1.0 Introduction

View the most recent version.

Shelter is the biggest expenditure most Canadian households make. Its affordability can have a big impact on their well being. For these reasons, housing affordability is closely watched by a wide range of stakeholders – from housing advocates to policy analysts - interested in the housing and broader welfare of Canadians.

Measuring housing affordability involves comparing housing costs to a household's ability to meet those costs. The indicator underlying this measurement, the share of household income spent on shelter costs, is known as the Shelter cost-To-Income Ratio (STIR). A benchmark of 30% for the STIR is commonly accepted as the upper limit for defining affordable housing. Those who spend 30% or more of their household income have been, and continue to be, the subject of intense study. In particular, it is of interest to determine if they do so out of choice, through having the means and preference to spend more than the norm for housing, or out of necessity, being of low income and possibly in housing need.

Housing affordability, as measured by the STIR, is also a critical input to the core housing need indicator¹ developed by Canada Mortgage and Housing Corporation (CMHC) to identify households in housing need in Canada. The core housing need indicator "identifies those households unable to obtain market housing that is in adequate condition, of suitable size and, at the same time, affordable."² The resulting information is used by all three levels of government to help design, deliver, fund and evaluate social housing programs.

Up to now, all household income and shelter cost data used to calculate STIRs in Canada were cross-sectional in nature, thus describing housing affordability at a particular point in time. No data source followed households over time, collecting both their incomes and shelter costs, to enable the tracking of how their housing affordability evolved. That is not to say that data sources which surveyed households over time did not exist. Indeed, Statistics Canada's Survey of Labour and Income Dynamics (SLID) was designed to do just that: to enable, for example, the study of the dynamics of low income.³ For the last five years CMHC has sponsored a module of questions on shelter costs on this survey and now information from this module enables us to review housing affordability on a longitudinal basis for the first time.

This report, written jointly by Statistics Canada and CMHC, focuses purely on the dynamics of housing affordability, and not on housing need. It examines the likelihood of spending 30% or more of household income on shelter, how often this occurs, whether it is occasional or persistent, and contrasts the characteristics of those spending 30% or more to those spending less.

The report demonstrates the analytical capability of the longitudinal housing data on SLID to enhance what we know about Canadians spending 30% or more of their total before-tax household income on shelter. SLID cross-sectional estimates show that around 20% of Canadians were living in households spending 30% or more of their income on shelter in any single year between 2002 and 2004. Looking at this same time period longitudinally, the numbers tell us that 9% lived in households that exceeded the affordability benchmark for all three years and 19% lived in households that exceeded the benchmark for one or two years. Thus, 28% ever experienced this situation (9% persistently and 19% occasionally) and 72% never experienced it over the three year study period. This paper exploits the longitudinal capability of SLID data to examine the socio-economic and geographic characteristics of those who ever (at least one year) spend 30% or more of income on shelter and those who persistently spend 30% or more during the three-year period between 2002 and 2004.

Section 2 presents a descriptive profile that uses both cross-sectional and longitudinal data to describe Canadians spending above the affordability benchmark. Section 3 models the likelihood of Canadians living in a household ever or persistently spending more than the housing affordability benchmark. Section 4 draws conclusions from the report and provides some suggestions for further research.⁴ Background information on SLID, and definitions of the study universe, shelter costs, household income, and the housing affordability standard employed in this research are found in Appendices A through C. Appendices D to F present information about the models.

The addition of the housing module on SLID by CMHC has resulted in a longitudinal data source for housing research which should be of interest to housing policy analysts. We hope that this report will give some indication of the possibilities and stimulate further research.

____________

1. Core housing need refers to those whose housing is over-crowded, in need of major repairs, or costs 30% or more of household income AND who could not afford to rent adequate, suitable and affordable housing in their local housing market for less than 30% of total before-tax household income. CMHC is conducting separate research using SLID to examine core housing need.
2. Core Housing Need in Canada, 1991, Canada Mortgage and Housing Corporation. Pg 1
3. Morrisette R. and X. Zhang's (2001) "Experiencing low income for several years". Perspectives, Vol. 2, No. 3, March 2001. Statistics Canada, Catalogue
   no. 75-001-X.
4. As noted in footnote 1, CMHC is conducting SLID core housing need research to develop estimates of, and track high level trends in, core housing need between censuses.