Dear Oxford City Council,

My name is Martina Radovanic and I am a research assistant for Professor Thiemo Fetzer at the University of Warwick who is also affiliated with the London School of Economics and the National Institute of Social and Economic Research, among others.

In the midst of the energy and climate crisis, retrofitting the building stock has become a priority. Since a significant part of Oxford's residential stock consists of private rentals, one specific challenge for our outreach strategy is whether landlords can be mobilised to act and retrofit their homes. It is essential to study to what extent natural person or legal entity ownership and foreign or non-local ownership may be an impediment, as it may require alternative policies to move “absentee” owners to invest in the properties.

I first request detailed information regarding the landlord data that is stored by the council for each rental property. This information should be provided as a comprehensive list of variables. For instance, if the council stores information on the residential address of landlords, the nationality of landlords, the type of landlord (such as type of organization or freeholder versus leaseholder), the type of property held (such as property characteristics), or the number of households in the rental property, please provide a detailed list of these information categories.

I also request an anonymized dataset based on the current landlord register that provides information on the landlord’s type (natural vs. legal entity) and resident location type (foreign vs. domestic) for each rental property in Oxford. The type of landlords can be deduced from the legal status of the landlord. For instance, Public Limited Companies, Private Limited Companies, Charity or Association, Limited Liability Partnership, are all different potential legal statuses of a legal entity. In contrast, an individual is a natural person. The resident location type can be deduced from the residential address of the landlord.
Specifically, I request the most granular dataset available.

1. Property level data
For each private rental property, I request access to the following property-level data:

UPRN (if available), Address (Street, house number, locality), Postcode, Indicator of natural person vs legal entity, Indicator of foreign vs domestic landlord residential address.

2. Postcode level data

If the information rights teams judges that data 1) cannot be provided, a justification should be provided. In that case, conditional on 1) not being satisfied, I request the information rights team to provide access to the dataset that is aggregated at the postcode level.

The tabular data layout should be:
Postcode, Indicator of natural person vs legal entity, Indicator of foreign vs domestic landlord residential address, Count of Properties.

For example:
Postcode / Landlord Type (Natural Person, Legal Entity) / Landlord legal residence (Domestic, Foreign) / Count of Properties
               CV4 7AL        /  Natural   /     Domestic   /   12
       CV4 7AL        /          Natural   /     Foreign     /   6
       CV4 7AL /  Legal     /     Domestic  /    8
       CV4 7AL / Legal     /            Foreign     /   7
 
3. Census Output Area level data

If the information rights teams judges that data 2) cannot be provided, a justification should be provided. In that case, conditional on 2) not being satisfied, I request the information rights team to provide access to the dataset that is aggregated at the Census Output Area level.
The tabular layout should be:
Output Area Code, Indicator of natural person vs legal entity, Indicator of foreign vs domestic landlord residential address, Count of Properties.
If data at the Census Output Area level is below a certain threshold to avoid disclosure these should be flagged as being omitted.

4. LSOA level data

If the information rights teams judges that data 3) cannot be provided, a justification should be provided. In that case, conditional on 3) not being satisfied, I request the information rights team to provide access to the dataset that is aggregated at the LSOA level.

The tabular layout should be:
LSOA Code, Indicator of natural person vs legal entity, Indicator of foreign vs domestic landlord residential address, Count of Properties.
If data at the LSOA level is below a certain threshold to avoid disclosure these should be flagged as being omitted.

5. MSOA level data

If the information rights teams judges that data 4) cannot be provided, a justification should be provided. In that case, conditional on 3) not being satisfied, I request the information rights team to provide access to the dataset that is aggregated at the MSOA level.

The tabular layout should be:
MSOA Code, Indicator of natural person vs legal entity, Indicator of foreign vs domestic landlord residential address, Count of Properties.

If data at the MSOA level is below a certain threshold to avoid disclosure these should be flagged as being omitted.

Qualifiers
* If the information rights team does not have data on some of the variables requested in the tabular layout, the dataset that comes nearest in spirit should be provided. An explanation of why said data is not held or shared should be provided and, in line with the spirit of FOI legislation, the Local Authority should attempt to help the requester of information about where or which government entity may hold said information.
* Data should be provided as early as possible
* Datasets should be provided, ideally in a tabular and accessible format.
* Datasets can be provided in formats as they are stored by the council. There is no need to combine them into the exact layouts requested to reduce processing costs.
* If any of the requests involve costs beyond the statutory limit for the council, then data that is of a similar spirit that can be extracted that involves a cost that is within the statutory limit should be provided.
* A justification of how the council determined that the cost of extracting said data would be beyond the statutory limit should be provided.

I am more than happy to provide technical assistance or input.

Yours faithfully,

Martina Radovanic