Tenure of household

Hello, Tēnā koutou!

Welcome to our discussion of Tenure of Household.

I’m Tina, from Statistics New Zealand. I look forward to open and inclusive discussion over the next six weeks to understand your Housing-related information needs.

Hi

In Waitara township the Council owns approximately 25 per cent of properties as leasehold land. These are generally located near each other, so it is easy to assess meshblock data. It appears approximately 57% of leaseholders say they own the property while the rest say they rent it (which could be possible due to sub-lease arrangements). The census questions don't appear to provide a logical place for leaseholders to mark their position. It could be useful to consider revise the tenure question to include better options for leaseholders.

For your information, the following meshblocks consist solely of leasehold properties:
1546600
1546500
1546700
1547701
1547900
1549501
1549302
There are other leasehold properties in that township, but they are not the sole tenure type in their meshblocks.

Hi @gregstephens thanks for kicking off this discussion! It would be great if you could visit our introductions page and share a little about your background and/or interest in this topic:
Welcome and introductions

Isn't the tenure already collected in the valuation roll in the TORAS flags for landuse? (Tenure, Ownership, Rateability,Assessed, Special). This is supposed to be maintained by the councils for rating purposes, and since money is involved there is an incentive to do this. Anyone can get access to the roll, although it is very very expensive. It is also very precise because each assessment has a unique ID, unlike the census address fiction.

Hi @gregstephens and @kimollivier , do you use Census data on tenure of households?
What changes would you recommend based on the issues and information gaps mentioned above?

Because the valuation roll has much more detail that is used instead of the census where access is possible such as work being done by a council. The worry is that this information is not being maintained in the valuation roll. Many of the valuation fields are 'optional' so they are set once and forgotten.
Also LINZ titles has much better information about tenure if you know who the landlords are such as Housing NZ Limited.
Similarly the mortgage details are registered against titles so I know exactly who has a mortgage at any time, not just every five years. The banks use that to work out their market share and how well their competitors are doing.

I don't see how valuation rolls provide greater data. Section 11 of the Local Government (Rating) Act 2002 sets out what information must be on valuation rolls. In terms of leases it is fairly limited - only registered leases of more than 10 years must be included. And those registered leases become liable for rates - which is not the case in residential tenancy agreements. As a local authority employee, we use census data for tenure - not any internal dataset. So I disagree with Kim Ollivier that councils have better information.

Mortgages being registered on the title do not actually mean anything as many mortgages remain on the title after it has been paid off (part laziness, part making it easier to borrow later on to do renovations). Also, LINZ don't know who are landlords when they are mum&dad investors with one or two extra houses.

It would be useful to get a better picture of social housing, particularly community housing providers. The last census had a massive undercount on HNZC properties, so will be interesting to see how you can address that problem.

I don't see any need to collect information on family trust ownership. To be honest, it just frustrates having this data as it means you need to add categories together to get an idea of ownership. I don't see what it serves, but it might do so for other people.

I've used tenure data a lot in working in both central and local government. The census is by far and away the most reliable source I have come across.

I have had a look at the valuation roll and GS is right about private tenure. The code for private tenure (2) has been dropped and is now empty. It does keep codes 3-5 for government, local body and Maori properties which I suppose is related to rateabilty exceptions.
The project I was doing needed to find the actual properties that were Crown owned which can be found from the title owners in the LandOnline titles. But there were a lot of properties with no title and some were simply gazetted, some owned by councils and the valuation roll indicated a lot more not accounted for.
The valuation roll may not have explicit codes for tenure, but it still has every property as a separate record with a lot of landuse and building characteristics.
It is very underused. For example I would have thought it would fairly easily identify potential old earthquake risk buildings to visit for an inspection, yet the councils claim that it is 'impossible' to quantify.

Since the bank is jointly liable for rates arrears if the is a mortgage on the property a lot more are closed down now. But my own mortgage paid off 20 years ago is still there and I am being asked for $120 to remove it by a solicitor!

The larger issue is raised by an obvious under count and clearly false responses. Does the post processing correct errors by cross referencing other public information?
How far can this go? Maybe correct the spelling of an address? But what about conflicting responses to different questions?

"The larger issue is raised by an obvious under count and clearly false responses" - I agree this is a substantial issue. The HNZC undercount is almost 20%!

The estimated undercount of people renting from Housing NZ Corp is approximately 18%. 2013 Census information here.

We are definitely looking into a few options to improve the quality of Sector of Landlord information for the 2018 Census, including collection of community-based housing providers.