Scoping the costs of homelessness in New Zealand

The cost of homelessness to society and to the individual is difficult to measure, which in turn makes it difficult to formulate and evaluate meaningful policy change to address rising homelessness and housing deprivation. This report reviews the international literature on quantifying the cost of homelessness, with a view to identifying methodologies (and the appropriate data) that may be useful in a New Zealand (NZ) context.
A recent report based on 2013 Census data estimated 41,705 people in NZ were severely housing deprived (Amore, 2016). Since 2006, the proportion of the population identified as homeless has increased from 0.8% to 1%. Several factors may have contributed to this increase – these potentially include the 2008/09 global financial crisis and housing shortages in both Christchurch and Auckland, leading to rising house and rental prices in recent years.
There are two components to this study. First, the relevant international literature is reviewed, with a focus on the different methods and measures used when quantifying the economic cost of homelessness. Second, we investigate the availability of data necessary for constructing an appropriate cost measure in NZ, as well as assess what data would be required when evaluating an intervention in this space
A total of 30 international studies on the cost of homelessness are reviewed. There were two types of studies – those that calculate the cost of homelessness to society; and economic evaluation studies that investigate the cost effectiveness of interventions. It was clear early in the review that there are various definitions of homelessness in the literature, making comparison across the studies difficult at times. Broadly speaking, definitions of homelessness fit across the following categories: Housing situation; temporal situation (eg chronic or temporary); social approach (such as primary, secondary or tertiary homelessness); and demographic group (such as youth, families). The definition provided by Statistics NZ is one we recommend adopting when assessing the cost of homelessness in NZ and it includes those in living situations “where people have no other options to acquire safe secure housing: are without shelter, in temporary accommodation, sharing accommodation with a household, or living in uninhabitable housing” (Statistics NZ, 2009, p.6).
Based on our survey of the literature we recommend that the average cost per homeless person in NZ be constructed based on their health, corrections, and government benefits usage – an integrated cost of homelessness approach focusing on the direct costs to institutional providers of support. As for the data scoping exercise within this report (part two), we propose (for the most part) following Amore et al’s (2013) methodology with respect to identifying homeless individuals in the Census. After which, given the newly available linked administrative data (Integrated Data Infrastructure) available from Statistics NZ, we have identified the key data sources and variables that should be linked with the Census, utilizing individuals’ unique identifier. This will allow both the economic cost of homelessness to be constructed, as well as permit cost/benefit analyses of intervention programs – both of which aid in growing the empirical evidence regarding the economic impact of this social issue. We thank Lottery Community Sector Research for funding this research.