Printer Friendly

Chapter 2: addressing the long-term challenges of fiscal policy.

Despite substantial income from petroleum wealth, Norway is nevertheless confronted with fiscal challenges in the long term. The number of older persons (aged 67 and above) is projected to almost double by 2060, making this age group as large as 40% of the working-age population. Because the bulk of net public transfers goes to older people, government expenditures will increase sharply in the absence of reform. On this basis, the Norwegian authorities currently estimate a long-term fiscal gap, defined as the difference between the structural non-oil deficit and the expected return on the Government Pension Fund Global, of about 6% of mainland GDP in 2060. In other words, by 2060, fiscal measures to increase revenues or reduce expenditures in the amount of 6% of GDP are needed to secure the sustainability of public finances. These estimates are surrounded with uncertainty and rely on a number of stylized assumptions, but taken at face value they imply that major policy changes are required, so as to avoid an undesirable increase in the tax pressure. Completing the pension reform and reforming the disability and sickness leave schemes, which are both crucial for achieving strong labour participation in the future, would make important contributions. In addition, this chapter argues that there is a large unexploited potential for providing effective public services at lower costs, notably in the areas of municipal services, such as education, health but also in tax expenditures.


Norwegian policymakers are confronted with the difficult challenge of keeping fiscal trends on a sustainable path in the long term as ageing puts public finances under strain. Although income from the petroleum wealth is substantial, this source of revenue will stop making growing contributions to the budget around 2025, while age-related spending will continue to rise well after Norway has extracted most of its petroleum reserves. Although there are currently large overall budget surpluses and financial pressure will not be felt for several decades, it would be a mistake to delay the day of reckoning, as this should just worsen the problem faced by future generations. As this chapter argues, it is necessary to complete a number of reforms in the social protection sector but also to improve public sector efficiency to meet these long term fiscal policy challenges. The Norwegian public sector is large and there is scope for delivering public services at lower costs. Potential savings in the sectors of education and health, where services are essentially provided by municipalities, are sizeable as well as the leeway for improving outcomes. This chapter also discusses several features of the tax system, which could be made more neutral, particularly through the reform of tax expenditures.

Long-term fiscal challenges require policy changes

The current fiscal framework was conceived with the idea of smoothing the spending of oil resources across generations. The fiscal rule has the merit of being robust in the face of uncertainties, as it does not depend on unknown parameters, such as future oil prices, asset prices and undiscovered natural resources. The fiscal rule is however not designed to ensure long-term sustainable public finances. Incomes drawn from the accumulation of savings in the GPFG will not be sufficient to finance increasing pension and long-term care expenditure when the Norwegian population will be getting older. Between 2010 and 2060 the number of people aged 67 and above will almost double, representing at the end of the period 40% of the working age population (Figure 2.1, first panel). As the bulk of net public transfers goes to this age group (Figure 2.1, second panel), government expenditures (including the effect of ongoing reforms) are estimated to increase by 12% points of mainland GDP by 2060, as compared to an already substantial increase of 5% of GDP in the EU15.

The Norwegian Ministry of Finance estimates a medium term fiscal gap, defined as the difference between the structural non-oil deficit and the expected return on the GPFG, of 6% of mainland GDP by 2060. These estimates are surrounded with uncertainties and rely on a number of stylised assumptions, such as unchanged participation rates and working time, stable productivity growth and unchanged standards and coverage of publicly financed services. Taken at face value, however, they imply that some major policy changes are required to meet the long term fiscal gap. Figure 2.2 shows that higher employment rates, as possibly implied by the completion of the ongoing pension reform (see Box 2.1), would almost suffice to close the fiscal gap. The gap would however drastically increase in case of lower average working hours (i.e. if hours worked were reduced by 16%, which corresponds to the trend reduction in average hours worked in line with the yearly average rate between 1990 and 2007). Assumptions on the effectiveness of health care spending and standards of services are also important to determine to what extent future revenues will fall short of expenditures, although other factors than policy may induce healthier ageing.



These illustrative simulations suggest that increasing labour-market participation will be extremely important to ensure the long-term sustainability of public finances. This requires a number of actions in many domains, including completion of pension reforms, rationalisation of the social protection system and reform of the sickness leave and disability systems. These issues are examined in turn.

The pension reform must be completed soon

The recent reform of old-age pensions in the National Insurance Scheme was a major step to encourage later withdrawals from the labour market (Box 2.1). After years of preparation, the pension reform was finally adopted in May 2009 and transition to the new system will begin in 2010. The enactment of flexible retirement age provisions has however been postponed to 2011. To complete the reform of the pension system there remain however some open issues, notably the early-retirement scheme and the occupational pensions in the public sector and the National Insurance disability pension.
Box 2.1. The Pension Reform

With the aim of making the National Insurance Scheme (NIS)
financially sustainable and raising the actual retirement age, the
pension reform is based on a notional (unfunded) defined
contribution system. The retirement age is flexible from 62 years
old onwards, based on actuarial neutrality at the margin. Pension
benefits are to be based on lifetime earnings (indexed on average
earnings) with a contribution rate of 18.1% up to a ceiling of NOK
517 455, and calculated on the basis of life expectancy at the time
of retirement. The accrual rate has been calculated as to
correspond to a replacement rate from 67 years of 1.35% per year in
employment, assuming life expectancy in 2010, giving a replacement
rate of 54% for constant income during 40 years. Before retirement
pension entitlements are indexed on wages, and after retirement on
wage indexation less 1/2 per cent. Pensioners will also be given the
option to continue working full time or part time with no reduction
in benefits. The phase in of the new system is gradual, 10% for
1954-cohort and 90% for 1962-cohort. As a result of the reform,
pension expenditures are estimated to be reduced from about 14% of
mainland GDP in 2050 to about 11%.

The new AFP-scheme in the private sector, currently examined by the
Parliament, covers people who are employed in AFP-companies at the
age of retirement (about half of the total). The current
AFP-benefit is transformed into a lifelong supplement to the new
old age pension in the NIS, based on the same principles as the new
NIS scheme, i.e. adjusted for life-expectancy, actuarially neutral
at the margin, indexed on wage until retirement and on wage minus
0.75% after retirement. The phase-in of AFP benefits is also
gradual. The AFP reform will cost a cumulative NOK 95 billion
(almost 6% of current annual mainland GDP) over the years 2011-50
and for all cohorts 1944-62.

The private sector early retirement scheme (AFP) used to provide a very strong incentive to retire early. Under an agreement in the 2008 wage round it was transformed into an income supplement for most people over 62, actuarially adjusted for the actual age at take-up and indexed on wages until retirement. The government agreed to finance one third of the scheme, employers financing the rest. Though reform of the AFP scheme is supposed to restore marginal incentives to work, the income effect is still likely to reduce participation among people above 62 years old. In addition, the government awarded two relatively costly concessions during the 2008 wage round, to smooth the transition for the oldest cohorts.

Reform of the remaining retirement pension schemes, essentially in the public sector, should be pursued. The absence of reform entails strong incentives to retire between 62 and 65 years old in the public sector. It is disappointing that, during the 2009 wage negotiations, neither the occupational pension scheme nor the early retirement scheme have been reformed along the lines of private sector reforms (with the exception of longevity adjustment and the introduction of a somewhat weaker indexation of benefits after retirement). This clearly shows that wage negotiations work asymmetrically over the cycle in terms of concessions made to the labour force (as noticed in the 2008 Survey, concessions to older cohorts were granted in a period of extremely tense labour market). No progress has been made on reform of the social security disability pension. A special commission issued a report in 2007, and a proposal to the Parliament is expected in 2010. The reform of the old-age pension scheme for former disability benefits recipients should incorporate the same principles as the National Insurance pension reform, such as longevity adjustment and indexation of benefits. Reforming the old-age pension scheme for disability benefits recipients should also go hand in hand with a more general reform of the disability schemes, as only actions on the latter could provide a consistent long-term solution to the trend increase in the number of disability recipients. However, little progress has been made on the reform of disability schemes, as discussed later. Finally, occupational pensions in the private sector are also under scrutiny, with the Banking Law Commission being due to present a report in spring 2010.

By OECD standards, net replacement rates for old-age pensioners remain relatively high for low income workers while they are less generous than the OECD average for average and high incomes (Table 2.1). However, as a result of the pension reform, net replacement rates have declined less in Norway than elsewhere, even for higher income workers. That, as well as the fact that the GDP per capita is rising fast, may still entail a high income effect and thus lower labour supply, as people replace work with leisure. Though the ultimate effect on social welfare is mixed, a reduced number of working hours will certainly add pressure to long-term finances because although the pension system itself is actuarially neutral at the margin, reduced average hours affect other tax revenues. It is thus urgent to reform the schemes that still imply strong disincentive to continue working after 62 without any further delay.

Administrative reorganization should lead to scale economies in the welfare system

Social protection has been a rapidly growing part of public spending, rising to 40% of general government spending. In July 2006, the new Employment and Welfare Administration (NAV) was created by merging the public employment service, the national insurance service and municipal welfare administration. The system of income support payments is being rationalised, in that the unemployed, sick and disabled and those in need of income support will in the future have just the one office to attend. Implementation of the reform at the central level--i.e. merging the two sectors--began in 2006, but implementation at the local level is taking place more gradually than initially expected. All NAV offices should be in place in 2010, which makes it premature to judge the overall effects. The strategic aim of the reform is to get more people into the workforce and off different types of benefits, with the secondary aim of increasing efficiency, which is to come from a rationalised structure. A major argument has been that economy-of-scale effects could be achieved by merging sectoral organizations and developing local partnerships. However, two other factors have run counter to this: firstly, that the merger has made the new organization very complex, and secondly, that strong pressures from the trade unions have ensured that all employees will keep their jobs after the reform.

A double dividend could come from reforming the disability benefit system

Reforming disability benefits would give a double windfall gain to long-term public finances, by on the one hand allowing sizeable expenditure savings and, on the other, increasing participation and hours worked, thereby boosting tax revenue. Norway spends a relatively high proportion of its resources on disability programmes: at just under 3.2% of mainland GDP, spending on disability benefits is higher than in any other OECD country and about 10% of the working-age population is registered as disabled, the second highest proportion in the OECD area. This has probably to do with generous replacement rates and loose eligibility criteria (OECD, 2009a).

Recent OECD work finds that, in Norway, the share of disability benefit recipients has increased by more than 2 percentage points in the last two decades, further widening the gap with respect to the OECD average (in Norway more than 10% of the working-age population receives a disability benefit versus less than 6% in the OECD area). Policy strongly influences the share of disability recipients: more generous disability and sickness benefits are associated with higher numbers of beneficiaries (OECD 2009a). Employment and rehabilitation programmers tend on the other hand to reduce the number of recipients. (1) Norway has made good progress in both directions since the beginning of the 1990s, by reducing replacement rates and by improving labor market integration opportunities for the disabled. But only around 1% of the disabled leave the benefit scheme each year due to successful rehabilitation or resumption of work, suggesting that more effort is needed to keep disabled people into the labor force. Other well-known problems are the almost sole responsibility given to family doctors to vet the qualification process, which is one of key factors making for exceptionally high benefit-award rates, since doctors find it difficult to act as gate-keepers where they are too "close" to patients, and a very low benefit rejection rate, while appeals by rejected applicants are more frequently successful than in other countries.

Another weakness of the system is that the entry gate to disability schemes is longterm sickness leave. Eligibility criteria for long-term sickness leave are relatively generous. In addition, there are few incentives for employers to discourage the longer-term increase in sickness benefits, since they only contribute for the first 16 days.

As recommended by the previous Survey and by the OECD Sickness, Disability and Work Review (2006), it is necessary to tighten the access to sickness and disability receivers, notably by introducing frequent checks of general practitioners' initial assessments and repeat assessments; these checks should be carried out by doctors accountable to the social security system; employers should co-finance sickness and disability benefits; in addition, a reduction in the rate of long-term sickness benefit should be considered.

Reducing the replacement rate of disability pensions is likely to have an impact on the number of recipients. However, that would most likely also lead to an increase of poverty among the disability pensioners and possibly their migration towards other welfare schemes. In the medium to long run, two types of additional actions are required. First, it is necessary to put in place pre-emptive policies which would lower the probability that individuals become dependent on dependent disability benefits. Since the disabled have lower skills than the average, education policies (both compulsory education and adult training) would be best placed in this respect. Second, medical and vocational rehabilitation benefits and the temporary disability benefit will be merged into a new benefit as from March 2010. That may facilitate resumption of work.

Improving the efficiency of public spending.

A labour resource-intensive government sector with scope for productivity increases

At around 40% of total GDP, Norway's public spending is internationally around average; it is however higher than average when measured with respect to mainland GDP, a better indicator of current Norwegian living standards (Figure 2.3). Relative to the general OECD pattern, the composition of general government spending in Norway reveals an above-average emphasis on social protection, health and education spending. Government service production (roughly equal to total government spending excluding debt interest, investment and transfers) is equivalent to 27.5% of mainland GDP compared with an average of around 20% for the OECD area.


Public services can be provided: i) directly, via government employees; ii) by contracting out services to private sector intermediaries; or iii) reimbursing private agents for expenditures incurred in obtaining such services. The last category includes, for example, refunds of expenses paid to private doctors or pharmacies, and is largely accounted for by health, housing, transport and education outlays. Efficiency is often improved where private provision is involved, reflecting a different incentive structure. However, Norway is relatively undeveloped in that respect: only two-fifths of public services are provided by the private sector, compared with around 50%, on average, in the OECD area (Table 2.2).

It is difficult to put a cost-saving figure on the efficiency gains which might be made if Norway were to raise the proportion of private provision to the international average. It would appear that compensation costs per employee in the Norwegian government service are relatively low, at 95% of private compensation rates compared with ratios of up to 130% in some OECD countries (Pilichowski and Turkisch 2008). Countries with high levels of government employment tend to have relatively similar or lower compensation costs per employee compared with those in the private sector, as in Norway and Sweden. This may reflect a more qualified and skilled government workforce in countries with smaller government sectors and vice versa, though Norwegian data on relative qualification of the work force do not exist to corroborate this. While government services employ nearly 30% of the Norwegian labor force, whereas in the OECD area the median (and average) proportion is near to 15%, the government wage bill, at 16 1/2 per cent of mainland GDP, is "only" 5% points or so higher than the OECD average (OECD, 2009). Moreover, this situation seems to be entrenched: since 1995, the proportion of the labor force working for the government has been rather stable (as it has been in many OECD countries), with the exception of 2007, government employment continuing to take up slightly under a third of total employment growth. No direct inference can be made about the relative efficiency of the Norwegian government sector from employment and national accounts value-added data (which in the case of the government sector measures mainly inputs); however, evidence on government outputs, where they can be measured, suggest that a low share in value added could be a function of low marginal government-sector manpower productivity, as well as low relative wage costs per se.

An important contributory explanation of the large government size in Norway is the requirement to provide standardised services in the whole country, including in sparsely-populated regions. To attract and retain households in more remote areas, the central government imposes a relatively demanding set of regulations and standards on municipalities for the provision of core public services. No comprehensive estimate of the costs of regional policies is available, but they are likely to be substantial. Public administration accounts for almost half of total employment in the northern part of Norway, compared with 32% overall.

There is however, evidence of organisational inefficiencies across a wide range of government activities which cannot just be ascribed to geography. Among central government spending functions, recognition of the clear potential for productivity improvement - both with respect to international best-practice and to evidence of marked domestic variations in efficiency--has led to significant reform, often aimed at exploiting economies of scale, in the area of police (2005), labour inspection (2006), justice (2008), defence (2008) and tax administration (2009). The relative performance of road toll companies provides an example of how efficiency can still vary within Norway because of unexploited scale economies (Odeck, 2008). Efficiency is greatest among larger road companies, as well as those who introduce new technologies the fastest, suggesting that mergers could allow scale economies to be exploited.

A better assessment of productivity in the public sector would require the collection of performance indicators. The government has initiated work on this in a number of areas: central government administration; universities and colleges; specialist health services; national child welfare; customs and excise tax; police and prosecution; transport and communication and the correctional service (Box 2.2).
Box 2.2. Implementing best practice in the public sector

Reform of the public administration: Increasing accountability

Agencies have been a major organisational form in the Norwegian
central government. Agencies outside the ministries represent the
largest share of the civil service. In 2003, only a small number of
civil servants were employed by ministries (about 3 900). In
comparison, about 120 000 civil servants (including those at the
regional and local levels) were employed by directorates/central
agencies, other ordinary public administration bodies, agencies
with extended authority, and government administrative enterprises.

Over the past 15 years, a process of structural devolution has been
going on in Norwegian central government, and the independent
agency model has become more differentiated (Christensen and
Laegreid, 2009). This development was partly inspired by New Public
Management ideas and solutions, but was also a part of Norway's
adaptation to the EU and the internal market. The model combines
greater agency independence, including control of human resource
management, with increased inter-organisational specialisation,
whereby the distribution of roles and tasks among agencies is more
differentiated and non-overlapping.

The increase in the range of management tools, such as performance
management, auditing and quality-measuring instruments, can be seen
as a response to political demands for greater accountability of
public service providers, as a product of increased concern with
the efficiency and quality of public sector services, and as a sign
of political determination to meet fiscal constraints. Norway is
particularly advanced in the use of performance reporting (annual
reports, budget documents, operational plans, and strategic plans)
and management-by-objective-and-results (MBOR). The MBOR concept
was put into practice through three reform measures from 1990
onwards: activity planning, budget reforms, and pay reforms, all of
which are now integrated parts of the governmental financial
regulatory system (Laegrid et al., 2006). In the process, Norway
has developed a government-wide performance budgeting system.
Performance targets are integrated into the budget process and are
used when developing the executive's budget. Each ministry and each
agency receives one lump sum appropriation for both wages and
operating expenditures and carry-overs are limited to a maximum of

Monitoring performance

Norway is also relatively advanced in respect of customer contact
and transparency. The Ministry of Government Administration and
Reform has initiated a new national Citizens Survey, to provide
systematic and regular feedback on the level of satisfaction with
public services. The aim is to enhance the quality and efficiency
of public services. At the same time, StatRes has been established
as a system for the development and dissemination of statistics,
with the aim of showing the level of resources the state uses and
what inputs provide in terms of activities and services. The final
outcome is shown by means of performance indicators. The
relationship between outputs and inputs illustrates productivity.
Statistics and indicators on local government are similarly
available in KOSTRA. The Ministry of Administration and Reform will
use StatRes data to initiate an assessment of public sector
efficiency across the various sectors where measurement is
considered possible: central government administration;
universities and colleges; specialist health services; national
child welfare; Norwegian customs and Excise; police and
prosecution; transport and communication and the correctional

Regulatory impact analyses

The 2005 and 2008 OECD Surveys of Indicators of Systems of
Regulatory Management identified the most important governance
practices to improve the quality of regulation in OECD countries as
the use of regulatory impact analysis (RIA) to assess the cost and
benefits of new regulation and the institutional oversight of the
quality of regulations, together with the design of programs to
reduce the costs of administrative activities on business and
citizens. The Norwegian government's Instructions for Official
Studies and Reports require that each matter shall include a
consequence assessment. However, ministries do not always follow
these requirements. The OECD has previously pointed out the
potential for improvement in this area (OECD 2003) and the findings
of this OECD report have since been followed up in Norway. A
working group submitted recommendations in 2004 and the measures
taken include the establishment in 2005 of a panel supporting the
ministries' work on impact analyses. The effectiveness of the panel
was evaluated in 2007 and it was subsequently made permanent,
meeting first in August 2009.

Management practices are improving

The Norwegian government has increasingly provided incentives for increased efficiency, in line with international best practice (OECD 2009), (Box 2.2). Human resource management is largely delegated to line departments and line managers, there is open recruitment and a high use of performance-related pay compared with the OECD average (OECD 2009b). The introduction of performance management has attracted attention to results and generally appears to have improved efficiency and performance. Norway has also taken steps to increase public scrutiny of government activities and to lighten the impact of regulation on the private sector. An assessment of the impact of new regulations on the budget, small businesses, regions, the public sector and gender is required for all new regulations, though this assessment is not always required to be quantitative.

It is necessary to rethink central-local relations

Local government covers much of education and health care, employment has grown rapidly

Government employment in Norway is decentralized, with almost two-thirds of general government staff employed at the local government level (counties and municipalities). At the same time, the central government plays a larger role in the determination of tax revenues than in most other OECD countries: almost 90% of total revenues come from taxes which are fixed at the central government level--though all non-oil taxes are in practice collected by municipalities. The system of local government financing seeks to offset differences in income and cost levels across jurisdictions, combined with specific grants for remote areas. At the same time setting taxes at central level has the advantage of taking into account the macroeconomic situation and labour market incentives. However, the separation of spending and revenue responsibilities gives rise to well-known principal-agent problems, which can impact negatively on allocational efficiency and on overall budget control, where local governments thwart central priorities by spending grants on their own programmes and the central government then has to reassert its initial commitments by increasing funding. Between 1986 and 2000, there was a general trend towards earmarking, so as to better ensure the implementation of central priorities without leakages towards other services (Borge 2009 and Box 2.3). The new schemes have reduced some of the political tension between central and government levels, perhaps at the cost, in some cases, of less local innovation and initiative (Borge 2009). However, judging by the strong expansion of employment in municipal activities since 2006, it is easier to ratchet up local spending than to reduce it.

To facilitate the merger of municipalities (which remains however done on a voluntary basis, not an explicit government policy), the government pays 50% of the preparatory work and refunds 40 to 60% of the administrative costs related to the merger process. Merged municipalities retain their former level of grants for ten years, thus reducing the disincentive to merge embodied in a grant system that involves a large fixed component independent of the size of the municipality. There have been, however, very few mergers during the past decade, despite that the fact that many municipalities are too small and many of them are losing population. To reap scale economies, municipalities frequently co-operate in some sectors such as waste disposal, water supply and auditing, and in the energy sector (through the joint ownership of power plants). However, in core spending domains, co-operation is limited, partly reflecting the absence of appropriate compensation schemes between jurisdictions and partly reflecting the importance attached to distance and commuting time. Primary and lower-secondary education provides an example. Each municipality has the obligation to deliver education to every resident child in the school nearest to his home but has no incentive to accept nonresident pupils since it is not entitled to a corresponding compensation from the central government grant system. Municipalities can enter into bilateral agreements but these have been rare. Absence of effective financial incentives hampers an efficient use of existing capacities.
Box 2.3. Reasserting control over municipal spending

Political economy factors make control of local government
budgeting difficult

Much of the political economy literature points to the importance
of governance institutions to deficit and expenditure outcomes; in
particular, systems that produce fragmented or minority governments
are more prone to deficit and/or expenditure bias. For instance,
cross-section evidence relates Norwegian municipal deficits to
party fragmentation at the local level (Borge, 2004). * Time series
and cross-section evidence suggests that political strength,
measured as stable government and low party fragmentation of
parliament, slows public sector growth, insofar as it--for
example--holds down teacher employment, while teachers' wages and
employment depends on which party is in power (Falch and Rattso,
1998). Educational efficiency at the lower secondary school sector
has been found to be conditioned by party fragmentation and local
government political affiliation (Borge and Naper, 2005). Cost
efficiency in the provision of longterm care is also found to be
negatively related to political fragmentation (Kalseth, 2003).

The grant system has resulted in central-local tensions...

Norway has been through phases of decentralization and partial
recentralization in order to improve the allocation and control of
local spending. The introduction of the block grant system in 1986
was meant to establish a more fair distribution of resources across
local governments, and to strengthen local democracy and improve
efficiency. Until 1994, the central government used sectoral block
grants to signal their strategic priorities (like education, health
care, etc.), while leaving local governments free to spend the
grant as they liked. Signals about how the money should be spent
could be ignored by local governments and their ability to thwart
the priorities set by central government led to a gradual increase
earmarked block grants (Borge 2009). However, where the local
government has two revenue sources, local taxes and block grants,
earmarked grants can still to some extent leak out to other
services, as local governments substitute grant money for their own
funds and spend those funds on other programmers.

... creating the need for more efficient ways of translating
central priorities into local policy implementation

Efficient earmarking should affect relative prices, i.e. they
should be of the matching type. Open-ended matching grants are more
stimulative of centrally prioritized services than a block grant,
because local authorities have a price incentive to comply with
central wishes. However, matching grants still do not ensure a
consonance of central and local priorities. From the late 1990s the
matching grant for child care was increased to improve coverage and
lower user charges, but the actual impact on both these was modest,
while the extra spending on child care was much lower than the
grant increase.

In the search for more efficient ways of translating central
priorities into policy implementation, in recent years the central
government has used so-called action plans to increase provision of
particular services. In this case matching grants are of the
temporary kind and affect only additional spending. Action plans
are used in care for the elderly, schools, and child care among
others. Even these schemes have their problems, however, as they
discriminate against those municipalities who have already expanded
the relevant service and may encourage municipalities to put off
spending in anticipation of such schemes (Borge, 2009).

* Statistical analysis of a large panel data set for Norwegian
local governments shows that party fragmentation has a very robust
effect on budget deficits. Survey data on politicians' spending
preferences indicates that party fragmentation favours underlying
"common-pool" problems of spending bias. Borge and Tovmo (2007)
also find that a high degree of party fragmentations is associated
with less forward looking behaviour on the part of local
governments. Using a measure of aggregate output based on
indicators of production for several service sectors developed by
the Norwegian Advisory Commission on Local Government Finances (Det
tekniske beregningsutvalg for communal og fylkeskommunal okonomi,
TBU), Borge et al. (2007) find that a Herfindahl-index measuring
political strength by party fragmentation has the most consistent
impact, and indicates that a strong political leadership
contributes to higher efficiency. High democratic participation
contributes to high efficiency, consistent with international
findings in the literature.

Reforms to raise human capital must be continued

Value for money can be increased in primary and secondary education

The last Survey focused on the performance of Norwegian primary and secondary education and analyzed the possible causes of relatively low value for money in this sector (OECD, 2008b; Boarini, 2009). The Norwegian public sector is almost the sole provider of education services and the government spends 61/4 per cent of GDP on primary and secondary education, compared with around 5% in the OECD at large (a figure which rises to just under 6% when private education is included). Expenditure per student at the primary and secondary levels is about 40% higher than in the OECD area. However, international assessment surveys have found that Norwegian pupils perform rather modestly in various core subjects: in particular, education outcomes as measured by the Programme for International Student Assessment (PISA) scores are around the OECD average. Poor performance also translates into a higher proportion of low achievers. Norway also has a problem with drop-outs beyond the age of 16: the rate is twice that in other Nordic countries and at nearly 21%, exceeds the OECD average of near to 18% (OECD 2008a).

The combination of relatively high spending and only moderate achievement could indicate either an excess of resources devoted to education or the possibility of improving the efficiency with which they are used. Sutherland et al. (2007) estimate that potential savings in compulsory education amount to over 1% of GDP, reflecting the fact that Norway transforms input in outputs in the least efficient way in the OECD area and devotes a sizeable share of GDP devoted to education.

Clearly, these are illustrative estimates. PISA provides only a snapshot of educational achievement at a certain point in time and from a selected academic viewpoint. Other international surveys show more mixed results on achievements of Norwegian students, though they generally point to a relatively poor performance (OECD, 2008b). Thus, there is a genuine concern that ample resources may not be translating effectively into educational performance and this has motivated a series of policy actions, including a new curriculum that clearly defined competence goals (the Knowledge Promotion reform launched in 2006), steps towards increasing performance assessment and local accountability (White Paper on Quality in Education, June 2008) and measures to improve the average competencies of Norwegian teachers (White Paper on Teacher Education, February 2009).

The emphasis is put on improving outcomes, rather than reducing spending, reflecting the traditional priority in Norway to maintain a comprehensive and inclusive education system. The view that more resources produce higher outcomes is still dominant, although this is now strongly questioned by empirical research (see Hanushek, 2007, for a review). Thus, Norway retains the lowest pupil-teacher ratios among OECD economies. In addition, on current policies spending could increase further, reflecting the goal to have additional teaching resources for following up unsatisfactory learning outcomes in reading and mathematics but also for longer teaching hours.

Norway has preferred to sustain employment in the education sector rather than carefully attract and select its teachers (see previous Economic Survey OECD, 2008b). This is a costly choice in terms of efficiency that needs to be reconsidered (see Boarini and Ludemann, 2009). The previous Survey argued that it would be better to introduce merit-based policies for managing the existing workforce, rather than adjusting the wage bill across the board. Merit-based policies have both advantages and pitfalls, but they are increasingly seen as effectively improving the performance of education system. In addition, it should be possible to further improve the quality of existing teachers by reallocate resources towards formal-accreditation training and away from informal training as well as replacing generic programs directed to mediocre schools with programmes conditional on outcomes and on effective school restructuring. The government has taken some steps to raise standards among new teachers, by increasing entry requirements for teacher training.

Small school size emerges as one of the principal factors explaining relative educational inefficiency within and among OECD economies (Sutherland and Price, 2007). In the Norwegian case, Bonesronning et al. (2008) conclude that increasing the average school size from the current level of 200 to 400 would reduce costs by 5-6%. There are a number of impediments to greater educational concentration, most of which are geographical. The school structure is the responsibility of local governments in Norway, which have the right to decide what structure is best in their municipality or region, and the government has stated that it wishes to uphold a decentralized school structure, a commitment which gives priority to avoiding the need for children to make long journeys to school. Hence, the recommendation made in the previous Survey, that the number of schools should be reduced to free resources for quality enhancement has not been pursued. Nevertheless, there are potential efficiency gains from enhanced co-operation. While the central government cannot force class or school merging on local governments, funding considerations should and increasingly do impinge on school density. Statistics indicate that the number of small schools is declining in Norway. This may demonstrate that local governments themselves find ways to minimize costs by changing their school structure, but it is a trend that should be further encouraged via suitable funding incentives. Since bilateral agreements between municipalities on compensatory transfers to municipalities which accept children from outside their catchment area have been rare up to now, it might be necessary that these transfers are financed at central level. In order for this not to increase the overall amount of spending, block grants to municipalities should be adjusted to take into account these compensatory transfers. Compensatory transfers should be carefully defined as to reflect the actual cost incurred by the municipalities to educate children coming from another catchment area.

Tertiary education performance could be improved

Norway's tertiary education system is relatively large in terms of resources absorbed. There is a heavy reliance (96%) on public funding, the budget resources devoted to it amounting to 2.1% of GDP compared with an OECD average of 1.3%. (2) The difference is largely due to free tuition and the support to students given through grants and loans, which is double the proportion in OECD countries as a whole (Figure 2.4). This has to be seen in connection with a relatively high degree of wage compression, which reduces the incentives that individuals have to invest in education. However, one third of the population receiving the public subsidy can afford to pay, so that the outcome in terms of equity is questionable. By contrast, early childhood education is not free. This is the reverse of what would be implied by the public externalities usually associated with the two different types of education. A suitably constructed tertiary fee system would have the simultaneous benefits of increasing equity, reducing excess demand for higher education, freeing up resources for uses with higher social returns and providing incentives for higher education institutions to develop programmes better suited to the economic needs of the country. The right to free tuition is, however, currently enshrined in legislation.


From the point of view of student throughput, the system has attracted some criticism as one reason for poor value for money. The tertiary student-teacher ratio is relatively low (11.9 compared to an OECD average of 14.9), and expenditure per student on educational institutions is almost a third higher than the OECD average. Small size implies a need to recruit and retain staff to teach specialised subjects which would, in a larger multi-faculty university, be provided by staff from other faculties. As with schools, some of this apparent inefficiency relates to Norway's dispersed geographical structure. It is not clear whether graduates from smaller, geographically dispersed institutions, achieve lower results, though there is some evidence that this is the case in some institutes, like teaching colleges. The case for closing small, cost-ineffective universities is even stronger than for primary and secondary institutions, since tertiary education is not a basic right and students' mobility across the country should be encouraged with the objective of maximizing labor market performance in the long-term.

As for the output, Norway does relatively well in terms of graduation ratios but it does less well in terms of completion rates and average duration of studies. Steps have been taken to improve incentives for both students and institutions to reduce non-completion rates and the total length of study time. The share of grants in total student assistance (to a maximum of 40% grants, 60% loans) depends on the student's progression and is awarded once the studies are completed. However, the opportunity costs of remaining in higher education are still quite low, especially when the labor market is weak.

Measuring the efficiency of higher education spending also involves weighting teaching and research outputs, which is rendered difficult by the absence of agreed international benchmarks for teaching outputs, as well as the problems of matching funds to outputs. Traditionally, a large part of the funding allocation has been made according to student throughput, universities receiving a block grant from the central government based, up to 2002, upon historical costs, admission capacity and standard unit costs per discipline. This funding system discriminated against science-based studies, which may explain part of the declining enrolment in these studies. Following the report of the Mjos Commission in 2000, Norway introduced a new performance-based funding model for higher education in 2002, aimed both at improving education outcomes as measured by the credits and student exchanges with foreign institutions and by research as measured by research publications. (3) The impact of the new system has been recently evaluated with satisfactory results. (4) Publication of evaluation reports and the institutional responses to them should be beneficial to system improvement, but has been criticized for a lack of reliable data addressing the issues concerning the learning outcomes--for the system to be fully effective, institutions should be required to publish (and not simply report to the Ministry) a range of performance indicators on a common basis, including course completion rates and the average time for completion. All in all, however, the increased accountability of the university system implied by the performance-based funding model seems to have increased the quality of higher education (Econ Poyry Report). One additional, unintended, consequence of this funding model may be an increasing concentration of funds on larger institutions, which would seem likely to enhance efficiency. (5)

The key challenge in the future is to improve the system's responsiveness to the needs of society and the economy (OECD, 2006a and b); this clearly implies increasing the number of graduates in scientific subjects, where Norway lags behind other OECD countries. Improving the management and leadership of institutions, including through increasing their legal autonomy and making their management structures clearer, should be central to this. There is also a need, in the interests of both improving the efficiency of the system and increasing its responsiveness to students and society, to improve the performance indicators and outcome measures by which it can be judged. This should be accompanied by a more transparent basis for the allocation of public funds to institutions. Most of these directions for the future are consistent with existing government thinking. As for free tuition in tertiary education, any reform would go against a long, and strong, tradition in Norway. However, the distribution of costs and benefits needs to be the subject of more rigorous public analysis and public debate. The system for supporting students in their studies through loans and grants could be improved through the introduction of an income-contingent system for the repayment of loans.

Health and long-term care spending is under pressure

The health sector should seek to achieve efficiency gains

In terms of the share of overall GDP spent on health and long-term care, Norway is near to the average of EU countries, and the pace at which that share has been rising--and will rise due to population ageing over the next half century--is not out of line with trends elsewhere. But, as already discussed in the Survey of 2005, Norway stands out in terms of health spending per capita--it is among the highest spenders of all OECD nations. While this may be partly explained by Norway's high living standards, this elevated level of health spending has not achieved a correspondingly high level of health outcomes. Life expectancy at birth is higher than the OECD average, but this is over-explained' by favorable socio-economic factors, leaving a considerable gap which may be ascribed to relative inefficiency in translating inputs into health outputs. (6) Norway also stands out for the large amounts of resources spent on care for the elderly and has the highest share of people working in the health and social sectors (Health at a Glance, 2009). Finally, a number of empirical papers have indicated that health inequalities seem to be at least as high as in other European economies, even though Norway's income structure is more egalitarian (which should lead to more evenly distributed health outcomes).

Data Envelopment Analysis (DEA), which compares the relationship of health outcomes to physical and costs inputs relative to the production frontier' of the most efficient economies, suggests that Norway has considerable potential to improve health performance without needing to increase inputs (Figure 2.5). Similar analysis using cross-sectional data on Finnish and Norwegian public hospitals with respect to output data on admissions grouped according to diagnosis related groups (DRGs), outpatient visits, day care and inpatient days revealed marked differences in cost efficiency, both within and between the two countries. A recent analysis of hospital performance in selected OECD countries, based on the system of DRGs suggests that savings in some areas could be around one-third on average. This is less than in some countries covered by the study but much higher than, for instance, in Finland and Denmark (Erlandsen, 2008; Table 2.3). An important factor on the cost side is the significantly above-average number of doctors and nurses relative to Norway's population, which makes for low apparent productivity. (7) The sparseness of the rural population is partly responsible for the generosity of inputs, but there is evidence of inferior performance for the resource used.


Reforms have increased technical efficiency

Policies have responded to health-sector inefficiency problems by instituting a succession of reforms (Box 2.4). The block-grant financing system for general hospitals was replaced by an activity-based financing system in 1997 (with reimbursement linked initially to 30% of costs, a share that has varied over time) to a Diagnosis Related Group (DRG) reference system relating to the number of patients treated. As a result of the reform, the majority of county councils introduced activity-based contracts with their hospitals. This was followed, in 2002, by the transfer of responsibility for the provision of hospital services from the counties to quasi-autonomous regional health enterprises, with independent boards and CEOs, answerable to the central government. This recentralisation process came after a period of increasing central government intervention at the county level regarding specialist care (such as the introduction of a waiting time guarantee and the introduction of free choice of hospital, by which the central authorities become partly responsible for the outcomes that were produced in the health care sector, but through which blame for not meeting targets became a matter of central-local dispute). Centralisation was meant to generate greater hospital efficiency with more equal medical outcomes across municipalities, via economies of scope (more specialised hospital production) and a more even distribution of specialist medical expertise. A further ingredient in the reform process was the enhancement of competition, through public choice (free choice of hospitals, including private hospitals) and increased use of the private sector inputs such as laboratory and radiology services (Hagen and Iversen, 2007).
Box 2.4. Reforms bearing on health-service efficiency

Since the late 1990s, Norway has implemented an impressive amount
of reforms aimed at greater efficiency of delivery of medical
services, in part by allowing a greater role for market focus,
while maintaining and where possible strengthening, quality and
equity. They include the following:

Introduction of a DRG system of funding in July 1997, substituted
part of the block-grant financing system for general hospitals with
a case-mix financing system. Reimbursement to general hospitals for
inpatient care is based on a block grant and an activity-based
component, the latter being calculated on the basis of the number
of patients treated and of the Diagnosis Related Group (DRG)
reference system. The reimbursement is based on a national
standardised cost per treatment, related to average costs. The
activity-based component was originally 30%, gradually increased to
60%; it was then reduced in 2004 and then again in 2006 to 40%,
since when it has remained unchanged. No out-of-pocket payments are
imposed for inpatient treatment. For hospital outpatient care,
activity-based financing will be included in the case-mix financing
system as from 2010. Psychiatric inpatient care and drug abuse
rehabilitation continue to be financed through block grants.
Ambulatory outpatient treatment is reimbursed by the National
Insurance Scheme.

Patient choice. In 1999, the Act on Patient Rights introduced free
choice of public hospitals for patients. This right has been
progressively extended to include the choice of private hospitals
contracted with a Regional Health Enterprise (see below) through a
GP referral. Patients also have the right to receive specialist
medical assessments within thirty days after GP referral, and if
the time limit is exceeded, then the NIS is given the
responsibility to provide treatment either through the private
sector or abroad with costs financed by the regional health

Hospital recentralization. The 2002 Health Enterprise Act
transferred hospital ownership from the counties to the central
government, establishing five (reduced to four in 2007)
geographically-based "Regional Health Enterprises" (RHE) each
reporting to the Ministry of Health and responsible for delivering
health services in their regions. The RHEs own the local "health
trusts" and are responsible for monitoring their costs and quality
of services. The central government still defines their main health
policy objectives as well as their financial means. Hospitals
continued to be publicly financed by general taxes, and, although
the hospital organizations gained a semi-independent status by
being organized as public enterprises with appointed boards, the
Minister of Health was still the ultimate guarantee for solvency.
At the same time, there was an element of decentralization, since
both the health regions and hospitals were organised as health
enterprises. These bodies are independent legal subjects with their
own responsibilities for personnel and capital. In this respect,
the Norwegian hospital reform is similar to reforms in England and
Wales, Sweden, Spain and Portugal that have focused on transforming
hospitals into more autonomous actors.

Prescriptions for a healthier Norway. In January the White Paper
Prescriptions for a healthier Norway--a broad policy for public
health was presented, putting forward four general principles for
public health: 1) Make it easier for people to take responsibility
for their own health; 2) Build alliances to promote public health;
3) Encourage more prevention and less cure in the health service;
4) Build up new knowledge.

The Coordination reform, announced in 2009, aims to encourage the
municipalities to assess whether positive impacts on health can be
achieved by using resources differently, for example through more
appropriate use of the hospitals. The most important financial
instruments are intended to be municipal co-financing of the
specialist health care services and municipal financial
responsibility for patients ready for discharge. Municipal
co-financing of the specialist health care services will entail
changing the financial parameters for the specialist health care
services. Activity-based financing should continue to be an
important element of financing, but the rate will be reduced from
40 to 30%, given the heavier emphasis on prevention and early
intervention efforts.

There is some evidence that recentralisation has been associated with improvements in technical efficiency. The number of in-patient stays has continuously increased, while bed-days have declined and bed-days per stay have fallen from 6 in 2000 to 4.8 in 2008; waiting times have been reduced. However, the observed increases in technical efficiency may have resulted mostly from the increasing role of activity-based funding (ABF) methods used in the allocation of health care resources (Magnussen et al., 2007). Both technical efficiency and patient satisfaction increased as a direct effect of the reimbursement reform, the latter being an effect of lower waiting time, which in its turn is an effect of the introduction of activity-based financing (Hagen et al., 2006). ABF itself has, nevertheless, been much less effective in enhancing cost-efficiency than technical efficiency (Biorn et al., 2006). Expenditure on the specialist health service has risen more than expected from budget plans, for both activity-based funding and the reimbursement of outpatient services and the deviations have tended to increase (DHSA, 2007). This has led to an increase in the proportion of total costs being met by supplementary funding (Magnussen et al., 2007). Hospitals have run large deficits and the state has always underwritten them via a soft ex ante budget constraint and financing assumption of hospital deficits when they emerged (Trond and Hagen, 2009). Deficits in 2009 are expected to be limited as compared with the past years.

Cost containment remains a goal to pursue

The focus of Norwegian health reform on obtaining better results from the existing ample resources applied to the health sector does not seem to have helped contain overall costs. The advantage of DRGs is that they can improve efficiency, by encouraging transparency and reducing patient stay. But there is evidence that setting the rate of reimbursement too high (relative to actual marginal costs) can lead to excessive increases in activity (DHSA 2007), hence the progressive reduction of the original reimbursement rates. ABF also provides incentives to maximise hospital income, via faster patient turnover--and even "upcoding" in order to receive higher reimbursement--which do not help the systemic containment of costs. To reduce the incentives for "upcoding" and improve budgetary control, an estimated "upcoding" is taken into account in the ABF-reimbursement to the RHEs at the national level. Hospitals have nonetheless had an incentive to discharge patients "quicker and sicker", which makes for multiple admissions, as well as leading to a focus of resources on acuity of illness and at the expense of disease prevention and health promotion.

In a system which is partly or fully activity based, the state bears part of the cost when activity over-expands, and the hospital ownership reform probably accentuated this budget bias, particularly insofar as managers came to rank boosting income as their most important goal (DHSA 2007). Moreover, the regional health enterprises become separate legal subjects and their accounts moved off the government budget (implying a softening of the ex ante budget constraint), although ownership is public and public debt guarantees obviously apply (implying an implicit on-budget liability when deficits actually emerged). Moreover, the hospital enterprises have lacked the political legitimacy (implicitly attributed to them in the 2002 reform) to make structural decisions about cost-saving rationalisation of the hospital structure (Trond and Hagen, 2009). For instance, decisions to shut down local hospitals, or delivery rooms and emergency rooms in hospitals, have been turned down by the Minister of Health. At the same time, economies of scope have proved difficult to engineer: structural decisions made by the boards have had low political legitimacy and decisions to shut down local hospitals, delivery rooms or emergency rooms in local hospitals have been difficult to implement because of political pressures.

User choice and co-payments should be used more

At the same time, between 1999 and 2005, a number of radical legislative moves expanded patient choice in Norway; in addition, complementary measures to expand capacity such as a11owing patient choice also to include contracted private hospitals and foreign hospitals have been developed in 2002-04. User choice can reduce costs only if the signalling power and capacity at the primary care level are good enough, elements which have not characterized the Norwegian reform (Ringard et el., 2006) and would thus be important to introduce in the future.

User choice may also need to be accompanied by increased price signalling via co-payments. Co-payments in Norway are not particularly low, but a wide range of health services are offered with no or a low co-payment: physiotherapy or elderly care, for example. Co-payments should be introduced or increased where there is evidence of excessive consumption relative to assessed care need and when the demand is price sensitive. The use of cost-sharing is still highly controversial as its use across the board can create problems of access for low-income people and high administration costs. However, it can work in cases such as to re-direct demand towards less expensive drugs, encourage healthy behaviour and reduce excessive expenditure driven by over-supply.

Care of the elderly is a particular case in point for possible use of co-payments. DEA analysis reveals substantial variation in efficiency across municipalities, with the potential to improve efficiency by around 10% (Borge and Haraldsvik, 2009). Importantly, in addition to the political economy factors such as local council fragmentation, low efficiency among municipalities is associated with a low degree of user charging. Furthermore, co-payments for nursing homes are based on income rather than wealth and it is paradoxical that the public sector should finance costly services for its wealthier citizens.

Co-ordination between municipalities must be reinforced

The municipalities, which provide a substantial proportion of public services, have been important players in the administrative reform process (see Box 2.4), and this has also affected the municipal health sector. The majority of the larger councils organise their care services on the lines of purchaser-provider principles and many local authorities have reorganised themselves and become two-tier systems in which individual undertakings (such as day nurseries or nursing homes) are turned into financially devolved units, are given a great deal of administrative freedom and subject to control by an administrative director. Hence, they are run more in line with private enterprises, becoming service delivery suppliers similar to those in the private sector. This could allows public services to be subject to market forces, if money effectively "followed the user"; in practice, however, the majority of care services is financed through block grants transferred to municipalities, without any direct user-related financing. In addition, municipal health budgets have remained under strain and spending has continued to rise, driven in part by a strong and continuous rise in the number of practicing physicians reaching a coverage of 3.9 per 1 000 inhabitants (as compared to 3.1 for the OECD average)--an increase which has largely been supply driven, but also related to growing areas of medical need, such as old-age and disability care, mental illness and rehabilitation.

In circumstances where municipal health commitments have already been growing rapidly, municipalities have had incentives to postpone the reintegration of hospital patients into the health care services they finance (in particular outpatient and elderly care). They have also resisted hospital rationalisation. The problem of many small hospitals derives not just from geographical remoteness, but applies also to populated areas where the social reasons for preventing rationalisation are not as strong as for schools; indeed, the merging of welfare services would often potentially increase the quality of treatment. Co-operation between municipalities has also been lacking. A person can in principle benefit from the services offered by another municipality and the providing municipality can receive some financial compensation from the municipality of origin if they have reached a bilateral agreement. But since, in most cases, this compensation does not cover the actual costs of providing the service, co-operation agreements across municipalities remain rare and scale economies largely unexploited.

Clearly, while health outcomes are improving and satisfaction with the health system is high, there is some considerable way to go before trends in health spending can be deemed sustainable. Four problems stand out as needing action:

* Economies of scale could be realised by centralising complicated surgery and closing down the smaller units, while at the same time reducing costs. Political economy problems need to be addressed. Dissemination of more information on hospital performance could inform public opinion, helping to reduce resistance to necessary changes.

* The percentage reimbursed under activity-based financing is to be reduced further, but there remains some ambiguity and grounds for misunderstanding as to the objectives and ambitions of the scheme.

* Patient choice remains ineffective without better signaling at general practitioner level and better information dissemination. It may also require an enhancement of co-payments.

* Financial incentives need to be established for municipalities to accommodate patients who are ready to be transferred from hospital to less costly follow-up treatment in their local communities. A recent White Paper proposes to change the incentives of municipalities to pass costs on to hospitals by charging them for inpatient care, while allocating greater resources to prevention and primary care, on the basis that there are higher marginal returns to additional resources in primary care than in hospitals.

There is some scope to improve tax neutrality

Though increasing taxes is not a solution that Norway should envisage for the moment, there is certainly room to redesign parts of the current system, so as to minimise the distortions and the impact on short and medium term growth. As noted in the Survey of 2000, Norway has a relatively high tax burden, though not out of line with the revealed preference of OECD economies for growth per capita to be more reflected in public than private spending (Figure 2.6). The tax mix shows a relatively even distribution of tax revenues between income taxes, corporation tax, taxes on goods and services and payroll taxes, but a much below average dependence on property taxes (Figure 2.7). This section examines the efficiency of the tax system from the point of view of the coverage of the various tax bases (or conversely the extent to which tax allowances and exemptions add to and substitute for direct spending) and of the neutrality of the tax system with respect to capital and labour income, which is a major determinant of the growth-friendliness of the tax system.



The incidence of tax expenditure should be reduced

Tax expenditures are often regarded as distortionary and costly elements of the tax system. Though some deviations from neutrality can make the tax system more efficient, the presumption is that the use of tax expenditures makes the budget susceptible to political pressure (rent seeking) and is distortionary by pushing up tax rates. In Norway tax expenditures are reported in the National Budget (all existing tax expenditures are included in the report, but the cost is not calculated for all of them), but there is no systematic evaluation of the effectiveness of tax expenditures and there is a lack of yearly evaluation and assessment as to whether the public support given through tax expenditures is increasing or decreasing, and how well the targets of the tax expenditures are achieved. Tax expenditures usually have no limits in yearly budgeting, as direct expenditures often do, and they tend to be or become permanent. Under the structural non-oil central government budget deficit framework, the government has a choice between tax expenditures and direct transfers, but it often seems like tax expenditures are regarded as easier to implement than direct expenditures and also are felt to he less costly. While a full integration of tax expenditures into the budget process is hardly feasible because of lack of data, computational methods, benchmark choice and other methodological issues, transparency would seem to require a more formal treatment than currently applied.

Comparing tax expenditures between countries and over time is not straightforward. Nevertheless, Norway has fewer tax expenditures (60) (8) than its Nordic neighbours--there being 115 in Sweden; France has around 400. However, the general trend is that the reported tax expenditures are growing. (9) A substantial part of this growth can be attributed to rising tax bases, but some tax policy changes have also contributed to this development. This particularly applies to tax expenditures in capital income tax. The taxation of imputed income of owner-occupied housing was removed in 2005. This policy change led to a considerable increase of the tax expenditure related to income taxation on own housing and vacation property. The tax expenditure related to the wealth tax on housing and vacation property also has grown significantly. Much of this is due to exceptional growth in market values on properties, without an equal growth in assessment values. In sum, the most important tax expenditures relate to lack of taxation of imputed income from, and low assessment values of, housing.

There has also been considerable growth in the tax expenditure linked to the tax relief on employee premiums and contributions to occupational pension schemes. Increasing employment is the cause of a substantial part of this growth, but the introduction of compulsory occupational pensions in 2006 has also been of importance. As well, the tax rules for shipping companies were changed in 2008, with effect from 2007. Shipping companies are now exempted from ordinary capital income tax, whereas they earlier had a tax deferral arrangement. The tax expenditure in personal income has been more stable, although some tax expenditures have grown and some have been reduced. Also, some tax expenditures have been abolished and new ones have been introduced.

The Norwegian VAT reform from July 2001 broadened the VAT base substantially by including services in the VAT base on a general basis. Several sectors have been included in the VAT base since 2002, e.g. accommodation, travel agencies, broadcasting and cinemas. The main tax expenditures in the VAT arise because some sectors are outside the VAT system or inside the system with a lower or zero-rate. The sectors implicated here are the reduced rate on food and zero rating of newspapers, books and periodicals. This preferential treatment leaves the revenue-raising effectiveness of the Norwegian VAT system at near to the OECD average. Tax expenditures from excise duties are mainly linked to environmental and health related items. The main sources are the excise tax on energy consumption, taxes on alcohol and tobacco, and a lower tax rate on diesel fuel than petrol. The variable-rate C[O.sub.2] tax is applied mainly to energy related activities, while none of the potential public revenue from sale of permits in the emissions trading scheme is yet being exploited (see Chapter 3).

Achieving capital income tax neutrality

The general corporate and capital income tax law in Norway is fairly neutral, though the progressive tax on incomes has caused problems with respect to the large gap in marginal rates on labour and capital income. The 1992 tax reform established a dual income tax system in Norway, motivated by the need to keep the capital income tax low in a small open economy faced with international capital mobility, the possibility of aligning the personal capital income tax rate with the corporate tax rate to reduce investment distortions and limit the scope for arbitrage and the political economy observation that it is easier to preserve a broad and fairly neutral capital income tax base when the capital income tax rate is not too high (Sorensen, 2005). However, tax treatment of small enterprises became a problem. Between 1992 and 2002, the wedge between the top effective marginal rate of tax on labour income and the tax rate on capital income rose from 28% to nearly 37%, increasing the incentive to transform labour income into capital income for tax purposes. By 2000, 80% of active shareholders subject to income splitting had negative calculated personal income. The tax reform of 2006 aimed to correct this, abolishing the existing rules for "active owners" and replacing them with a tax on above-risk-free return on the share income of personal shareholders and partnerships. At the same time, the top marginal tax rate on labour income was reduced to 54.3% and the tax on dividends rose to 48.2%. This remains high by OECD standards, however. Reducing further marginal tax rates at high income levels could diminish the incentive to misreport labour income as more favourably treated capital income and at the same time increase the private returns to tertiary education (Boarini and Strauss, 2007), which are relatively low.


The main deviation from neutrality which emerges from the system described above is the virtual tax exemption given to income from housing property, through personal income tax deductibility of mortgage interest payments without any associated tax on imputed income from ownership, and the bias in favour of house ownership implicit in the wealth tax (Figure 2.8; see also Annex 1.A1). The taxable value of wealth varies strongly between its different components. Wealth held in housing has, up to now, benefited both from a much reduced rate applied to it, and from the absence of any systematic revaluation of houses themselves, though the 2010 budget does provide for a programme of revaluations. As well as distorting investment and savings incentives, this tax treatment of housing can also contribute to volatility in house prices and increase the likelihood of bubble behaviour, as discussed in Annex 1.A1.
Box 2.5. Summary of longer-term fiscal policy recommendations

* To ensure fiscal sustainability in the long run, the pension
reform should be completed soon. This includes aligning the
disability pensions schemes with the principles of the old-age
social security pension reform. The disability and sickness leave
schemes need to be improved more in general, by strengthening
eligibility criteria and follow-up of recipients. In addition,
employers should co-finance sickness and disability benefits and a
reduction in the rate of long-term sickness benefit should be

* To increase the efficiency of public spending, the use of
existing tools, such as regulatory impact analysis and cost-benefit
analysis should be used systematically and given more weight in
policymaking. Wide dissemination of information on performance of
public services can also be useful in securing support for
expenditure rationalisation.

* Manpower increases in the public sector, especially at local
level, need to be better planned and co-ordinated. Further
encouragement needs to be given to merging and rationalising
municipal jurisdictions by strengthening financial and political
incentives to rationalisation. Strategic correspondence between
overall central government priorities and local preferences should
be encouraged through appropriate financial incentives.

* To reduce expenditure growth in primary and secondary education,
the trend towards reducing the number of schools should be
encouraged by financing incentives that reward efficiency so that
cost-ineffective schools are merged or closed. To increase the
effectiveness of public spending in primary and secondary
education, the current policy effort towards increasing the average
competencies of Norwegian teachers must continue. Similarly,
teachers and schools have to be made accountable for their results,
using appropriate performance incentives.

* In tertiary education, a more transparent basis for the
allocation of public funds to institutions is needed. The
distribution of costs and benefits attached to free higher
education needs to be the subject of more rigorous public analysis
and public debate with a view to reform. Students should pay at
least part of the cost of their studies, and the incentives in the
student loan system could be improved by reducing the grant element
while introducing an income-contingent repayment scheme.

Box 2.5. Summary of longer-term fiscal policy recommendations

* Budget constraints on health care institutions must be made more
stringent, and give incentives for the hospital structure to be
rationalised. Major cost savings could be realised by centralising
complicated surgery and closing down of the smaller units.
Dissemination of more information on hospital performance could
help convince public opinion of the reasons for some unpopular

* Activity-based financing needs to be adjusted so as to eliminate
distortions and prevent over-supply of health services. The share
of resources provided through such finance is to be reduced
further, but the objectives and ambitions of the approach need to
be set out clearly, and better communicated.

* Patient choice needs to be made more effective. This entails
better signalling at GP level and better information dissemination.
It may also involve an enhancement of copayments. Greater use of
co-payments should be considered, in particular in the elderly care
sector, to enhance efficiency and equity.

* Co-ordination between municipalities and hospitals needs to be
improved. Financial incentives need to be established for
municipalities to accommodate patients who are ready to be
transferred from hospital to less cosily follow-up treatment in
their local communities.

* Tax allowances and reliefs should be reviewed more systematically
and integrated more closely into the budget process.

* Housing taxation should be reformed with the objective of
re-establishing neutrality with respect to other types of
investment. Reform of housing taxation should include measures to
phase out the asymmetries resulting from the deductibility of
interest payments on owner-occupied dwellings without taxing
imputed rent, and from the remaining substantial discount applied
to housing for the wealth tax. Higher fiscal revenues from reformed
housing taxation could be used to finance a reduction of marginal
tax rates at high income levels, thereby reducing the incentive to
misreport labour income as more favorably treated capital income
and also increasing the private returns to tertiary education.


Biorn, E., T. Hagen, T. Iversen, J., Magnussen (2006), "Heterogeneity in Hospitals' Responses to a Financial Reform: A Random Coefficient Analysis of the Impact of Activity-Based Financing on Efficiency", Norwegian University of Science and Technology.

Boarini, R. (2009), "Making the most of Norwegian schools", Economics Department Working Paper, No. 661.

Boarini, R. and E. Ludemann (2009), "The Role of Teacher Compensation and Selected Accountability Policies on Learning Outcomes: an Empirical Analysis for OECD countries", OECD Economic Studies, Vol. 2009.

Boarini, R. and H. Strauss (2007), "The private internal returns to Tertiary Education: New Estimates for 21 OECD countries"; OECD Economics Department Working Paper No. 591.

Borge, L.-E. (2004), "Strong politicians, small deficits: evidence from Norwegian local governments", Department of Economics.

Borge, L.-E. and L.R. Naper (2005), "Efficiency potential and efficiency variation in Norwegian lower secondary schools", Working Paper Series No. 15/2005, Department of Economics, Norwegian University of Science and Technology.

Borge, L.-E. and P. Tovmo (2007), "Myopic or constrained by balanced-budget-rules? The inter-temporal spending behavior of Norwegian local governments", Working Paper Series No. 5/2007, Norwegian University of Science and Technology,

Borge, L.E., T. Falch and P. Tovmo (2007), "Public sector efficiency: the roles of political and budgetary institutions, fiscal capacity and democratic participation", Working Paper Series, No. 1/2007 Department of Economics & Centre for Economic Research, Norwegian University of Science and Technology.

Borge, L.-E. and M. Haraldsvik (2009), "Efficiency potential and determinants of efficiency: An analysis of the care for the elderly sector in Norway", International Tax and Public Finance, Vol. 16, No. 4, August.

Borge, L.-E. (2009), "Block grants and earmarked grants: the Norwegian experience", Department of Economics, Norwegian University of Science and Technology, Paper prepared for the Copenhagen Workshop on Intergovernmental Grants, 17-18 September 2009.

Christensen, T. and P. Laegreid (2009), "Coordination and hybrid governance--theoretical and empirical challenges", Working Paper 7-2009, Stein Rokkan Centre for Social Studies.

Directorate for Health and Social Affairs (2007), Activity-Based Funding Of Health Services In Norway: An Assessment and Suggested Measures.

Erlandsen, E. (2008), "Improving the Efficiency of Health Care Spending: What Can be Learnt from Partial and Selected Analyses of Hospital Performance?", OECD Economic Studies No. 44, Vol. 1.

Falch, T. and J. Rattso (1998), "Political economic determinants of school spending in federal states: Theory and time-series evidence", Department of Economics, Norwegian University of Science and Technology, Available online.

Hagen, T.P., T. Iversen and SAC Kettelsen (2007), "Increased privatization and lower costs? Laboratory and diagnostic radiology after the Norwegian hospital reform", Oslo University, Health Economics Research Programme, On line Working Paper Series 2007:2.

Hagen, T.P., Veenstra, M., Stavem, K. (2006), "Efficiency and patient satisfaction in Norwegian hospitals", Working Paper 2006:1, Health Organization Research Norway.

Joumard, I., Andre, C., Nicq, C. and O. Chatal, OECD (2008), "Health care determinants: Lifestyle, environment, health care resources and efficiency", OECD Economics Department Working Paper No 627.

Kalseth, J. (2003), "Politics and resource use in local government service production: Empirical analyses of administration and long-term care in Norway", Dr.polit.-avhandling 2003 Institutt for samfunnsokonomi Fakultet for samfunnsvitenskap og teknologiledelse Norges teknisknaturvitenskapelige universitet, NTNU Trondheim.

Loegreid, P., P.G. Roness, K. Rubecksen (2006), "Modern management tools in Norwegian State agencies: regulation inside government or shopping basket?", Working Paper No. 13, Stein Rokkan Centre for Social Studies.

Linna, M., U. Hakkinen and J. Magnussen (2005), "Comparing hospital cost efficiency between Norway and Finland", Health Economics Research Programme at the University of Oslo.

Magnussen, J., T. Hagen, O.M. Kaarboe, (2007), "Centralized or decentralized? A case study of Norwegian hospital reform", Social Science and Medicine, 64(10), May.

Ministry of Finance (2009), Long-Term Perspectives for the Norwegian Economy.

Ringard, A., T.P. Hagen and A. Rico (2006), "Introducing patient choice of hospital in National Health Systems--A comparison of the UK and Norway", Working Paper 2006:2 Horn--Health Organization Research Norway.

Odeck (2008), "How efficient and productive are road toll companies?: Evidence from Norway", Transport Policy, Vol. 15, Issue 4, July 2008, Pages 232-241.

OECD (2003), Norway, Preparing for the Future Now, June 2003.

OECD (2006a), Directorate for Education, Education and Training Policy Division, Thematic Review Of Tertiary Education: Norway Country Note, September, OECD.

OECD (2006b), "Funding Systems and their Effects on higher education systems", National Study-Norway, November, OECD.

OECD (2006c), Sickness, Disability and Work Review, OECD.

OECD (2008a), Jobs for Youth: Norway, OECD, Paris.

OECD (2008b), "Making the best of Norwegian schools"; OECD Economic Survey Norway, OECD.

OECD (2009a), OECD Employment Outlook 2009, OECD.

OECD (2009b), Government at a Glance, 2009, OECD.

Pilichowski, E. and E. Turkisch (2008), "Employment in government in the perspective of the production costs of goods and services in the public domain", OECD Working Papers on Public Governance No. 8.

Sorensen, P. (2005), "Neutral taxation of shareholder income", International Tax and Public Finance, 12.

Sutherland, D., Price, R., Joumard, I. and C. Nick (2007), "Performance indicators for public spending efficiency in primary and secondary education", OECD Economics Department Working Papers 546, OECD, Economics Department.

Sutherland, D. and R. Price (2007), "Linkages between performance and institutions in the primary and secondary education sector", OECD Economics Departement Working Paper No. 558.

Torp H., D.F. Edvardsen and S.A.C. Kittelsen (2000), "Evaluering av formidling En effektivitetsanalyse av arbeidskontorenes samlete virksomhet basert pa DEA", Institutt for samfunnsforskning og Stiftelsen Frischsenteret for samfunnsokonomisk forskning, Oslo.

Trond, T. and T.P. Hagen (2009), "Soft budget constraints and bailouts: budgeting after the Norwegian hospital reform", Scandinavian Political Studies.


(1.) This work also finds that some groups of the population face a higher probability to enter a disability benefit scheme or a lower probability to stay in employment after experiencing a health problem. There is however some variability across countries across patterns of disability receivers (in some countries for instance women and young individuals are more likely to stay in disability schemes, but not in others). In other countries people working with temporary contracts or in manual occupations tend also to remain less in employment after health problems.

(2.) The Norwegian funding system for higher education institutions (HE) includes both public and private institutions. The public HE institutions have a high share of public funding than the private institutions. The most common private income sources for the private HE institutions are tuition fees and other sources (e.g. external consulting, gifts, donations and R&D-activities).

(3.) Explicit links between the funding system and national higher education policies have been established as a result of the "Quality Reform" process, involving the establishment of a new agency for quality assurance, the Research Council of Norway, which allocates funds on a competitive basis, designating Centres of Excellence so as to give other institutions a benchmark for comparing their own achievements. Performance assessment is based on student assessments of courses, annual departmental or faculty self evaluations of academic performance and administrative efficiency, as well as on periodic external reviews of academic units.

(4.) The funding system was evaluated in 2009 by the Ministry of Education and Research based on input from the institutions and an external consultancy review (Econ Poyry). The results from the evaluation were presented to the Parliament in the state budget for 2010. The main conclusion from the evaluation was that the impact of the funding model has been satisfactory (credits have increased as well as grades; research activity has also improved). Some minor adjustments to the funding system will be implemented in 2010.

(5.) A currently trend is for students to move to the big cities, a trend which may be accentuated by the fact that regional higher education institutions may be disadvantaged by the new funding system. When the number of students and "credit production" are the most important components of the financing system, this favours institutions with many students in each programme, making it difficult for institutions with a small number of students to maintain a sound financial position.

(6.) Life expectancy at birth is 1.5 years higher than the average, but high spending, high levels of education, low tobacco and alcohol consumption and a high standard of living can more than account for this difference, according to cross-section regression analysis reported in Joumard et al. (2008). Allowing for all these factors, life expectancy should be 1 1/2 years higher than it actually is--a residual which is unexplained but which may be ascribed to inefficiencies.

(7.) While general practitioner density is not unusual, that of specialists is high. There does not seem to be a significant relationship between physician density and health outcomes (OECD, 2007b).

(8.) There are 60 official tax expenditures, plus 9 additional tax expenditures which are mentioned in the National Budget but not calculated.

(9.) Some examples of such allowances are the childcare expense deduction and the tax allowance for commuters' daily work travel and visits to main residence, which were not regarded as tax expenditures until 1999. The same applies to the tax expenditure related to the geographically differentiated employer's social security contribution. These tax expenditures are now some of the most significant tax expenditures in Norway.
Table 2.1. The Norwegian pension system before and after the NIS reform

                              Gross replacement rate
Individual               Pre-reform              Post-reform
earnings:           0.5       1     1.5     0.5       1     1.5

Australia          46.2    23.1    15.4      67    41.6    33.1
Austria              90      90    85.9    80.1    80.1    76.4
Belgium            54.8    40.4    31.4    58.1      42    32.5
Czech Republic     72.1      45    32.9    79.2    49.7    36.4
Finland            69.9    66.2    65.2    66.5    56.2    56.2
France             64.7    64.7    58.4    61.7    53.3    48.5
Germany            47.9    47.9    46.5      43      43    42.6
Hungary            69.9    57.7    53.6    76.9    76.9    76.9
Italy                90      90      90    67.9    67.9    67.9
Japan              56.5    40.6    35.3    47.1    33.9    29.4
Korea               100    69.3      56    64.1    42.1    33.6
Mexico             72.5    72.5    72.5    55.3    36.1    34.5
Norway             68.5    56.7    45.4    64.1    53.5      42
New Zealand        77.5    38.7    25.8    79.3    41.1      29
Poland             81.2    62.9    56.8    61.2    61.2    61.2
Portugal           91.3    89.9    88.5    63.0    53.9    53.1
Slovak Republic      65    58.9    39.3    56.4    56.4    56.4
Sweden             82.5    78.6    76.5    76.6    61.5    75.6
Turkey            107.6   107.6   107.6    86.9    86.9    86.9
United Kingdom     41.1    29.7    20.6      51    30.8    21.3
OECD               72.5    61.5    55.2    65.3    53.4    49.7

                               Net replacement rate
Individual               Pre-reform              Post-reform
earnings:           0.5       1     1.5     0.5       1     1.5

Australia          55.3    30.4    21.8    80.2    53.1    41.8
Austria            98.4    99.2    95.1    90.5    90.3    86.3
Belgium            74.2    62.1    50.6    78.7    63.7    51.7
Czech Republic     86.7    58.1    44.6    95.3    64.1    49.4
Finland            75.9    71.4    72.4    73.2    62.4    63.8
France             79.7    78.2    70.8    76.2    65.7    60.2
Germany            56.4    66.6    66.4    59.2    61.3    60.3
Hungary            85.9    83.2    79.1    94.3   105.5    99.2
Italy              99.1    99.1    99.2    74.8    74.8    77.1
Japan              55.8      41      37    51.4    38.7    33.9
Korea             105.9    74.9    61.6    68.8    46.6    38.7
Mexico             73.4    76.5    83.2      56      38    39.6
Norway             85.1    65.7    55.3    81.8    62.5    51.7
New Zealand        77.5    38.7    25.8    79.3    41.1      29
Poland             97.1    76.9    69.7    74.4    74.9      75
Portugal          106.1     112   110.8    73.2    69.6      72
Slovak Republic    76.4    75.9    52.2    66.3    72.7    74.9
Sweden             84.5    80.3    81.9    79.3    64.1    81.2
Turkey              150   154.4   157.9   121.2   124.7   127.1
United Kingdom     51.9    39.8    28.3    63.8    40.9    29.2
OECD               83.8    74.2    68.2    76.9    65.7    62.1

Note: Replacement rate is expressed as the ratio of pension to
final earnings (just) before retirement. Under the baseline
assumptions, workers earn the same percentage of economy-wide
average earnings throughout their career. The gross replacement
rate is defined as gross pension entitlement divided by gross
pre-retirement earnings. It is shown here at 0.5, land 1.5 times
average earnings levels underlying the newly defined OECD
"average worker" concept. The net replacement rate is defined as
the individual net pension entitlement divided by net
pre-retirement earnings, taking into account personal income
taxes and social security contributions paid by workers and
pensioners. The table shows the impact of recent pension reforms
in OECD countries, by comparing gross (net) replacement rates of
a worker entering the labour market before the reform with those
of a worker entering after the reform. For Norway the post-
reform replacement rates include both the reform of the National
Insurance Scheme and the fact that private pensions were made
mandatory. OECD average refers to the arithmetic average of the
30 countries (not population weighted).

Source: OECD, Pensions at a Glance (2009). Norwegian figures were
provided by the national authorities to take into account a
reform approved after the publication of PAG (2009).

Table 2.2. Government sector production and employment

                                        % GDP
                                                        % total
                                  Total      Mainland

Government production of which:     20.9       27.6
Compensation                        12.4       16.4       59.4
Contracting out                      6.4        8.5       30.7
Provided by the private sector       2.1        2.7        9.9
DI, inv. and transfers              21.3       28.1
Total government expenditure        42.2       55.7
Employment in government            29.0
  (% total labour force)
Compensation costs per employee      1.0
  (ratio of public to private)

                                        OECD average

                                     % GDP       % total

Government production of which:       19.9
Compensation                          10.6         52.6
Contracting out                        6.3         31.1
Provided by the private sector         3.3         16.3
DI, inv. and transfers                42.7
Total government expenditure          42.7
Employment in government              15.0
  (% total labour force)
Compensation costs per employee    0.9 to 1.3
  (ratio of public to private)

Source: Pilichowski and Rirkisch (2008).

Table 2.3. Potential for hospital cost reductions
based on standardised DRGs (1)

Per cent, 2006

Australia        42
Denmark           5
Finland          13
France           44
Germany          32
Iceland          38
Norway           34
Sweden           42
United Kingdom   12
United States    48

(1.) Based on cross-country comparisons of hospital unit costs
for seven DRGs, with lowest unit costs used as a benchmark.

Source: Erlandsen (2008).
COPYRIGHT 2010 OECD Publications and Information Centre
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Publication:OECD Economic Surveys - Norway
Date:Mar 1, 2010
Previous Article:Chapter 1: emerging from the crisis.
Next Article:Chapter 3: sustainable development: climate change and fisheries policies.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters