Child Poverty Action Group Incorporated v Attorney-General HC Wellington CIV-2009-404-273 [2011] NZHC 1228 (25 October 2011)

	Home \| Databases \| WorldLII \| Search \| Feedback High Court of New Zealand Decisions

Introduction ........................................................................................................................................................ [1] The 31 March 2004 Cabinet Paper.................................................................................................................. [14]

2005 amendments ............................................................................................................................................. [26]

Terms of the legislative provisions under challenge ....................................................................................... [33] Relevant human rights provisions ................................................................................................................... [43] International treaties/covenant obligations .................................................................................................... [47] The Tribunal’s decision .................................................................................................................................... [54] What is required to constitute discrimination? .............................................................................................. [56] How to identify the comparator group............................................................................................................ [85] Does the different treatment constitute discrimination? ............................................................................. [104] Section 5 – The issues on justification ........................................................................................................... [130] Deference ......................................................................................................................................................... [133] What the limitation on the right is measured against .................................................................................. [147] CPAG’s criticisms of the IWTC and WFF package ..................................................................................... [167] Objectives sufficiently important to justify curtailment of the right? ........................................................ [184] Is there a rational connection between exclusion of beneficiaries and the identified objectives? ............ [194] Is the exclusion of beneficiaries more than is reasonably necessary? ......................................................... [201] Proportionality ................................................................................................................................................ [212] Conclusion ....................................................................................................................................................... [225]

[1] This appeal relates to a claim pursued originally before the Human Rights Review Tribunal (the Tribunal), to the effect that a component of the Working for Families (WFF) package of tax credits discriminates against those who are on State benefits, and in particular against the children of those families. One objective of the WFF package was to alleviate child poverty and the claim is that the exclusion of those on State benefits from the caregivers who qualify for a component of the WFF package discriminates against the children of those in receipt of State benefits, to their material disadvantage.

[2] Child poverty is recognised as a social ill with significant long-term adverse social and economic consequences. For a so-called developed economy and in a society that aspires to be “caring” (which in the formal sense is manifested in our accession to various international covenants on human rights1), New Zealand has a poor record on child poverty. One source of government statistics on child poverty, pleaded in the Second Amended Statement of Claim in December 2006, referred to a

Ministry of Social Development report of that year. That report identified that, as at

2004, 38 per cent of dependent children‟s living standards had been characterised by low living standards which included some in severe or significant hardship categories. That report also included the statistic that families enduring low living standards included 51 per cent of families whose main source of income was an income-tested benefit, compared with 17 per cent of families whose main source of income was market income.2 More recent data cited in evidence before the Tribunal suggests that the position has not significantly improved.

[3] Accordingly, a claim that steps being taken on this important front involve prohibited discrimination is to be taken seriously. Indeed, this challenge has been pursued with the commitment of significant resources. The Tribunal heard

1 Relevant international covenants to which New Zealand has acceded are reviewed in [47] to [53]

2 Second Amended Statement of Claim at [2] and [43], CPG.161.0002 and CPG.161.0010. Such statistics are measured by comparison with median household incomes, so are comparative and not by reference to any absolute measure, either domestically or internationally.

[4] The appellant, Child Poverty Action Group Incorporated (CPAG), is an incorporated society that was formed in 1994 to advocate for better informed social policy to support New Zealand children, in particular those living in poverty. CPAG conducts research, publishes information and lobbies for changes to government policy. It works with other child-focused organisations and has, among its management committee, Dr Susan St John, an Associate Professor in the Economics Department of Auckland University and Professor Innes Asher, Head of Paediatrics at Auckland University School of Medicine. Both of them gave evidence before the Tribunal.

[5] Since 2002, CPAG has pursued complaints that forms of State assistance to families with children that are not available to families where the parents are in receipt of benefits constitute prohibited discrimination. The present complaint was pursued in respect of measures introduced in the Income Tax Act 2004 and CPAG was granted legal representation by the Office of Human Rights Proceedings to pursue the case before the Tribunal. Those representation arrangements have continued for the appeal.

[6] The Crown challenged the Tribunal decision that CPAG had standing as a “complainant” but its status as a complainant was upheld.3 This Court subsequently found that the Crown had waived the time prescribed for bringing the appeal when it was inadvertently filed eight days late.4

[7] As to the nature of the present appeal, the parties are agreed that the appellate approach settled by the Supreme Court in Austin, Nichols & Co Inc v Stichting Lodestar should apply.5 This approach requires the Court to form its own view on the relevant issues, whilst having appropriate regard to the conclusions reached by

_{the Tribunal. Although the Tribunal heard extensive evidence, it does
not appear that}

4 _{Child Poverty Action Group Inc v
Attorney-General [2009] NZHC 740; (2009) 19 PRNZ 689 at [31].}

5 Austin, Nichols & Co Inc v Stichting Lodestar [2007] NZSC 103; [2008] 2 NZLR 141.

any material issues of credibility arose, and certainly statistical and opinion evidence is well able to be assessed by us without the benefit of seeing and hearing the witnesses.

[8] As to the scope of the appeal, CPAG‟s pleaded challenge focused on the provisions in the Income Tax Act 2004 by which the WFF package was initially implemented. Before any of those measures came into effect, a relatively significant amendment to the package was introduced in 2005. At the hearing before the Tribunal, the parties joined issue on the effect of the package as introduced, including the 2005 amendments. However, the Tribunal decided that it was constrained by the terms of the pleading, and also by the lack of complete submissions on the WFF package in its amended form, from ruling on the WFF

[9] Three of CPAG‟s grounds of appeal challenged the decision of the Tribunal not to deal with the scheme as it had come into force. It is accepted for the Crown that the appeal ought to deal with the scheme as introduced, which includes consideration of the 2005 amendments to it.

[10] In those circumstances, it is not necessary for us to decide whether it was in error for the Tribunal not to deal with the form of package including the 2005 amendments. It seems that nothing turns on whether CPAG can make out an error by the Tribunal in this regard for costs purposes, and nor is it likely to be applicable as any form of precedent for subsequent proceedings. It is sufficient that the appeal deal with the challenge to the scheme as it exists, given that the Court was certainly adequately informed on the existing scheme, and that there would be an artificiality in dealing with the arguments as they related to the WFF package in a form other than as it is operating.

[11] CPAG‟s challenge before the Tribunal extended to a parallel provision that was ss KD to AAA(8)(a) of the Income Tax Act 2004, excluding from those eligible

_{for a component of the WFF package, persons in receipt of compensation
payable}

under the Injury Prevention, Rehabilitation and Compensation Act 2001. That aspect of the challenge was not pursued on appeal.

[12] The formal relief sought before the Tribunal was a declaration pursuant to s 92J of the Human Rights Act 1993 (HRA) that the relevant provisions in the Income Tax Act 2004 “are inconsistent with the right to freedom from discrimination affirmed by s 19 of the New Zealand Bill of Rights Act 1990”.

[13] We will first review the genesis of the WFF package, and relate that to the statutory provisions that provided for it. We will then briefly describe the outcome of the Tribunal‟s decision, before listing the sequence of legal considerations in respect of which we have to settle on the tests to be applied. That leads to an application of those tests to the factual circumstances as we find them to be.

[14] The WFF package was introduced as a 2004 budget initiative. The major rationale for it was recorded in a 47 page Cabinet Paper dated 31 March 2004. It recorded that initiatives had begun in December 20027 when Cabinet had invited the Minister of Finance and Revenue and the Minister of Social Development and Employment to report on major reform of family income assistance (including child care assistance) and simplification of the benefit system. This work had expanded to

include analysis of the accommodation supplement, third tier hardship provisions and the invalid benefit.

make work pay by supporting families with dependent children, so that they are rewarded for their work effort. This involves better alignment of benefits and in-work support (including Family Income Assistance, Childcare Assistance and Accommodation Supplement) so that people are better off as a result of the work they do

7 CPAG‟s analysis also referred to earlier work: “Social Assistance Strategy: Goals and Work

Programme”, a paper for Cabinet Social Policy and Health Committee, SPH (00) 128,

15 September 2000 (CPG.054.0020) and a further paper: “Pathways to Opportunity: Social

ensure income adequacy, with a focus on low and middle income families with dependent children, to significantly address issues of poverty, especially child poverty. The package also addresses housing affordability problems by responding to the increased cost of private housing for low income people, and

achieve a social assistance system that supports people into work, by making sure that people get the assistance they are entitled to, when they should, and with delivery that supports people into employment. This involves steps to streamline the social assistance system so that it is easier for people to understand and access, and initiatives to improve take-up and enhance the effectiveness of delivery.

[16] The Cabinet Paper proposed staged introduction of various components of the WFF package, on a number of dates between 1 October 2004 and 1 April 2008. The components of the package on each of the dates of introduction had discrete objectives, with those on 1 October 2004, 1 October 2005 and 1 April 2006 all having the objective of “strengthening the returns from work”. In contrast, the initiatives to be introduced on 1 April 2005 and 1 April 2007 were “improving income adequacy”. Whilst the bullet points for the initiatives on these various dates did not make any reference to addressing issues of poverty or child poverty, it is fair to infer that the initiatives listed for the purposes of improving income adequacy were the ones that would address that aspect of the overall objectives. For instance, the key initiatives on 1 April 2005 included increasing family support rates, introducing a new main benefit rate structure for families with children, and increasing orphan and unsupported child benefits and foster care allowances. The

1 April 2007 initiatives constituted a further increase of family support rates.

[17] The final milestone, on 1 April 2008, had the objective of “maintaining income adequacy” with the initiative being to protect payments against the effects of inflation by legislating for a policy of periodic adjustments to family support rates and thresholds, as well as child care assistance thresholds and other components.

...using a poverty measure of 60% of the median household income, there is expected to be a 30% reduction in child poverty by 2007/08. Using a 50% measure, the expected reduction is 70%.

[19] The paper recognised a need to bring social assistance systems up to date so that they no longer act as a barrier to people moving from benefits to employment, and that they actively support families in work. The policy objectives also recognised that child poverty hampers long-term economic performance as well as resulting in poor social outcomes.

[20] The Cabinet Paper did not address the conflict that could arise in advancing those two objectives. This would arise because maintaining a gap between the net income available to beneficiaries, and the net income for the lowest paid employed parents, was perceived as necessary to create the incentive for people to move off benefits and into work. Maintaining such a gap potentially limits the extent of assistance that could be provided to beneficiary parents.

(a) family income assistance and the in-work payment initiatives; (b) child care assistance improvements;

[22] The Cabinet Paper reported under the heading “Human Rights Implications”

1994 and the Social Security Act 1964 currently contain provisions that discriminate on the grounds of family status, employment status,

Acts to implement the changes proposed for the Working for Families package will raise New Zealand Bill of Rights issues and any continued or new discrimination will need to be justified.

78 Before the Working for Families Package is introduced the Ministry of Justice will vet the Bill for compliance with the New Zealand Bill of Rights 1990. MSD and IRD will work closely with the Ministry of Justice to provide justifications for any continuing or new discriminatory provisions that may be contained in the Working for Families Bill. However, both Acts currently discriminate on the grounds of sexual orientation, and the Social Security Act discriminates on the grounds of sex, and in neither case is it likely that a justification for either discriminatory provisions will be accepted. The package does not introduce any additional discrimination on these grounds. It is therefore likely that a Section

[23] In the Regulatory Impact Statement appended to the Cabinet Paper, in reflecting on the magnitude of the problem and the need for government action, it observed:11

209 At any one time, more than one quarter of all dependent children in New Zealand are supported by benefits. Families have a higher likelihood of restricted living standards than single people or childless couples, and beneficiary families are more likely to have restricted living standards than other families. There is increasing international evidence that the negative effects of poverty on children, particularly younger children, intensify the longer a family is poor.

[24] The WFF package was announced in the Budget on 27 May 2004, and on the same day the legislation effecting it was introduced and passed. The Bills containing the measures were not referred to a Select Committee. CPAG criticised the absence of that step in the Parliamentary process, as well as the absence of any Green Paper or White Paper to canvass for views about the proposed changes.

[25] The Attorney-General did report to Parliament pursuant to s 7 of the New Zealand Bill of Rights Act 1990 (BORA) in relation to the Bill. That report concluded that the effect of the Bill in treating same sex couples differently from opposite sex couples appeared to be inconsistent with s 19(1) of BORA and did not appear to be a justified limitation under s 5 of that Act. The Attorney-General‟s

_{report did not raise issues relating to employment status
discrimination. However,}

CPAG had raised concerns of this type as early as October 2002. The Tribunal commented on the lack of any consideration of discrimination on the ground of employment status in the Attorney-General‟s report as “surprising” and “unfortunate”.12 We were similarly troubled by the absence of any analysis of the potential discrimination, particularly in light of the commitments made by signing international instruments such as the International Covenant on Economic, Social

[26] The 2004 initiatives had been costed at approximately $1.14 billion per annum once fully implemented from 2007. In the course of 2005, the government became aware of the potential availability of significant further funding that could be committed to income assistance or tax relief.

[27] Prior to any of the measures that had been enacted in 2004 coming into effect, the government decided to extend the In Work Tax Credit (IWTC) component of the WFF package to higher income earners. This would substantially increase the level of earnings at which taxpayers would qualify for the IWTC, and also raise the levels at which the IWTC would gradually abate. Under the 2004 WFF package, the IWTC would begin to abate at $27,500 per annum in income, and thereafter there would be a 30 per cent abatement rate. Under the 2005 alteration, the threshold became $35,000 and thereafter abated at 20 per cent. The altered abatement rates were such that a family with three dependent children would only lose the last of the IWTC at an annual income of $101,000.

[28] As the case had been argued before the Tribunal, and initially before us, there were no cabinet papers or other departmental research documents analysing the policy justifications, or implications, of the 2005 change to the WFF package. In reviewing the facts from the Crown‟s perspective, Ms Foster said that this was

_{because officials did not give advice because the amendments to the
scheme were “a}

manifesto issue” by which we took her to mean that it was a party political initiative in anticipation of the 2005 General Election. She characterised the change as one driven by the then Minister of Finance, Dr Cullen, who was, in her terms, “very hands on” in pursuing the 2005 amendment.

[29] However, in response to questions from us, we were provided on the last day of the hearing with two documents that had been disclosed on discovery by the Attorney-General, but not previously included in the documents either before the Tribunal, or in the bundle for the Court.

[30] The first of those documents, dated 27 October 2005, was prepared by Mr Bryan Perry, a principal adviser with the Ministry of Social Development. The document was described as an assessment of the likely impact on income poverty of the tax relief package proposed by the Labour Party in that year‟s election campaign. The summary at the outset was as follows:14

2 The short answer to the impact question is that in contrast to the WFF package which is expected to reduce income poverty by a significant amount, the proposed [tax relief] package will have no impact on income poverty measured using constant value thresholds, and only a small upward impact when using relative-to- contemporary median thresholds. The difference in impact between the two packages arises because of the quite different income ranges that each package targets.

[31] The second document provided to the Court was an undated Treasury report that seems most likely to have been produced in late October or November 2005. It repeated the observation from Mr Perry‟s paper that the proposed tax reform package would have no impact on income poverty measured using constant value thresholds, and a small upwards impact when using relative-to-contemporary median thresholds. It observed that no families below the standard poverty threshold would receive any extra assistance, but also that the very large proportionate poverty reduction expected from the WFF package would not be compromised by the impact on the median of

2005, Dr Cullen described it as a Bill that delivered on “key election pledges”, one of those being to extend the WFF package. He described that part of the measure in the following terms:

1 April next year as part of the phased implementation of that package. As a result of the changes in the bill, about 100,000 working families who were to

60,000 families will become eligible for family assistance under the Working for Families programme.

assistance begins to abate, from $27,500 – actually, at the moment it is still

$20,000 – to $35,000, and by reducing the rate at which assistance abates for income that is over the new threshold, from 30 per cent to 20 per cent,

[33] The structure of the relevant provisions in the Income Tax Act 2004 was retained in the subsequent enactment in the Income Tax Act 2007 and it is convenient to describe the relevant provisions by reference to the latter statute, which continued to apply.

[34] There have been a number of somewhat confusing changes in the names given to various components of the WFF package. For the sake of simplicity, we will adopt the current names, and where referring to the differently named predecessors of the various components, will still give them the names that are current now, even although they were referred to by different names at various points in time.

[35] The contentious part of the WFF package for present purposes is the IWTC (ie In Work Tax Credit),15 the objective of which was to supplement the income of those who were in work, with one consequence being that a positive margin was maintained between the amounts earned by those at and towards the bottom of the

[36] Part M of the Income Tax Act 2007 provides for tax credits paid in cash. The IWTC is provided for in ss MD4 to MD10. Section MD4 provides that a person is entitled to an IWTC for a child if, in respect of an entitlement period, the person meets the five requirements in ss MD5 to MD9.

[37] Sections MD5 to MD9 are headed respectively as the first to fifth requirements for entitlement. These sections provide that the person has to be:

[38] The concept of a full-time earner is the subject of its own definition in s MA7, which provides that it is a person who is employed for 20 hours or more per week if single, or a total of 30 hours‟ work per week between a couple. “Income-tested benefit” is separately defined in s YA 1 of the 2007 Act. That term means the seven forms of primary benefit payable under the Social Security Act

1964, including domestic purposes and unemployment benefits. The requirement that recipients of the IWTC not be receiving an income-tested benefit is referred to as “the off-benefit rule”.

(FTC), payable to all principal caregivers, irrespective of their employment status.16

That was an adaptation of relatively long-standing family support, paid per dependent child.

16 Previously called the Family Support Credit, provided for in s KD2(3) of the Income Tax Act

[40] In addition, for those in work but earning at low levels, a supplement to their earnings was also payable by way of the Minimum Family Tax Credit (MFTC).17 It is only payable for the weeks in which a person is a full-time earner.18

[41] In the evidence before the Tribunal, Donald Gray, the Deputy Chief Executive, Social Development Policy and Knowledge in the Ministry of Social Development, provided a table to reflect the different application of these various forms of benefit for both a sole parent and a couple, contrasting the situations in which a sole parent worked for 20 hours (or a couple for 30 hours) with a situation where a sole parent or a couple were not working and claiming the domestic purposes or unemployment benefits. The chart in evidence described the various tax credits by their previous names, and in the form below the current names for the

_{benefits are added in italics as they are being referred to in this
judgment.}

	Sole Parent				Couple
Tax credit and benefit examples	Working 20 hours			Not working	Working 30 hours			Not working	6 hours work
As at April 2007	DPB claimed (net)	Wages (gross)	IWP claimed (net)	DPB claimed (net)	UB claimed (net)	Wages (gross)	IWP claimed (net)	UB claimed (net)	IB claimed (net)
Benefit abated for income	194.15		0.00	255.65	117.21		0.00	297.46	371.84
Family Support Credit (2 children under 12) Family Tax Credit	139.00		139.00	139.00	139.00		139.00	139.00	139.00
Family Tax Credit Minimum Family Tax Credit	0.00		158.29		0.00		60.13	0.00	0
In Work Payment In Work Tax Credit	0.00		60.00		0.00		60.00	0.00	0
Support from the State	333.15		357.29	394.65	256.21		259.13	436.46	510.84
Income from employment (secondary tax when on benefit)	174.83	225.00	185.79		262.24	337.50	282.49		52.45
Net total income	507.98		543.08	394.65	518.45		541.62	436.46	563.29
Gain from moving off benefit			35.10				23.17

[42] The components of the WFF package were designed so that the MFTC was means tested and only available to those on incomes at the lower end of the scale. Thereafter, the FTC abated from $35,000 in family income, until earners became

_{ineligible at salaries of $56,000. Thereafter, the IWTC began abating
from $56,000}

17 Previously called the Family Tax Credit, and provided for under s KD3 of the Income Tax Act

and, depending on the number of dependent children being cared for, continued to be available until the family‟s income reached substantially higher levels. For two dependent children, it cut out at $86,000, for three, $101,000 and for four at

$119,000. The evidence included an extract from a WFF tax credits registration document, which set out a schedule of the rates of abatement. A copy of that schedule is Annexure A to our judgment.

[43] Section 21 of the HRA lists 13 forms of discrimination that are prohibited under the HRA. Relevantly to these proceedings, s 21(1)(k) prohibits discrimination on the ground of employment status, in the following terms:

(1) For the purposes of this Act, the prohibited grounds of discrimination are –

(ii) being a recipient of a benefit under the Social Security Act 1964 or an entitlement under the Injury Prevention, Rehabilitation, and Compensation Act

[44] A further provision in the HRA effected an amendment to BORA, by introducing the current form of s 19 of BORA as follows:

(1) Everyone has the right to freedom from discrimination on the grounds of discrimination in the Human Rights Act 1993.

(2) Measures taken in good faith for the purpose of assisting or advancing persons or groups of persons disadvantaged because of discrimination that is unlawful by virtue of Part 2 of the Human Rights Act 1993 do not constitute discrimination.

[45] The extent of rights and freedoms recognised in BORA is subject to s 5 of that Act, acknowledging “justified limitations”. It is in the following terms:

Subject to section 4 of this Bill of Rights, the rights and freedoms contained in this Bill of Rights may be subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.

[46] Most recently in terms of the relevant human rights provisions, the HRA was amended in 2001 by the insertion of a new Part 1A which introduced a limited jurisdiction in respect of discrimination by government or persons when performing a public function or duty conferred or imposed on that person by or pursuant to law. Part 1A provides for the prospect of a declaration by the Tribunal, or by the Courts on appeal from the Tribunal, to the effect that an act or omission by government, including provisions in statutes, is inconsistent with s 19 of BORA. At an earlier stage of the present proceedings, Miller J characterised the effect of Part 1A as

By admitting claims of discrimination in respect of enactments, however, Parliament has made available a cause of action and a forum in which such claims may be publicised and to some degree vindicated, if not actually remedied. Armed with a declaration, the plaintiff may press its case for a remedy in the community and in the legislature.

[47] New Zealand is a signatory to a number of conventions and international covenants that articulate standards of conduct for States in relation to their citizens and others resident in countries that become signatories to them. CPAG cited the provisions of the International Covenant on Civil and Political Rights (ICCPR), ICESCR and UNCROC as relevant to the interpretation and application of New Zealand‟s domestic legislation protecting human rights.

[48] At the core of such international documents are broadly expressed standards for State parties. For instance, the ICCPR provides:20

Each State Party to the present Covenant undertakes to respect and to ensure to all individuals within its territory and subject to its jurisdiction the rights recognized in the present Covenant, without distinction of any kind, such as

19 _{Attorney-General v Human Rights Review
Tribunal [2006] NZHC 1661; (2006) 18 PRNZ 295 at [65].}

race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status.

[49] One of the primary general commitments expressed in the ICESCR is as follows:21

The States Parties to the present Covenant undertake to guarantee that the rights enunciated in the present Covenant will be exercised without discrimination of any kind as to race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status.

[50] The long title to BORA acknowledges one of its purposes is to affirm New Zealand‟s commitment to the ICCPR. The Crown submissions acknowledged that interpretations of the ICCPR by the United Nations Human Rights‟ Committee (HRC) are to be treated as having “considerable persuasive authority”.22

Accordingly, the approach adopted by the HRC may influence the application of the provisions of BORA in the current context.

[51] ICESCR also contains other commitments on State Parties to assist families in aspirational terms such as in recognising the right of everyone to an adequate standard of living for them and their family, including adequate food, clothing and housing, and the continuous improvement of living conditions.23

[52] Except to any extent that such commitments are reproduced in domestic legislation, they do not create obligations that are enforceable in judicial proceedings in New Zealand. However, they operate as an influence on the approach to interpretation of human rights‟ provisions in New Zealand statutes.24

[53] Although we are mindful of the international commitments made in the various covenants, we have not found it necessary to rely on any of the content that was drawn to our attention, in settling on the appropriate interpretation of the

22 R v Goodwin (No 2) [1993] 2 NZLR 390 at 393, Nicholls v Registrar of the Court of Appeal

24 See Yee v Minister of Immigration [2010] 1 NZLR 104 at [24] re provisions of the Immigration

[54] The Tribunal‟s decision was that the exclusion of those on benefits from the IWTC, as reflected in the form it was introduced in 2004, amounted to discrimination on a prohibited ground, for the purposes of s 19 of BORA. However, the Tribunal‟s evaluation under s 5 of that Act resulted in a finding that the extent of discrimination involved was justified, so that CPAG was not entitled to the declaration it sought that the statutory provision involved was inconsistent with BORA. It is from that outcome that CPAG has appealed. In response, the Crown argued that the provision does not constitute discrimination for the purposes of s 19 of BORA, but that if it does, then it is discrimination that is justified in terms of the s 5 analysis.

[55] To the extent that the Tribunal‟s reasoning is relevant to various of the issues that we analyse, it is more appropriate to refer to those passages as we address each topic. We intend no disrespect to the thorough review of all issues undertaken by the Tribunal in describing its decision in cursory terms. It is unnecessary at this stage to review the Tribunal‟s decision in more detail, in part because the argument before us had moved on, in the sense that it focused on the scheme including the 2005 amendments, which was not determined by the Tribunal.

[56] The parties took different approaches to what will constitute discrimination for the purposes of s 19 of BORA. It is clear that it means more than different treatment on a prohibited ground, but the Crown argued for a substantive requirement for disadvantageous effect, in circumstances where it perpetuated historic disadvantage or is predicated on prejudicial stereotypes. In contrast, CPAG submitted that proof of discrimination was a relatively straightforward test, involving reliance on a distinction accompanied by a disadvantage that was more than trivial.

[57] Similarly opposed views were argued before the Tribunal, which preferred the approach contended for by CPAG. The Tribunal recognised as legitimate a concern for the Crown that if relevant discrimination could be made out for the

purposes of s 19, even where the consequences of differentiation were innocuous, then the Crown would be put to considerable and potentially unwarranted cost. This would arguably arise in having to defend differentiations that should not require the commitment of resources, to contest a claim of discrimination and then seeking to establish that any limitation on the right to be free from the particular form of discrimination was justified under s 5.

[58] That point was also pressed before us by Ms Gwyn, and we agree that it is a valid concern. There may well be forms of discrimination that constitute a distinction between otherwise comparable groups, with one group purportedly being disadvantaged, even although the context in which the distinction arises makes it relatively innocuous. Further, for a claimant, the prospect of a finding of breach of s 19 could excite an anticipation of some vindication when, on any balanced view, that is not warranted.

[59] The issue is what standard ought to be applied in undertaking the threshold analysis as to the existence of discrimination. This is to be assessed having regard to the statutory structure of a two stage test of whether (under s 19) a practice constitutes discrimination on a prohibited ground and, if so, then (under s 5) whether it is demonstrably justified.

[60] The Crown urged adoption of a substantive meaning of discrimination, as advocated by Professor Huscroft in Professor Paul Rishworth‟s text on BORA.25

Professor Huscroft‟s concern is that a neutral interpretation of the concept of discrimination “may end up trivialising it”. He suggested that discrimination is best understood as involving invidious treatment as it is understood in common parlance.

[61] The Crown acknowledged that the Butlers‟ text on BORA recommended a lower threshold, with discrimination arising simply from different treatment.26

CPAG‟s submissions criticised the Crown for portraying these different approaches

as “substantive” versus “formal”, when those were not labels used by the respective

_{authors. We consider that it is less helpful to treat the two
approaches as starkly}

25 _{Rishworth (ed) The New Zealand Bill of
Rights (OUP, Melbourne, 2003) at 376.}

26 Andrew Butler and Petra Butler The New Zealand Bill of Rights Act: A Commentary

defined alternatives, and more helpful to see them as points towards opposing ends of a continuum. As is invariably the case, context is everything. The assessment of what will constitute discrimination for the purposes of s 19 should involve consideration of the real effects of what is complained of. That is in one sense a “substantive” analysis, but it does not mean that certain forms of disadvantage are, by definition, insufficient.

[62] The Crown cited a 1989 General Comment of the HRC on non- discrimination. That stated that the term “discrimination” should be understood to imply a distinction, exclusion or restriction based on grounds of distinction which “has the purpose or effect” of impairing the enjoyment of rights available to others. Given the considerable persuasive authority of such determinations,27 it was argued for the Crown that any analysis into both purposes and effects must contemplate a substantive analysis of the impact of the alleged discrimination complained of. That is consistent with the view that the impact of the alleged discrimination must be

[63] The Crown‟s formulation that discrimination constitutes a distinction that is disadvantageous in that it perpetuates historic disadvantage, or is predicated on prejudicial stereotypes, reflects the approach of the Supreme Court of Canada.28 The right not to be discriminated against is expressed differently in s 15(1) of the Canadian Charter of Rights and Freedoms. That states:

Every individual is equal before and under the law and has the right to the equal protection and equal benefit of the law without discrimination and, in particular, without discrimination based on race, national or ethnic origin, colour, religion, sex, age or mental or physical disability.

[64] The New Zealand Legislature did not adopt this approach. Rather, s 19 of BORA confirms the right to be free from discrimination based on the defined categories that are prohibited under the HRA. That amounts to a materially different

_{context from the inquiry necessary under the Canadian
provision,}29 _{which will}

28 See, for example, Withler v Canada (Attorney-General) 2011 SCC 12, [2011] 1 SCR 396 at

generally begin with an analysis of whether the form of discrimination complained of falls within one of the grounds cited as examples in s 15(2), or an analogous ground. When the range of grounds is potentially open-ended, a different rationale arises for defining common characteristics that will be required to qualify the form of differentiation in issue as prohibited discrimination.

[65] The Crown also argued that the right to be free from discrimination is a negative right in the sense that it does not create obligations to remedy the inequality. For instance, in the present circumstances, making out the discrimination does not trigger an obligation to extend the IWTC to those previously excluded from it. The negative nature of the right does not influence our approach to what constitutes discrimination in the first place. CPAG‟s case is not overreaching, in that the relief sought is confined to a declaration of inconsistency. Miller J‟s earlier judgment recognising the entitlement to seek such relief referred to the potential utility of it in

[66] Another argument for the Crown was that the enumerated grounds of discrimination in s 21 of the HRA should be given a ranking of importance and that employment status is one of the least “suspect” grounds of discrimination. This approach reflected one that has applied in decisions in the United Kingdom where it has been recognised that not all possible grounds of discrimination are equally

potent.31 The Crown cited the lack of permanence about employment status and the

prospect for a person to change it, which does not arise for immutable grounds such as age and gender, as rendering discrimination on the basis of employment status in some way less serious.

[67] We are not persuaded that there should be any ranking by reference to the inherent importance of various of the enumerated grounds of discrimination. The notion of a hierarchy of grounds reflects the approach in other jurisdictions where any list of potential grounds of discrimination is open-ended, leading to the prospect of an evaluation on the relative significance of a ground of discrimination alleged in

_{a particular case. In contrast, the enumerated grounds in s 21 of the
HRA are all to}

be taken as creating the jurisdiction for claims of prohibited discrimination and the status of a claim invoking any of the grounds is to be assessed on its own merits.

[68] The issue of what constitutes discrimination for the purposes of s 19 was considered by the Court of Appeal in Quilter v Attorney-General.32 That case involved claims by lesbian couples in long-term relationships that, if the Marriage Act 1955 was interpreted to exclude marriage licences being granted to same-sex couples, then such restriction would be discrimination contrary to s 19 of BORA. The Full Court was unanimous in the outcome, namely that the Marriage Act 1955

was to be interpreted as providing only for marriage between persons of different sexes. However, the reasoning in the separate judgments reflected very different approaches to what would constitute discrimination.

Discrimination generally is understood to involve differentiation by reference to a particular characteristic (classification) which characteristic does not justify the difference. Justification for differences frequently will be found in social policy resting on community values.

[70] That approach contemplates an analysis of potential discrimination that evaluates whether the differentiation between the purportedly similar groups was justified. On Gault J‟s approach, that was part of the definition of discrimination and was to be considered before any possible application of s 5 of BORA.

...I would prefer to define the right (that is to be free from discrimination) with the purpose of anti-discrimination laws in mind, and then consider whether any suggested limitation is justified or otherwise lawful rather than circumscribe the content of the right at the outset. This accords more with the spirit and purpose of the Bill of Rights. In this kind of case it is better conceptually to start with a more widely-defined right and legitimise or justify a restriction if appropriate, than to start with a more restricted right. Of course any such restriction or legitimation will, as is its purpose, pro tanto abrogate the right; but if restrictions which may be legitimate or justified in some circumstances are built into the right itself the risk is that they will apply in other circumstances when they are not legitimised or justified.

[72] It was submitted for the Crown that this passage from Tipping J‟s judgment is frequently relied on in misconstruing the test his Honour intended to pose. The argument for the Crown contrasted the observation cited above with earlier passages in Tipping J‟s judgment that required an evaluation of the subject matter of the circumstances in which discrimination is alleged to have occurred:35

....of course difference of treatment will not necessarily in itself amount to discrimination; and not all discrimination will be unlawful. In considering whether there is discrimination, it is necessary to define two things: first, the subject-matter of and, second, the basis for the alleged discrimination.

[73] That contemplates an assessment of the context in which the alleged discrimination arises and the rationale for the distinction that is relied on by the alleged discriminator. That is not a matter of justification, but rather an evaluation of the context in which the differentiation occurs.

[74] The approach of Thomas J in Quilter is reflected in the following passage:36

... is not whether there is a distinction but whether the distinction which exists is based on the personal characteristics of the individual or group and has the effect of imposing burdens, obligations, or disadvantages on that individual or group which are not imposed on others.

[75] In the context of that judgment, it is implicit that Thomas J had in mind burdens, obligations or disadvantages that had some real or material effect, not merely notional or theoretical.

[76] More recently in Ministry of Health v Atkinson, in another appeal from the Tribunal to this Court, the Court had to rule on comparable differences of approach urged in relation to what constitutes discrimination:37

Section 20L(2) of the HRA gives an indication. It specifies a two-stage approach, first a consideration of discrimination and secondly, if there is discrimination, a consideration of whether s 5 applies. Section 21 of the HRA sets out “prohibited grounds of discrimination” without qualification. No evaluative process is indicated. It goes against its unqualified language for there to be an enquiry into the quality and nature of the prohibited discrimination at the s 19 stage, after a prima facie breach is established.

[77] With respect, if that approach was taken to suggest that any degree of disadvantage suffered by one group, distinguished from another on one of the prohibited grounds for discrimination would be sufficient to trigger s 19, then it may enable complainants to make out the first stage of the inquiry under s 19 at too low a threshold. On the approach in Atkinson, the essence is what constitutes a “prima

... It is necessary if discrimination is to be established for disadvantage to be

[78] A relatively early indication that the New Zealand anti-discrimination provisions would not be limited to traditionally disadvantaged groups arose in Northern Regional Health Authority v Human Rights Commission.39 In that case, Cartwright J rejected the prospect that the focus should be on remedying adverse outcomes for traditionally disadvantaged groups. Such a constraint would be inconsistent with the recognition in that case that a discriminatory effect may be recognised as against any group, and that an historical analysis was not necessarily required.40

[79] Ms Joychild disputed that adoption of the threshold for discrimination that CPAG contended for might lead to a flood of claims of discrimination against the government. She submitted that the Human Rights Commission, the Office of Human Rights Proceedings and the Tribunal have adopted the same approach to discrimination as contended for on behalf of CPAG, without there being any flood of claims. She attributed this to procedural constraints that she characterised as an effective filter and protector for the government against unmeritorious claims. That procedure involves complaints of non-compliance with Part 1A of the HRA first being submitted to the Human Rights Commission, which has a dispute resolution

function.41 Persons wishing to bring civil proceedings that arise from a complaint

must have pursued the matter first with the Commission.42 In practical terms, if a complainant wishes to have the support (including financial) of the Office of Human

39 _{Northern Regional Health Authority v
Human Rights Commission [1998] 2 NZLR 218 (HC).}

Rights Proceedings, then the complaint must have some basic credibility, as tested in that preliminary stage.

[80] Ms Joychild cited statistics that had been distilled from the annual reports of the Human Rights Commission showing that it had received approximately 3,000 complaints against the government under Part 1A, but only five of those have been substantively considered by the Tribunal. Accordingly, CPAG argued that adoption of the standard for identifying discrimination that it contends for will not open the floodgates to a substantial number of proceedings required to be defended by the Crown. We are inclined to agree.

[81] We also agree with the submissions for CPAG that the separation of the two issues addressed in ss 19 and 5 of BORA (and as reflected in s 20L of HRA) confirms that the inquiry at the s 19 stage as to the existence of relevant discrimination ought not to intrude on the second stage that addresses whether any relevant discrimination found to exist can be justified on the test under s 5. Accordingly, any evaluation of the “real” or “more than trivial” nature of the discrimination under s 19 should not intrude into matters of justification. Rather, a case-specific inquiry is required as to the materiality or sufficiency of the disadvantageous treatment (or, on the term used in Atkinson, “discriminatory

impact”43) when its adverse effects are assessed in the context in which it arises,

before a finding can be made that it constitutes discrimination for the purposes of s 19. It is not necessary that the differential treatment arise from prejudice or stereotyping, or that it perpetuates the disadvantage for an already disadvantaged group.

[82] This test we have settled on is not materially different from that applied by the Tribunal.44 However, unlike the Tribunal, we would not be inclined to reduce the requirement for a meaningful extent of disadvantageous treatment by relying on the terms of the rights protected by the ICCPR where rights are expressed as being

[83] The Tribunal considered that little would be gained by adding adjectives to the idea of “disadvantage”.45 However, if the concept of disadvantage was entirely unqualified, it would raise the prospect of theoretical, innocuous or de minimus disadvantages qualifying as prohibited discrimination, and that indeed would risk trivialising the right protected by s 19. We note that in another part of its decision, the Tribunal observed that some “real disadvantage...needs to be established”.46

[84] Further, we respectfully disagree that the s 19 threshold can be triggered at any level of discrimination on the basis that any such conduct should be capable of justification on a s 5 analysis. That would risk creating the additional burden for the alleged discriminator, when the inquiry ought properly not to advance beyond the s 19 stage. To the extent that this approach differs from that in Atkinson, where the Court preferred that “consideration of the qualitative nature of the discrimination” be

deferred until the “flexible balancing context of s 5”,47 then the analysis we propose

at the earlier, s 19, stage is the more appropriate to the disputed existence of discrimination in the present case.

[85] The next contested element of the legal tests to be applied is the identity of the group that is most appropriate as the comparator, for the purposes of measuring whether the group cast as complainants are discriminated against by the off-benefit rule.

[86] It was argued for the Crown that the comparator ought to be the group most closely resembling those complaining of discrimination, except for the characteristic on which the alleged discrimination depends. That would involve comparing two groups who are both in “full-time” employment (that being another of the qualifying criteria for the IWTC) where one is also on a benefit and therefore does not qualify for the IWTC, and the other group are in paid work but not on a benefit so they do

_{qualify for the IWTC. Statistics showed that, at the time of the
Tribunal hearing,}

approximately 1,270 families were in this first category. Counsel‟s assumption was that the number is likely to be more or less constant, although those making up the group are likely to change. Because the benefits they are on are income-tested (and therefore abate by the extent of earned income), it is inferred that the majority of people in this category choose to continue on the benefit because of uncertainty about the on-going availability of paid employment.

[87] CPAG rejects the notion that the comparator should necessarily be the next most similar group on any narrow analysis. Instead, CPAG urges that as a matter of logic and common sense, the comparison that is relevant for a complaint of discrimination here is between those who are on a benefit and not working, and those who are working and not on a benefit. Again, using the numbers from the time of the Tribunal hearing, this contrasted approximately 165,000 families who were dependent on benefits, against 180,000 families who were in paid work and not in receipt of a benefit, and who would qualify for the IWTC.

[88] Given this difference on what was perceived by the parties to be the identity of the comparator, we heard detailed submissions on what the test should be for identifying the relevant comparator in cases such as the present. However, the argument for the Crown sought to redefine the class who constituted the complainants, not the comparator. Both parties accepted that those with whom the complainants should be compared were those who qualified for the IWTC, ie those who were full-time earners and were off benefit. What the Crown disputed was CPAG‟s entitlement to advance a complaint for a group that failed to qualify on one requirement, whilst ignoring another disqualifying characteristic. However, it is not for the Crown to redefine the characteristics asserted on behalf of complainants. We do not therefore see this as a case involving claims of different comparators. Both parties agree on the comparator.

[89] Against the prospect that we are wrong in characterising the identity of the complainants and the appropriate comparator, we will briefly summarise the arguments put to us on what the test for a comparator should be.

We therefore accept the plaintiff‟s submission that the substantive reason for the differentiation which is at issue in this case is employment status – in the sense of either being in receipt of an income-tested benefit, or being in paid work. When it comes to assessing comparative disadvantage, we think that it is appropriate to look at the situation of those families who receive an income-tested benefit and who are not eligible for the IWTC as a result, as against those families who are in paid work and who are therefore eligible for the IWTC (within the parameters of the relevant thresholds and cut-offs that apply to the IWTC).

[91] In Air New Zealand v McAlister, the Supreme Court considered the approach to comparators under the Employment Relations Act 2000 (ERA).49 That case involved an Air New Zealand pilot who brought a personal grievance claim, alleging he had been discriminated against by reason of age, contrary to s 104 of the ERA. Mr McAlister had effectively been demoted on reaching the age of 60 because of provisions under United States law preventing pilots over the age of 60 being in control of large aircraft in that country.

[92] The Court of Appeal had held that the appropriate comparator was other pilots who could not fly to the United States, for example because their visas were not current, and that there had been no discrimination on the grounds of age. In contrast, a majority of the Supreme Court held that the comparator was between pilots otherwise similarly qualified, but younger in age.

[93] The judgment of Elias CJ and Blanchard J (concurred in by Wilson J) recognised the importance of the choice of a comparator, and the extent to which it can be influenced by context:50

In cases of alleged discrimination the choice of a comparator is often critical. We were referred to a number of decisions from senior courts in different jurisdictions which were said to provide guidance. For the most part, we did not find them especially helpful. Unless there are distinct similarities in the statutory scheme and in the type of discrimination which is being alleged, what is said in another jurisdiction about how to arrive at a comparator is of limited assistance. The task of a court is to select the comparator which best fits the statutory scheme in relation to the particular ground of discrimination which is in issue, taking full account of all facets of the scheme, including particularly any defences made available to the person against whom

discrimination is alleged. A comparator which is appropriate in one setting may produce a completely inapt result in another. It will certainly do so if it effectively deprives part of the statutory scheme of its operation.

[94] The importance of context is illustrated by their Honours‟ observation that the appropriate comparator may have been different, if Mr McAlister‟s personal grievance had been pursued under another sub-section of s 104 of the ERA. They accepted counsel‟s proposition that “the comparator must allow the surrounding

[95] Although in the minority as to the outcome, the observations on choice of the comparator from the judgment of Tipping J are instructive:52

... The approach of the Court to the comparator issue should be guided by the underlying purpose of anti-discrimination laws and the context in which the issue arises. Anti-discrimination laws are designed, as I have said, to prohibit employment and other relevant decisions from being influenced by any feature which amounts to a prohibited ground of discrimination. Exceptions allow what would otherwise be a discriminatory feature to be taken into account if there is good cause for doing so. A comparator is not appropriate if it artificially rules out discrimination at an early stage of the inquiry. By artificially I mean that the comparator chosen fails to reflect the policy of the legislation, which is to take a purposive and untechnical approach to whether there is what I will call prima facie discrimination, while allowing the alleged discriminator to justify that prima facie discrimination if the case comes within an exception.

... Subject to any applicable statutory provision, the most natural and appropriate comparator is likely to be a person in exactly the same circumstances as the complainant but without the feature which is said to have been the prohibited ground.

[96] The Crown relied on the second paragraph of the quotation from Tipping J cited in [95] above, and submitted that the comparator needs to be alike in all respects that are material to the issue in hand, other than on the prohibited ground of discrimination.

[97] Notwithstanding the Supreme Court‟s caution about the utility of the approach adopted in overseas jurisdictions where the statutory schemes have different features, both parties invited comparison with observations on how to

Canada. Of the United Kingdom decisions, the Crown relied on Shamoon v Chief Constable of the Royal Ulster Constabulary.53 Counsel for both parties also cited AL (Serbia) v Secretary of State for the Home Department54 and A v Secretary of State for the Home Department.55

[99] Those decisions illustrate that analysis of the appropriate comparator will reflect the context in which the impugned provision or conduct occurs. There is recognition that arguments over the appropriate comparator can be an “arid” exercise and overall, they do not support a requirement for the closest possible comparator necessarily being the most appropriate. Of the Canadian decisions, the more recent

ones move away from a search for the so-called “mirror comparator groups”.58

[100] The Crown characterised the Tribunal‟s finding of prima facie discrimination in respect of the wider complainant group as necessarily implying a finding that all beneficiaries ought to receive the IWTC, regardless of whether they met all of the other statutory criteria. However, that is a point relevant to the nature of the disadvantage caused by the off-benefit rule, and it does not render the definition of the complainants or the comparator groups wrong.

[101] If this was a dispute over the comparator, then our analysis would begin with the context of the complaint. It is of discrimination on the ground of employment status, the definition of which, in s 21(1)(k) of the HRA, extends to being a recipient of a benefit under the Social Security Act 1963. Therefore the relevant context is that discrimination is claimed to be occurring on behalf of recipients of such

_{benefits, by virtue of that status. Within this context, they are
contrasted with those}

53 Shamoon v Chief Constable of the Royal Ulster Constabulary [2003] UKHL 11, [2003] 2 All ER

54 AL (Serbia) v Secretary of State for the Home Department [2008] UKHL 42, [2008] 1 WLR

55 A v Secretary of State for the Home Department [2004] UKHL 56, [2005] 2 AC 68 at [53] per

57 _{Withler v Canada (Attorney-General)
2011 SCC 12, [2011] 1 SCR 396 at [55] to [60] and [63].}

58 For example, Withler at [55]-[60]. See also the comments of the Court in Atkinson at [82].

who are employed. The essence of the difference is that those on the other side of the line receive their income from earned wages.

[102] As to the relevance of the difference, CPAG‟s issue is that the IWTC is a component of the WFF package, one objective of which is to address child poverty. Within a scheme that includes that objective, one group of those who would benefit to the greatest extent from assistance arguably intended to address this objective has been excluded because of the source of the rest of their income. Therefore the relevant aspect of the difference is the source of income.

[103] Applying these tests of context and relevance, we would be satisfied that the appropriate comparator is between those who earn income from wages for sufficient hours to entitle them to IWTC, when compared with those who are excluded from qualifying for IWTC because they are in receipt of benefits. This latter group includes, as a subset of it, those who do work sufficient hours to otherwise qualify for IWTC but are disqualified because they continue to receive a benefit, abated to whatever extent is required in the periods when they receive earned income.

[104] When we first confronted the issues raised by CPAG‟s challenge to the WFF package, the members of the Court shared an intuitive doubt that the provision of additional funding to parents of dependent children who were under pressure in a range of other socio-economic respects would necessarily result in alleviating child poverty for those children. In terms of targeting the money given to parents, there is no way of enforcing a requirement that it be spent on the children.

[105] The evidence before the Tribunal did address this concern. An analysis for the Ministry of Social Development (MSD) by Susan Mayer had concluded that parental income makes a contribution to many other aspects of children‟s well-being, and that income gains have the potential to make a big difference in the lives of

primary vehicle for getting income to children to assist with their living standards is through income to their parents.60 Again, the Cabinet Paper stated:61

Income support is a key instrument for poverty alleviation and for improving living standards. Given the large investment through the Working for Families package, we would expect a significant reduction in measured income poverty.

[106] Whilst none of these sources establish the proposition empirically, we accept the practical reality that to create the opportunity for lessening poverty among children, it is necessary to provide funding that is identified as being for their benefit, to those caring for them. In any event, the risk of monies made available for improving the lot of children living in poverty being diverted by their caregivers arises in a similar way for caregivers dependent on benefits, as for caregivers earning income whose financial position is supplemented by the IWTC.

[107] The Tribunal considered the notion of disadvantage by evaluating the overall effect of the WFF package. Its decision cited a number of respects in which there would have been merit in extending to beneficiary families the components of the WFF package that were targeted only to families in employment. The Tribunal was satisfied that a lack of comparable gain could be sufficient.62 The Tribunal found that the WFF package as a whole, and the eligibility rules for the IWTC in particular, treated beneficiary families less favourably than it does families in work, and that constituted a disadvantage in a real and substantive way.63

[108] As to the existence of disadvantage, the Crown argued that legislation promoting a change in employment status is not necessarily, or even usually, invidious. Where the distinction was made to encourage people off benefits and into work, it arguably has advantages for those excluded, in the sense of an incentive to move from that group into the group that does qualify for IWTC. However, we are not persuaded that the incentive to move into employment applies either sufficiently compellingly or sufficiently broadly to neutralise the potential disadvantage that

_{exists. This is not a situation where s 19(2) of BORA applies to treat
measures taken}

in good faith to assist disadvantaged persons as not constituting discrimination because of that motive, when the measures might otherwise come within s 19(1).

[109] The Crown also argued that substantive disadvantage to those on benefits should not be recognised in the present context because they receive more financial support from the State overall than those who are not on benefits but who do qualify for support by means of the IWTC.64 The Crown‟s point was that whilst beneficiaries cannot receive payments in the form of IWTC, they do receive payments constituting a substitute for their absence of employed wages

(unemployment benefit), or recognising other circumstances precluding their obtaining paid employment (domestic purposes benefit (DPB), invalid and sickness benefits). In addition, they receive a per child payment by way of the FTC. Implicitly, the Crown was arguing that where policy objectives have led to the design of various benefits, those targeted for the receipt of one or more forms of benefit cannot complain of their exclusion from entitlement to others.

[110] CPAG denied that the nature or amount of other benefits paid to beneficiaries is relevant to their claims of discrimination by virtue of their exclusion from IWTC. CPAG‟s analysis distinguished what are categorised as “first tier assistance” (ie unemployment benefit and DPB) from “second tier payments” (such as the FTC and the IWTC). It was argued that the receipt or otherwise of a first tier benefit ought not to be relevant to consideration of complaints of discrimination in the entitlement to second or third tier benefits.

[111] In support of its argument that the position of beneficiaries should reflect their overall financial position as recipients of State benefits, the Crown invited an analogy with the approach adopted in certain Canadian decisions. In Withler v Canada (Attorney-General),65 widows who were entitled to supplementary death benefits in relation to their husbands‟ superannuation, challenged the amount calculated for the widows‟ supplementary death benefits having been reduced by reference to the age of their husbands at the time of the husbands‟ death. The

_{widows claimed that this provision constituted discrimination on the
basis of age.}

65 Withler v Canada (Attorney-General) 2011 SCC 12, [2011] 1 SCR 396 at [33] to [39].

All Courts dismissed the claim. In the analysis of whether it was discriminatory, the

In cases involving a pension benefits program such as this case, the contextual inquiry at the second step of the s. 15(1) analysis will typically focus on the purpose of the provision that is alleged to discriminate, viewed in the broader context of the scheme as a whole. Whom did the legislature intend to benefit and why? In determining whether the distinction perpetuates prejudice or stereotypes a particular group, the court will take into account the fact that such programs are designed to benefit a number of different groups and necessarily draw lines on factors like age.

[112] A similar approach in analysing the attributes of a benefit scheme overall was adopted in the earlier decision of Gosselin v Quebec,67 in which it was claimed that lower levels of payments to unemployed persons under the age of 30, unless the recipient was participating in a work experience programme, was discriminatory on the grounds of age. The majority of the Canadian Supreme Court held that it was not discriminatory on reasoning that assessed the broader objectives of the benefit programme. Incentivising those under 30 to undertake job training was seen in the context of the scheme as a whole as supported by logic and common sense.68

[113] The analogy that the Crown sought to draw with these Canadian decisions is that the effect of targeting an aspect of a benefit programme should take into account the broader perspective of government in all its dealings with categories of beneficiaries claimed to be disadvantaged.

[114] The approach in these Canadian decisions can be distinguished. They rely on a narrower interpretation of what constitutes relevant discrimination, as instanced in the extract cited from Withler at [111] above, which approach we have rejected.

[115] We do not accept that the extent of other benefits paid to beneficiaries should be taken into account in assessing the extent of disadvantage claimed for the purposes of making out discrimination. The IWTC is “targeted assistance” to supplement the income available for families. Conceptually, it could be paid

_{irrespective of the source of other income. The essence of CPAG‟s
complaint is that}

qualifying for assistance intended to help families in this respect depends on employment status, and that the grounds for qualifying for the IWTC operate to the disadvantage of children in beneficiary families.

[116] CPAG‟s argument for discrimination by exclusion did not explicitly assert that all beneficiaries ought to be paid the IWTC. However, in a number of respects that was the effect of it, and many of CPAG‟s arguments lost their relevance if the complaint was not that beneficiaries should get the IWTC. The relevance of the differentiation between employed people qualifying for the IWTC and those on benefits who do not, is that the latter group are disqualified on account of their status as beneficiaries, from the opportunity of qualifying for the IWTC. However, in the vast majority of cases, those excluded could not capitalise on that opportunity because they are also excluded on the separate criterion that they are not “full-time earners”. CPAG conceded that the “full-time earner” requirement for the IWTC was not one that it could challenge as prohibited discrimination. We consider that it is a relevant and valid criterion for assistance targeted at those in full-time work.

[117] It is artificial to characterise the absence of opportunity to qualify for the IWTC as a real disadvantage, when the statistics reveal how small a number might then qualify. CPAG‟s submissions made much of the point that, on their analysis, between 95 to 98 per cent of beneficiaries will not move into work, so that that very substantial majority could never take advantage of the IWTC, even if their status as beneficiaries did not disqualify them. We incline to the view that more beneficiaries than two to five per cent may well have realistic prospects of moving into work, and the bald statistics take no account of the turnover in those presented with opportunities to move into work. Nevertheless, the full-time earner requirement will still exclude a very substantial majority of beneficiaries from qualifying for the IWTC.

[118] The “lack of comparable gain” focused on by the Tribunal has not been caused by the off-benefit rule. Absent that requirement, beneficiaries who do not work 20 hours per week would still not qualify for the IWTC unless they became “full-time earners”. Although the concept of lack of comparable advantage focused

on by the Tribunal might constitute an adequate disadvantage where that is caused by the prohibited ground of discrimination, the present is not such a situation.

[119] We are also unable to identify a qualifying disadvantage by separately considering the predicament of the children of beneficiaries. The same difficulty arises in that their indirect exclusion from the benefits of the IWTC would not be cured if the off-benefit rule was removed. That is, except in those cases where their primary caregiver worked 20 hours, or their parents worked 30 hours per week. It is the discrete requirement of minimum hours of work that would continue to exclude them from enjoying the indirect advantage of a payment of the IWTC to the caregiver.

[120] The WFF package was designed on the basis that no one would be worse off by virtue of the changes it introduced. CPAG accepted that that was literally correct on the day the scheme was introduced, but argued that for certain beneficiaries whose entitlements at that time were “grandfathered”, maintenance of the position equivalent to that pertaining before the WFF package came into existence depended on their not subsequently moving off benefits for any period. CPAG‟s concern was that in that situation, beneficiaries would lose the grandfathered benefits, and have to apply again for a different form of payments, on criteria that might apply more strictly against them. In assessing disadvantage, that concern is not relevant to the direct impact of the off-benefit rule, as a requirement for the IWTC. More generally, CPAG‟s concern is that the exclusion of children in beneficiary families from entitlement to the IWTC would, in relative terms, see them getting further and further behind in terms of alleviating child poverty. However, that concern would not be addressed by removal of the off-benefit rule.

[121] In other respects, CPAG has argued that the financial difference is not all-important, certainly in attempts to incentivise beneficiaries into work. That raises the prospect that the analysis of disadvantage should not focus entirely on the financial consequences of exclusion from the IWTC, and that the stigma of exclusion on account of beneficiary status may be sufficient. We do not consider that could objectively constitute a sufficient disadvantage for the purpose of making out a form of prohibited discrimination. Given CPAG‟s status as a notional complainant, and

the absence of any claims by identified individuals, there is no point in assessing the likely nature of subjective disadvantage perceived by particular beneficiaries.69 The IWTC focuses on getting beneficiaries into work by creating a gap in net earnings. That does draw attention to the status of beneficiaries, but not in a context that could realistically be said to stigmatise them.

[122] The small subset of approximately 1,270 families who are full-time earners, but elect still to be beneficiaries, are the only ones who can claim a direct financial disadvantage because, absent the off-benefit rule, they would qualify for the IWTC.

[123] The extent of difference in income for that small group will vary, depending on the availability of other possible sources of assistance and their level of earned income. However, we proceed on the assumption that it is more than trivial, or real, in terms of the financial position of those families.70

[124] Accordingly, whilst we rejected the narrower scope of the complainant group contended for by the Crown,71 the different attributes of that subset of families who are full-time earners but still remain as beneficiaries does become relevant when considering whether a real disadvantage is made out on behalf of the group complaining of discrimination. It was appropriate to accept CPAG‟s definition of the comparison as one between all of those excluded from entitlement to the IWTC on

the ground that they are beneficiaries, when compared with the non-beneficiaries who do qualify. However, on our analysis CPAG has failed to make out a real disadvantage for all of the group it claims has been discriminated against, except to the extent that the subset within that group would otherwise have qualified for the IWTC.

[125] We are wary of drawing analogies with Canadian decisions, because of the different context in which analyses under s 15 of the Canadian Charter occur in the

_{context of an open-ended presumption of equality. We have previously
rejected an}

69 The Canadian jurisprudence has ranged across subjective and objective analyses of disadvantage: Hogg Constitutional Law of Canada (5th ed supplemented, Carswell, 2007) at [55.10(c)].

analogy with Canadian Supreme Court decisions which suggested that we might analyse the impact on beneficiaries only by reviewing their total benefits received.72

[126] However, some support for our approach on whether beneficiaries suffer a real disadvantage by virtue of the off-benefit rule arises from the approach of the Canadian Supreme Court in the decision in Thibaudeau v Canada.73 In that case, a mother who was paid alimony by her estranged husband exclusively for the benefit of her children challenged a provision in the relevant Income Tax Act requiring her to include the alimony within her taxable income. Because the money was received exclusively for the benefit of her children, she argued that the taxing provision infringed her right to equality guaranteed by s 15 of the Canadian Charter. The

relevant taxing provisions provided that the payment by the father to the mother was a deductible expense in the assessment of his income, and receipt constituted assessable income in the mother‟s hands.

[127] The majority of the Supreme Court of Canada was not prepared to analyse the constitutionality of the taxing provision by focusing just on the disadvantage for the mother. Rather, in the relevant sense the parents remained a family unit, and where the taxability to the recipient of child support constituted a disadvantage, that was matched by the advantage of the deductibility of the amount paid by the parent who did not have day-to-day care of the children.74 That approach suggests that the consideration of disadvantage to a group claiming discrimination cannot focus on a narrow or conceptual analysis, and that the inquiry into a more than trivial or real

disadvantage requires a practical analysis of the context in which the statutory provision applies, and its effect on “both sides of the line”.

[128] Accordingly, we find that the off-benefit rule does not involve a sufficient disadvantage to constitute prohibited discrimination under s 19 of BORA in respect of beneficiaries who do not work 20 hours, or in the case of beneficiary couples a total of 30 hours, per week. However, a real disadvantage does exist for the small subset of beneficiaries who do work 20 hours per week (or 30 hours in the case of a

_{couple) and who elect to remain on benefit. For that group, the real
disadvantage is}

72 _{See discussion of the Withler and
Gosselin decisions at [111]-[114] above.}

the financial difference between their total net income if it included the IWTC, when compared with the total income as earned from wages plus abated benefits.

[129] If we had found a sufficient disadvantage, then we would have found that the off-benefit rule constituted prohibited discrimination. Against the contingency that our analysis of disadvantage is wrong, and in any event to consider the position of the small subset of beneficiaries who do work sufficient hours to qualify as full-time earners, it is necessary to carry out the second stage of the analysis, namely whether, under s 5, the discrimination found to exist for the purposes of s 19 is demonstrably justified.

[130] The onus to establish discrimination under s 19 is on the claimant. However, once that stage has been reached, the onus shifts to the alleged discriminator to establish that the limitation on the right to be free from discrimination is justified in terms of the s 5 analysis. The Crown accepted that the burden on the government under s 5 is an onerous one. That acknowledgement was made for a number of purposes, including support for its earlier argument that there ought to be a rigorous assessment of the real extent of discrimination before a finding that it exists is made under s 19 because of the onerous consequences that follow from doing so.

[131] Another feature of s 5 is that the prospect of reasonable limitations on the rights and freedoms that are protected renders them contextually reasonable rather than absolute or doctrinaire. As McGrath J observed in Hansen, BORA is a “bill of reasonable rights”.75 Further, without acknowledging a hierarchy in their importance, there is scope for fluctuating levels to the qualifications on various of the rights. As Blanchard J observed in Hansen:76

... In the case of some rights, no limitation could be justified. The overarching rights not to be tortured or tried unfairly, for example, can have no meaningful existence as anything less than absolute protections. ... And no one would dispute that many of the freedoms enumerated in Part 2, for

75 R v Hansen [2007] NZSC 7, [2007] 3 NZLR 1 at [186], adopting the characterisation of Professor Rishworth “Interpreting and Invalidating Enactments under a Bill of Rights” in R Bigwood (ed) The Statute: Making and Meaning (LexisNexis, Wellington, 2003) 251 at 277.

example freedom of expression, are in practice routinely limited to a greater or lesser extent by other concerns, both within and external to the Bill of Rights, which are demonstrably justified in a free and democratic society.

[132] The sequence of considerations to evaluate whether a limitation on a right is justified under s 5 were conveniently summarised by Tipping J in Hansen. That litigation involved a challenge to the application of a statutory presumption in s 6(6) of the Misuse of Drugs Act 1975. The presumption provided that a person found in possession of more than specified amounts of various drugs was deemed to possess them for the purpose of supply or sale “until the contrary is proved”. In an appeal against a conviction for possessing more than the specified quantity of cannabis with intent to supply it, it was argued that the meaning that had been applied to the presumption was inconsistent with the presumption of innocence as guaranteed by BORA. The Supreme Court found that the reversal of the onus of proof was inconsistent with the presumption of innocence. The Court went on to consider whether it was an intrusion into that right only to an extent that was justified. The sequence of issues adopted those used by the Supreme Court of Canada in

(a) does the limiting measure serve a purpose sufficiently important to justify curtailment of the right or freedom?

(ii) does the limiting measure impair the right or freedom no more than is reasonably necessary for sufficient achievement of its purpose?

[133] An influence on the analysis of each of these issues is the concept of the extent to which, if any, the Court ought to defer to Parliament. In Hansen, Tipping J

_{followed his adoption of the list of issues from Oakes with an
analysis of deference.}

His Honour began with the notion of the need to allow some discretion to Parliament in its determination of whether a limit on a freedom or right is reasonable and justified.79 Notwithstanding that, by virtue of s 5 and in particular the requirement that a limit on a right or freedom be demonstrably justified, Tipping J still considered that there was a place for giving some latitude to Parliament, the extent of which may be greater or lesser according to the circumstances.80

[134] His judgment treated the English authorities as supporting the view that the Courts perform a review function rather than one of simply substituting their own view. How much latitude the Courts give to Parliament‟s appreciation of the matter will depend on a variety of circumstances.81 Tipping J used the analogy of whether Parliament had “hit the target”, the target representing the objective of the measure in question. Parliament could, in the view of the Court, hit the bull‟s eye of a target

when addressing a particular objective or, if it missed it, still land in the outer circles of a target. That was an analogy for the fluctuating stringency with which the Court should test the quality of Parliament‟s outcome, by reference to the nature of the limitation on a right, and the importance of that limitation in its context.82

[135] The parties were broadly agreed that the Court has to assess the extent to which it should defer to Parliament. However, they invited analogies with different observations in other cases, as to the level of deference that is appropriate in the present circumstances. The arguments for CPAG on deference urged that the Court ought to apply a relatively more rigorous test to the extent of success achieved by the government in respect of the objectives it committed to when introducing the WFF package.

[136] For its part, the Crown invited an analogy with the approach adopted by the

House of Lords in R (Carson) v Secretary of State for Work and Pensions, which related to alleged discrimination in not extending indexed increases in the level of a

81 At [116], relying on the approach of Lord Hoffman in R (SB) v Governors of Denbigh High School [2007] 1 AC 100 and Ghaidan v Godin-Mendoza [2004] 2 AC 557 per Lord Nicholls at [19].

United Kingdom pension scheme to non-UK residents.83 In that case, Lord Walker stated that the relevant approach was:84

... In the field of what may be called macro-economic policy, [where] the decision-making power of the elected arms of government is all but at its greatest, and the constraining role of the courts, absent a florid violation by government of established legal principles, is correspondingly modest.

[137] CPAG submitted that the Tribunal was wrong to draw an analogy with the House of Lords decision in Carson, to the extent that it required greater deference when the objective reflected macro-economic policy. Ms Joychild argued that the outcome for the majority of their Lordships in Carson (namely that there was no prima facie discrimination) turned on rulings that the claimant‟s position was not analogous to the situation of a United Kingdom pensioner, and that cost of living increases were based upon living in the United Kingdom. Observations that related the extent of deference to the breadth of the policy initiatives being addressed were therefore obiter, and arguably should not deter the Court from a robust analysis of the merits of the measure that creates a limitation on a protected right or freedom.

[138] For the Crown, in addition to arguing that relatively more deference is due when Parliament has enacted legislation to further a broad macro-economic policy, it was also argued that the thorough extent of policy analysis and development in relation to the WFF package meant that the output as reflected in the legislation should be treated as thoroughly considered and therefore entitled to a larger measure of deference. CPAG‟s rejoinder was to the effect that the quality of the policy analysis should be tested against the relative success that the WFF package enjoyed in achieving the objectives that drove it. From that perspective, in light of the alleged failings of the package it was argued that it must have been poor quality analysis and development, counting against substantial deference.

[139] In arguing against the need for any substantial degree of deference, Ms Joychild disputed that the development of the legislation reflecting the policy had been thoroughly prepared. She criticised the 2004 Bill as not having been

_{referred to a Select Committee, nor subjected to a commentary by
interested parties}

83 _{R (Carson) v Secretary of State for Work
and Pensions [2005] UKHL 37, [2006] 1 AC 173.}

in any other context. So far as the 2005 measure was concerned, she cited the absence of any thorough policy work, and although the Bill was referred to a Select Committee, she treated it as a token reference in that interested parties were still not afforded an opportunity to contribute.

[140] CPAG cited academic commentaries encouraging more robust consideration by the Courts, and a relative lessening in the need for deference to Parliament. Passages cited included one from Professor Jowell:85

... In the course of some of the steps in the process of this assessment the courts may properly acknowledge their own institutional limitations. In doing so, however, they should guard against a presumption that matters of public interest are outside their competence and be ever aware that they are now the ultimate arbiters (although not ultimate guarantors) of the necessary qualities of a democracy in which the popular will is no longer always expected to prevail.

Traditional deference to the legislature or executive on matters of public interest is pitted against the newly acquired judicial role of protecting fundamental rights. ... The result has been a new approach to deference, no longer based on “abasement” of courts to legislatures, but on a view of the courts as active participants in upholding a rights-based democracy.

Most legislation is passed to advance a policy. And frequently it involves in one way or another the allocation of resources. ... What I am saying is that there cannot be a legal principle requiring the court to desist from making a judgment on the issues in such cases. ... There is in my view no justification for a court to adopt an a priori view in favour of economic conservatism.

In common law adjudication it is an everyday occurrence for courts to consider, together with principled arguments, the balance sheet of policy advantages and disadvantages. It would be a matter of public disquiet if the courts did not do so. Of course, in striking the balance the courts may arrive at a result unacceptable to Parliament. In such cases Parliament can act with great speed to reverse the effect of a decision. ... But there is no need to create a legal principle requiring the courts to abstain from ruling on policy matters or allocation of resource issues. ...

86 Sandra Fredman “From Deference to Democracy: The Role of Equality under the Human Rights

[143] In addition, CPAG argued that less deference was due when the legislation breached obligations assumed by the New Zealand Government in ratifying the ICESCR and UNCROC human rights instruments. Allied to this is the criticism that the Attorney-General‟s report on the implications under BORA of the proposed measure did not recognise the potential for incompatibility with the right to be free from discrimination on grounds of employment status, as had been raised by CPAG some time before introduction of the 2004 measures.

[144] Lastly, CPAG argued that less deference is due when the government has developed a package of benefits that were aimed at alleviating child poverty, but had denied those benefits to one category of children.

[145] We have had regard to the range of views on deference, and recognise, as Tipping J predicted in Hansen, that it may be necessary to consider the appropriate extent of deference differently, on different aspects of the policy justification for any limitation on protected rights. We acknowledge that our task is different from judicial review where (except for unreasonableness) the Court focuses on the lawfulness of the process adopted. Here, we are concerned with the substantive quality of the measure that has limited the freedom from discrimination, but that

does not make it a matter of substituting our own view,88 and does not transform it

into a merits review.89 Our task is to consider the proportionality of the limitation on

[146] In forming our view about that, the extent of deference will depend on the level of abstraction at which we undertake the analysis, so that the perceived quality of the measure is likely to depend on whether we measure it narrowly against the objective of the off-benefit rule, or more broadly against the objectives of the whole

89 R (Daly) v Secretary of State for the Home Department [2001] 2 AC 23 at [28] per Lord Steyn.

[147] It was argued for the Crown that the limitation on the right to be free of discrimination on grounds of employment status is reflected in the off-benefit rule that excludes those on benefits from qualifying for the IWTC. Further, that matters of justification for that exclusion are to be measured against the objective of the off-benefit rule which was to incentivise those on benefits to move off the benefit if they could, and for those in low paid employment to stay there. Allied with these was the objective of signalling that work pays more than benefits.

[148] In contrast, CPAG argued that whether the exclusion of those on benefits from the families entitled to the IWTC was justified, should be measured against the objectives of the WFF package overall, and certainly against the objectives of the IWTC. This difference could be material because CPAG relied on the requirement that earners have dependent children to qualify for IWTC as demonstrating that IWTC was a measure addressing child poverty. CPAG cited the characterisation given to the IWTC by Suzanne Mackwell, a senior officer with the Ministry of

This does not imply that the purpose of the [IWTC] is to assist with the costs of raising children, but rather that it was introduced as a work incentive measure for the long term objectives of reducing child poverty and improving outcomes for children.

[149] In addition, CPAG cited a number of references in the WFF Cabinet Paper where one of the two primary objectives of the package was cast in terms of ensuring income adequacy, to significantly address issues of poverty, especially child poverty.91 However, the content of the Cabinet Paper does not justify a submission that the two primary objectives, namely making work pay as an incentive to move people off benefits, and ensuring income adequacy, were combined, or necessarily

being pursued together by each of the components of the package. We make this point notwithstanding CPAG‟s reference to the opinion of the OECD experts, which

_{was to the effect that such schemes of in work benefits “...appear
to achieve both}

employment and distributional objectives at the same time...”.92 The potential for some forms of such schemes to do so does not mean that both must necessarily be attributed as objectives of all aspects of the scheme in New Zealand. The opinion was expressed in a paragraph that referred to the nature of such schemes in nine different jurisdictions, which appear to have wide variations in their features.

[150] On the Crown‟s approach, the issue would be whether the exclusion from the IWTC of beneficiary families was justified in relation to an objective of incentivising people into work. On CPAG‟s approach, the issue would be whether exclusion of beneficiary families from the IWTC was justified in light of the objectives of IWTC, which it treats as including that of alleviating child poverty.

[151] The Crown cited two House of Lords decisions in support of the proposition that justification analyses should focus on the direct or narrow objectives of the provision that limited the relevant right.

[152] First, in R (RJM) v Secretary of State for Work and Pensions,93 State payments to disabled persons could include a disability premium unless the person was “without accommodation”. The objective for excluding that group was to encourage the homeless to seek shelter and help, as well as there being recognition that disabled persons would be less likely to need a financial supplement if they were not incurring housing costs. This analysis focused on the rationale for excluding the homeless, and did not take into account the wider rationale for an accommodation supplement to disabled persons who had accommodation.

[153] Secondly, in R (Reynolds) v Secretary of State for Work and Pensions,94 the issue was the lower level of a job seeker‟s allowance paid to people under the age of

25, relative to the level paid to those over 25. The rationale for excluding the under

25s was that they could be expected to both earn less and have lower living costs, and that paying them a smaller benefit would encourage them to live with others

_{rather than independently, which was seen as desirable in terms of
social policy. The}

93 _{R (RJM) v Secretary of
State for Work and Pensions [2008] UKHL 63, [2009] 1 AC 311.}

justification for paying a job seeker‟s allowance at higher rates to those over 25 was not part of the analysis.

[154] Neither of the payment arrangements in those cases involved a “package” of payments by the State, intended (whether globally or discretely by reference to individual components) to address a range of policy objectives, which is the case with the WFF package. They are, however, comparable to an analysis that focuses only on the IWTC, and the exclusion from it of those on benefits. The analogy is that the Court need not analyse why the IWTC is being paid, but rather only whether the exclusion of those on benefits is justified.

[155] CPAG‟s argument required the objectives to be identified as at least those it attributed to the IWTC, and arguably the WFF package overall. CPAG‟s reply submissions cited two decisions of the Supreme Court of Canada in which claims of breach of Charter rights caused by “underinclusion” involved evaluations of the claimed justification for exclusion that began with the objectives of the legislative measure as a whole, then the objectives of the impugned provisions, and finally the

objectives of the disputed exclusion.95 Each justification exercise that arises under

s 5 will have to be undertaken in its own particular context. It is likely to be helpful orientation to consider each of these three layers in most cases. However, whether the rationale argued to justify discrimination has to sustain an analysis on a wider basis than that motivating the specific limitation on the right seems unlikely in most situations. Certainly in the present case, where there were a number of objectives, two of which would come into conflict with each other at some point, the justification analysis appropriately focuses on the narrower rationale for the limitation of the right.

[156] CPAG argued that an inevitable adverse consequence arises for the government from choosing to combine child poverty alleviation with a work incentive. It was argued that having elected to address both objectives together, and having made the former conditional on the latter, the government had thereby created

_{unlawful discrimination.}96 _{It is
implicit in the approach described in the preceding}

paragraph that we reject this argument. To accept CPAG‟s argument on this would impose a constraint on government that is at odds with the statutory purpose of the Court‟s jurisdiction to declare a statutory measure inconsistent with BORA. The government should be free to justify its actions on their substantive effect, and not be trapped by the form in which it has pursued discrete objectives.

[157] CPAG argued that the range of objectives was to be discerned from the terms of the Cabinet Paper and the Minister‟s speech on introduction of the relevant provisions. Ms Joychild urged that “it is important in a free and democratic society that objectives are transparent”. On CPAG‟s analysis of the Cabinet Paper, creating a gap between the total of payments made by the State to various forms of beneficiary and total receipts for those on lower levels of earned income was not an explicit objective. CPAG criticised the Tribunal for concluding that “creating and enhancing a gap” was the objective against which the WFF package was to be assessed.

[158] CPAG‟s argument that the government‟s objectives had to be “transparent” was, in essence, an argument that the government should be held to account by reference to the terms in which it identified the objectives at the time it proposed the measure that arguably included discrimination. However, in an analysis of whether a discriminatory provision is demonstrably justified, any inadequacy in the terms in which the policy objectives were recorded at the time is not appropriately held against the government. An exception would arise if there were grounds for concern that there has been an attempted ex post facto justification for the discrimination by proposing an objective that did not in fact motivate the introduction of the provision in issue. We consider that the nature of an objective and its relative importance should be evaluated on a substantive basis, and not be determined by the form in which the government either publicly explained policy objectives or recorded them at the time.

[159] CPAG‟s arguments did not admit of the prospect that within the WFF package, it was legitimate for the Crown to justify a component part of the measures introduced, by reference to a discrete part of the objectives identified in the development of the policy and as reviewed in the Cabinet Paper. On the other hand,

the arguments for the Crown did not seek to justify the off-benefit rule as if it furthered the objective of addressing child poverty. To this extent, the parties‟ arguments did not engage directly.

[160] The Tribunal concluded that it should test the various components of the Crown‟s claim under s 5 to justify any limit on the relevant right by reference to the objectives overall of the WFF package.97 The Tribunal found that the IWTC was seen as a method for alleviating poverty, so that “making work pay” was an integral part of achieving income adequacy.98 However, it would be illogical to attribute to the government an intention that the IWTC be used to alleviate poverty for beneficiaries when it was so obvious that they were excluded from it. It is only after a beneficiary caregiver has achieved full-time earner status that his or her entitlement to the IWTC could then contribute to this form of poverty alleviation. The Tribunal was evaluating the WFF package objectives relative to the scope of the scheme as originally introduced in 2004. The entitlement of the Crown to have the evaluation undertaken, specifically comparing one identified objective with one aspect of the

package, becomes even more important when the s 5 analysis is undertaken in relation to the expanded scope of the scheme after the 2005 amendments were incorporated.

[161] The objectives of incentivising people into work, and reducing child poverty, are complementary when considered over the longer term, in that success in moving people into work will result, over time, in lifting income levels and thereby reducing poverty. We take Ms Mackwell‟s description rationalising the different objectives99 to reflect this longer term complementarity when she saw the work incentive contributing to the long term objective of reducing child poverty. That longer term

complementarity does not avoid the prospect of there being short-term conflict between these two objectives, as we have recognised at [20] above. The incentive to move people into work uses the gap between total net income for those in work, compared with those on benefits, as the incentive and that structural incentive cannot

_{exist if the IWTC is paid indiscriminately to both groups. For the IWTC
to achieve}

the objective, it must first convey the message that work pays, and the extent of prominence that objective must have renders it inappropriate to assess the IWTC also against the objectives of ensuring income adequacy across the board.

[162] If there was a realistic expectation that the IWTC would move all, or substantially all, beneficiaries into full-time employment, then the objectives would remain complementary because payment of the IWTC addresses poverty alleviation for those who receive it. However, success with such initiatives is necessarily measured far more modestly. For as long as substantial numbers cannot be moved into full-time employment, pursuit of the work incentive objective is at odds with the income adequacy/poverty alleviation objective because to be effective, non- employed persons have to be excluded. The relevant objective is therefore the first one.

[163] At an earlier stage of our analysis, when assessing whether there was prohibited discrimination, we rejected an argument for the Crown that the analysis of whether beneficiaries‟ exclusion from those entitled to the IWTC should be assessed in the broader context of the total extent of payments to them by the State.100 There is a certain complementarity between rejecting that broader approach to the earlier analysis urged by the Crown, and accepting the narrower approach the Crown

contends for in analysing the scope of the objectives of the off-benefit rule in this later stage of the analysis. If it is irrelevant that the Crown pays other amounts to those allegedly discriminated against, then it is appropriate to confine the analysis of objectives to the particular circumstance in which the alleged discrimination arises, and not to have regard to the broader objectives of the package of which that is a part.

[164] We are satisfied in the circumstances of this case that the s 5 analysis should be resolved on the narrower consideration of objectives being pursued in imposing the limitation on the right, ie the off-benefit rule. The appropriate approach to such issues will inevitably be case-specific. In this case, relatively broad objectives of encouraging existing and potential earners to pursue work rather than depend on

progressed together. One justification for that is the extent of infrastructure required for monitoring eligibility and providing delivery mechanisms for such significant numbers. The WFF package was described as the “biggest change to New Zealand‟s social assistance system in over a decade”,101 and as “the centrepiece of the 2004 budget”.102

[165] In this case, the design of components of an overall package intended to address both work incentive and income adequacy objectives could separate their delivery by one or more components addressing income adequacy, and other components addressing the incentive to work. The Crown argued that such a separation did occur, evidenced, for example, by the different objectives sought to be advanced by different initiatives planned for introduction on various of the 1 October

and 1 April dates between 2004 and 2008.103 Essentially, the objective addressed

was either “Strengthening the Returns from Work” or “Improving income adequacy”, with the introduction of the IWTC on 1 April 2006 described in the former category.

[166] We are satisfied that the objective of the off-benefit rule was to incentivise people with relatively low earning capacity to pursue work and, once employed, to remain in work, instead of resorting to State benefits. A predictable component of initiatives to address that objective would be to ensure a meaningful and readily identified gap between returns from work, and the total of payments that could otherwise be received from the State, in the absence of employed work.

relation to its objective of alleviating child poverty, when considering the plight of children whose caregivers are beneficiaries. We review those criticisms next, before

101 Minister‟s speech on introduction of the Future Directions (Working for Families) Bill, Hansard

considering their relevance to each of the sequence of issues relating to a s 5 claim of justified limitation.

[168] CPAG‟s criticisms were strongest when the relative success of the identified objectives of the WFF package was assessed against the outcomes of the WFF package as a whole. Further, the impact of CPAG‟s criticisms was heightened when the relative success in achieving the objectives was assessed in relation to the scheme as introduced in 2005. Once the WFF package was extended to provide tax cuts for working families whose incomes were well above the level at which any of the objectives could be advanced, the criticisms of success overall relative to the objectives were said, on Tipping J‟s analogy in Hansen, to have “completely missed the target”.

[169] As to the objective of incentivising people into work, there was no logic in providing a supplement to the earnings of those who were, say, 15 or 25 per cent or more above the level of payments made to beneficiaries. So, too, payment by way of generous tax credits to those, say, at or above the average wage do nothing to alleviate child poverty for those at the bottom end of the range of earnings and levels of payments to beneficiaries.

[170] Although the impact of the 2005 amendments was ultimately not relevant to the Tribunal‟s outcome, its decision cites one illustration of the effect of the 2005 amendments from CPAG‟s arguments with which we agree:104

In 2006/07 187,870 families received the IWTC at a cost of $480.3 million. It is not clear how many of these families really needed the payment as an incentive to stay in work, but it is clear that a significant proportion of the spending went to families outside of the stated objectives of the WFF package (by way of illustration, if one accepts $45,000 per annum as a point beyond which the payment is no longer needed as a work incentive, then in the 2006/07 year some $229 million – or about 47% of all that was paid out as IWTC – went to families that did not really need it).

[171] CPAG argued that the allocation of a further $500 million per annum in the

_{$1.1 billion, rendered the initiative a failure. CPAG argued that far
more could be}

done to alleviate child poverty with the amount of money committed overall to the WFF package, so that it could not be claimed to be successful to an extent that justified the limitation on the right not to be discriminated against. Accordingly, CPAG‟s criticism was that the government had “missed the target” by allocating such a substantial part of the benefits to be delivered by the WFF package to those not in the categories that would, on any analysis, be recipients in order to achieve the identified objectives.

[172] CPAG separately argued that the analysis of projected outcomes in respect of those within the categories who might be helped relative to the stated objectives was, in any event, misconceived. CPAG relied on a projection in the Cabinet Paper that only two per cent of sole parents could be expected to move into paid employment, and therefore move off benefits, as a result of the availability of the IWTC, with no net change for “couple families”.105 Our impression is that the projections in the Cabinet Paper were relatively conservative. Although it is impossible to isolate the impact of the IWTC from other influences in the economy, we are satisfied that the

[173] In addition, CPAG argued that a financial incentive to move into work was not the only, or a compelling, incentive when a range of other pressing circumstances prevented beneficiaries moving into paid employment. The evidence for CPAG cited the mismatch between skills and available paid work, the type of work available, availability and quality of child care, poor health of beneficiaries, circumstances of their parenting roles and discrimination against those on benefits by

[174] Further, whilst financial incentives may work, CPAG‟s evidence was to the effect that non-financial reasons can also motivate beneficiaries to seek employment. CPAG‟s evidence cited the stigma attached to receipt of the DPB, beneficiaries not wanting to deal with Work and Income New Zealand (WINZ) and the achievement

_{of personal goals as factors influencing beneficiaries in their attempts
to move into}

paid employment.107 Mr Gray, a senior Ministry of Social Development officer who was cross-examined before the Tribunal, acknowledged:108

A lot of people are on a benefit because they are unable to work or are unable to find work and they work very hard to try and find work at some stage. Their choices are very limited, I agree with that.

[175] So far as the effect of the WFF package on alleviation of child poverty was concerned, CPAG argued that the exclusion of those on benefits from the families who qualified for the IWTC, when those on benefits comprised a disproportionately high portion of the families living in poverty, meant that the WFF package was flawed and could not achieve that objective. This was particularly so when compared with the prospect of an alternative provision for support to families in financial need that did not so discriminate.

[176] CPAG argued that various alternative packages applying similar amounts of State aid to those most in need could be devised to better address child poverty. At the level of abstraction involved, these aspirational claims were virtually inarguable. Those arguments were supported by CPAG‟s submissions that government has an imperative to meet that priority because of the commitments it has made at international law.

1. States Parties recognize the right of every child to a standard of living adequate for the child‟s physical, mental, spiritual, moral and social development.

2. The parent(s) or others responsible for the child have the primary responsibility to secure, within their abilities and financial capacities, the conditions of living necessary for the child‟s development.

3. States Parties, in accordance with national conditions and within their means, shall take appropriate measures to assist parents and others responsible for the child to implement this right and shall in case of need provide material assistance and support programmes, particularly with regard to nutrition, clothing and housing.

3. Special measures of protection and assistance should be taken on behalf of all children and young persons without any discrimination for reasons of parentage or other conditions. Children and young persons should be protected from economic and social exploitation. Their employment in work harmful to their morals or health or dangerous to life or likely to hamper their normal development should be punishable by law. States should also set age limits below which the paid employment of child labour should be prohibited and punishable by law.

[179] The response on behalf of the Crown to these wide-ranging criticisms was, first, that they are advanced on an unjustifiably generic basis. The second response was that once the discrete objectives that were targeted by various components of the WFF package were measured within their own terms, then the limitation on the right to be free of discrimination that was in issue was justified on a proper analysis of the sequence of issues raised by s 5 of BORA.

[180] A first significant exclusion argued for the Crown was the lack of relevance of the terms and effect of the 2005 additions to the WFF package.

[181] The Crown denied that the then government had sought to further the 2004 objectives of the WFF package, by grafting onto it the discrete initiative pursued as an election strategy in 2005. The argument before us implicitly accepted that the then availability of substantial further funding was targeted to middle and some higher income families as a political strategy to win votes in the 2005 election. Because the theme of that “manifesto initiative” revolved around tax credits for working families, the logical mode of administering entitlements and payment of benefits under it was the structure that had been devised between the Ministry of Social Development and Inland Revenue for administering the WFF package.

much funding to mid to high income families “under the guise of funding for social

assistance”.109 However, this appears to be a matter of inference and there was no evidence stating that the government of the day claimed that the fresh 2005 initiative would further enhance the objective of the 2004 package. CPAG‟s argument is, in effect, that because the 2005 initiative was added to the WFF package as originally designed, then interested parties analysing it were entitled to attribute to government that the amendments to the scheme had necessarily to be considered as advancing the original objectives. We reject that notion. The boost to middle and some higher income families provided by the political initiative taken in 2005 was just that. The two documents produced for us on the last day of the hearing bear out that the then government was not seeking to make the original objectives of the WFF package work better. Rather, so long as those objectives were not harmed by bolting on a different initiative, the government wished to utilise the delivery mechanism that was in place, to distribute tax credits to working families.

[183] Turning then to the sequence of questions to be addressed under s 5, as set out in [132] above, we apply them to the off-benefit rule which excludes beneficiary families from the IWTC.

[184] The objective of incentivising work by supplementing the returns for the lower paid is indeed a significant one for New Zealand, both in societal and economic terms. Welfare dependence is debilitating in a range of senses for those in that predicament and for the children of families dependent on benefits. The pursuit of skills, enhanced self-esteem and the prospect of more positive contributions to be made by individuals once in work are all important attributes. In addition, improved productivity for the economy follows from a fuller state of employment. The

26 Improving New Zealand‟s economic performance is our top policy priority. Through the growth and innovation framework, investment in skills development and careful macro-economic and fiscal management we are putting into place the conditions for better economic performance. To complement these strategies, we need to

bring our social assistance system up to date so that it no longer acts as a barrier to people moving from benefit to employment and actively supports families in work.

[185] Before the Tribunal, the Crown called two experts who worked with the Organisation for Economic Co-operation and Development (OECD), whose expertise was in the effects of tax and benefit systems on employment and poverty. The joint views of Mr Mark Pearson and Dr Herwig Immervoll traversed the importance of government schemes to provide benefits to those in lower paid work, the policy reasons for adopting such schemes and a reflection on the international experience of their effect. Those experts opined that had New Zealand not already introduced such a policy, then the OECD would recommend that it do so. Although the form of schemes varies widely, over half of the 30 member countries of the

[186] The international experience is that schemes for paying in work benefits have significantly positive employment effects. The experts cited the effect of a scheme in the United States of America between 1984 and 1996 being attributed with a four to seven percentage point increase in employment rates of single mothers,112 and a five percentage point increase in employment rates of solo parents over a three year period in the United Kingdom from its in work payment scheme. The United Kingdom scheme was one of the models closer to New Zealand‟s IWTC.113 It is recognised that movement of a small number of people from benefits into earned income is a significant achievement.114 The Crown cited the evidence of other witnesses that consistently recognised non-statistical benefits as well, in terms of getting people off benefits and into work, helping self-esteem, intellectual stimulation and social connectedness.115

[187] The arguments for CPAG did not dispute these attributes identified for in work benefit schemes. Indeed, CPAG accepted more generally that the aims of

_{making work pay and ensuring income adequacy are legitimate aims that
justify}

some discrimination.116 Rather, the arguments criticised the lack of positive outcomes achieved by the IWTC. We accept that the outcomes achieved have been affected by the state of the New Zealand economy. At the time of the Tribunal‟s decision, it cited statistics that the number of DPB dependent families had decreased from 109,700 in August 2004, to 97,200 at August 2007, and that the number of children dependent on the benefit system had fallen from 261,527 in September 2002 to 199,006 in March 2008.117

[188] In an up-dating affidavit since the Tribunal decision, completed in December

2009, Mr Gray stated that sole parent employment had increased from 48.3 per cent in 2004 to 57.7 per cent in 2007, with six percentage points being estimated to be due to the WFF package changes.118

February 2010. She stated that the sole parent employment rate had fallen again to

53.9 per cent in June 2009, calculating that the sole parent employment rate had declined by some 8.4 percentage points between 2007 and 2009, whilst in the same period the single adult employment rate had increased by 0.3 percentage points.119

Dr St John‟s evidence also cited statistics that by September 2009, the number of recipients of the DPB had increased again to 107,658, with an increase also in the recipients of the unemployment benefit, which jumped from 23,158 in September

[190] The essence of CPAG‟s argument was not to relegate the importance of moving beneficiaries into paid employment, with the consequential impact on lessening poverty that would follow from that, but rather that the IWTC had not achieved that outcome.

[191] Given the international experience that such schemes are important and can improve outcomes, and given the likely heavy impact of broader fluctuations in the

New Zealand economy, we are satisfied the IWTC is addressing a very important objective. We are not inclined to discount its importance by virtue of the fluctuating levels of success, as reflected in the range of statistics cited to us.

[192] Because the IWTC is targeted at incentivising people to move from being beneficiaries, to being at least in part self-sufficient by undertaking paid work, it is sensible that an incentive targeted at those going into lower paid work should not also be available to those who are on benefits. Such an extension would destroy the incentive effect of conferring a benefit on those in low-paid work. Accordingly, we are satisfied that the purpose of incentivising paid employment (coupled with the objective of incentivising those in lower paid employment to remain there) is sufficiently important to justify the extent of discrimination on the grounds of employment status. This analysis applies both to the exclusion of all beneficiaries, in the event that we are wrong in deciding that they are not subjected to prohibited discrimination, and to the predicament of the subset of people who are full-time earners, but are excluded because they continue to receive benefits.

[193] We have also reflected on the importance of the full extent of the objectives of the WFF package, relative to the extent of curtailment on the right to be free from discrimination on grounds of employment status, against the contingency that we are wrong in focusing on the narrow objective of incentivising people into work. Notwithstanding CPAG‟s arguments about the failure to progress the objective of alleviating child poverty for children in beneficiary families, we would also find that the extent of infringement on the right is justified, if it had to be compared with the broader objectives of incentivising work, and alleviating child poverty. They are hardly objectives on which any miracle cures could be expected. Structurally, the prospects of successfully incentivising work and ultimately making progress in the alleviation of child poverty do justify the discrimination that has been complained of.

Is there a rational connection between exclusion of beneficiaries and the identified objectives?

[194] The Tribunal found that there was a rational connection in relation to the

2004 form of the WFF package,121 but not so if the original objectives were tested against the outcomes from the 2005 version, given the lack of relevance of those objectives to families in mid to high income brackets.122

[195] Given that the sequence of issues in Hansen is adopted from the Canadian Supreme Court decision in Oakes, it is appropriate to refer to other Canadian decisions on such analyses, including the need for a rational connection. Both parties cited the Supreme Court of Canada decision in Hutterian Brethren of Wilson Colony v Alberta, in which McLachlin CJ, for the majority, observed:123

To establish a rational connection, the government “must show a causal connection between the infringement and the benefit sought on the basis of reason or logic”: RJR- MacDonald Inc v Canada (Attorney General), [1995]

3 S.C.R. 199, at para 153. The rational connection requirement is aimed at preventing limits being imposed on rights arbitrarily. The government must show that it is reasonable to suppose that the limit may further the goal, not that it will do so.

[196] The major thrust of CPAG‟s argument as to the absence of rational connection depended on its proposition that a rational connection had to exist with all three of the objectives it distilled from the first page of the Cabinet Paper.

[197] However, if the rational connection had to be established between the off-benefit rule as the limit on the right, and the objective of incentivising people into work and keeping them in work, then CPAG still argued that there was no rational connection when the incentive would not work for 95 to 98 per cent of excluded parents on benefits. CPAG‟s arguments, which we have reviewed above,124 drew on the lack of justification for the 2005 amended form of the WFF package, when assessing the prospects of successfully achieving the objective of getting those

123 _{Hutterian Brethren of Wilson Colony v
Alberta 2009 SC 37, [2009] 2 SCR 567 at [48].}

[198] In reliance on our earlier analysis, we disregard the group advantaged by the

2005 amendment. Notwithstanding all of the arguments raised by CPAG against this form of financial incentive succeeding in getting beneficiaries into work, and those in lower paid work remaining in it, the essential objective of pursuing that incentive, and doing so by restricting one form of financial benefit to those who are in work, is sufficiently rationally connected for the purposes of this test.

[199] The arguments for CPAG cited numerous references to the evidence before the Tribunal, the effect of which was that the incentive intended by the IWTC to encourage beneficiaries into work would not be effective for a substantial majority of beneficiaries. On that ground, and because the exclusion of beneficiary families so seriously impaired the achievement of the objective of alleviating child poverty, CPAG submitted that there was no rational connection in the sense required on this threshold issue as to whether a limitation on a right to be free from discrimination was demonstrably justified.

[200] These points go to relatively how successful the IWTC could be, whilst the off-benefit rule comprised a part of it. They do not attack the rational connection that does exist between the work incentive objective and providing an incentive for people who would earn at relatively low levels, but not for those who instead are recipients of State benefits. This obvious connection is “rational” in the appropriate sense.

[201] In Hansen, Tipping J described the task of the Court on this issue as follows:125

... The Court must be satisfied that the limit imposed on the presumption of innocence is no greater than is reasonably necessary to achieve Parliament‟s objective. I prefer that formulation to one which says that the limit must impair the right as little as possible. The former approach builds in appropriate latitude to Parliament; the latter would unreasonably circumscribe Parliament‟s discretion. In practical terms this inquiry involves the Court in considering whether Parliament might have sufficiently

achieved its objective by another method involving less cost to the presumption of innocence.

[202] Also in Hansen, McGrath J described this issue in the following terms:126

... The inquiry here is into whether there was an alternative but less intrusive means of addressing the legislature‟s objective which would have a similar level of effectiveness. This aspect of the analysis of proportionality has rightly been described as both crucial and difficult.

[203] It is not appropriate to deal with this consideration on a precise quantitative basis. The objective of the IWTC was to create a sufficient margin between net payments to beneficiary parents on the one hand, and the net income paid to parents working for relatively low wages on the other. Given the influence of non-financial incentives to pursue employment rather than remain in receipt of a benefit, and the variation in individual circumstances, it is not reasonable to determine how much of a gap between the two categories is justified.

[204] The approximate extent of difference in the financial position of the two groups is reflected in the chart at [41] above. Those numbers suggest a gain from moving “off benefit” of some $23 for a couple and $35 for a sole parent. By far the largest State payment made in recognition of child care responsibilities is the FTC, which is paid indiscriminately to all principal caregivers (included in the chart at

$139 per week for two children under 12). For those earning at the lowest levels, the biggest contributor to the difference between earned income and beneficiaries is the MFTC, which abates first.

[205] We are certainly not persuaded that in the quantitative sense, there is a greater disparity and therefore a greater impairment on the right than is reasonably necessary to achieve the objective of incentivising those who can to move from benefit status to being employed.

[206] The Tribunal reviewed CPAG‟s criticisms of the failings of the IWTC before

_{concluding that the limitation on the right not to be discriminated
against was not}

more than was reasonably necessary to achieve the purpose. Having made that finding, the Tribunal commented:127

Our conclusion under this heading might have been different if, for example, there had been some other burden cast upon those on benefit income as well as the implementation of this financial gap. But the off benefit rule of eligibility for the IWTC does no more than to put the intended gap into place. In our assessment the limiting measure is not of a kind that impairs the right to freedom from discrimination any more than is reasonably necessary for sufficient achievement of the legislative purpose.

[207] CPAG‟s challenge to this aspect of the Tribunal‟s decision argued that continuation of the most severe levels of child poverty for those in beneficiary families amounted to continuation of an existing burden and therefore did cast a burden on them.128 This is another respect in which the essence of CPAG‟s arguments depends on the premise that beneficiary families should not have been excluded from the IWTC, irrespective of their inability to qualify as “full-time

[208] An allied criticism of the Tribunal‟s decision was that it had undertaken its analysis on whether the limitation was more than reasonably necessary only by taking into account the objective of making work pay. On CPAG‟s analysis, the effectiveness of the measure had also to be assessed against the objective of ensuring income adequacy. Again in this respect, our own analysis is the same as the Tribunal‟s: it is appropriate to measure the extent of intrusion by reference to the relevant objective of the limitation on the right that is under challenge.

[209] CPAG‟s arguments compared the extent of limitation on the right caused by beneficiaries‟ exclusion from the IWTC with alternatives such as some that had been proposed in Dr St John‟s evidence, and the discrete forms of incentives provided in other OECD countries. Again, these criticisms depended largely on measuring the effectiveness of the IWTC, by reference to both work incentive and poverty alleviation, rather than just one of the relevant objectives. They involved a particular

[210] To complete the analysis on this issue, we have reflected on the assessment required if we are wrong not to have measured the extent of the intrusion into the right by reference to all of the objectives of the WFF package. That was in effect the analysis urged on behalf of CPAG. In this context, the off-benefit rule advances all the objectives of the WFF for those who are able to move into full-time employment. The IWTC incentivises their moving into work and remaining there, and receipt of it addresses income adequacy for its recipients. The fact that beneficiaries are excluded from the IWTC to the extent that it is an income adequacy initiative is justified because that is a consequence of the IWTC necessarily establishing a gap between those in full-time employment and those in receipt of benefits. Accordingly, if we had to address this issue by reference to the wider objectives of the WFF package, we would still be satisfied that the intrusion is not to any greater extent than is justified.

[211] Finally on this issue, we note another approach might be to consider in a more qualitative way whether the scope of those excluded from the IWTC is wider than would be necessary to achieve its purpose. Another qualifying requirement for the IWTC is that the recipient be the principal caregiver of dependent children. Predictably, no issue was raised that that is discriminatory against those without children. Once the relevant category is confined to those caring for dependent children, then a distinction can appropriately be drawn between those in receipt of benefits, and those in work. From that perspective, the limiting measure does not impair the right to be free from discrimination on the grounds of employment status more than is reasonably necessary to achieve the objectives of the IWTC.

[212] The essence of the task on this last issue was described in Hansen as follows:129

Ultimately, the balance to be struck is between social advantage and harm to a right.

[213] In evaluating the proportionality between social advantage and harm to the right in a social policy context, the Supreme Court of Canada has observed:130

In summary, the salutary effects of the universal photo requirement for driver‟s licences are sufficient, subject to final weighing against the negative impact on the right, to support some restriction of the right. As discussed earlier, a government enacting social legislation is not required to show that the law will in fact produce the forecast benefits. Legislatures can only be asked to impose measures that reason and the evidence suggest will be beneficial. If legislation designed to further the public good were required to await proof positive that the benefits would in fact be realized, few laws would be passed and the public interest would suffer.

[214] CPAG‟s submissions accepted that it is legitimate for the government to want to create a gap between income paid to beneficiary parents, and total net income received by working parents. CPAG described its complaint as being about the way in which the IWTC was structured, ie the use of one instrument to achieve two objectives with consequent discrimination against beneficiary families with children.

[215] However, when assessing the proportionality of the advantages that can reasonably be expected from a work incentive measure such as IWTC, against the disadvantage for beneficiary families of being excluded, the form in which that exclusion occurs should not determine the decision on proportionality.

[216] In the end, CPAG‟s complaint is that the government is not doing enough to alleviate child poverty for families dependent on benefits. That complaint can be justified statistically, and is strengthened by the 2005 addition to the WFF package being used to deliver support to families that are better off. That addition demonstrated the availability of further funding that, with different priorities, the government might have applied to re-design those parts of the WFF package most important to achieving the objective of alleviating child poverty.

[217] The fact that the government did not do that does not render the off-benefit rule excluding beneficiaries from IWTC as being out of proportion to the social

130 Hutterian Brethren of Wilson Colony v Alberta 2009 SC 37, [2009] 2 SCR 567 at [85] per

[218] The evidence before the Tribunal included observations from an experienced economist, John Yeabsley. At the time of giving his evidence, Mr Yeabsley was an economist working at the New Zealand Institute of Economic Research and was a member of the Ministry of Social Development Social Policy Reference Group. Mr Yeabsley had 25 years‟ previous experience working for the New Zealand Government as an independent researcher and consultant. Mr Yeabsley deposed to the reality of the rather imprecise process of governments developing social policy. The impression we have is that it is certainly more an art than a science, and Mr Yeabsley listed a number of factors that can influence the accuracy and success of predicting the impact of measures intended to change social or economic policy.

that relatively small adjustments in policy settings can radically alter the choices that most suit (for instance, in this situation, families).

[219] There is also a realisation that it is difficult for government to pursue “pure” goals and that social adjustment often has to be prioritised.131 Mr Yeabsley‟s observations suggest that whilst deference is due to the expertise and deliberation committed to developing relevant measures, it would be unrealistic to expect too high a standard in achieving the objectives, at least without a period of fine tuning. We have had regard to that consideration in weighing CPAG‟s criticisms of what

could have been done better, and its analysis of the failures in relation to the relevant objectives, against what might reasonably be expected in such macro- economic/social policy areas. It is not an area in which government should be marked down for not “hitting the bull‟s eye” in Tipping J‟s terms. Deference in this context requires a reasonably wide outer circle around the bull‟s eye before the

[220] Although the following observation made in Atkinson was in a different context, we nonetheless agree when considering proportionality:132

We have already recognised that governments must make distinctions between people if they are to govern effectively. Governments must be free to target social programmes so that those whom they consider should benefit from them do so, and delineate boundaries between those who will benefit and those who will not.

[221] The element of CPAG‟s argument on proportionality having the most impact was the relatively modest levels of success on the objectives of encouraging beneficiaries into work, and alleviating child poverty, when such significant government resources were committed to the WFF package overall. On the view we have taken, that argument for a lack of proportionality does not deal with the more confined analysis evaluating the proportionality between exclusion of beneficiaries from the IWTC, and the aim of encouraging beneficiaries into work. Within that issue, the extent of impairment of the right is not out of proportion with the objective of the IWTC.

The reality, however, is that any scheme that includes financial incentives to encourage those who are on benefit incomes to move into work necessarily involves conferring financial advantage on those who can make the move. To the extent that people in that class have families, it follows that their children have the chance to access the advantages as well. At the same time the children of those who cannot or will not make the move will miss out. That is so whether or not there is a formal connection between the incentives and children (as there is in the case of the IWTC). Unless one takes the extreme view that the Government cannot create financial incentives to work at all – because they infringe the right to freedom from discrimination on grounds of employment status, and are inherently unjustifiable – then a level of harm to that right (here, experienced in the sense that those who do not qualify will be „left behind‟) is inevitable.

[223] Here, the proportionality analysis focuses on the extent of the intrusion into the right to be free from prohibited discrimination, relative to the importance of the

[224] The sequence of issues suggested for the s 5 analysis in Hansen places this question at the final stage of the justification analysis. It is a form of “standing back”, and in that light requires the Court to balance the relative importance of the positives and the negatives involved. Here, we are well-satisfied that the negative impact of the discrimination is not out of proportion to the objective pursued.

[225] The outcome of our analysis is that for the vast majority of beneficiaries there is no prohibited discrimination because the differentiation of beneficiaries in the off-benefit rule does not create a real disadvantage to them. The exception is the small subset of full-time earners (for the purposes of s MD9 of the Income Tax Act

[226] Whilst that is our conclusion on that limb, the discrimination (whether just for the subset described, or if relevant for all beneficiaries) caused by exclusion of beneficiaries from the IWTC is demonstrably justified. It follows that CPAG is not entitled to the declaration of inconsistency sought.

High Court of New Zealand Decisions

Child Poverty Action Group Incorporated v Attorney-General HC Wellington CIV-2009-404-273 [2011] NZHC 1228 (25 October 2011)