "cost rental" is to offer rentals that cover the cost of capital, management and maintenance costs so tenants don't pay any more than required.
Its targeted at the segment of the market who won't qualify for social housing (which is why we don't allow HAP tenants in) but with capped earnings (so it doesn't end up being used by those who could afford to buy).
There's no requirement for it to be 75% of "market rent", which as you correctly say isn't something easily calculable, its well worth reading the full act. The entire point is to link the rent charged to the cost of delivery of the home over a 40 year period. That's why they are getting rent reviews - that might not be a fixed figure as there are management/maintenance costs to consider as well as the build and initial fit out.
If we don't review rents for cost rental, which is the big whinge this week, then you end up with flats that resemble the horrific condition most Dublin city flats have ended up in after 30 plus years.
Correct. There has rarely been a level of misinformation, deliberate disinformation and outright fear-mongering from every side, including lots of politicians who are paid to actually read changes to the law, as there is about this bill.
Likewise the nonsense that "all rents are going up by 25% nonsense" the opposition was peddling the week before last.
GitHub - landlawd/rtbquerytool: RTB Rent Register Query Tool
RTB Rent Register Query Tool. Contribute to landlawd/rtbquerytool development by creating an account on GitHub.
github.com
V2 has two enhancements:
* it is now possible to define multiple eircodes for comparable properties. Note the eircode-related changes in the template config file.
* Looping logic is optimized to reduce the number of queries (break out of the "floor space" iterative loop when there are no new results, and move to next criteria)
I'm going to make a large survey of houses in one Dublin ED. If it succeeds, I will upload to the github repo and post here.
Curious about this. Does S128 not say the rent for a tenancy can’t appear on the published register? The previous register had the address which meant the tenant or landlord were easily identifiable so was that a violation of (a) and possibly gdpr personal data. Now (a) is gone from the new register but (b) is included. If the list of comparable properties is for a particular road in an estate where dwellings are the same, or a rural area with very few properties, could these details (a + b) be easily discovered.
S128. (4) The published register shall not contain any information, as respects a particular dwelling, that discloses or could reasonably lead to the disclosure of—
(a) the identity of the landlord or the tenant or tenants of the dwelling,
(b) the amount of the rent payable under a tenancy of the dwelling.
rtbquerytool/data at main · landlawd/rtbquerytool
RTB Rent Register Query Tool. Contribute to landlawd/rtbquerytool development by creating an account on GitHub.
github.com
5 data files uploaded
| Electoral Area | Dwelling Type | BER | Beds | Sq.m. | Result count |
| Dublin 8 | Apartment | A3-D2 | 1-3 | 30-70 | 450 |
| Firhouse/Bohernabreena | House | B1-D2 | 3-5 | 70-140 | 201 |
| Clontarf | House | B1-D2 | 3-5 | 70-140 | 187 |
| Donegal | House | B1-E1 | 3-5 | 70-170 | 103 |
| Donegal | House | B1-E1 | 1-2 | 45-70 | 34 |
Last edited:
Some observations on collecting this data:
- The property address (from the eircode in step 1) is used by the step 2 search. The location of the property does appear to influence the 10 results. You can reproduce this in the RTB register by using a different address within the same EA and all other criteria the same.
- While running the script, the rent register often returned "Invalid API Response" instead of 10 hits. This was most common when the search criteria contained a large square meter value or 5 bedrooms. I suppose this is because no (or <10) matching properties are found, or the match score is very low. I guess the RTB will realize this and and fix it sometime.
- I could not say I notice any bias towards returning lower-rent properties, since the overall data contains a wide range from low to high. That's not to say the register isn't "rigged" in some other way such as including cost rentals, and obviously it contains many properties in RPZs.
- I do wonder if there is some other bias in the RTB's match score algorithm. Despite the wide search criteria I used (including multiple eircodes in each EA), the total number of properties found seems quite small, at least in the case of houses. I cannot imagine EAs like Clontarf and Firhouse having only around 200 rental houses each.
Possibly while it does produce some higher-rent properties, the bias is to produce mostly lower-rent properties?
That is what I have been seeing. Some higher-rent properties turn up, but likely not those at the highest rents. The 10 produced are 70% lower-rent or thereabouts.
I spoke to an estate agent last night who I have used in the past and found astute. His advice is, if you can, hold off renting for a month or two and see how this develops. He said estate agents in Cork have been talking about this and also think the Register is skewed to produce lower-rented properties.
Most of the register will consist of rents relating to properties that have only been allowed 2% rises for years. If you wanted to design a system to provide examples approximating to current market price of rent you would have to exclude them.
If the system designer couldn't be bothered to, or had no available data field to enable selection of the rental prices of properties recently rented at a free market price (so excluding re-lets of RPZ restricted properties) then the next most sensible thing to do would be for the 10 examples presented to the user to the ten highest rents, since that would occasionally achieve the same thing.
How recently a property was first introduced to the market would be a far more important factor in selection than BER, floor space etc.
However because new entrants to the rental market have been so uncommon in most areas, there's a high probability that for many properties the sample of 10 comparisons supplied will not include any similar properties which have been recently introduced to the market (so unencumbered by RPZs) in the relevant area - ie no sensible indication of market rent will appear in the sample given by the RTB.
Another possibility which may or may not be happening is that sampling could actually involve chopping off the highest and lowest x% (say, 5% or 10%) of the data before presenting it to the user. This is a technique often used to clean 'outliers' out of data in order to present a more generally representative picture.
The problem (or helpful consequence) arising were they to do that is that the top 5% or 10% of any given sample is likely to include the only properties on the register actually showing anything close to the real market rent (ie recently introduced to the market & not suppressed by being leased in an RPZ).
They shouldn't be doing that, and it's a bit ridiculous that you have to even consider the possibility, but if they are then the effect would be to systematically fail to present examples indicating market rent.
You could try to establish whether that is happening if you know of individual properties recently let for the first time - which you know represent actual market price - and see if they often get selected as examples.
If the system designer couldn't be bothered to, or had no available data field to enable selection of the rental prices of properties recently rented at a free market price (so excluding re-lets of RPZ restricted properties) then the next most sensible thing to do would be for the 10 examples presented to the user to the ten highest rents, since that would occasionally achieve the same thing.
How recently a property was first introduced to the market would be a far more important factor in selection than BER, floor space etc.
However because new entrants to the rental market have been so uncommon in most areas, there's a high probability that for many properties the sample of 10 comparisons supplied will not include any similar properties which have been recently introduced to the market (so unencumbered by RPZs) in the relevant area - ie no sensible indication of market rent will appear in the sample given by the RTB.
Another possibility which may or may not be happening is that sampling could actually involve chopping off the highest and lowest x% (say, 5% or 10%) of the data before presenting it to the user. This is a technique often used to clean 'outliers' out of data in order to present a more generally representative picture.
The problem (or helpful consequence) arising were they to do that is that the top 5% or 10% of any given sample is likely to include the only properties on the register actually showing anything close to the real market rent (ie recently introduced to the market & not suppressed by being leased in an RPZ).
They shouldn't be doing that, and it's a bit ridiculous that you have to even consider the possibility, but if they are then the effect would be to systematically fail to present examples indicating market rent.
You could try to establish whether that is happening if you know of individual properties recently let for the first time - which you know represent actual market price - and see if they often get selected as examples.
Thanks @polecon
I suspect that that is what they are doing and based on their 'form' so far, I think it deliberate and not just clearing out outliers.
Properties at very low rents still come up, just now ones at high rents.
We can all speculate about what's behind this algorithm but the data I gathered appears to show a reasonable spread of rent levels from low to quite high. From the 201 results for Firhouse, 27 are 3000+EUR (max 3775), 60 are 2500-2999, 42 are 2000-2499, 60 are 1500-1999, and just 12 under 1500.
FWIW I have a house in Dublin which rents for under 1500 and I was not able to see it in any search (even trying the eircode next door). So there is some bias or filtering in the algorithm, but I cannot identify what it might be.
FWIW I have a house in Dublin which rents for under 1500 and I was not able to see it in any search (even trying the eircode next door). So there is some bias or filtering in the algorithm, but I cannot identify what it might be.
Article 'written for the RTB' in today's Independent here
Heavy emphasis on using the Rent Register. That and filing forms is all that is really talked about. Nothing about Daft.ie and independent valuation etc.
From the RTB's standpoint, if you can't justify the increase based on the Register, it is not 'market rent'.
I really get the impression that in the government's mind, 'market rent' means what the RTB think is the correct rent.
Heavy emphasis on using the Rent Register. That and filing forms is all that is really talked about. Nothing about Daft.ie and independent valuation etc.
From the RTB's standpoint, if you can't justify the increase based on the Register, it is not 'market rent'.
I really get the impression that in the government's mind, 'market rent' means what the RTB think is the correct rent.