The social condition of the New Zealand people: a pre-election review of social indicator information

ABSTRACT

Many social commentators have considered that alongside the fiscal transparency enjoined by contemporary New Zealand governments, there should be a complementary social responsibility reporting. This task is usually assigned to social indicator frameworks. However, at present (as the 2017 election looms) there is a faltering in the provision of social indicators which have been in place in New Zealand for almost two decades, with the exception of the recent 2016 survey data from Statistics New Zealand and Ministry of Social Development that were made available within a month of writing this article. Having commented on the current status of the New Zealand social indicator system, we present data from the General Social Survey and the Quality of Life survey to at least convey recent trends in subjective social well-being and reported behaviours and experiences. References are also made to the accumulating literature on social well-being in New Zealand, followed by suggestions for more systematic indicator development and underpinning research.

KEYWORDS:

• Social indicators
• subjective well-being
• quality of life
• New Zealand/aotearoa
• pre-election commentary

The task of providing information on the social condition of the people

In the 1980s and early 1990s, widespread public dissatisfaction with successive out-going governments’ handling of information about the economy and the government’s own finances led to the passing of the Fiscal Responsibility Act 1994. This Act required regular and official, and especially pre-election, presentation of Treasury’s official views on the state of the economy and particularly the government financial situation. The 2017 pre-election report was issued on Wednesday, 23 August – a month out from the election.

Fiscal policy involves decisions about government spending and taxation which are to be made with a view to goals, such as the optimal allocation of resources, economic stabilization and the longer term sustainability of public finances. Part 2 of the Public Finance Act has two key aims: increased transparency and greater accountability which are achieved by requiring:

• Governments to be explicit about their long-term fiscal objectives and short-term fiscal intentions and to assess them against principles of responsible fiscal management;

• Governments to report on a wide range of economic and fiscal information.

However, the ‘coldness’ of economic indicators leaves many gaps in the information which should be in the hands of voters, and civil society more generally, particularly relating to the condition of the people or the collective health of society. Economic indicators do not tap subjective information on the state of people’s quality of life (QOL), nor examine the social distribution of well-being, let alone how QOL varies across ‘life domains’. Reliance on GDP as a measure of progress suffers from a well-known litany of limitations. Economic activity by itself is not necessarily ‘good’ economic activity: so, increasing GDP today may mean depleting resources for tomorrow. Reparation of the GDP measure using approaches such as ‘Genuine Progress Indicators’ is one strategy which needs to be further explored: an important local example is the Treasury’s Living Standard Framework (see special issue edited by Crothers and Fletcher 2015; Weijers and Mukherjee 2016). However, these still need supplements. In short, an inquiry about the health of the economy needs to be supplemented with an inquiry about the health of society, while the GDP and other economic measures need to be complemented with indicators that cover other important domains in order to measure well-being.

Unfortunately, the non-economic ‘social’ sector is more inchoate, with more complex issues and difficulties in rendering judgements as the social sector has never had clearly authoritative agencies who can make these judgements. Although from time to time, attempts have been made to resolve this, such as the social monitoring group of the New Zealand Planning Council, the Royal Commission on Social Policy and even most recently the Families Commissioner (currently scheduled for dissolution). Once the Ministry of Social Development (MSD) ‘Social Report’ began to be embedded circa 2005, a ‘Social Responsibilities’ Bill was drafted to sit alongside the Fiscal Responsibilities Act, but this initiative then faded. Interestingly, the Local Government Act was amended in 2002 to require local authorities to be concerned with social, economic, environmental and cultural well-being, but this provision was removed by the National Government in 2012. Despite the lack of a broad over-arching framework, it is still possible to assess progress across social sectors: the obvious measurement framework is provided by ‘social indicators’, which have already established within institutionalized practices and a considerable body of the academic literature.

Compared to economic indicators, social indicators have even more methodological difficulties, particularly since there are major problems in developing adequate measures. There are core or generic aspects of social well-being, with different aspects of this pointed out by the near-synonyms used in the literature: e.g. subjective well-being (SWB), happiness, QOL, purpose in life and resilience. In addition, it is also important to consider both objective and subjective aspects of people’s experiences across a wide range of ‘life domains’. So, a further step is to conceptualize a set of life domains, work out the relative ranking of these, and how they might contribute to an overall measure of social well-being. The social characteristics across which SWB is to be examined need to be specified. Additionally, a set of measures need to be chosen and updated over time. Finally, any social indicator framework needs to be well-institutionalized and protected from political or other sways.

Several conceptualizations have been developed that provide a rational framing for sets of indicators: such as the Organisation for Economic Co-operation and Development’s (OECD’s) well-being conceptual framework, Statistics New Zealand’s 2002 framework for measuring sustainable development, and the Treasury’s Living Standards framework: see also Statistics New Zealand social indicator development papers. Conceptualization of the core of social well-being often is considered to involve three sub-dimensions (e.g. OECD guidelines on Measuring Subjective Well- Being):

• Life satisfaction (cognitive appreciation);

• Affect (a person’s feelings or emotional states, both positive and negative, typically measured with reference to a particular point in time);

• Eudaemonics (a sense of having meaning and purpose in one’s life, or good psychological functioning).

In addition, there are various schema outlining a set of domains which should be considered. Eurostat refers to an 8 + 1 dimensions/domains, and other schema similarly indicate some 10 domains; however, the typology is somewhat arbitrary around the edges. Because of the linkages, overlaps and potential trade-offs between domains, it is important that they are analysed simultaneously. Multiple disadvantages need to be added together as often they far exceed the sum of the effects of individual characteristics on the QOL (for a useful New Zealand summative index of family risk factors using General Social Survey (GSS) data and deserving up-date see Stats NZ 2012).

The Eurostat schema is (others are similar):

• Material living conditions (income, consumption and material conditions);

• Productive or main activity;

• Health;

• Education;

• Leisure and social interactions;

• Economic and physical safety;

• Governance and basic rights;

• Natural and living environment;

• Overall experience of life.

For each of these domains, a range of sub-concerns may be specified, and an array of both objective and subjective indicators is used in their measurement. A range of social groupings across which outcomes are to be tracked must be established. In carrying out our study, we are alert to four different dimensions for evaluation (as discussed in the social indicator literature):

• The distribution between ‘good’ and ‘bad’ outcomes: preferably higher proportions of the former;

• Change over time: preferably towards better outcomes;

• The social distribution of outcomes amongst social groupings;

• International comparability: preferably with New Zealand doing better than comprador countries (although this is not explicitly pursued in this article).

Since sample data are often drawn upon, attention to confidence ranges is important, particularly when interpreting changes between years. For example, whether a rise from 10% to 15% is actually a rise depends on the margin of error.

A research frontier in the social indicator movement is to ensure that SWB is not merely seen as the quality of individual’s life, but also considers the effects of various levels on QOL: e.g. families, communities and regions.

Establishing and maintaining social indicator frameworks are now common across advanced countries. Some were propelled by French initiatives, where the need for national statistics agencies to collect and publish measures of SWB was a key recommendation of the Report of the Stiglitz–Sen–Fitoussi Commission on the Measurement of Economic Performance and Social Progress in 2009. The UK government also famously mandated a measurement of happiness. More recently, international agencies have been active in developing cross-national social indicator frameworks (e.g. OECD better living) and the UN has sponsored the Social Progress Index.

New Zealand social indicator frameworks

There was an early phase of New Zealand interest in Social Indicator work in the late 1970s/early 1980s when a few studies were assayed, before the momentum faded. From the turn of the millennium, New Zealand picked up this trend in public policy and had developed two major social indicator frameworks: the MSD’s Social Report (2016b) and a consortium of big cities’ QOL project. (For an historical review see Cotterell and Crothers 2011, and for an overview of assessment frameworks see articles in Crothers and Fletcher 2015.) In turn, other social indicators have been developed in New Zealand.

General argumentation about social indicators was most articulately advanced with the Social Report, repeated in some part by Statistics New Zealand indicators webpages. Social indicators were defined as ‘measures of social well-being which provide a contemporary view of social conditions and monitor trends in a range of areas of social concern over time’. Indicators for the report were chosen on the basis that it ‘should always be possible to interpret changes in indicators quite clearly as an improvement or deterioration in the quality of life … (and)  … should focus on the outcomes of social processes or policies, rather than inputs’. The detailed and well-displayed material is a strong feature too. Frameworks are discussed in Dalziel and Saunders (2015) and Duncan (2015); and the GSS supporting documents.

However, unfortunately as the 2017 election looms, it coincides with something of a hiatus of the armament of New Zealand social indicators. Although the QOL project continues, it is confined to six major cities, and the extensive MSD Social Report, which lasted from 2001 through to 2010 and had a brief reprise in 2016, is currently not intended to be further updated. Further work is scattered, for example:

• Statistics New Zealand provides both social indicators and a small set of social indicators within its progress indicators, but these are only partly updated and are in the process of being more systematically reworked (Statistics NZ, no date);

• Other more specifically targeted surveys can contribute SWB data: e.g. Statistics New Zealand’s Disability Survey and Te Kupenga; and there are also crucial data-series: e.g. MBIE (2017) and Perry (2017a, 2017b);

• Social Policy Evaluation and Research Unit/Families Commission have explored measures of Family/Whānau Well-being: e.g. 2017;

• Alan Johnston produces an annual set of social indicators for the Salvation Army with the most recent published in the early months of 2017;

• DeLoitte’s Regional Social Progress indicators (Jacobs 2015);

• The former Family/Whānau Well-being Project (FWWP) project is confined to census data so is currently awaiting an updating after 2018 (Von Randow et al. 2014) (had the Christchurch earthquake not upset its timing this would have been an excellent source for a partial but recently updated set of indicators based on a 2016 census);

• Sovereign Life Index: Prendergast et al. (2016); Soverign (2015).

• There are (and have been) several indicator frameworks which are specific to particular sub-populations.

Some overseas social indicator frameworks cover a range of countries, including New Zealand. These not only provide useful collections of local data but set them within a cross-national comparative frame, for example:

• OECD Better Living Index;

• OECD Regional Indicators;

• Gallup World Poll (2017, Clifton 2017);

• Material Well-being Index (Grimes 2015);

• Also the Social Progress Index (Social Progress Imperative, 2017: see also Jacobs 2015) and an array of other both general and more specific indices.

The area of SWB has also been investigated in longer-term exercises built around GSS, Census or QOL survey data: e.g. FWWP or MSD’s Living Standards studies. In addition to these indicator series, there is valuable backup from the academic literature provided by various one-off studies (for summary see Crothers and Yeung 2017):

Groupings which allow disaggregation have to be a pragmatic decision with the published GSS tables. As compared to the range of social characteristics inquired about in a full-scale academic survey (or the GSS), a far sparser coverage limits more official statistics: usually age, gender, marital status, ethnicity, region (or type of settlement) and possibly an indirect measure of social class (such as the deprivation index). Type of household and other household/whānau characteristics may be important.

Other policy-relevant groups that might be considered could include: disability or mental health condition, iwi, religious groups, and the gay, lesbian, bisexual and transgender community.

It is important that there are some attempts to round-up existing measures of the social condition of New Zealanders, but this is a large task. So rather than coming up with yet another social indicator framework, it is better to extend previous efforts. In this commentary, we do not attempt to construct an up-to-date indicator system from scratch. Two tasks are attempted, beyond a review of the state of social indicator reporting in New Zealand:

• Examining the results of the 2016 Social Report, which should be almost up-to-date;

• Developing tables on over-time trends for key GSS (and QOL) well-being survey items. The well-being indicators deployed here are a pragmatic selection provided by Statistics New Zealand without any explicit theoretical rationale.

Overviews of the characteristics of available social indicator frameworks are given in the supplementary tables. Consideration of the Comparison of Social Indicator Framework Characteristics supplementary table points to the need to draw on a wider range of indicators than are readily available in any one framework. The Social Report is the local standard against which others might be assessed, and it would seem important to endeavour to regain the capacity it provided.

The Comparison of Domains supplementary table reviews the differential coverage of the different social indicator frameworks across potential sets of domains. A wide coverage is important for ensuring that important areas of life are included. The data deployed in this article from the GSS cover 8 out of 15 domains in the full set, which imposes limitations.

Lessons from the social report, 2016

The MSD Social Report has been a particularly well developed and clearly articulated Social Indicator Framework. Its demise is unfortunate for the New Zealand Social Indicators movement. Fortunately, the 2016 Social Report is recent. Although, given the publication delays involved, the data are often at least out of date by 2–3 years. In this section, we provide a quick examination, drawing on any evidence at hand, and highlight if the indicators deployed in the MSD report have changed since it was published in 2016.

Over time, most social well-being outcomes for New Zealanders have improved or remained unchanged. In its summative comparison of change between 2016 and 2010 the Social Report found:

• Outcomes have particularly improved for the Knowledge and Skills and Safety domains.

• The Health, Paid Work, Civil and Political Rights, and Leisure and Recreation domains showed a mixed picture, with some areas showing improvements and others showing no change or a worsening situation.

• Other domains, such as Economic Standard of Living and Social Connectedness, generally show a steady, unchanging picture.

• Outcomes have generally worsened for the Cultural Identity domain, though this was based on a small number of indicators.

• Overall the various indicators showed that relatively few indicators were worsening.

Further the report had a useful summary of the demographic distributions of social well-being outcomes:

• Māori and Pacific peoples are performing less well across a number of measures, although improvements are occurring over time.

• Females continue to fall behind males in some domains, such as Economic Standard of Living, while they are ahead of males in other domains, such as Health.

• Sole-parent households consistently have poorer outcomes, particularly in the areas of Economic Standard of Living and Social Connectedness.

• Having a low income, low material well-being and/or living in an area of high deprivation result in relatively poor social well-being outcomes across most domains.

Finally, carrying out cross-national comparisons, the Social Report found that New Zealand is performing well against OECD countries:

• New Zealand is performing better than the OECD median for comparable measures in the Civil and Political Rights and Social Connectedness domains;

• New Zealand is generally performing worse than the OECD median for comparable measures in the Safety domain;

• There is a mixed picture for the remainder of the domains, with some measures showing that New Zealand is performing better than the OECD, some showing a worse picture, while some showing no difference between New Zealand and the OECD median;

• Of measures compared in the report, New Zealand did better on 14, no difference on 6 and worse on 7 domains: see Table 1.

Table 1. 2016 Social report: typology of trends of indicators since 2010.

	Recent change	%	Medium term change	%
Improving	21	40	20	42
No change	26	49	19	40
Worsening	6	11	9	19
Not applicable	8	–	12	–

Other New Zealand social indicator frameworks

Statistics New Zealand (n.d.) have two relevant scoreboards, but the data are limited and imperfectly updated, and is currently under review.

The Big Cities project (Colmar Brunton, 2016) uses a mix of subjective and objective indicators to paint a comprehensive picture of the state of each of the six cities, providing a good example of social reporting. The predominant limitation in the Big Cities project is its use of large number of indicators which will preclude an easy analysis of any change. In addition, examining differences across so many cities is another analytical challenge.

The Family and Whānau Well-being Indicators project (FWWP) was part of a five year research programme supported by the Social Science funding pool of the Foundation for Research, Science and Technology (see Cotterell and Crothers 2011). The principal goal of this programme was to develop ways to examine and monitor the social and economic determinants of Family and Whānau Well-being. Additionally, how these have changed over the period 1981–2001, although extensions bringing the data up-to-date with the 2013 census have been published (von Randow et al. 2014). Indicators of family well-being were constructed to explore the viability of using Census data to reveal trends in family well-being outcomes across various social categories, including cohorts. Advantages of this data source are that it allows analysis at the household/family level, allowing for examinations of long-term trends, and provides information on small population groupings. However, an updating of this work awaits the 2018 census results, and the 2013 results are older than needed for the current study.

MSD living standards research

The 2017 updates of the ongoing MSD series on inequality trends include both their annual Household Incomes Report and its companion report using non-income measures, using data from Statistics New Zealand’s 2015–16 Household Economic Survey (2016 HES). The former report is important in the very careful sifting of complex evidence on changing household economic circumstances. The latter report supplements this by covering aspects of quality of living issues, such as the ability to keep the home warm, pay the bills, repair appliances, have a couple of decent pairs of shoes, a good meal each day and the purchasing of commonly desired non-essentials (Ministry of Social Development 2016a). Headline findings from these two reports include:

There has been strong real income growth of 11–13% for most income groups from HES 2008–09 to HES 2015–16, apart from most of the bottom decile, where Beneficiary incomes were flat or declining in real terms. (This is better than many OECD countries whose middle incomes have been relatively flat since the Global Financial Crisis (e.g. Australia, UK, Spain, France, Italy, Germany and the US).) Since housing costs now make up a much larger proportion of the household budget for most households, after housing costs (AHC) incomes have become less favourable. There is no evidence of any sustained rising or falling trend in before housing costs household income inequality over the last two decades using either the Gini or the top 1% share measures. Housing affordability issues remain for the lower two income quintiles (not affordability for purchasing homes, just how affordable the accommodation is once in the accommodation).

In terms of material hardships, although based on a much shorter time-span of data, there are some housing quality issues, especially for children, who are concentrated among those already experiencing material hardship on other basic items. On average over three surveys, HES 2013–HES 2015: For children, their household reports a major problem with dampness and mould (10%), heating/keeping it warm in winter (13%) and for both (7%).

Low income (poverty) and material hardship trends for children are flat or falling depending on the start date or measure used. The material well-being of the vast majority of older (65+), New Zealanders continues to be good to very good with lower AHC income poverty rates and lower hardship rates than other age groups. (Internationally, older New Zealanders have very low hardship rates in league tables for the richer countries.) The small group (4–8%) that do have financial challenges are, unsurprisingly, those who rent and have little other than New Zealand Superannuation for income. Declining mortgage-free home ownership for the cohorts approaching ‘retirement’, and elevated low income rates (AHC) for older working-age adults living on their own suggest that this small group (4–8%) may grow in coming years. The ‘working poor’ remains an issue for New Zealand as for other OECD countries. An estimated one-third to one-half of children in hardship are from working families, even with the assistance from working for families and child-care subsidies.

OECD better living and regional indicators

Over recent years, the OECD (2016, 2017) has been heavily involved in the debate on measuring well-being. It argues that 11 topics reflect what is essential to well-being in terms of material living conditions (housing, income and jobs) and QOL (community, education, environment, governance, health, life satisfaction, safety and work-life balance). Each topic is built on 1–4 specific indicators (results can also be compared for gender, and for social and economic status). User-weighting drives any weighting amongst domains and in turn generates some useful information. However, while there are similar broad weights according to each of the domains, there is a slight emphasis on SWB, health status and education and skills and a downgrading of civic engagement/governance and social connections (see Table 2). Although the sample generating this data is small (n = c700) and self-selected it may provide a useful rule of thumb in considering the relations amongst domains in other studies as well. OECD data are also provided in relation to children, volunteering, regional and ranking of domains. The regional data cover 11 topics: income, jobs, housing, health, access to services, environment, education, safety, civic engagement and governance, community and life satisfaction (see attachment).

Table 2. Ranking of domains.

	Mean	Rank
Civic engagement and governance	0.06558	11
Social connections	0.07987	10
Education and skills	0.09672	3
Environmental quality	0.09419	4
Health status	0.10098	2
Housing	0.08569	7
Income and wealth	0.08378	9
Jobs and earnings	0.08507	8
SWB	0.10713	1
Personal security	0.08971	6
Work-life balance	0.09407	5
Responses	813

Another way of assessing domain importance is from explicit rankings of importance. The 2012 New Zealand Values Survey found that respondents ranked the following as important areas in their lives: family, friends, leisure-time, work, politics and religion. A ‘revealed preference’ can also be established by investigating which social background characteristics have the biggest impact on overall life satisfaction. Using 2008–2012 GSS data, Jia and Conal (2016) found that health, social network support and housing had a greater impact while education, unemployment and incomes all had medium weights, and work-life balance scored extremely low (although this may be a methodological artefact).

General social survey

Every two years since 2008, the New Zealand GSS has provided snapshots of the well-being of people in New Zealand. The survey face-to-face interviews over 8000 respondents across a wide range of life domains. Its objectives are to:

• Provide a picture of (changes in) well-being in New Zealand;

• Provide data to understand and monitor distributional issues with regard to well-being;

• Contribute to an understanding of the interrelationships between different aspects of well-being.

Since 2014, rotating supplementary modules have been added. In 2014, the module focused on social networks and support, and in 2016 the module’s topic is on civic and cultural participation, while the prospective 2018 module will be on housing and the physical environment. The tables reported here are augmented by those in the supplementary tables.

Life satisfaction and purpose of life

GSS includes a question on respondents ‘How do you feel about your life as a whole right now?’, but unfortunately, two different versions of response categories are supplied for the 2008–2012 GSS and the 2014–2016 GSS. Table 3 is a plausible reshuffling of the data to better provide a trend. For question wordings and a full set of tables see attachment. There appears to be little difference between 2008 and 2012, similar to 2014 and 2016. However, Jia and Conal (2016) found that ‘ … subjective well-being increased modestly by some three to four percent between 2008 and 2012, after controlling for variation in individual characteristics’. Characteristics associated with high levels of SWB include:

Table 3. Self-rated life satisfaction by year (2008–2016).

	Scale 1	2008	2010	2012	Scale 2	2014	2016
How do you feel about your life as a whole right now?	Neither satisfied nor dissatisfied	14.1%	12.60%	13.30%	0–6	17.4%	17.1%
	Satisfied	53.60%	54.30%	54.20%	7	19.3%	18%
	Very satisfied	32.20%	33.00%	32.50%	8–10	29%	31%
Total		100.0%	100.0%	100.0%		100.0%	100.0%

However, education does not appear to affect SWB once other factors, such as income, are controlled for. Brown, Julie, and Conal (2012) found that mental health, income, being employed, owning a home and a range of social life and community relationships were all correlated with life satisfaction.

A model by Ussher and Walker (2015) showed that health, money, relationships and housing had the strongest relationships with the life satisfaction of New Zealand adults. Ussher and Walker (2015) report that for Māori, connecting with their culture is also associated with life satisfaction. The more important the characteristic is to aspects of the Māori culture, the stronger the association between that characteristic and the individual’s level of life satisfaction. However, this linkage is not strong; relationships, health and income remain the most important factors for Māori.

For the 2014–2016 GSS, an additional question about the worthwhileness/purpose of life was asked. This correlates very highly with the QOL measure and has similar social distribution of responses. Furthermore, Crothers (2015) shows that all the QOL measures fell into a single factor. There was little change, with the majority finding life worthwhile. Given the importance of this measure, data from QOL and Gallup were used to triangulate these findings. Both these sources also report the pattern was steady.

Income adequacy

More people now consider themselves as being able to adequately support their everyday needs financially than in 2008. Gallup has several trend-lines with some suggesting more difficulty over time while others are steady (see Table 4).

Table 4. Perceptions on financial adequacy by year (2008–2016).

		2008	2010	2012	2014	2016
How well does (your/you and your partner’s combined) total income meet your everyday needs for such things as accommodation, food, clothing and other necessities	Not enough money	14.5%	15.6%	15.3%	12.2%	10.9%
	Only just enough money	32.4%	32.4%	33.1%	25%	24.2%
	Enough money	39.4%	38.5%	36.5%	45.6%	46.5%
	More than enough money	13.7%	13.4%	15.1%	17.2%	18.4%
Total		100.0%	100.0%	100.0%	100.0%	100.0%

Health

Overall self-rated health seems to have worsened slightly since 2008, with a decline in those reporting excellent health from 24% to 19%.

Housing quality

Several questions about housing quality were asked across the 5 rounds of GSS (see Table 5). One issue is coldness (at 15–20%) which has become more common since 2008. The reported condition of housing has been constant over the period (e.g. proportion of dwellings needing immediate/extensive repairs). However, problems with dampness have been decreasing steadily since 2008.

Table 5. Housing issues by year (2008–2016).

		2008	2010	2012	2014	2016
How would you describe the condition of your house or flat?	No/minor repairs or maintenance	93.4%	94.2%	93.9%	92.8%	93.4%
	Immediate/extensive	6.6%	5.8%	6.1%	7.1%	6.7%
Does your house or flat have a problem with dampness or mould?	No/minor problem	90.5%	90.6%	90.9%	93.9%	94.7%
	Major problem	9.5%	9.4%	9.1%	6.2%	5.3%
In winter, is your house or flat colder than you would like?	No	83.9%	85.3%	85.2%	78.9%	79.3%
	Yes: Always or often	16.1%	14.7%	14.8%	21.2%	20.7%

Table

• Higher incomes	• Lower time spent commuting
• Being employed	• Good social connections
• Good health	• Satisfaction with democracy
• Extraversion	• High levels of generalized trust
• Low neuroticism	• Higher quality environment
• Good work-life balance	• Lower crime

Safety and security

The GSS, and many other surveys, asks about respondents’ feelings of safety across various situations. Feelings of safety are high for being at home at night (85%), and using the internet (low 70s) and moderate for waiting for/using public transport at night (50%) and walking alone in neighbourhood after dark (60%). There has been little change in the level of safety felt by respondents in the four situations examined. Feelings of safety in other situations are included in GSS 2008–2012.

Discrimination

Respondents were asked whether ‘In the last 12 months, have you been treated unfairly or had something nasty done to you because of the group you belong to or seem to belong to?’ with the 2014–2016 question altering the wording to ‘differently’ which seemed to lead to an observable increase. There was little change between 2008 and 2012 when some 10% reported discrimination, and similarly between 2014 and 2016 (when the rate was 17%). Unfortunately, there does not seem to be any other source which would indicate whether the jump can be correctly interpreted as an effect of the wording change.

Social contact

Social connectedness with family members (that you do not live with) and friends is important as is the ability to draw on such networks in times of emergency when support is needed. The proportions interacting with family (either face-to-face or not) increased markedly from 2014 to 2016 while contact with friends decreased. Data from earlier GSS rounds are not readily comparable. A further question on isolation (up to 2012) and loneliness (2014 and 2016) over the last four weeks indicates a considerable rise from 4% most/all of the time through to 2014 to 6% in 2016, while those never experiencing loneliness/isolation fell from 67% to 60%. (However, the 2014 data-points do not fit a trend.)

Trust

In addition to some questions on trust in particular institutions, respondents were asked in general how much they trusted most people for 2014 and 2016 with 10% rising to 15% in 2016 indicating high and increasing levels. Trust is asked about in a considerable range of New Zealand surveys.

Cultural diversity

The GSS also asked ‘How easy or hard is it for you to be yourself in New Zealand?’ which showed that Ease of expression is steadily decreasing: with those answering ‘easy’ declining from 17% in 2008 to 13% in 2016. Another way of measuring acceptance of diversity is to ask about how comfortable a respondent would be with neighbours of varying characteristics. Overall, New Zealanders appear to have become more comfortable with having diverse neighbours on the grounds of religion, sexual orientation and ethnicity, but still have reservations about people with mental illnesses. Conversely, New Zealanders are comfortable with people with disabilities or long-term health conditions. The QOL also asks a related question about diversity which shows marginally increasing support. Thus, findings seem mixed.

Conclusions

The recent social indicators and social reporting initiatives in New Zealand have been too diverse and insufficiently viable long-term. In general, earlier issues of data availability and timeliness have been reduced with the advent of the GSS and the continuance of other important data collections and data analyses. But the current hiatus points to continuing challenges, not least of which includes the increasing financial constraints and the interest in using ‘big data’ which usefully links to a wider range of government official statistics although these have a strong lean towards measuring ‘deficits’. Over- time, preferably longitudinal, study using comparable data is needed to check out if there have been recent improvements in New Zealanders social conditions. In the meantime, the interim social indicators reported here based on a component of the GSS suffice to provide some useful current information.

Social conditions have tended on the whole to show general improvement and are particularly improved for the Knowledge and Skills, and Safety domains, while the Health, Paid Work, Civil and Political Rights, and Leisure and Recreation domains showed a mixed picture. The Economic Standard of Living and Social Connectedness generally show a steady and unchanging picture, while outcomes have generally worsened for the Cultural Identity domain.

In terms of international comparisons (albeit mainly with OECD and similar wealthy countries) Grimes (2015) reports:

Most of the OECD indicators, plus several indicators from the Grimes et al. (2014) study and the mean material well-being indicator of Grimes and Hyland (2015) indicate moderately high levels of objective and subjective well-being on average within New Zealand – even when measured against advanced country comparators. However, when it comes to inequality, New Zealand is one of the most unequal of the 24 comparator nations. This is shown to be the case for inequality of income (the Gini coefficient of income), inequality of household consumption (the Atkinson Inequality Measure of material well-being) and inequality of subjective well-being (the standard deviation of life satisfaction).

Jia and Conal (2016) report that Better Life Index shows that New Zealand compares favourably with the average for the OECD countries in all aspects of well-being except income and work-life balance. In terms of social distributions, those doing less well include Māori and Pacific peoples, solo-parent households and those households with low material conditions, especially where living in areas of high deprivation.

Without access to adequate indicator systems, it is impossible to track the complex ongoing social conditions of New Zealanders and their households/Whānau and communities. This lack is particularly problematic in an election period when such information should be readily available. Provision of some social indicators needs substantial measurement investment (e.g. housing affordability, household living standards and subjective satisfactions) and this task inevitably must be spread amongst several responsible agencies. It is not enough to merely provide social indicators, but they must be underpinned by a research programme testing for validity, reliability and the causal social mechanisms creating the observed patterns. In turn, social measurement needs to be anchored in politically secure long-lasting governance arrangements. Arguably, the Government (People’s!) Statistician should exercise their overall statutory statistical responsibility to ensure that there is a coherent and comprehensive, theoretically based and empirically populated indicator framework. But Statistics New Zealand has yet to equivocally ‘put their hand up’ for this task and should be encouraged to do so. Perhaps social indicators should be linked to some social overview panel which would draw on non-political governmental, voluntary/welfare sector and academic expertise to provide authoritative commentary on social conditions but also to supply the technical expertise in social measurement. While the now-to-be-disestablished Families Commission might have been one setting for such a panel, perhaps Royal Society Te Apārangi might turn its attention to setting up a panel for this task.

Disclosure statement

No potential conflict of interest was reported by the authors.