WP4. 1. Land resources
Land is a key, but limited resource, in the development of a bio-economy. The optimal allocation
and management of different land uses is a major challenge (deGroot et al 2010). With the 2011
White paper (LMD, 2011) Norwegian agriculture got a renewed incentive to increase agricultural
production. A remaining question, however, is the location, quality and extent of available land
resources, in particular with respect to other landscape functions (see e.g. Bastian 2000, Bastian &
Lutz 2006). Further, a stronger focus on efforts to increase production and efficiency in farming can
result in a less sustainable production or loss of ecosystem services, e.g. in “High nature value
farmland” (Fjellstad et al. 2012, Henle et al. 2008) unless it can be ensured that recommended
growth and change do not jeopardisze sustainability (see e.g. Firbank et al. 2013).
At present there is an on-going polarization of the agricultural landscape in Norway, as more
intensive use of land and land abandonment take place simultaneously and sometimes even within
the same regions. Both these processes may have a negative effect of production of public goods
(Cooper et al. 2009). Farming activity within areas with such contrasting trends requires different
measures to increase or maintain the provision of public goods (Westhoek et. al. 2013). Land
use/land cover is a key issue in this context, and the influence of on-going and possible future
changes needs to be analysed and communicated, e.g. through the use of well-functioning indicators
(see e.g. EEA 2006, Müller & Burkhard 2012). An understanding of the spatial variation is thus
essential for a sustainable management of cultural landscapes in a growing bio-economy.
In addition to providing input to the other thematic areas and WPs, this WP will focus on the
following key question; How can spatial patterning of land resources in Norway and potential
consequences of changes in the use of these resources best be mapped and what indicators best
describe the situation and can be used to communicate these findings to different user groups?
Underlying themes are:
1. What is the potential for and possible consequences of renewed intensification of
2. What does abandonment mean in terms of loss of production potential, what is the speed of
3. To what extent can we use map-based indices as a proxy for public preferences and their
perception of the cultural landscape?
4. How important are various landscape elements to selected public goods in the agricultural
landscape, how does this vary with spatial context and can we relate this to the above
mentioned indices and map based evaluations?
Analyses related to the first two questions are based on use of AR5 and SLFs producer database.
The first will be analysed using statistical analysis, while in question two the use of GIS-analysis
is essential. Question 3 and 4 will be addressed through calculated indices from existing, new
and revised maps as well as a new survey of the perception of landscape elements by different
groups of stakeholders related to an assessment of indicator performance.
Work package leader: Wenche Dramstad, NFLI. Additional participants: CRR & David Miller
WP4.2 Types of production
Conditions for types of production vary across Norway. Still, Norwegian agriculture has survived
despite marginal conditions and a small-scale structure, due to a protectionist setting with the
support and cooperation of the public, the state and agricultural actors (Bjørkhaug and Richards
2008). The cultural, historical and political importance of the small-scale, independent self-owning
farmer has been pointed out (Almås, 2004; Daugstad et al. 2006; Rønningen et al 2012). However,
recently the policy has shifted from ‘preservation’ to ‘neo-productivism’ (Bjørkhaug et al, 2012,
Almås et al 2013) with renewed focus on sustainable growth and prospects of a new bioeconomy.
Given the high production costs Norway cannot compete on an international agricultural market,
except for products with specific qualities. While an overarching WTO agreement on agricultural
trade has yet to be agreed, the latest drafts propose the categorisation of so-called ‘sensitive
products’ that may be allowed higher tariffs. For Norway, dairy and beef production based on
home-grown feed and grazing resources are most likely to be defined as sensitive products. The
outfields, accounting for more than 80 percent of Norway’s land area, represent valuable feed
resources and qualify, to a large extent, as organic. Yet a short growing season in large parts of the
country set some strict limitations. In this sense, the principle of “relative comparative advantages”
which was the basis for the policy of agriculture from the 1950s onwards, with dairy and beef
production in fjord and mountain areas and Northern Norway and grain in central and highly
productive areas (Almås 2004; Borgan 1978), may possibly live on under more liberal political
conditions, but very likely with fewer producers. Similar distinctions exist in Austria between dairy
livestock in the Alps and grain and wine production in the east, but current challenges arise with
abandonment of the milk quota system in the EU.
Easy access to imports of soybeans and other concentrated feedstuff, chicken and hog production
represent in principle productions independent of domestic natural resources and a step towards a
globally integrated bioeconomy. However, current changes in the food supply chains through
vertical integration and changing consumer behaviour (Bjørkhaug et al 2012, Richards et al 2013)
as in a developing “eco-economy” can affect localisation of agricultural productions (cf. WP4.3).
Hence, natural resources, framework conditions (prices on input factors and products and
policy), value chains and markets, (food, fuel, energy) affect development and localisation of landbased
bio-production. In addition, the community (e.g. socio-cultural conditions for production) and
actors making informed choices have influence. WP 4.2 asks:
1. How do types of production vary with spatial localization (what is being produced and
how it is conducted)?
2. What opportunities and challenges will be created under different forms of bio-economic
transitions in land-based productions, and how can these be exploited and mitigated at the
3. What lessons can be learned about types and localisation of agricultural and bio-based
productions through comparison with naturally similar but politically and socio-culturally
different systems & regions in Austria?
Data and methods for investigating the questions will be multilevel analysis of quantitative register
(Economic and demographic, direct payment base) and survey data (e.g Trender i landbruket)
coupled to e.g. Norwegian Institute of Forest and Landscapes database on landscape characters
(Landscape models in 5x5km grids) (See e.g. Heggem et al N.D.) and other spatial indicators in
close relationship to WP4.1 and WP4.3. These will be complemented with qualitative data analysis
such as interview data from selected case regions with farmers and farmers’ families, landowners,
authorities, and other stakeholders, and document (policy documents, other strategic documents)
reviews. The comparative component will be built on Austrian data.
Work package leader: Hilde Bjørkhaug, CRR. Additional participants: NILF, Arild Blekesaune
(NTNU), Markus Schermer (Austria) & Maureen Kilkenny (US).
WP4.3 Value chains
Due to natural conditions Norwegian agriculture is characterised by particularly vulnerable value
chains and, as a result, production at the farm gate cannot be analysed independently of
production at the processing stage. For example, because of the difficulties in shipping raw
commodities form Southern Norway; farms and agricultural industries in Northern Norway
depend on the production of raw materials in their own region. This has strong implications for
the development of the bio-economy in the region. For this reason, a study of spatial
heterogeneity in Norwegian agriculture needs to apply a broad perspective on value chains. In
this WP, the agri-food value chain includes markets for agricultural inputs (e.g., land, capital,
labour), agricultural outputs (e.g., raw milk, animals for slaughter, crops), food products at
several levels of the food industry as well as final food products demanded by consumers at the
Research questions addressed in this WP are therefore:
1. How large is the (spatial) heterogeneity of farms of similar types in similar regions?
2. What causes the observed heterogeneity?
3. How can the observed heterogeneity be rationalized and modelled?
4. How does heterogeneity affect the impact of a policy change on the spatial distribution of
The coherent and consistent framework for the spatial analysis of agricultural value chains will
be the Norwegian agricultural sector model Jordmod which is frequently used for policy
analysis. The model assumes about 400 different programming models at the farm level (i.e.,
farm types) representative of Norwegian agriculture as a whole. Farms within each farm type are
simply duplicated during the model’s optimization procedure that leads to market equilibrium.
Hence, the model’s equilibrium solution is characterized by a large number of identical farms
with homogenous technology. This feature leads typically to regional overspecialisation.
Jordmod also includes dairy firms and two types of meat processing firms, one for slaughtering
and one for processed meat products. The structure of the food industry in the model is
determined by a trade-off between transportation costs between farms and processing plant on
the one hand and economies of scale for farms and processing plants on the other; Large and few
processing plants and farms lower per unit processing costs, but increase per unit transportation
costs. Similarly, many small processing plants and farms give rise to higher per unit processing
costs, but lower per unit transportation costs. In its current form, the model is not suited for the
spatial analysis of policy change.
The objective of this WP is therefore to enable in Jordmod the spatial analysis of a policy
reform or a change in the sector’s exogenous framework conditions on the performance of the
agricultural sector at the regional level. As spatial heterogeneity is more important at the farm
level than the food industry level, this WP will focus at the farm level. This requires the
introduction of empirically specified regional programming models at the farm level that
represent the observed spatial heterogeneity. The scientific challenge consists of developing
convincing methods that explicitly aim at rationalizing farmers’ observed choices based on
heterogeneous farming conditions and farmers’ behaviour.
A common approach to overcome the overspecialisation problem is Positive Mathematical
Programming (PMP) introduced by Howitt (1995). PMP uses the dual solution of a constrained
optimization problem to specify a quadratic cost term that allows exact model calibration to
observed base year values. Although appealing, Heckelei and Britz (2005) argue that the
theoretical and empirical validity of the cost function is questionable as it may hide both data
errors and important unobserved spatial heterogeneity. Moreover, the function is treated as a
black box in the model. Instead, we will rationalize observed farm management and farmers’
observed behaviour using primal solutions to calibrate agricultural supply in Jordmod. Contrary
to PMP which was developed for an environment where little empirical data is available, our
proposed method will make use of rich empirical data sets from various sources.
Data will be taken from mainly three sources: (1) the Direct Payment database that contains
the activity levels (acreage and herd size) for all farms applying for direct payments in the years
1995-2013, (2) the Norwegian farm accounts comprising about 1,000 farms for the same period,
and (3) the income tax accounts of all farms applying for direct payments for the years 1999 and
2010 (and possibly for the years therein). The first and the last data source are especially useful
as we hypothesize that heterogeneity between similar farms can be rationalized through the
farmers’ opportunity costs of labour and farm household characteristics. The novelty and
methodological challenge of this WP will be that the calibration procedure will be applied at the
farm level (instead of the activity level).
Work package leader: Klaus Mittenzwei, NILF. Additional participant: Thomas Heckelei,
WP4.4 New policy for a new bio-economy
Agricultural policy strongly influences both agricultural production and its spatial distribution. In
Norway policy pays particular attention to regional differences, for example, by providing
economic support to compensate areas less favoured by climatic conditions, topography or transport
distances. This is manifested in the canalisation policy (also see WPs 4.2 and WP5) whereby
subsidies have been directed towards supporting grain production in optimal grain producing areas
and animal production in more marginal environments (Almås, 2004). Other spatially distributed
policies include Rural Development Funds (“Bygdeutviklingsmidler”) (Almås, 2002) and Regional
Environmental Schemes (Brandtzæg et al., 2008). In contrast to the canalisation policy these
schemes are administered at a regional level (by the County Governor) and aimed at supporting
diversification and the provision of public goods.
As discussed in Section 2.1 the development of a bio-economy has two potentially conflicting
visions. One is of a technologically advanced industry based on globalised supply chains (the “bioeconomy”)
and the other is of the realignment of production-consumption chains to capture local
and regional value (the “eco-economy”). The possibility of a pluralistic approach involving both
(Levidow et al., 2012) sets an interesting policy challenge for Norway: how to develop policies that
promote Norway’s integration into a bio-technologically driven globalised future while, at the same
time, preserving the ‘eco-economy’ that persists in much of the country? Both are likely to be
required to meet Norway’s overall policy goal of increasing agricultural production to match
agricultural productivity growth with expected population increase (LMD, 2011). As these goals
may conflict in some regions of Norway, new policies for the bio-economy must have a strong
This work package sets out to identify the challenges to developing spatially considered polices
that promote balanced territorial development in the wider bio-economy. Here we will draw on the
spatial analysis of WPs 4.1 to 4.3 and combine these with an assessment of the policy environments
that have contributed to these distributions (conducted in this work package). The scenarios
developed in WP5 will then be employed for a final analysis of the policies required to meet the
goals established in the scenarios. The main tasks of this Work Package are therefore:
1. To outline and analyse the spatial impact of Norway’s current agricultural policies.
2. To analyse the impact policies have had on the spatial outcomes of WPs 4.1 to 4.3.
3. To investigate the spatial conflicts between the goals of increasing production via
integration into the bio-economy and the need to maintain agriculture in all regions of
4. To make recommendations for policies that may achieve desirable outcomes for
Norwegian landbased bioproductions.
To assist in policy development, comparisons will be made between policy environments and their
spatial outcomes both within the European Union (i.e. Austria – as an example of a regulated
policy-led approach in a country with similar agricultural challenges and a predominantly ecoeconomic
approach) and New Zealand (as an example of a free market approach with a strong bioeconomic
Studies of documents (policy documents, regulations, statistics, research reports etc.) and
interviews with key informants (politicians, authorities, agro-industrial firms, researchers and experts) will be employed in the analysis. Investigations will be conducted through fieldwork in
both of the comparative regions. Final policy recommendations will be developed in close
collaboration with the Norwegian Government.
Work package leader: Magnar Forbord, CRR. Additional participants: NFLI, NILF, Markus
Schermer (Austria) & Hugh Campbell (New Zealand)