Cover photo of Social Policy Journal

Race and Ethnicity in Public Policy: Does it Work?

Mason Durie
Professor of Maori Research and Development
Te Mata o te Tau
Academy for Maori Scholarship and Research
Massey University


Race-based policies have a lengthy history in New Zealand. Nineteenth century statutes relating to land, governance, public health and justice, for example, were essentially premised on the values and philosophies of the European races. In contrast, policies specific to Māori were usually introduced to encourage conformity to Western preferences. By the 20th century, although the rationale was inconsistent and frequently unclear, minority ethnicity reporting had become an accepted marker of social wellbeing. Sometimes it was used for political purposes and often as a proxy measure for socio-economic disadvantage. In that regard policies of equality between individuals and needs-based policies have tended to assume that ethnicity and race are significant only in as much as they can be subsumed under universal indicators such as social class, life expectancy and educational achievement. Twenty-first century research, however, has demonstrated that not only is socio-economic status distinguishable from ethnicity, but that universal indicators are insufficient measures of need and outcome for members of different ethnicities. Because race and ethnicity are closely aligned to world views, culture and lifestyle it is inconsistent with the evidence to exclude them from social and economic policies. Increasingly, race and ethnicity are visible characteristics of New Zealand society, and unless policies reflect that reality, diversity will be masked, best outcomes will be compromised, and assimilation will be fostered – as it was in the 19th century.


In keeping with the theme of the 2004 Social Policy, Research and Evaluation Conference, “What works?”, this paper asks a single question: Do policies based on race or ethnicity work? It is unlikely to produce a straightforward or unequivocal answer, not because there is a dearth of research about the impacts of policies on race and ethnicity, or any lack of experience with race-based policies in New Zealand, but because the answer to “What works?” depends as much on who asks the question as who answers it. How should a good result be measured? Does it “work” if it meets the objectives of the policy? Or should it be assessed according to a set of higher-order principles capable of transcending political ideologies and good intention? Or is it best to decide what works by focusing on results, using a set of outcome indicators that may be quite remote from the policy’s immediate influence?

Although “race” and “ethnicity” are used in similar contexts in this paper they are not identical in meaning. Whereas race has connotations of biological variation and genetic determinism, ethnicity emphasises social and cultural distinctiveness and places greater importance on world views, lifestyles and societal interaction. In addition, a particular type of both race and ethnicity is indigeneity. There are some 5,000 indigenous groups around the world with a total population of about 200 million, or around 4% of the global population. A long-standing bond with the land and the natural environment is the fundamental feature of indigeneity, and arising from that relationship it is possible to identify five secondary characteristics of indigeneity: time, culture, an indigenous system of knowledge, environmental sustainability, and a native language.

Before attempting to answer the question about the effectiveness of race-based policies, I will discuss briefly the history of race-based policies in New Zealand.

The English Acts Act 1854

It is worth recalling that 2004 is a significant year for New Zealand. It marks the 150th anniversary of the opening of Parliament. After the signing of the Treaty of Waitangi in 1840, when Britain assumed sovereignty and tribes ceded the right to govern to the Crown, New Zealand initially became a Dependency of New South Wales. But the following year the constitutional position of the country changed from a Dependency to a Crown Colony, governed now by the British parliament. Further constitutional change was heralded in a British statute, the New Zealand Constitution Act 1852, which provided for New Zealand to establish its own legislature and act as a self-governing colony. Two years later, in 1854, Parliament opened in Auckland and in 1865 it was relocated to Wellington.

One of the first pieces of legislation passed by the new settler parliament was the English Acts Act, sometimes know as the Imperial Statutes Act. In a single statute the Act made all English laws applicable to New Zealand. It was an economic use of parliamentary time that spared the colonial politicians the task of developing a whole raft of laws specific to the new colony. Instead, it was taken for granted that if the laws worked in England, they should work in New Zealand. Part of the Crown’s rationale for assuming sovereignty over New Zealand had been expressly to institute British law so that Māori tribes would be protected from unruly settlers and settlers would be forced to live up to their obligations as law abiding British subjects. As it transpired, British law was less protective than well-intentioned humanitarian officials in the Colonial Office had contemplated; if anything, the law was to be used as a mechanism to advance settler interests regardless of impacts on Māori.

But when Parliament opened in 1854, the prospect that Māori understandings of justice and fairness would be different in any way from those held by the English did not enter parliamentary conscience. The English Acts Act represented a peculiar mixture of patronage and arrogance. On the one hand it implemented a goal identified in the preamble to the Treaty of Waitangi to:

establish a settled form of Civil Government with a view to avert the evil consequences which must result from the absence of the necessary Laws and Institutions, alike to the native population and to Her [Majesty’s] subjects.

Yet on the other hand there was no indication that tribal lore might be based on alternative concepts of right and wrong, or different approaches to ownership, civil responsibility and societal decision-making.

English laws are founded on notions of the common law, and the common law is simply an expression of community regard for right and fair. In that sense the common law is a measure of English culture as it has evolved over centuries. English common law today differs from what it was in 1840, so that the death penalty could hardly be regarded now as an expression of common law or contemporary English culture. But the point is that law and culture are intimately linked, and English law in 1854 was as much a product of an ethnic-English culture as Māori lore was a product of tribal world views. From that perspective the English Acts Act 1854 was New Zealand’s first race-based policy. Built on the presumption that English common law had a universal dimension, the culture, customs and conventions of Britain were imposed on all New Zealanders to the benefit of a few (at that time Maori outnumbered settlers).

This might be a good point to return to the question I asked earlier: Do policies based on race and ethnicity work? From the perspective of the coloniser the English Acts Act worked very well. It introduced a series of racially inspired reforms into New Zealand and laid the foundations for a policy environment within which English common law was the norm and Māori common law (culture) was the problem. Land tenure, criminal law, taxation policies, fishing policies and the authority of the Crown had more or less worked in Britain and were now to work in New Zealand. Even before a decade after the introduction of the Act, however, Māori had concluded that the new policies were not working for them. They protested that their understandings of land ownership, customary fishing, and tribal authority were at odds with the new laws. But their protest was interpreted as defiance of the very law they opposed. It was not entirely surprising, therefore, that war should break out, which it did in 1860.

Māori-Specific Policies

In order to address Māori custom that was at odds with English custom/common law, successive parliaments introduced legislation and policies that were race-based. Māori-specific legislation can be categorised according to the objectives of the policy and the impacts on Māori. Whether they worked or not depends on whether they are measured against the achievement of parliamentary objectives or against the impacts experienced by Māori. Three major objectives and three domains of impact can be identified. Broad objectives of Māori-specific policies have included the limitation or extinguishment of Māori interests, the restoration of Māori interests (either through compensatory payments or the return of resources), and the protection of Māori interests. The domains of impact on Māori encompass impacts on property, culture and a Māori policy.

Table 1 Māori-Specific Legislation: Domains of Impact and Objectives

Objectives Domains of Impact
Property *eg land, forests, waterways, fisheries) knowledge and social arrangements Culture (i.e. Māori values, custom, language) Policy (i.e. Māori tribal and political organisation)
Provisions that limit or extinguish Māori interests Māori Affairs Amendment Act 1953
Coal Mine Act 1903
Oyster Fisheries Act 1866
Tohunga Suppression Act 1907 Māori Representation Act 1867
Provisions that restore or compensate for losses Treaty of Waitangi (Fisheries Claim) Settlement Act 1992 Māori Language Act 1987 Te Runanga o Ngai Tahu Act 1999
Provisions that protect and develop Māori interests Ture Whenua Māori Act 1993 Children Young Persons and their Families Act 1989 Resource Management Act 1991 Runanga Iwi Act 1990
Electroal Act 1993

An analysis of Māori-specific policies and legislation based on an objectives/impact matrix shows that inconsistent political priorities for Māori have resulted in oscillations between policies of assimilation and policies that support the retention and development of Māori interests (Table 1).

However, by far the greatest impact of Māori-specific provisions in legislation, mostly enacted in the nineteenth century, has been to limit or extinguish Māori interests. As a result, a range of compensatory mechanisms became necessary more than a century later. Some of the motivation for limiting Māori interests can be traced to different understandings of customary rights and the relative bluntness of a system of law derived from English cultural experience to address Māori systems of tenure and organisation. Even in modern times there is a great deal of uncertainty as to whether a determination of Crown ownership over natural resources based on the English common law is consistent with interpretations of indigenous property rights.

A number of social policy statutes – including the Education Act 1989, the Broadcasting Act 1989, and the Mental Health (Compulsory Assessment and Treatment) Act 1992 – make specific provisions for Māori, and the Public Health and Disability Act 2000 contains a Treaty of Waitangi provision. But the inclusion of a Treaty clause into legislation or the addition of another Māori-specific reference is not generally based on granting additional rights to Māori individuals, but rather on ensuring that the same rights (such as the right to receive a sound education that does not sideline Māori perspectives, or to enjoy television programmes in one’s own language, or to receive an adequate psychiatric assessment) can be guaranteed, taking into account Māori cultural values, processes and protocols. For the most part, the majority population takes those rights as givens.

Socio-Economic Disadvantage and Ethnicity

Recent debate about race-based policies in New Zealand has revealed a general lack of understanding about the objectives of policies, their application, and measures of effectiveness. Two sets of policies linked to social service delivery and affirmative action programmes illustrate some of the misunderstandings.

First, policies that provide for Māori – or other ethnic groups – to deliver social services to their own people or to target ethnic groups have been criticised on the grounds that they lead to a form of advantage that other New Zealanders do not have. The argument against specific ethnic provision is based on the goal of equity as between individuals and makes a case for a needs-based approach that is racially and ethnically neutral. Within the needs-based approach universality is emphasised and contextual variables are minimised or dismissed. Each person is to be treated equally according to “need”, regardless of wider societal associations.

In practice, however, the distinctions between individual needs, wider environmental contexts and ethnic affiliation are not so clear. The association between material disadvantage and ethnicity, especially among some ethnic minorities, has been well established in a number of studies. Compared to other New Zealanders, Māori and Pacific Peoples have higher rates of unemployment, lower household incomes and lower participation rates in early childhood and university education; and their children are more likely to live in a lone-parent family, not be immunised, have no parent in paid work and live in a household in the lowest income quintile. In addition, life expectancy is significantly lower and mortality rates are higher. However, the strong relationship between ethnicity and adverse socio-economic circumstances has sometimes led to an assumption that one is a proxy measure for the other. Being Māori, for example, is often seen as a synonym for being poor and being poor is sometimes seen as the distinguishing characteristic of Māori and Pacific peoples.

While there is a significant correlation between the two measures – ethnicity and socio-economic status – they do not measure the same phenomena. Needs-based policies and policies of equity between individuals have tended to regard ethnicity and race as significant only insofar as they might be subsumed under universal indicators such as social class, life expectancy and educational achievement. Recent research, however, has demonstrated that not only is class distinguishable from ethnicity, but that universal indicators by themselves are insufficient measures of need and outcome.

Based on an analysis of socio-economic and ethnic data, three types of ethnic inequalities in health have been described:

  • the distribution gap (Māori are not distributed evenly across all deprivation deciles and are overly represented in the very deprived neighbourhoods [deciles 8–10])
  • the outcome gap (Māori health outcomes are worse even after controlling for deprivation)
  • the gradient gap (socio-economic hardship impacts more heavily on Māori) (Reid et al. 2000).

Māori who live in the most affluent areas, for example, have health outcomes that are similar to non-Māori living in the most deprived areas. The study confirms that quite apart from social class, ethnicity is a determinant of health outcome. An intervention framework to improve health and reduce inequalities therefore recommended structural interventions that affirm power relationships as well as Māori health provider development, and health and disability services that recognise cultural needs and improved ethnic data collection (Ministry of Health 2002:18-22).

In a report on mental health outcomes, it was also shown that deprivation (socio-economic disadvantage) did not entirely explain the greater severity of mental disorders among Māori. Despite having similar levels of deprivation, Māori consumers were more likely than other groups to have higher levels of severity and lower levels of functioning. Further, in contrast to the general population, Māori who were living in areas of least relative deprivation were more likely to have higher levels of severity and lower levels of functioning than those living in areas of greater deprivation. Although bias on the part of researchers could have contributed to that unexpected finding, it might also have reflected a greater sense of cultural dislocation by Māori living in more affluent areas where there was less close contact with family networks and community support agencies (Trauer et al. 2004:83-86).

The relative roles of material circumstances and ethnicity have also received attention with respect to Māori educational outcomes. Family income and associated social and economic factors are significant determinants of outcomes, and many researchers have concluded that once socio-economic differences are taken into account there are no differences between Māori and other New Zealanders. However, instead of focusing on socio-economic differences, other researchers have examined the role of culture and language in outcomes and have concluded that there is often a mismatch between the culture of the school and the ethnic cultures of learners (Bishop and Berryman 2002). Both learners and teachers may make assumptions about “normal” that implicitly exclude Māori, while processes such as assessment can provide legitimisation for deficit views, effectively “disabling” minority children (Cummins 2001).

Evidence therefore suggests that difference in the educational outcomes of Māori children cannot be explained entirely on the basis of family incomes or class; the centrality of ethnicity and culture to outcome is a factor in its own right (Durie 2002). Deficit assumptions by teachers towards Māori have hampered progress, but when they have been addressed higher levels of achievement have been demonstrated even in low-decile schools. In other words, while family income, poverty and social class have a confounding effect, ethnicity cannot be dismissed as a relevant determinant of outcome (Biddulph et al. 2003:62-63).

Quite apart from social and/or economic deprivation, therefore, explanations for disparities may also be found in ethnic-specific causes such as genetic predisposition, customary beliefs and cultural practices; or, alternatively, discriminatory behaviour in the provision of services and access to economic opportunities, culturally inappropriate design of goods and services, and cultural differences in values and aspirations (Jacobsen et al. 2002:11-12).

Affirmative Action

A second area of contemporary debate concerns the maintenance of affirmation action programmes based on race and ethnicity. There are a number of programmes that provide targeted assistance to Māori and Pasifika students, either through government scholarships and bursaries, operational grants to tertiary education institutions (e.g. the Special Supplementary Grant) (Tertiary Education Commission 2003) or preferential entry into academic programmes. As a matter of interest it is worth noting that 2004 is the centennial year of the graduation of Te Rangi Hiroa (Peter Buck) who was the first Māori to graduate from the University of Otago. Along with Tutere Wirepa he was a recipient of a government grant made available specifically for Māori who wished to study medicine. The purpose of the grant was primarily to improve Māori health. Both Otago and Auckland universities still have an entry scheme that enables a limited number of Māori and Pacific students to enter Medical School without necessarily having the same academic profiles as other students.

Criticism of programmes such as these has been made on two grounds. First, there has been a suggestion that Māori and Pacific students who enter tertiary education under a preferential scheme are allowed to graduate with lesser standards. Clearly that view represents a gross distortion. While different criteria might be used to justify admission, once admitted, students undertake similar course work, sit the same examinations and meet the same qualifying standards.

Second, the case has been made for all students to be admitted on “merit”. Merit appears to mean that academic criteria should be the sole determinant of admission. The need for a non-Māori student with high grades to forfeit a place to a Māori student with lower grades seems wrong to those who associate academic performance with academic right. However, successful educational outcomes depend on many factors apart from earlier academic achievement. Moreover, the purpose of ethnically based preferential entry schemes is not simply to have more Māori or Pacific doctors, but for educational institutions to make a contribution to society. Education has both personal and public benefits, and according to the charters of New Zealand universities it is the public good which is to be accorded high priority.

While it makes sense to ensure that students accepted into a programme are going to be able to meet the required academic standards, it may be more meritorious to admit students who will help institutions achieve their public goals and meet charter obligations to provide for future societal leadership. It is both simplistic and short-sighted to define merit solely on the academic merits of individual students in isolation of other students or the institution’s broader social goals. In that respect it may be perfectly fair to reject a student because too many others like him or her have already been enrolled at the expense of diversity and institutional goals for a better society. Merit must also be defined according to the institution’s mission, and taking account of race helps institutions achieve their aims of having diversity on the campus and attending to long-term societal needs (Bowen and Bok 1998).


There are two main reasons why, alongside other factors (such as socio-economic status, government goals, equity and fairness), race and ethnicity should be identified as rationales for policy in their own right. First, there have been recent suggestions in New Zealand that a needs-based formula centred on individuals and their socio-economic status will suffice to meet policy requirements in health, education and social policy generally. Clearly that approach is inconsistent with the evidence and tends to assume that ethnicity is a function of economic need rather than a determinant of lifestyle, culture and social organisation. Second, an increasing diversity of ethnic affiliations is a characteristic of modern New Zealand. Although race-based policies in the past have been used to disadvantage Māori more often than to create advantage, race-based policies need not be unfair. Instead, while race and ethnicity play such large roles in societies like New Zealand, it is nonsense to act as if they were non-existent.

To return to the question asked at the beginning of this paper – Do policies based on race or ethnicity work? – the answer largely depends on the identification of policy goals and the instruments used to measure impacts. A framework for considering race-based and ethnic-based policies can be shaped around goals and indicators (Table 2). Three broad goals can be identified in current ethnic-based and race-based policies:

  • full participation in society, education and the economy (the participatory goal)
  • certainty of access to indigenous culture, networks and resources by indigenous people (the indigeneity goal)
  • fairness between members of society (the equity goal).

Table 2 - Goals and Indicators

Individual Indicators Population Indicators Comparative Indicators
Universal Specific Universal Specific Inter-population Intro-population
Participatory Goal
Indigeneity Goal
Equity Goal

In practice, indicators tend to be based on aggregated individual measures and often use the Pakeha population as a benchmark for inter-ethnic comparisons. However, three shortcomings arise from those approaches. First, while many indicators such as life expectancy have universal application, some are specific to particular populations or groups. Health outcome measures, for example, should not only reflect clinical indicators derived from evidence, but also the health perspectives arising from specific ethnic world views. Second, while measurements based on individual circumstances such as educational experience are in common use, less use has been made of collective measures whether they are linked to groups such as families or to whole ethnic populations. Third, comparisons between Māori and non-Māori populations may not be the most useful set of measures since they do not take into account the significance of ethnicity and race. Instead, comparisons over time or comparisons between urban migrants and rural Māori communities may be more informative. Comparing the health of Pacific peoples in New Zealand with health standards on Pacific islands may also provide more useful indicators of adaptability than comparisons with non-Pacific New Zealanders.

In short, indicators should be able to capture both the individual and the group, they should include universal measures and population-specific measures, and the comparative indicators should be capable of reflecting the significance of ethnicity.

Political ideologies that promote individual freedom as the foundation of modern society fail to acknowledge that societies are based on individuals who belong to groups – families, iwi, communities and races. Socialists, on the other hand, see society through different eyes. Though more inclined to recognise that groups are foundational to society, they have placed greater emphasis on class than either race or ethnicity.

But for whatever reason, it is illusory to develop policies, programmes and practices that purport to be “blind” to race and ethnicity when for an increasingly large number of people an ethnic orientation underlies both personal and collective identity, provides pathways to participation in society, and largely influences the ways in which societal institutions and systems respond to their needs. Unless ethnicity is reflected in policies, diversity will be masked, best outcomes compromised, and an assimilatory approach fostered.


Biddulph, Fred, Jeanne Biddulph and Chris Biddulph (2003) Best Evidence Synthesis: The Complexity of Community and Family Influences on Children’s Achievement in New Zealand, a report prepared for the Ministry of Education, Ministry of Education, Wellington.

Bishop, R. and M. Berryman (2002) The Experiences of Indigenous Māori Students in New Zealand Classrooms, University of Waikato, Hamilton.

Bowen, William G. and Derek Bok (1998) The Shape of the River: Long Term Consequences of Considering Race in College and University Admissions, Princeton University Press, New Jersey.

Cummins, J. (2001) “HER Classic – Empowering minority students: A framework for intervention” Harvard Educational Review, 71(4):649-675.

Durie, Arohia (2002) Te Rerenga o te Ra Autonomy and Identity: Māori Educational Aspirations, Ph D thesis, Massey University, Palmerston North.

Jacobsen, Veronica, Nicholas Mays, Ron Crawford, Barbara Annesley, Paul Christoffel, Grant Johnston and Sid Burbin (2002) Investing in Well-being: An Analytical Framework, Working paper 02/23, The Treasury, Wellington.

Ministry of Health (2002) Reducing Inequalities in Health, Ministry of Health, Wellington.

Reid, P., B. Robson and C. Jones (2000) “Disparities in health: common myths and uncommon truths” Pacific Health Dialogue, 7:38-48.

Tertiary Education Commission (2003) Review of the Implementation and Effectiveness of Special Supplementary Grants for Māori and Pasifika Students at Tertiary Education Institutions From 2001–2002: Māori Report, Tertiary Education Commission, Wellington.

Trauer, Tom, Kathy Eagar, Phillipa Gaines and Alison Bower (2004) New Zealand Mental Health Consumers and their Outcomes, Mental Health Research & Development Strategy, Health Research Council, Auckland.

Cover photo of Social Policy Journal


Social Policy Journal - Issue 24

Race and Ethnicity in Public Policy:Does it Work?

Print this page.