The Norwegian policy to reduce health inequalities : key challenges

The Norwegian strategy for reducing health inequalities from 2007 has been recognised as one of the most ambitious and encompassing in Europe. By proposing action on the social determinants of health, such as income structure, employment opportunities and affordable child-care, the strategy was able to approach the entire social gradient rather than just the socially disadvantaged. In this article, we present the main features of the health equity strategy, and discuss possible obstacles to a successful implementation and a prolonged commitment to reducing health inequalities in Norway. We raise three major concerns: 1) a stubborn fundamental inequality structure, 2) a lack of focus on the gradient in the implementation of cross-sectoral reforms and 3) a possible re-orientation of policy away from redistribution and universalism.


INTRODUCTION
Norway used to be a laggard in recognising health inequality as a social problem in need of political solutions (Dahl, 2000, p. 10;Dahl, 2002).However, once this was establishedafter much debate following the now famous paper by Mackenbach et al. (1997), which showed that the figures for Norway were no more favourable than the figures for most European countries -Norwegian policy caught up fast (Giaever & Torgersen, 2009).According to Whitehead andPopay (2010, p. 1235), Norway was probably the first country to adopt a social gradient approach in its national strategy to reduce social inequality in health in 2007 (Report No. 20 (2006-2007) to the Storting, 2007).The 'social gradient approach' acknowledges that health inequality is not a dichotomy between the haves and have nots, but a feature of society running across the entire socio-economic structure.
Being a laggard probably was beneficial to the swift and ambitious turn of policy, as Norway could build on extensive research and experience (Whitehead & Popay, 2010, p. 1235).Furthermore, compared to most European countries the Norwegian welfare state had a longstanding commitment to redistribution, work-centred social policy and universalism.Thus, in a sense Norway had already been taking action on the social determinants of health for many years, although not framed as such.The national health equity strategy has a tenyear perspective, thus implying that the current Government is also expected to follow it.
The process of policy formation can be viewed as a linear process from measurement, recognition and awareness-raising, to concern, will to action, isolated initiatives, structured developments, and finally into comprehensive coordinated policy (Whitehead, 1998, p. 471).Critical points in this ' Action spectrum on inequalities in health' are awarenessraising, which may lead to denial/indifference instead of concern, and concern may be followed by mental block instead of will to action.In his 2002 appraisal of the Norwegian policies on health inequalities, Dahl (2002, p. 72) noted that 'Norway is located in the lower end of Whitehead's action spectrum (...).It would be an exaggeration that the attention is on the move to concern' .Yet, five years later an ambitious strategy was launched.According to Strand and Fosse (2011, p. 380), this was made possible in part by an unpredictable change of government in 2005 -a window of opportunity -to which policy entrepreneurs within the Directorate of Health and the Ministry of Health responded by 'hooking a proposal to a political momentum' .The process thus cannot be understood as a simple linear development, Strand and Fosse argue, although the entrepreneurs had been pushing the issue for many years, but rather as a mix of linear and non-linear processes.Furthermore, as underscored by Whitehead's (1998) analysis, these processes depend on the policy context and on influential external factors.
Policy formulation is not the same as policy implementation, however (Sutton, 1999).Non-linearity may also imply retraction.If a policy initiative does not have the necessary support across the political spectrum, or insufficient public awareness, or if bureaucrats and policy makers at different levels ignore or modify the policy during implementation, the development may reverse.Just as political winds and external economic and demographic factors may have fuelled the initial agenda to reduce health inequalities (Whitehead, 1998), they may at some point turn against it.
The question we address in this article is whether the commitment to pursue a policy that aims at reducing health inequalities will last.Now that an awareness of the problem and a strategy to solve it is ready in place, will policy efforts to reduce health inequality be further strengthened, stagnate or even be reduced?In this article, we present the Norwegian policy to reduce health inequality -and discuss whether recent policy reforms and signals represent continuity or divergence from the intentions and principles underpinning the original strategy launched almost ten years ago.As such the article provides an analysis of policy documents and government documents describing recent reforms that are relevant for the topic of health inequalities and their social determinants.Hence, our data sources are policy documents (e.g. Report No. 20 [2006-2007] to the Storting, 2007and Report No. 19 [2014-2015] to the Storting, 2015), and scholarly assessments of relevant policy initiatives such as Giaever (2013) and Giaever and Torgersen (2009).

THE NORWEGIAN STRATEGY FOR REDUCING HEALTH INEQUALITIES
In 2007, the left-centre coalition government, formed in 2005, launched the Norwegian strategy to reduce social inequalities in health (Report No. 20 [2006-2007) to the Storting, 2007).Three important characteristics of the Norwegian strategy were 1) a holistic, broad cross-sectoral approach, 2) an explicit focus on the gradient and 3) the principle of 'proportional universalism' , i.e. population-covering policies in combination with more targeted measures (Dahl & van der Wel, 2015;Giaever, 2013).
The cross-sectoral approach implied that the aim of reducing health inequality was integrated in several policy fields and ministries (Dahl & Lie, 2009, p. 517).The strategy targeted five selected domains of actions to reduce health inequalities: childhood/adolescence and education, work and working conditions, income, health services and health behaviour, as well as social inclusion.This approach was judged to be rather successful because all ministries had a focus on redistribution; inter-ministerial organisation structures were available; and there was an articulated political will to act on the problem (Dahl et al., 2014, p. 304).The cross-sectoral approach was facilitated by the perspective of social determinants of health that helped identify common goals and interdependencies between sectors.
In contrast to most previous efforts, the Norwegian strategy targeted the 'gradient' rather than disadvantaged groups.Consequently, universal measures were preferred.General welfare schemes were perceived as less stigmatising, more effective in preventing people from ending up in high-risk positions, and at the same time capable of protecting disadvantaged groups.A targeted approach only has the latter advantage.However, the strategy also acknowledged that additional efforts directed towards vulnerable groups might be necessary, hence the notion of 'proportional universalism' .
By these approaches, the Norwegian strategy to reduce social inequalities in health adopts the perspective of the WHO Commission on Social Determinants of Health (Dahl & Lie, 2009, p. 520).

KEY CHALLENGES
We now raise issues that we consider to be major challenges in the policy to tackle health inequalities since 2007.There have certainly been areas that have been successful, such as the vastly increased availability of kindergartens (Sibley et al., 2015), tax reforms (Giaever, This article is downloaded from www.idunn.no.© 2016 Author(s).This is an Open Access article distributed under the terms of the Creative Commons CC-BY-NC 4.0 License (http://creativecommons.org/licenses/by-nc/4.0/).2013, p. 15) and efforts to reduce drop-out (Høst, 2009(Høst, , p. 2011)).In this short article, however, we highlight the following possible obstacles to a successful levelling out of health inequalities; 1) a stubborn fundamental inequality structure, 2) implementation of cross-sectoral reforms and 3) a possible re-orientation of policy away from redistribution and proportional universalism.

Stubborn inequalities
Together with the other Nordic countries that belong to the Social Democratic camp, Norway is considered to have a "passion for equality" (de Toqueville, 1835).Nevertheless, welfare institutions and policies are far from identical across all Nordic countries.In Dahl and van der Wel (2016), we have discussed similarities and differences in public health policies in the Nordic region.Yet, common features are that the welfare state is comprehensive and offers free or cheap health care and education for all.The system of social protection is universal and rather generous (Dølvik et al., 2015).The welfare state thus has a distinct levelling effect on the distribution of disposable income (Dahl et al., 2014).
Despite the egalitarian ethos and redistributive institutions in Norway, the distribution of many goods and evils are uneven in favour of the better off.Dahl et al. (2014) have summed up some of the hard facts: Living conditions among families with children vary quite a lot, and poverty rates, in relative terms, have increased over the past decade to reach the average level of the entire Norwegian population, i.e. around 8 per cent.Despite full coverage, the use of kindergartens are still skewed in the favour of small children from well off families.If anything, socioeconomic inequalities in academic achievement in the school system have widened over the past decade.Socioeconomic differentials in drop-out rates from high-school have been large and remarkably stable since the right to 12 years of schooling was introduced in 1994.Income inequalities, as measured by the Gini-coefficient, have increased over the past decades, but are still at comparatively low level, i.e. at 0.24.Although Norway is renowned for a compressed wage structure, inequalities in earnings and wages have also grown over the past decades.As in other nations, wealth is very unequally distributed: More than half of the wealth accrues to the 10 per cent on the top of the income distribution.
The Norwegian economy and labour market were mildly affected by the 2007 financial crisis and had a quick recovery.However, due to current low oil prices and reduced activity in the petroleum sector, unemployment has risen to almost five per cent, i.e. a doubling over a couple of years (Statistics Norway, 2016).

Implementation of large policy reforms
A risk of cross-sectoral policy initiatives, like the Norwegian policy to reduce health inequalities, is that there will often be multiple and possibly conflicting aims.When reforms are being launched in areas such as education or work, overall effect expectations might override a focus on inequality, if at all considered in programme and evaluation design.Furthermore, knowledge about health inequality among actors, as well as the resources and power available to them, may differ significantly across sectors and governance levels.In this section, we will review some key reforms initiated or implemented in the aftermath of the Norwegian strategy to tackle health inequality: the Public Health Act, the Coordination Reform, and the Inclusive Working Life Agreement (IA-avtalen).
This article is downloaded from www.idunn.no.© 2016 Author(s).This is an Open Access article distributed under the terms of the Creative Commons CC-BY-NC 4.0 License (http://creativecommons.org/licenses/by-nc/4.0/).

The new Public Health Act and the Coordination Reform
A new Public Health Act was introduced in 2012.Reducing social inequality in health was an important aim: 'The purpose of this Act is to contribute to societal development that promotes public health, including the levelling out of health inequalities' (Folkehelseloven, 2012 §1, our translation).
The act underscores the need for action on the social determinants of health, and the 'health in all policies' idea (Fosse & Helgesen, 2015, p. 330).The new Act shifted more responsibility for public health back to local communities by assigning Norway's 428 municipalities an important role.The municipalities are expected to monitor social inequalities in health and its main social determinants, and to take coordinated action across sectors and administrative levels.The municipalities were advised to employ a public health coordinator to initiate and secure coordination at the local level (Ibid, p. 331).
To strengthen the cross-sectoral anchoring and preventive potential of public health work, the ambition to dampen health inequalities were also implemented in the Planning and Building Act of 2008, where section 3-1 f) enforces plans to 'promote public health and counteract social inequalities in health (...)' (Ministry of Local Government and Administration, 2008).This strategy secured that ways to improve health inequalities now has to be part of the crucial societal planning at the regional and local level (Higdem, 2015).This way, municipalities, in co-operation with national and regional government and civil society, may act on health inequalities through municipal services (e.g.kindergartens, schools, social services and primary health care), as well as in local planning and transport, agriculture, environmental issues, and culture and business development (Tallarek née Grimm et al., 2013, p. 229).
The new Public Health Act was part of the Coordination Reform.However, in the underlying government report (Report No. 47 [2008-2009] to the Storting, 2009), health inequality was barely mentioned.This was also the case in the evaluation programme documents (Dahl et al., 2014, p. 270).
To a certain extent, this lack of consistence between intentions and implementation has also been observed in the municipalities' subsequent work.Based on analyses of documents and expert interviews in 2011, Tallarek née Grimm and colleagues (2013) find that many municipalities have a rather individualistic and sectoral-oriented focus in their approach to health inequalities.The municipalities defined drug abuse, mental health, nutrition and physical activity as the biggest challenges.They gave little notice to factors outside the domain of the health sector, such as housing, education, youth living conditions, economic circumstances and the labour market.The sectoral focus was still prevalent in 2014 (Fosse & Helgesen, 2015).However, 40 per cent of the municipalities confirmed that living conditions was the main challenge in their health promotion work (Hagen et al., 2016, p. 5), and this was associated with larger municipality size.
Another recurrent concern is the financing of coordinated actions to reduce health inequalities.These have mostly been temporary ring-fenced budgets, which serve to protect financing from local cuts and create important incentives (Tallarek née Grimm et al., 2013, p. 231).However, ring-fencing public health budgets may be counterproductive in promoting a strategy that is supposed to be cross-sectoral and directed towards social determinants, which means that money should be allocated and spent where they can be expected This article is downloaded from www.idunn.no.© 2016 Author(s).This is an Open Access article distributed under the terms of the Creative Commons CC-BY-NC 4.0 License (http://creativecommons.org/licenses/by-nc/4.0/).
to have the largest impact (Hunter & Marks, 2016, p. 143).Fosse and Helgesen (2015, p. 343) argue that funding for measures targeting the social determinants of health should be long-term in order to 'secure sustainability and ensure that the objectives of levelling out the social gradient are met' .On the other hand, general stable funds may result in large differences in resources, measures, competencies and effects across Norwegian regions and municipalities.
Finally, the localisation and knowledge of the public health coordinator may be crucial.For the coordinators to influence important planning processes across municipal sectors, they need to be placed with the executive staff not with health deputies (Fosse & Helgesen, 2015, p. 342).In 2014, less than a third of the public health coordinators were located among the executive staff (Ibid, p. 337).

The Inclusive Working Life Agreement
Employment is an arena by which many fundamental living conditions become available: income, social relations, opportunities to develop one's skills, a sense of cohesion and the enjoyment of being appreciated and useful.As the risk of not being employed have increased among those with poor health, and even more so if combined with low education (van der Wel et al., 2010), helping more people to find a job may be considered a sensible measure, particularly if finding a job means that a family with children will escape poverty.
Over the years, many initiatives have been launched to increase employment among disadvantaged people.The Inclusive Working Life Agreement was a three-partite agreement between the Government, the trade unions and the employers' unions (Hagelund & Bryngelson, 2014).The agreement had three overall aims: 1) Reducing sickness absence rates, 2) Increased employment among people with disabilities, and 3) Increased employment in the 50+ age group.
Even though the agreement has been renewed several times, an explicit aim to reduce social inequalities in the employment consequences of ill health is yet to be included (Dahl et al., 2014, pp. 179-180).Neither has the agreement focused much on improving working conditions to prevent health-related exclusion from the labour market (Dahl et al., 2014, pp. 205-6).The most recent evaluation of the agreement found that sickness absence rates have been reduced, whereas there was hardly any effect on the employment rates of disabled people (Ose et al., 2013).We are not aware of any evaluation of possible health effects of the agreement.Mehlum (2011) has criticised the agreement for caring too much about getting sick people back to work, rather than making sure that people do not get sick from work in the first place.On the basis of presented evidence, she argues that this latter measure may be a much more effective way of reducing sickness absence than the current alternatives.
This lack of focus on social inequality in the planning, implementation and evaluation of large reforms in the aftermath of the strategy to reduce health inequalities is paradoxical, particularly given the fact that the government that launched the strategy remained in office until 2013.What we have seen for the Coordination Reform and the Inclusive Working Life Agreement was also true of the Welfare and Administration Reform, another major public sector reforms.
This article is downloaded from www.idunn.no.© 2016 Author(s).This is an Open Access article distributed under the terms of the Creative Commons CC-BY-NC 4.0 License (http://creativecommons.org/licenses/by-nc/4.0/).

A reorientation of policy? White paper on public health
In 2015, the Government presented its white paper on public health called 'Coping and Opportunities' (Report no. 19 [2014(Report no. 19 [ -2015] ] to the Storting, 2015).The priority areas are mental health, healthy life style, active elderly, children and young people, and cross-sectoral public health work.These are all important components of public health policies, but it is unclear how these areas are related to initiatives to tackle social inequalities in health.Nonetheless, in the white paper the Government re-emphasizes the goal to reduce health inequalities.Reducing mortality differentials are also seen as a means to increase life expectancy in general.Yet, these bold equity ambitions are not translated into concrete policies.To the degree that the white paper addresses policies to tackle inequality, it is not the gradient that attracts attention, but the gap.One prominent example is the concern with child poverty.The white paper also focuses on health related behaviours rather than the social determinants of health, a term hardly mentioned.

Admission of temporary work
In 2015, several amendments in the Working Environment Act were implemented.The most controversial issue was the ruling to admit temporary employment (Røed Steen, 2016).The change in the Act implies a general admission to hire employees for up to 12 months on temporary contracts.The work tasks can thus be limited in time or permanent.Three constraints to this general admission apply, among them a quarantine of 12 months on the actual work tasks, and the imposition of an upper limit of 15 per cent of the staff that can be hired temporarily (Ministry of Labour and Social Affairs, 2015).Advocates of the amendments in the Act argued that they would ensure greater flexibility in the working life and give more people, in particular vulnerable groups like young people and the disabled, a better chance to enter working life (Innst. 207 L [2014(Innst. 207 L [ -2015(Innst. 207 L [ ], 2015, p. 4), p. 4).However, research carried out by the OECD suggests that introduction of temporary work does not lead to more jobs and recruitment of more vulnerable workers, but rather to a larger proportion of temporary jobs in the labour market (Røed Steen, 2016).A literature review concludes that temporary workers have poorer mental health than permanent employees (Virtanen et al., 2005).

Cuts in the wealth tax
The Government has for three years in a row initiated cuts in the tax on net wealth.For the period 2014-16, the total cut is estimated at 5.4 billion NOK (NTB, 2016).The justification for these tax cuts is to allow small and medium size companies to free means for investment in economic activity, enable innovation and increase employment (Prop. no. 41, 2014).The critics point out that other tax cuts are much more effective in creating new investments and that such tax cuts favour the rich and super rich.For example, the tax cuts in 2016 imply that the 1400 richest in the country on average will receive 43,200 dollars more in return on their wealth and profit assets.The average tax cut for ordinary people with much more moderate wealth is 24 dollars (Mosveen et al., 2015).
This article is downloaded from www.idunn.no.© 2016 Author(s).This is an Open Access article distributed under the terms of the Creative Commons CC-BY-NC 4.0 License (http://creativecommons.org/licenses/by-nc/4.0/).

Cut in the child allowance in the disability scheme
In 2016, the Government introduced a cut in the child allowance to parents who receive disability benefit.According to the new rule, child allowance will be reduced if the sum of disability benefit and child allowance exceeds 95 per cent of pre-disability income (Ruud, 2016).A likely consequence is that low income earners with many children will lose, for them, significant amounts of money: for example, a single disabled and low paid parent with three children will lose more than 2400 dollars per year.The more children the parents have, the more they will lose (Ruud, 2016).

CONCLUDING DISCUSSION: WHAT FUTURE FOR THE NORWEGIAN HEALTH EQUITY STRATEGY?
The current Government's recent white paper on public health maintains that reducing social inequalities in health is a high-priority goal.However, there is an inconsistency between the bold equity goals and meagre, or even absent, means.Little attention is paid to policies to tackle health inequalities despite existing knowledge both on the causes and on effective interventions to reduce such inequalities.Therefore, there is a danger that the Government will not be able to achieve its own equity goals, and of a hollowing-out of the ambitious 2007 strategy.In this paper, we have pointed out three key challenges: stubborn inequalities; a lack of equity focus in the planning and implementation of important reforms; and a possible reorientation of policy.
The most recent national policy reforms do not in general reflect equity issues (Giaever, 2013) or seem to have the potential to even out existing inequalities.Comprehensive policy initiatives such as the Work and Welfare Administration Reform (Askim et al., 2010), the Coordination Reform (Romøren et al., 2011), and the Inclusive Working Life Agreement (Hagelund & Bryngelson, 2013) were never designed to affront social inequalities, and were never evaluated in terms of their potential effects on health inequalities (Dahl & van der Wel, 2015).Furthermore, and although the Public Health Act is an important step forward, there is a long way to go both at the national and the local level before smooth crosssectoral cooperation becomes a reality.It is still too early to tell with certainty, but these policy reforms will probably lead to larger rather than smaller inequalities in key social determinants of health, e.g.work, income and wealth, and hence in the longer run probably to larger health inequalities.It also still remains to be seen whether the transferring of more responsibility to the municipalities in The Public Health Act will strengthen or weaken efforts to reduce social inequalities in health.Most of the municipalities (82 per cent) believe they are capable of reducing inequalities in health, and the larger municipalities more often than smaller ones see living conditions as a main challenge in health promotion (Hagen et al., 2016).
Policies to reduce health inequalities do not seem to follow a linear development (Whitehead, 1998).Just as Strand and Fosse (2011, p. 379) point to the change in government coalitions in 2005 as an illustration of a non-linear movement boosting the agenda on social inequalities in health in Norway, the shift in government in 2013 may be a similar illustration of a break in agenda setting.This non-linear development is partly seen in new priorities in national public health policy, and partly in recent national reforms in the