net household income and adjusting this figure by the OECD scale. The new poverty rates are now
calculated on a reduced sample that only consists of renting in-work and non-work households.
We know from Figure 11 that the share of homeowners is especially low among single parents with
child(ren), two or more family households, and other multi-person households. Unsurprisingly, the
share of these groups in the restricted sample (with only renters) increases compared to our initial full
sample, as shown in Table 15. For example, while about 9.6 percent are one parent with children
households in the full sample, this share increases to 12.3 percent in the restricted sample.
The BHC in-work poverty rate for the restricted sample of renters equates to 8.1 percent. When
accounting for housing costs, the AHC in-work poverty rate for the restricted sample increases to 12.8
percent, which is a lift of about 50 percent. This is a common finding when examining AHC measures
of in-work poverty (McGuinness, 2018, p. 7). For example, Bourquin et al. (2019) find for the UK that
in 2018 the BHC in-work poverty rate was 13 percent, and this figure rose to 18 percent after housing
costs (+38 percent). Furthermore, this increase is reasonably equally distributed across different
household structure types.
Table 15. Before and after housing costs and in-work poverty rate, for renters
Restricted sample
Before housing
costs
After housing
costs
In-work
poverty rate
4.5%
In-work
poverty rate
8.3%
Couple with child(ren)
One parent with child(ren)
7.6%
12.7%
Two or more family household
Full sample
(from Table 8)
Share
Before housing
costs
21.5%
In-work
poverty rate
4.8%
25.7%
10.4%
23.2%
30.1%
12.3%
6.3%
12.3%
36.8%
9.6%
10.3%
12.1%
8.4%
9.6%
6.6%
One-person household
5.7%
11.0%
17.1%
6.4%
15.3%
Other multi-person household
14.0%
20.7%
10.6%
11.6%
6.0%
Total
8.1%
12.8%
100%
7.0%
100.0%
Couple without children
Share
Source: IDI 2019. Notes: In-work poverty as defined in Section 3. Full sample = 725,313 working households. Restricted
sample = renters with information on rental expenditure = 244,854 working households. Share = proportion of respective
household type from sample of working households.
4.10 Poverty duration
The current definition to determine where a household sits in relation to the poverty threshold is
based on information in the month of March 2013 when the Census was conducted. To investigate
poverty duration within a year for both in-work and non-work households, we present data on the
number of months that a household is poor in the preceding year. The literature on poverty over time
focusses largely on annual observations of poverty over a multi-year period (e.g., Vandecasteele &
Giesselmann, 2018; Buddelmeyer & Verick, 2008); in this analysis we go further to examine the
within-year duration of poverty, as shown in Figure 12.
Page 41