KEY ISSUES: CONSTRUCTION WORKERS, OWNER BUILDERS AND RENTERS OF IMPROVISED DWELLINGS

SUMMARY OF KEY POINTS

• The Review analysed the characteristics of 'persons living in an improvised dwelling', and based on the characteristics of some of those identified in this category, it was determined that some people may be in a shed or partially constructed dwelling on their own property while they build their own homes, or they were construction workers. As a result, they were not included in the reviewed homeless estimates. This chapter provides some background data on the characteristics of these populations.
• Submissions identified that consideration needs to be given to people who are living on their own properties in improvised dwellings as a result of natural disasters. The ABS will develop a strategy to analyse areas affected by recent disaster events in the treatment of homeless estimation for 2011. One submission recommended tracking natural disaster areas to enable an understanding of the impact on homelessness. The ABS will utilise the expertise of the Homelessness Statistics Reference Group in the formulation of the strategy.
• There are differing views on the treatment of these populations in homeless estimates ranging from the inclusion in the homeless estimates because these people were likely to be living in marginal housing for some time regardless of whether or not they owned a home elsewhere, to excluding those who owned the dwelling with a mortgage but to track these populations separately.

INTRODUCTION

In the Census, the dwelling structure is determined by the Census collector. There are a number of broad categories used to classify housing and identify the characteristics of housing and how the housing stock changes over time. This includes distinguishing between different types of houses (i.e. separate or semi-detached), apartments etc. There is a category 'Improvised home, tent, sleepers out'. This category includes sheds, tents, humpies and other improvised dwellings occupied on Census Night. It also included people sleeping on park benches or in other 'rough' accommodation (ABS 2006a).

In CTH, people staying in 'improvised dwellings' who reported no usual adderss or being at home on Census night were operationalised as being homeless. However, analysis of those who lived in this dwelling type showed that some of these households have characteristics that suggest they were living in a shed or in a partially constructed dwelling on their own property that had features beyond what would be considered an improvised dwelling. Analysis of selected characteristics from the Census suggested that some people in this dwelling type had the characteristics that would be consistent with those people who were working in mobile construction crews building suburbs or highways, or people in the process of renovation or building their own new homes.

Submissions raised concerns as to whether the rules applied to identify homeless people in 'improvised dwellings, tents and sleepers out' will exclude those living in temporary housing as a result of recent disaster events. They may have high mortgages on their dwelling but may be unable to afford to rebuild their home. They may have the same characteristics in the Census as owner builders. One submission suggested that this group be tracked separately to understand the impact of natural disasters on overall homeless figures, and to see the impact on homelessness as a result of the recent natural disasters. The ABS is developing a separate strategy for identifying and classifying natural disaster areas.

There was a view in one submission that those who were living in improvised dwellings such as owner builders, hobby farmers and construction workers were still people who were living in marginal housing that was sub-standard, and should remain in the homeless population even though they may own a home elsewhere.

One submission requested constructions workers, owner builders and hobby farmers be reported separately, regardless of whether they were classified as homeless.

There were some other groups that were not included in the Reviewed homeless estimates, including 656 records that were imputed, but the Census collector had made no contact with the occupants. No details about the people who lived in the dwelling could be established, nor if in fact anyone actually lived in the dwelling. The Review also removed some dwellings that were incorrectly classified as an improvised dwelling: about 500 in manufactured homes in Tweed Heads were removed in the compilation of CTH and a further 200 people living in their new homes in Wanneroo (North metropolitan Perth) were removed in the Review. There was no disagreement in the submissions about the exclusion of imputed or incorrect records in the Review.


USING THE CENSUS TO IDENTIFY CONSTRUCTION WORKERS, OWNER BUILDERS AND RENTERS OF IMPROVISED DWELLINGS

The Census dwelling type variable identifies people who are residing in improvised dwellings which, when used in conjunction with additional variables such as tenure type, mortgage payments, rental payments, weekly income and employment status provide further information about the occupants of these dwellings. The ABS Review used such data to conclude that some occupants of improvised dwellings were construction workers, including agricultural staff and miners who could be in an 'improvised home' for employment reasons, or people residing in improvised dwellings on their own property while they build their own homes and as a result were unlikely to be homeless.

Persons reporting a tenure of ‘owned outright’, ‘owned with a mortgage’, ‘being purchased under a rent/buy scheme’, ‘being rented’ or ‘being occupied under a life tenure scheme’ where at least one person in the dwelling was classified as ‘employed – worked full time’ were considered in the Review to be outside the homeless population. This was based on both their tenure and employment status. This population consisted of 5,054 persons. The Review concluded that they were most likely to be residing in improvised dwellings such as sheds, garages and shacks, which could be on one of the properties that they own while building a house, or alternatively, the employment status of persons in this group indicated that the nature of their accommodation could be due to employment reasons, such as being in an on-site or transportable dwelling away from their own home (construction workers).

Analysis of the characteristics of high concentrations, in particular geographic locations, of people enumerated at home in the dwelling type 'improvised dwellings..." has shown concentrations of people on their mortgaged properties. Specific sites have been examined to determine the green fields nature of the location, or the 'hobby farmer' nature of the area. For example, of the 130 people enumerated in "improvised dwellings..." in the Bega Valley area in 2006, over half had a mortgage on their dwelling, while in the Mackay area the proportion was 40% of the 214 people so classified.

For the group considered to be 'construction workers' (persons in rented improvised dwellings where at least one person was working full-time) (725 persons), the Review analysed the occupations for occupants who were working either full-time, part-time or away from work. The majority of occupations were consistent with the assumption that they were construction workers: 23% were labourers, 18% technicians and trades workers and a further 12% worked as machinery operators and drivers. Over half of persons in this group (53%) lived in dwellings with Internet connection (either broadband, dial-up or an other connection) and 67% of dwellings had two or more bedrooms. This group were generally younger with 45% of people aged between 20 and 39 years. Specific locations, such as new suburbs being built in Port Stephens, were identified in this process. The Review concluded that the persons in such dwellings were not homeless.

After considering all those dwellings where at least one person had an employment status of ‘employed – worked full time’, the Review analysed similar dwellings where no one in the dwelling reported working full-time. The following shows analysis of three distinct groups namely: 'owners', 'owners with a mortgage' and 'renters' in improvised dwellings where no one reported working full time, and provides data to show which persons were not included as homeless in the estimates.

Owners: People who are not working full-time

People who reported being 'at home' in an improvised dwelling who also reported that they owned the dwelling outright, and where no person was employed full time (2,356) were not included as homeless in the Review. Analysis conducted during the review concluded this group were most likely residing in temporarily accommodation while building a home on their own property. The analysis below shows more detailed data about this group.

Of the 2,356 persons reporting being 'at home' in an improvised dwelling, tent or sleeping out where there was no one in the dwelling employed full time who reported owning the dwelling outright, 42% were in dwellings which had total reported household incomes greater then $500 a week; with 15% having incomes greater than $1,000 a week (Table 1).

	no.
Negative income	16
Nil income	38
$1–$149	96
$150–$249	328
$250–$349	319
$350–$499	346
$500–$649	434
$650–$799	106
$800–$999	98
$1,000–$1,199	154
$1,200–$1,399	64
$1,400–$1,699	33
$1,700–$1,999	21
$2,000–$2,499	35
$2,500–$2,999	22
$3,000–3,499	3
$3,500–$3,999	6
$4,000 or more	7
All incomes not stated	74
Partial income stated	156
Total	2,356
(a) Cells in this table have been randomly adjusted to avoid the release of confidential data. (b) Although persons temporarily absent were excluded from this table, the calculations for household income includes incomes of those persons temporarily absent from the dwelling.

In addition, half of the 2,356 persons in this group lived in dwellings which had two or more bedrooms (51%) and 34% lived in dwellings with an Internet connection (either broadband, dial-up or an other connection) (Table 2). For those living in a dwelling with none or one bedrooms, 28% were connected to the Internet.

		TYPE OF INTERNET
	No Internet connection	Broadband connection	Dial-up connection	Other connection	Not stated	Total
	no.	no.	no.	no.	no.	no.
None (includes bedsitters)	280	32	68	4	5	389
1 bedroom	492	74	118	16	10	710
2 bedrooms	345	67	118	9	4	543
3 bedrooms	263	80	94	–	3	440
4 bedrooms	49	49	46	–	3	147
5 or more bedrooms	46	–	26	–	–	72
Not stated	49	–	3	–	3	55
Total	1,524	302	473	29	28	2,356
– nil or rounded to zero (including null cells) (a) Cells in this table have been randomly adjusted to avoid the release of confidential data.

In addition to the group discussed above, there were 225 people who reported 'no usual address' who were in an improvised dwelling, tent or sleeping out where there was no one in the dwelling employed full-time and where the dwelling was reported to be owned outright. Of these, 205 were in dwellings with no or 1 bedroom. The ABS Review concluded that these people should be included in estimates of homelessness because they could be homeless and are possibly living in their car or other shelter.

Owners with a mortgage: People residing in improvised dwellings, no one in the dwelling working full time

The Review identified 379 persons who reported owning their 'improvised dwelling' with a mortgage, with mortgage repayments of $1,050 or over per month and no one in the dwelling was employed full-time (excluding persons temporarily absent). Based on their characteristics they were unlikely to be homeless and were not included in the homeless population. The Review concluded that the characteristics of this group indicate they could be persons who are repaying a mortgage on land they are purchasing, and who could be in an ‘improvised dwelling’ while building or waiting to build their home.

In these improvised dwellings, 79% of people were living in dwellings which had 2 or more bedrooms (Table 3) and 63% lived in dwellings with an Internet connection (either broadband, dial-up or an other connection) (Table 3). Over 60% of people in the group lived in a household with reported incomes of $650 or more per week.

None of the 379 persons reported having 'no usual address'.

		TYPE OF INTERNET
	No Internet connection	Broadband connection	Dial-up connection	Other connection	Not stated	Total
	no.	no.	no.	no.	no.	no.
None (includes bedsitters)	14	24	3	–	–	41
1 bedroom	17	9	11	–	–	37
2 bedrooms	27	21	46	–	–	94
3 bedrooms	35	24	20	–	3	82
4 bedrooms	40	32	25	–	–	97
5 or more bedrooms	–	25	–	–	–	25
Not stated	3	–	–	–	–	3
Total	136	135	105	0	3	379
– nil or rounded to zero (including null cells) (a) Cells in this table have been randomly adjusted to avoid the release of confidential data.

Renters in dwellings where no one reported working full time

The Review identified 57 people who reported that they were in an improvised dwelling which they rented and reported rent payments of $300 per week or more, and where no persons in the dwelling reported working full-time (excluding persons temporarily absent) (Table 4). Mean weekly housing costs, as measured in the 2005-06 ABS Survey of Income and Housing, for households renting in the private rental market, was $223 per week. No state, and no capital city recorded an average weekly rent in the private rental market that was above $300 per week. Median weekly rentals in the private rental market were lower than the means, in total and for all states and territories, except the ACT (ABS 2007b).

As shown below the Review examined the characteristics of the households, including number of bedrooms, type of Internet connection as well as household income before determining they were unlikely to be homeless. There were another 960 people who rented an improvised dwellings and had no household members who worked full-time where it was determined that they were likely to be homeless, and were included in the Review homeless estimates.

Of the 57 people who were unlikely to be homeless 70% lived in dwellings with an Internet connection (either broadband, dial-up or an other connection) and 63% lived in dwellings with 2 or more bedrooms. For those who reported their household income, over a third (36%) reported incomes above $1,000 per week and a further 21% had nil or negative incomes. The people in this group tended to be younger, with 60% aged under 30 years. The Review concluded that these characteristics indicated that these people should not be considered to be homeless and they were removed from the estimates of homelessness.

	no.
$0–$49	62
$50–$74	111
$75–$99	144
$100–$119	134
$120–$139	134
$140–$159	63
$160–$179	51
$180–$199	52
$200–$224	45
$225–$249	24
$250–$274	17
$275–$299	12
$300–$349	24
$350–$449	16
$450–$549	6
$550+	11
Not stated	111
Total	1,017
(a) Cells in this table have been randomly adjusted to avoid the release of confidential data.

FUTURE CENSUS IDENTIFICATION

The ABS will seek advice from the Homelessness Statistics Reference Group and explore the groups discussed above in more detail. The ABS will then determine if there are other characteristics, collected in the Census of Population and Housing, which could be used to further refine these populations and how the data should be separately presented alongside homeless estimates in the future.

For the 2011 Census the ABS will undertake additional analysis to identify geographic areas where construction workers may be generally concentrated or in mobile accommodation and following construction programs. The ABS will consider ways to present information about 'construction workers' as part of those marginally housed alongside estimates of homelessness.

As a result of the feedback received through consultation and submission processes, the ABS plans further analysis to gain a statistical picture of residents living in improvised dwellings on their own properties as a result of natural disasters. The ABS will, along with advice from the Homelessness Statistics Reference Group seek expertise in developing methods for analysing areas affected by recent disaster events.

Previous Page

Next Page