3  Further Considerations

A number of considerations that are additional to the analysis above are discussed below. Accommodation and childcare costs are discussed in section 3.1 and the length of time that people tend to spend on social welfare benefits is discussed in section 3.2.

3.1  Accommodation and Childcare Costs

Due to the difficulty of accurately incorporating issues surrounding accommodation and childcare costs into the type of analysis above, these two factors have been treated separately in this paper. However, this should not be seen as reducing the importance of these factors. Accommodation costs have been identified as an important indicator of poverty and low living standards [Stephens et al, 2001]. Childcare costs are an important cost associated with working and have particular implications for the labour market participation of caregivers.

Issues surrounding accommodation and childcare costs are difficult to accurately incorporate into the analysis above for two reasons. Firstly, considerable variability exists among the families who face these costs. Accommodation costs, for example, vary widely among and within different regions of the country.[13] Families also differ in the degree to which they can access informal childcare provided by other family members (such as grandparents). Modelling these costs accurately requires information that is more detailed than that captured by the HES or during routine programme administration. Secondly, the nature of accommodation and childcare costs are such that they are both largely endogenous, which means that their consumption to some degree reflects other factors, such as the degree to which people choose to participate in the labour market.

The following discussion largely focuses on those people who currently receive assistance for accommodation, childcare, or both types of cost. However, the assumption that receipt of income support is an indicator of need is limited as take-up of assistance among the needy population is likely to be less than 100% [Atkinson, 1989, p. 197]. For instance, not all low-income families in need of accommodation assistance receive public assistance for their accommodation costs (Income Related Rentals and the Accommodation Supplement). Further, not all those who take-up assistance may not be in need, due to fraudulent behaviour and overpayments due to administrative error. Receipt of assistance may also reflect the potential for moral hazard associated with the provision of such assistance. For example, some people may alter their circumstances in ways viewed as undesirable by policymakers (e.g., place themselves in positions of need) in order to remain eligible for assistance [Sen, 1995, p. 11].[14]

The two main areas of publicly funded accommodation assistance are the Income Related Rentals and Accommodation Supplement programmes.[15] These programmes address different needs (with Income Related Rentals only being available to state house tenants while Accommodation Supplement payments are available for both public and private accommodation) and differ in terms of their generosity (with Income Related Rentals generally being the more generous of the two programmes). The Accommodation Supplement is the main form of accommodation assistance for non-beneficiaries.

For recipients of main benefits the Accommodation Supplement reduces by 25 cents per dollar for the first $80 of non-benefit income. Above the $80 threshold the Accommodation Supplement is not reduced until the recipient is no longer a beneficiary. The abatement then follows the rules for non-beneficiaries. For non-beneficiaries, the supplement abates at 25 per cent when income exceeds the rate of the applicable gross Invalids Benefit plus an add-on of $17.92.

Both forms of accommodation assistance tend to be targeted towards beneficiaries. As at March 2002, the largest numbers of people benefiting from Income Related Rentals were beneficiaries (as these people make up 88% of tenants in state houses) and the large majority (91%) of Accommodation Supplement recipients were beneficiaries [Roper et al, 2002, pp. 55-56]. Disincentives associated with the receipt of these programmes therefore mostly face beneficiaries.

Further, a majority (56%) of the recipients of the Accommodation Supplement were in families without children (of this group 89% were single and 11% were in partnered relationships). A large majority (81%) of the families with dependent children who were receiving the Accommodation Supplement were sole parents [Roper et al, 2002, p. 56]. The majority of the families that face any disincentives associated with the receipt of the Accommodation Supplement are childless. Of the families with children who face these disincentives the majority are sole parents.

The Childcare Subsidy provides financial assistance to low-income families with a dependent child under the age of five to obtain access to childcare services. The OSCAR Subsidy helps low-income families to pay for before and after school programmes and school holiday programmes for children aged five to 13.[16]

As with accommodation assistance, the majority of childcare assistance goes to recipients of main benefits. At August 2002 the majority (61%) of caregivers receiving the Childcare Subsidy were in families who also received main forms of income assistance and of these families the majority (59%) were Domestic Purposes Benefit recipients and a small proportion (2%) were Unemployment Benefit recipients. Disincentives associated with the receipt of these programmes therefore mostly face beneficiaries (and Domestic Purposes Beneficiaries in particular). The majority of families who receive childcare assistance do so for formal childcare arrangements.[17]

Overall the disincentives associated with the Accommodation Supplement are likely to mostly face childless beneficiaries but a significant proportion of these recipients are also sole parents (particularly Domestic Purposes Benefit recipients), who are the group that account for the majority of (and thus may mostly face any disincentives associated with) childcare assistance.

3.2  The Time People Spend on Social Assistance

The probability of a person’s spell on a social welfare benefit ending decreases as benefit duration increases. This could be due to a wide range of factors, such as reservations employers may have about hiring the long-term unemployed workers, discouragement that may arise when a person has been unable to move off a benefit for a long period, or a composition effect (where, as the length of time on benefit increases, people with low probabilities of employment account for greater proportions of those who receive assistance) [Wilson, 1999, p. 66; Wilson, 2002, p. 48].

Wilson [1999] examines administrative data on the duration of receipt of and numbers of spells on main working-aged social welfare benefits (excluding supplementary benefits and New Zealand Superannuation) between 1993 and 1998 for a cohort of around 250,000 people who were granted a working aged benefit in 1993. For the majority (an estimated 54%) of recipients the duration of receipt of the first observed spell on a benefit was less than 20 weeks. For an estimated 79% of recipients the duration of receipt for the first spell on a benefit was less than one year. Further, for an estimated 93% of recipients the duration of receipt for the first spell on a benefit was less than three years. An estimated 4% of recipients spent at least five years on their first spell on a benefit [Wilson, 1999, p. 66].

The duration of the first spell on a benefit varied among the benefit types. It was estimated that approximately 3% of Unemployment Benefit recipients spent all of the five years from 1993 to 1998 receiving a benefit. In contrast, it was estimated that approximately 26% of Domestic Purposes Beneficiaries, 39% of Widows’ Benefit recipients, 58% of Invalids’ Benefit recipients, and 15% of Sickness Benefit recipients spent all of the five years receiving a benefit [Wilson, 1999, p. 68].

For those people who transfer to another benefit or return to a benefit after some period off a benefit, the duration of the first spell on a benefit clearly understates the length of time that they spend on a benefit. In terms of the total time on a benefit, an estimated 8% of the cohort spent all of the five-year period on a benefit, 33% spent three or more of the five years on a benefit, and 62% spent at least one of the five years on a benefit [Wilson, 1999, p. 67].

An alternative view on the duration of benefit receipt is provided by comparing the number of people in the cohort that received a benefit at different points of time. Of those people who entered a benefit in 1993, an estimated 47% of Sickness Benefit recipients, 52% of Widows’ Benefit recipients, 57% of Domestic Purposes Benefit recipients, and 69% of Invalids’ Benefit recipients who entered a benefit in 1993 were receiving a benefit (whether continuously or not) at the end of the five-year period. Unemployment Benefit recipients were the least likely to have remained on or be back on benefit at the end of the five-year period. However, due to variations in seasonal employment, receipt of the Unemployment Benefit was strongly cyclical [Wilson, 1999, p. 71].