Case Study | Optimising Tenants: What Can We Learn from Administrative Data?

Testimonial | Signpost 2 Grow
21 January, 2019
Show all

Case Study | Optimising Tenants: What Can We Learn from Administrative Data?

Research overview

This case study illustrates how an analysis of a Council’s routine administrative data may generate information for service planning. The data in question concerned the Council’s Introductory Tenants. The analysis provided a more rigorous basis for the Council to target their limited resources on those Introductory Tenants most likely to fall into rent arrears, be the subject of anti-social behaviour allegations and/or have a relatively high number of contacts with the Council. Factors associated with each of these three outcomes are summarised in Figure 1.

How the research helps

The analysis has also prompted ideas for further research and analysis to underpin how a Council may best support its Introductory Tenants. The case study should be of interest to other Councils particularly those who retain a substantial stock of Council housing.

Background

Introductory Tenancies were introduced as a discretionary tool for local government in 1996[1]. An introductory tenancy is essentially a one-year trial during which the council can see whether the tenant can keep to the terms of their tenancy agreement. Where a tenant has difficulties doing this, the tenancy may be extended for six months or the council may take legal action to evict. This English city council wanted to better understand the drivers of demand for its services. It holds data for the administration of its tenancies and the analysis reported here focuses on the Council’s housing tenants in their first year, i.e. its ‘Introductory Tenants’.

Data sources and sharing arrangements

A data sharing agreement was drawn up between the Council and UEA members of the Business and Local Government Data Research Centre who conducted the analysis. No information was passed to UEA that could identify individual tenants. The data from the Council were stored on a secure server, accessible only to the researchers working on the study. The data consisted of: information on the lead tenant and their household, for example demographic characteristics, gross rent, Housing Benefit, and letting category (e.g. ‘emergency’) for each Introductory Tenancy during the period July 2016 –July 2017; and for the properties occupied by these tenants, records of contacts made by or to the Council for a similar time period.

Method

The analysis explored which Introductory Tenant characteristics were associated with: rent arrears; being the subject of anti-social behaviour allegations and/or having a relatively high number of contacts with the Council. A statistical model was estimated in which each of the outcomes was related to a set of tenant characteristics within a structural equation model, allowing for the possibility that the same unobserved factors influence each of the outcomes. A simpler approach to the statistical analysis was found to produce similar results to the structural equation modelling.

Potential for further analysis

A number of limitations to the analysis could be overcome if data could be assembled prospectively rather than retrospectively so that e.g. changes in tenants’ circumstances could be identified. However, some limitations were inherent in the administrative data available e.g. information on income sources was available only for those claiming Housing Benefit.

Report Author(s): Marcello Morciano , Ruth Hancock, Amanda Burke: Norwich School of Medicine, UEA