ARTICLE IN PRESS
Land Use Policy 24 (2007) 562–575 www.elsevier.com/locate/landusepol
Economic and ecological consequences of four European land use scenarios B. Eickhouta,, H. van Meijlb, A. Tabeaub, T. van Rheenenb a
Netherlands Environmental Assessment Agency (RIVM/MNP), P.O Box 303, 3720 AH, Bilthoven, The Netherlands b Agricultural Economics Research Institute (LEI), The Hague, The Netherlands Received 4 March 2005; received in revised form 5 October 2005; accepted 2 January 2006
Abstract The impact of globalization on trade, production and land use is key to the Doha development round. This paper deals with the complex interaction between agricultural trade, production, land-use change and environmental consequences on the basis of four different scenarios. In these scenarios, major uncertainties from trade liberalization to maintained regional trade blocks are considered. Although economic growth is apparent in liberalizing scenarios, we also found that environmental threats of climate and nutrients to the sustainability of the global agricultural practices pose new challenges to future food production. Since most of the environmental threats will be experienced in tropical regions where most of the increase in population and food and feed demand is expected, an indirect pressure on the European agricultural market is likely. For the coming decades European agriculture is expected to decrease slightly, especially in liberalizing worlds. New demand for land for biofuels, carbon plantations and the global food market, will prevent the European agricultural sector from being eliminated. Moreover, current EU policies already result in less vulnerable farmers to additional liberalizing policies. Therefore, we conclude the global context is important for future European land use, especially in futures where environmental policies are ignored. Therefore, we conclude that environmental and trade agreements must be sufficiently integrated or coordinated, to assure they work together to improve the environment and attain the benefits of free trade. r 2006 Elsevier Ltd. All rights reserved. Keywords: Trade liberalization; Economic benefits; Environmental consequences; Climate change; Land use; Scenarios
Introduction On the 1st of May 2004 the European Union (EU) was extended with 10 countries from Central and Southern Europe. Until then the European Union spent most of its budget (42 billion of Euros) on its common agricultural policies (CAP). A dominant feature of the CAP was to offer European agricultural producers guaranteed prices above market level. Price support, however, created distortions to production as farmers were encouraged to expand supply and produce large surpluses of agricultural commodities which implied huge storage costs and costs to dispose surpluses on the world market (export subsidies). In addition, it requires import protection and market Corresponding author. Tel.: +31 30 274 29 24; fax: +31 30 274 44 64.
E-mail address:
[email protected] (B. Eickhout). URL: http://www.mnp.nl/en. 0264-8377/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.landusepol.2006.01.004
access barriers to restrict imports. This development has led to much criticism on the European CAP in the international trade liberalization or WTO rounds. Furthermore, one of the key motives for agricultural support is to provide and maintain a sufficient farm income. However, empirical research by the OECD (2001) as well as the World Bank (2003) demonstrates that price support is neither effective in serving this objective, nor is it very efficient. These pressures led to shifts in the European CAP from market price support to income support improving the income transfer efficiency and declining the distortion on markets. Good examples of such a shift from market price support to income support are the McSharry and Agenda 2000 CAP reforms of the EU (Van Meijl and Van Tongeren, 2002). In the 2003 Mid-Term Review of the CAP the support changed from coupled income support to decoupled income support, which is more effective and efficient. With the accession of the Central European
ARTICLE IN PRESS B. Eickhout et al. / Land Use Policy 24 (2007) 562–575
countries there is an additional driving force to adjust the European CAP within the coming decade to prevent an ever-increasing budget pressure on the European budget (OECD, 2003). Given these developments, many shifts in the European agricultural and trade policies are expected in the coming 30 years. Moreover, in the same period world population and world food demand will also continue to increase (UN, 2002; Bruinsma, 2003). In combination with expected increases in economic growth (World Bank, 2003), these demographic shifts will have a major impact on the global food supply market and therefore on the European agricultural market. Hence, future land use scenarios need to be considered in the light of changing global conditions, agricultural policies and the impact of these on the environment. So far, the discussion on consequences of changes in OECD agricultural policies has been dominated by economic analyses. Most of these studies indicate that trade liberalization will be beneficial for most of the developing countries (Anderson, 1999; Hertel et al., 1999), with optimistic analyses claiming that trade liberalization will add 1.7% to the GDP of developing countries (World Bank, 2003). In scientific literature the discussion mainly focuses on the disputed positive assumptions of the World Bank study (Van Meijl and Van Tongeren, 2002; Francois et al., 2005). However, most of these studies ignore the objectives of sustainable development and environmental protection that are stated in the preambule to the Agreement Establishing the World Trade Organization (WTO, 2003). On the environmental side, most of the European land studies only looked at environmental consequences of different long-term baseline scenarios with exogenous driving forces of population, economic growth and agricultural food supply (Rounsevell et al., 2006). Some studies touched upon the consequences of land and water shortage for the amount of food demand, but only used one reference scenario without implementing specific changes in the agricultural trade regime (Bruinsma, 2003; Rosegrant et al., 2002). Moreover, only few studies touched upon issues like agricultural intensification because of increased pressure on arable land and the consequences for the nitrogen cycle (Cassman et al., 2003; Bouwman et al., 2005). Here, we present a study that combines an economic analysis of the consequences of agricultural (CAP) and trade policy (WTO) reform with an environmental analysis on changes in European land use. By combining these two analyses, a coherent insight is obtained in the economic and environmental consequences of changing global drivers and changes in agricultural and trade policies. To take uncertainties in global drivers into consideration, we used four storylines that are based on the storylines of the Intergovernmental Panel on Climate Change (IPCC: Nakicenovic et al., 2000). By downscaling these global storylines to the European level, we can assess both the environmental and economic consequences for Europe,
563
imbedded in four global storylines. This study was part of the EURURALIS study (Klijn et al., 2005), which was designed to cover the integrated future of European’s agriculture and rural areas, starting from four contrasting scenarios (Westhoek et al., 2006) and desiring outlooks for socio-economic developments in Europe and the European environment. Of the environmental indicators, we focused on the land use results since trade policies have an immediate impact on land use. To assess economic consequences and landuse changes, a modeling framework capable of capturing both economic and environmental consequences is needed. The Integrated Model to Assess the Global Environment (IMAGE; Alcamo et al., 1998; IMAGE Team, 2001) is one of the most frequently used global land models, to simulate land use emissions (Strengers et al., 2004), environmental impacts (GEO3, 2002; Eickhout et al., 2006) and impacts on ecosystems (Leemans and Eickhout, 2004; MA, 2005). However, the IMAGE model is less suitable for economic analyses, given the scarce elaboration of the agricultural economy model (Strengers, 2001). Therefore, the IMAGE model needed to be expanded with an agricultural economy model capable of implementing different agricultural policies. The adjusted version of the Global Trade Analysis Project (GTAP; Hertel, 1997; Van Meijl et al., 2006) model offers a viable framework taking into consideration the impact of non-agricultural sectors on agriculture and a full treatment of factor markets with special features concerning land modeling. Hence, the combination of IMAGE and GTAP offers a modeling framework that can achieve the some of the key objectives of the EURURALIS study (Klijn et al., 2005). In ‘Methodology’, we present the methodology of the IMAGE–GTAP modeling framework. Thereafter, the scenarios that were used for EURURALIS study, including the policy assumptions that were implemented, are being introduced in ‘Scenarios and policy assumptions’. In ‘Consequences for agricultural production, trade and income’ we will elaborate on the economic results of the modeling framework IMAGE–GTAP. The consequences for the European agricultural practices are described in ‘Change in land use and agricultural practices’, and the environmental impacts are given in ‘Consequences for the European environment’. We conclude with our major findings in ‘Discussion and conclusions’. Methodology To analyze the economic and environmental consequences of changes in global drivers and agricultural policies, we developed a global economic-biophysical framework with focus on Europe, by combining the GTAP model (Van Meijl et al., 2006) with the IMAGE model (Alcamo et al., 1998; IMAGE Team, 2001). The standard GTAP model is characterized by an input–output structure, based on regional and national input–output tables. It explicitly links industries in a value added chain from
ARTICLE IN PRESS 564
B. Eickhout et al. / Land Use Policy 24 (2007) 562–575
primary goods, over continuously higher stages of intermediate processing, to the final assembling of goods and services for consumption (Hertel, 1997). For this analysis an extended version of the standard GTAP model was developed that improved the treatment of agricultural production and land use (Van Meijl et al., 2006). Since it was assumed that the various types of land use are imperfectly substitutable, the land use allocation structure was extended by taking into account the degree of substitutability of types of land differs between types (Huang et al., 2004). Therefore, OECD’s more detailed Policy Evaluation Model (OECD, 2003) structure was used. Moreover, in this extended version of the GTAP model the total agricultural land supply was modeled using a land supply curve which specifies the relation between land supply and a rental rate (Van Meijl et al., 2006). Through this land supply curve an increase in demand for agricultural purposes will lead to land conversion to agricultural land and a modest increase in rental rates when enough land is available, whereas if almost all agricultural land is in use increases in demand will lead to increases in rental rates. Fig. 1 shows the methodology of iterating the extended version of GTAP with IMAGE. Macro-economic drivers like population and economic growth are used as input in both the GTAP and IMAGE model. In the extended GTAP model yield depends on a trend factor and also on prices. The production structure used here implies that there are substitution possibilities among production factors. If land gets more expensive, the producer uses less land and more other production factors such as capital. The impact of a higher land price is that land productivity or yields will increase. Consequently, yield is dependent on an exogenous part—the trend component—and on an endogenous part with relative factor prices, which is the management component. The exogenous trend of the yield was taken from the FAO study ‘Agriculture towards 2030’ (Bruinsma, 2003) where macro-economic prospects were combined with local expert knowledge. However, many
studies indicated this change in productivity are enhanced or reduced by other external factors, of which climate change is mentioned most often (Rosenzweig et al., 1995; Parry et al., 2001; Fischer et al., 2002). These studies indicated increasing adverse global impacts on crop yields because of climate change will be encountered with temperature increases above 3–4 1C compared to preindustrial levels. These productivity changes need to be included in a global study. Moreover, the amount of land expansion or land abandonment will have an additional impact on productivity changes, since land productivity is not homogeneously distributed. The economic consequences for the agricultural system are calculated by GTAP. The outputs of GTAP include, among others, sectoral production growth rates, land use, and an adjusted management factor describing the degree of land intensification. This information is used as input for the IMAGE simulations, together with the same global drivers as used by GTAP. Since the IMAGE model performs its calculations on a grid scale (of 0.51 0.51) this heterogeneity of the land is taken into consideration on a grid level (Leemans et al., 2002). Here, we only present land use results on country level, since the geographical explicit results are simulated on a finer resolution of 1 1 km by the CLUE model (Verburg et al., 2006). This procedure delivers an amount of land needed per world region and the coinciding changes in yields, because of changes in the extent of used land and climate change. Next, these additional changes in crop productivity are given back to GTAP, therefore correcting the exogenous trend component of the crop yield. A general feature is that yields decline if large land expansions occur since marginal lands are taken into production. In the near term, these factors are more important than the effects of climate change. Through this iteration, GTAP simulates crop yields and production levels on the basis of economic drivers and changes in environmental services. This combined result is once more used as input in IMAGE to calculate the environmental consequences consistently. The environmental impact models of IMAGE involve specific models for sea-level rise and land degradation risk and make use of specific features of the ecosystem and crop models to depict impacts on vegetation and crop growth (Bakkenes et al., 2006; Leemans and Eickhout, 2004). The details of the modeling framework of GTAP and IMAGE are documented in Van Meijl et al. (2006). Scenarios and policy assumptions
Fig. 1. The modeling framework of GTAP and IMAGE.
To deal with uncertainties in future global drivers and changes in agricultural policies, four scenarios have been developed on the basis of four narratives. These narratives are an elaboration of those developed for the Special Report on Emission Scenarios (SRES) of the IPCC (Nakicenovic et al., 2000). The quantitative translation of these SRES narratives is updated with a focus on Europe given the objective of this paper. The SRES narratives have
ARTICLE IN PRESS B. Eickhout et al. / Land Use Policy 24 (2007) 562–575
been used, since these narratives can be linked relatively easy to current agricultural trade discussions (protection versus liberalization) and societal developments (globalization versus regionalization; see also MNP, 2005, for a selection of societal concerns on the basis of a public opinion survey). Other scenarios like the scenarios from the Global Environment Outlook (GEO3, 2002) and the Millennium Ecosystem Assessment (MA, 2005) are more difficult to link to such issues. For example, within the GEO3 scenarios a regionalized, protectionist scenario is lacking, whereas in the MA framework a pure marketoriented, globalized scenario is missing. Therefore, we decided to use the SRES narratives as a basis for our study (see also Klijn et al., 2005, for the objectives of the EURURALIS study). As mentioned above, the narratives for this study are developed along the line of two axes. One axis distinguishes a world that is focused on economic welfare values and market orientation (‘low regulation’) versus one that is focused on social and environmental values with a strong role for governments (‘high regulation’). The other axis distinguishes a world being further globalized versus a regionalized world. These two axes lead to a world that can evolve in four different directions, depending on people’s dominant drivers and world visions (see Fig. 2). The rationale for these four scenarios is based on the acknowledgement that different worldviews exist and that it is impossible to predict which world vision will become dominant. The rationale of these scenarios is elaborated further in Westhoek et al. (2006). For this analysis, we laid down the possible outcomes of future trade policies in these four narratives. The Global Economy (GE) scenario; equivalent of A1 of SRES) assumes multilateral cooperation on economic issues and successful WTO negotiations. Global trade will be fully liberalized and a successful economic integration in Europe results in further eastwards EU enlargement. Global integration puts poor countries on a path of catching up and high growth. Technological change is high and driven by economic profit and not directed to or hampered by
Fig. 2. Four scenarios along the two axes of globalization versus regionalization and efficiency versus equity.
565
planet and people considerations, e.g. current environmental restrictions are not increased and genetically modified organisms (GMOs) are accepted. Similarly, to the GE scenario, the Global Co-operation (GC) scenario (equivalent of B1 of SRES) assumes that international trade will be liberalized. However, societal values not only consist of profit but also of people and planet. This implies that contrary to the GE scenario domestic support in agriculture will partly be sustained because subsidies will be linked to environment and nature conservation. International agreements on environment, biodiversity, animal welfare, working conditions and food safety play an important role. Successful climate policies are implemented leading to a stabilization of greenhouse gas concentrations at a level of 550 ppmv (Bollen et al., 2004). This all leads to a lower economic development and a slightly lower productivity growth in economic terms than in the GE scenario. In the Continental Markets (CM) scenario (equivalent of A2 of SRES) the focus is on markets and economic incentives, combined with national interests prevailing. Multilateral trade liberalization fails and the USA and EU are pursuing their own interest by creating a transatlantic internal market. Important goals are ensuring food security and food safety on the internal market. Technological change is lower in developing countries whose markets become more segmented and separated. This yields welfare gains in EU and the USA in contrast with poverty in Eastern Europe and developing countries. This is accompanied by slow population growth in industrial countries and fast population growth in developing countries due to continuing poverty. In Regional Communities (RC) scenario (equivalent of B2 of SRES) the focus is on both economic and non-economic values, but national interests prevail. Trade and agricultural policies remain almost unchanged, except for export subsidies that are abolished because this kind of dumping is socially not accepted. Income subsidies are accompanied with strict environmental regulations and preserving nature and landscapes. Consumers prefer products from their own region, regional self-sufficiency is high. Agricultural production is dominated by small-scale farms (partly organic farms), producing at relatively low intensity. EU integration is only partial and technological change is limited because of segmented markets and the focus on non-economic issues (GMOs not allowed, environment important). The resulting economic growth is lower than in other scenarios. Social values lead to catching up of developing countries because they can adopt existing technologies from developed countries. In the analyses the population numbers (Table 1) are taken from IPCC (Nakicenovic et al., 2000). The GDP numbers (Table 1) are taken from the report Four Futures of Europe (CPB, 2003), in which comparable four narratives have been used with focus on the EU. Just like most of the economic studies, the GDP numbers have been derived using market exchange rates (MER). Within the
ARTICLE IN PRESS B. Eickhout et al. / Land Use Policy 24 (2007) 562–575
566
Table 1 Macro assumptions, growth rates in per cent per year (CPB, 2003)a Gross Domestic Product (GDP)
Western Europe Eastern Europe Canada USA Oceania Japan Eastern Asia South-east Asia Southern Asia Central America South America Former Soviet Union Turkey Middle East North Africa Central Africa Southern Africa
Population
GE
GC
CM
RC
GE
GC
CM
RC
2.6 3.7 2.8 3.0 2.5 1.6 6.1 4.8 7.1 4.0 3.7 3.7 4.2 4.2 5.0 7.9 4.8
1.8 3.6 2.2 2.4 1.9 0.9 5.3 4.2 6.6 3.8 3.5 3.4 4.2 4.2 5.0 8.5 5.4
2.2 3.0 2.2 3.0 1.9 1.2 3.7 2.9 4.1 2.0 1.7 3.3 2.6 2.6 3.2 6.6 3.9
1.0 2.1 1.8 2.0 1.5 0.7 4.8 3.8 5.4 3.1 2.8 1.9 3.3 3.3 3.9 5.5 2.9
0.33 0.05 0.58 0.80 0.29 0.04 0.43 0.91 1.32 1.09 1.09 0.07 1.78 1.78 1.70 3.53 2.23
0.33 0.05 0.58 0.80 0.29 0.04 0.43 0.91 1.32 1.09 1.09 0.07 1.78 1.78 1.70 3.53 2.23
0.13 0.03 0.62 0.91 0.39 0.01 1.07 1.34 1.57 1.61 1.61 0.37 2.14 2.14 2.11 3.78 2.42
0.05 0.12 0.66 0.72 0.37 0.02 0.61 1.06 1.37 1.12 1.12 0.08 1.65 1.65 1.65 3.86 2.48
a In contrast to the original CPB projections, in this study Latin America was assumed not to be part of the EU-US trade block, leading to lower economic growth than in CPB (2003). These data were kindly provided by Arjan Lejour (CPB).
economic community there is a debate whether economic numbers have to be corrected with respect to purchasing power parities (PPP). Since in our analysis no convergence of economic level is assumed, most of the critique on using MER is not applicable here.1 Along these most important driving forces and along the line of the four story lines the four scenarios have been developed. In Table 2, the specific assumptions on implementation of trade liberalization, agricultural policies and consumer preferences are summarized. The basic technological change is taken from Bruinsma (2003). To make a distinction between the scenarios it is assumed GE and GC are on the high-side of this FAO-projection and CM and RC on the low-side (Eickhout et al., 2004).
Consequences for agricultural production, trade and income In this section we present results of simulation experiments. Since in our analysis differences between scenario outcomes are very important, we focus more on scenario comparison than on the description of the individual scenarios. We pay special attention to the importance of macro versus policy effects and their impact on agricultural sectors in developing and EU countries (distinguishing results for the EU-15 and the 10 accession countries; hereafter EU-10). We comment the world agricultural production, the world trade development, the self-suffi1 The discussion has been mainly focused on the SRES scenarios of the IPCC in a critique from Castles and Henderson (2003a), replied to by Nakicenovic et al. (2003). The discussion is still continuing (see for example the re-iterating discussion by Castles and Henderson, 2003b and Gru¨bler et al., 2004), although current consensus seems to point to using MER because of lack of existing projections in PPP.
ciency of EU countries and the development in farmers’ income under different scenarios.
Global and European agricultural production In all four scenarios, the highest crop and livestock production growth is observed in the developing regions Africa, Asia and Latin America (Figs. 3 and 4). Because of the high economic growth in these regions in the liberalizing scenarios, the largest increases in production are observed in GE and GC. Crop production growth in the EU countries is relative low compared to the other world regions (Fig. 3). Lower economic growth in combination with a low-income elasticity of demand is important in this respect. For countries outside the EU-US market in the Continental Markets (CM) scenario crop production is lower than in other scenarios due to low economic growth and no enhanced access to the transatlantic market. Livestock production growth is also low in the EU countries relative to other regions (Fig. 4). This is explained by lower demographic and economic growths (Table 1). In the EU-15 and other high-income countries meat consumption declines due to preference shifts in diet in GC and RC (Table 2). For the EU-10 and developing countries this effect is smaller, because of the higher income that dominates the increase in consumption. For the EU it is important to analyze the influence of different policy options on these production results. Therefore, the effects of changes in domestic support (domestic subsidies) and border support (export subsidies and import tariffs) on production growth are distinguished in Figs. 5 and 6 for the EU-15 and -10 countries, respectively. The figures describe production growth for
ARTICLE IN PRESS B. Eickhout et al. / Land Use Policy 24 (2007) 562–575
567
Table 2 Policies and consumer preferences in the scenarios (Van Meijl et al., 2006)
Border support Export subsidies Import tariffs Trade blocks
Domestic support Domestic subsidies
Milk and sugar quota Consumer preferences Preference for regional products Consumption of animal protein from meat
All scenarios
Global Economy
Global Co-operation
Continental Markets
Regional Communities
2003 CAP reform 2003 CAP reform Enlargement to EU27
Abolished Abolished Rumania, Bulgaria, FSU accede EU
Abolished Abolished Rumania, Bulgaria, FSU accede EU
No change No change EU-USA
Abolished No change Manufacturing: FTAA (North+South America), TURMiddle East and North Africa, Rest Africa, FSU
2003 CAP reform (incl. decoupling)
Abolished
No change
+10%, linked to env. and social targets
2003 CAP reform
Abolished
67%, rest linked to env. and social targets Abolished
Self sufficient EU
Self sufficient EU
No
No
Endogenous outcome
Meat consumption 10% lower
Preference for products from own region (5%) Endogenous outcome
Preference for products from own region (5%) Meat consumption 10% lower
Fig. 3. Annual growth of crop production, 2000–30 (GE ¼ Global Economy; GC ¼ Global Co-operation; CM ¼ Continental Markets; and RC ¼ Regional Communities).
products under CAP (grains, oilseeds, sugar, beef and dairy) and other agricultural products (non-CAP, such as horticulture, pork and poultry) and the contribution of specific changes in policy instruments (Table 2). Fig. 5 shows that European production growth of products with protection of CAP is lower than for other agricultural products in all scenarios. This is caused by a lower-income elasticity of demand and by reduction in CAP support in some scenarios. Reduction of border support has a significant negative impact in the global scenarios (GE and GC) where all border support is abolished. The impact is less severe in the RC scenario where only all export subsidies are abolished. However, because of the low economic growth in RC this abolishment of export subsidies result
Fig. 4. Annual growth of livestock production, 2000–2030. (GE ¼ Global Economy; GC ¼ Global Co-operation; CM ¼ Continental Markets; and RC ¼ Regional Communities).
in a decrease of the production of CAP products. Logically, the changes in production of non-CAP products (horticulture and pork and poultry) are less susceptible to changes in agriculture policies and follow the economic growth (higher income elasticity; see Table 1). For the EU-10 countries the impact of changes in border and domestic support is relatively limited in the liberalization scenarios compared to the impact in the EU-15. The policy effects are slightly positive for CAP commodities and slightly negative for non-CAP commodities due to EU accession effects. The reallocation from resources from non-CAP to CAP sectors causes the negative impact for non-CAP commodities. For the regionalization scenarios the impact of changes in border and domestic support is
ARTICLE IN PRESS 568
B. Eickhout et al. / Land Use Policy 24 (2007) 562–575
more important since benefits of accession to the EU15 are less eroded by liberalization. CAP payments are not reduced or even increased (RC scenario) and access of other countries to the EU is not enhanced. Agricultural trade World trade growth increases in all scenarios, especially in the liberalized scenarios (GE and GC). The higher
growth in trade mainly follows the trends of increases in income growth (see Table 1). Fig. 7 shows the contribution of changes in policies to the world trade growth of products. Reducing border support by eliminating import tariffs and export subsidies has an important impact on the growth of world trade in the GE and GC world. The contribution of abolishing of border support is highest in the most protected sectors (sugar, beef, grain and processed food). For the industry sector the contribution of eliminating border support is only 10% because the market has already been substantially liberalized in the former WTO rounds. In the CM scenario the contribution of border support is marginal because only the market between the US and the EU is liberalized. Reducing agricultural domestic support in the GE and GC scenarios has a minor negative impact on world trade because it leads to higher world prices and therefore less demand. The impact is only negative for grains, oilseeds and to a lesser extent for cattle because these are the sectors where domestic support is highest.
Fig. 5. Contribution of domestic and border support to yearly production growth of CAP-products and non-CAP products for the EU15.
Importance of agriculture in the economy
Fig. 6. Contribution of domestic and border support to yearly production growth of CAP-products and non-CAP products for the EU10.
Fig. 8 shows that the historic trend of a decreasing share of the agriculture-food complex in total economy continues in all scenarios. This trend is due to a low-income elasticity of food demand (i.e. people spend relatively less money on food as their income gets higher) and a low price elasticity of food demand (i.e. people do not buy more food when the price is decreasing because of the relative high productivity growth in the agricultural sector). Therefore, a faster economic growth leads to a lower share of the agriculture-food complex. For the EU-15 and the USA this implies that the share is lowest in the GE and CM scenarios, whereas the share is highest for the developing countries in the CM scenario. In the non-liberalizing scenarios CM and RC, the share of the agriculture-food complex only decreases slightly in Africa, indicating a
Fig. 7. Growth in trade of the different agricultural products between 2000 and 2030: The contribution of domestic support, border support and other factors (GE ¼ Global Economy; GC ¼ Global Co-operation; CM ¼ Continental Markets; and RC ¼ Regional Communities).
ARTICLE IN PRESS B. Eickhout et al. / Land Use Policy 24 (2007) 562–575
Fig. 8. Share of the agriculture-food complex in total GDP per world region (numbers for 2000 and 2030 for the four scenarios; GE ¼ Global Economy; GC ¼ Global Co-operation; CM ¼ Continental Markets; and RC ¼ Regional Communities).
Fig. 9. Yearly growth in real farm income in the EU-15 for CAP commodities, non-CAP commodities and the total economy. The contribution of changes in domestic and border support on farm income of CAP products is also visualized (GE ¼ Global Economy; GC ¼ Global Co-operation; CM ¼ Continental Markets; and RC ¼ Regional Communities).
remaining dependency on the agricultural sector. This dependency holds Africa vulnerable to external factors like climate change. Macro-economic growth has an important impact on real farm income of EU-15 countries: consumers have higher incomes and consumption patterns. Market-oriented scenarios (GE and CM) lead to highest real income growth of European farmers (Fig. 9). In the other two scenarios the farmer income growth is also lower due to preference shift in diet from animal to crop proteins. Reductions of border support and domestic support (in GE and GC) have a significant negative impact on real agricultural income of CAP commodities. Real farm income growth for these commodities is low compared to growth of rest of economy. A comparison of the impact of reducing domestic and border support on production (Fig. 5) and on real farm income (Fig. 9) in the liberalization scenarios clearly shows that border support has an impact on both production and farm income and
569
Fig. 10. Yearly growth in real farm income in the EU-10 for CAP commodities, non-CAP commodities and the total economy. The contribution of changes in domestic and border support on farm income of CAP products is also visualized (GE ¼ Global Economy; GC ¼ Global Co-operation; CM ¼ Continental Markets; and RC ¼ Regional Communities).
that domestic support has a limited impact on production and a substantial impact on real farm income. Our findings confirm the conclusions of OECD (2001) that border support is more distorting. However, our results also show European farmers are more susceptible to changes in economic growth and food demand than shifts in agricultural policies. Similarly to EU-15, overall economic growth has an important impact on real farm income in the EU-10 countries (Fig. 10). However, accession to the common EU market (reduction of border support) and implementation of domestic support schemes in EU-10 have a positive impact on real agricultural income for CAP commodities in these countries. The contribution is highest in the regionalization scenarios (GC and RC) where domestic support is sustained and not all countries get equal access to the EU-15 market (preferential access sustained). Change in land use and agricultural practices The changes in crop and livestock production and trade will not only impact the real farm income, but also the environment. A direct impact will be through changes in the size of arable land and pastureland.2 Indirectly, intensification or extensification of agricultural practices will also impact global biodiversity and the nitrogen cycle. In this section, we touch upon the consequences for the European land use. Land-use changes in EU-15 Most of the changes in land use are expected in the developing countries, since the economic growth and 2
See Van Meijl et al. (2006) for the description of simulated changes in feed conversion affecting the size of arable land and pastureland needed for livestock.
ARTICLE IN PRESS 570
B. Eickhout et al. / Land Use Policy 24 (2007) 562–575
consequently, the growth in food and feed demand is highest in these scenarios (see Figs. 3 and 4). Moreover, liberalization will have a major impact on exporting regions like South America, the USA and Australia. Focusing on the EU-15, the size of arable land decreases in the liberalizing scenarios (GE and GC in Fig. 11). Although not many changes in CAP occur in RC, the increase in crop productivity outruns the slow increase in crop production (Fig. 3) and therefore, the size of arable land also decreases in RC. Only the CM scenario shows a slight increase in arable land because of high production growth due to high economic growth, an unchanged CAP, an increase in high caloric food demand and protection of the transatlantic market keeping cheaper imports outside. Moreover, land productivity increase is moderate due to segmented markets, which temper technological progress. The negative impact of liberalization of agricultural policies on production of CAP products in the EU-15 countries (Fig. 5) is relatively small because of the high increase in total production volume, due to higher macroeconomic growth (Fig. 12). In the liberalizing scenarios the land productivity and land management are high because of optimistic assumptions on technology improvements (Eickhout et al., 2004) and an endogenous intensification that occurs since land prices increase relative faster than
other factors. The increase in land price for the farmer is partly caused by decoupling of area premiums in the MidTerm Review and the reduction of domestic support in the liberalization scenarios (Fig. 12). The decrease in arable land in RC is slightly hampered by an extensification in land management, because of the reduced pressure on land demand. These responses of farmers cause less drastic abandonment of arable land in Europe (Fig. 12). In all analyzed scenarios, more land is needed for livestock in EU-15, which means the land moves from crop to pasture for livestock production (Fig. 13). In the market scenarios (GE and CM) we observe intensification and in the regulated scenarios (GC and RC) extensification of production (see influence of yield management in Fig. 13). The increase of land use for livestock production is highest in the market scenarios (GE and CM) due to high production growth caused by high-income growth (see influence of output in Fig. 13). The high growth in demand for land puts an upward pressure on the price of land and leads to intensification (the yield trend and yield management together imply that less land is used to produce a certain production quantity; Fig. 13). In the regulated scenarios (GC and RC) land use growth is positive despite the low (GC) or negative (RC) production growth. Production growth is low due to low-income growth and diet changes. Land use is positive due to extensification, which is caused by decoupling of animal premiums. The reduction of domestic support for livestock (cattle) in the EU15 in the liberalization scenarios (GE and GC) leads to extensification and not to land abandonment. Land-use changes in EU-10 For the EU-10 countries, an increase in the size of arable land until 2010 in three of the four scenarios can be seen (Fig. 14). This mainly comes from the fact the accession countries will benefit from the EU CAP support measures for farmers in the first decade. The increased production will lead to an expansion of arable land in the first 10 years, after which CAP reform in the GE and GC scenario will force the EU-10 countries to intensify their agricultural
Fig. 11. Change in arable land in the EU-15 countries (in kha) between 2000 and 2030.
Fig. 12. Changes in factors influencing the size of arable land in EU-15 between 2000 and 2030.
Fig. 13. Change in pasture land and factors influencing this size of pasture land in the EU-15 countries (in kha) between 2000 and 2030.
ARTICLE IN PRESS B. Eickhout et al. / Land Use Policy 24 (2007) 562–575
571
Fig. 16. Change in pasture land and factors influencing this size of pasture land in the EU-10countries (in kha) between 2000 and 2030.
Consequences for the European environment Fig. 14. Change in arable land in the Eastern European countries (in kha) between 2000 and 2030.
Fig. 15. Changes in factors influencing the size of arable land in EU-10 between 2000 and 2030.
practices and consequently, the size of arable land decreases again. However, this decrease is fairly small compared to the decrease in the EU-15 countries, since the economic growth in the EU-10 countries is catching up, resulting in a high increase in production volume. In Fig. 15 the different factors influencing the size of arable land are depicted, showing the steep increase in crop production. A difference with the EU-15 is that the intensification process that takes place in the EU-15 due to decoupling and/or reduction of initially high area premiums is not present because these premiums do not exist in the EU-10 before accession (new premiums are assumed to be decoupled). In the liberalization scenarios the reduction of border support (mainly for non-CAP products; Fig. 6) dominates and it has a negative impact on production because EU-10 countries loose their preferential access to the EU-25 and other countries compete successfully on their home market. More land is also needed for livestock in EU-10 in all scenarios (Fig. 16). The increase in land use follows the pattern of production and income growth (GE4GC4CM4RC; see influence of output in Fig. 16). The increased demand for land due to higher production increases the price of land and leads to further intensification (see influence of yield management in Fig. 16).
Besides land-use change, climate change and eutrophication are considered the most important environmental threats to ecosystem services (MA, 2005). Future climate change strongly depends on emissions of CO2, CH4 and N2O, mainly emitted by the energy system. The energy emissions are taken from the CPB/RIVM-study ‘‘Four futures of Energy’’ (Bollen et al., 2004) in which GE shows the highest emissions and GC is the only scenario where climate policy is successfully implemented. Here, it is assumed that the greenhouse gas concentration will stabilize at 550 ppmv CO2-equivalents. This stabilization level has a good chance to coincide with the EU climate policy objective of a maximum temperature increase of 2 1C over its pre-industrial level (Bollen et al., 2004). In RC climate policy is implemented on a local scale through regional initiatives like implementation of wind and solar energy systems. In the timeframe of this study, benefits from climate change polices as implemented in Global Cooperation cannot be expected, although several studies indicate beneficial impacts of climate change policies for European ecosystems in the long-term (Bakkenes et al., 2006). Because of the low economic growth in the CM scenario, this scenario shows low emissions until 2030 (Bollen et al., 2004). However, the expected growth of population in CM will probably result in high temperatures by the end of the century. For European land use, the option of modern biofuels as energy carrier is an important competitor for land. In scenarios where climate policies are implemented on a European level (e.g. GC and RC) biofuels play a major role in using the land that is abandoned because of liberalization (GC) or small increase in food demand (RC). In the CM scenario modern biofuels can also play an important role in Europe because of the desire to be self-reliant on energy resources. However, since in CM no land abandonment takes place (Figs. 11 and 15), this scenario returns most of the competition for land in Europe, leading to conflicts of interests. Because of inertia in the climate system the consequences for the global-mean temperature change are very similar in the four scenarios (left panel of Fig. 17). The GC scenario even shows the highest temperature in the first decades. This result is related to the climate policies that are
ARTICLE IN PRESS B. Eickhout et al. / Land Use Policy 24 (2007) 562–575
572 °C
in °C per decade
1.60 1.40 1.20
0.40 GE CM GC RC
0.35 0.30
1.00
0.25
0.80
0.20
0.60
0.15
0.40
0.10
0.20
0.05
0.00 1970
GE CM GC RC
0.00 1980
1990
2000
2010
2020
2030
2000
2010
2020
2030
Fig. 17. Global-mean temperature change (left) and rate of temperature change (right).
implemented in the energy system: less coal not only decreases the CO2 emissions, but also the SO2 emissions. And since SO2 aerosols have an instant cooling effect compared to a long-lasting warming effect of CO2 concentrations, the decrease of SO2 particles increases the temperature immediately (right panel of Fig. 17). The effect of CO2 reductions is visible after 2030 (not shown). Results for GE show that this high-consumption scenario will lead to high temperature levels by 2030, having a major impact on the agricultural system through CO2 fertilization and changes in temperature level and precipitation. Effects of climate change policies are not effective by 2030. Another threat to future agriculture is the amount of nutrient that are depleted from the soils and the use of fertilizer that coincides with intensification (Cassman et al., 2003; Bouwman et al., 2005). Moreover, the air quality is heavily influenced by use of nutrients through emissions of NH3. Logically, in the GE scenario with highest animal production and large areas of agricultural land, the NH3 emissions are highest. In the GE scenario fast economic development causes a shift towards more protein-rich food consumption and higher human waste production. At the same time the N removal in wastewater treatment will be higher than in FAO Bruinsma (2003), because of higher technological improvements. The river N export from agricultural systems is based on the total N fertilizer use and animal manure production (Bouwman et al., 2005). N fertilizer use is assumed to be correlated with total crop production in dry matter, while animal manure production is assumed to be related to total livestock production in dry matter. In the environmentally oriented GC scenario meat consumption per capita is less than in the GE scenario, while wastewater treatment has a high policy priority for the prevention of eutrophication of surface water. Atmospheric deposition is much less than in the GE scenario, causing a reduction of the N loading in the coming three decades (see Fig. 18). Concluding, only the GC scenario shows a decrease in global river N export compared to the current situation, whereas the GE scenario returns an increase in global river
Fig. 18. Changes in global N pressure on the land use system for each scenario between 2000 and 2030. GE is top left, CM is bottom left, GC is top right and RC is bottom right.
N export, especially in the regions where animal production is expected to increase (China, India and Latin America). Regionally, these increases in river N export will have major impacts on the quality of soil and nature, especially in regions where the increases will occur. Bouwman et al. (2005) showed the regions that will experience the most increases are China, India and some regions in Latin America. These results indicate that the pressure on agricultural systems is about to increase in the regions that are expected to become more vulnerable because of land expansion. From the results in N export (Fig. 18), it is clear that the increase in animal production in GE and CM (and therefore the increase in crops for feed use) will cause large pressures on the land system, especially in the arid and semi-arid tropics and subtropics where the lands experience a high risk of land degradation due to soil erosion, soil nutrient depletion and overgrazing (Sere´ and Steinfeld, 1996). It is uncertain how important the role of shifting cultivation is in maintaining the soil productivity in these regions. Nevertheless, this global context has to be taken into account when the different scenarios are judged for Europe.
ARTICLE IN PRESS B. Eickhout et al. / Land Use Policy 24 (2007) 562–575
Discussion and conclusions In this paper, we presented indicators covering economic aspects of trade liberalization and environmental issues that will influence future agriculture. This combination of economic and environmental aspects is crucial when conclusions on the benefits of trade liberalization need to be drawn. Especially, the environmental consequences of trade liberalization are rarely investigated, although these consequences can have adverse effects on the sustained productivity of land through climate impacts and soil depletion. Eickhout et al. (2004) concluded that the effect of trade liberalization will lead to less agricultural pressure within Europe, while major food exporting regions like South America will experience an additional need for agricultural land on top of default increases as a consequence of population growth. Additional analyses on the consequences of nutrient use and climate change are still needed to fully understand the impact of trade liberalization on the environment. Our analysis shows that although the economic growth in the liberalizing scenarios (GE and GC) is the highest globally, environmental impacts may offset these beneficiary results. Especially in Global Economy, where environmental concerns are not considered, the impacts of climate change and eutrophication increase, probably affecting the sustainability of ecosystem services like food production (see also MA, 2005). Trade liberalization has a large influence on the growth in trade of current protected products like sugar, beef and processed food. This growth in trade will lead to a world that is more and more interrelated, resulting in dependencies which may not be preferred by all people and could lead to societal discontent. However, in a world where self-reliance is cherished (i.e. the CM scenario) an interesting conflict of interests might occur: the formation of powerful trade blocks might lead to an increasing poverty gap between developed and developing countries, which in our analysis led to an increase in trade due to very low wages of farmers from developing countries. An increase in import tariffs would be needed to prevent this competition and therefore, an increasing budget on agricultural subsidies. For European land use, trade liberalization does not lead to massive abandonment of agricultural land. On the one hand, global growth in food and feed demand will remain a factor of importance for the European farmer, offering opportunities of new regions to export to. On the other hand, new environmental challenges, especially because of climate policies resulting in an increasing demand for biofuels and carbon plantations, can lead to new options for the European farmer. However, on a sectoral level, farmers can experience major impacts of trade liberalization, especially in sectors that are expected to collapse because of global competition (i.e. the sugar sector). Moreover, current EU policies shifting support from coupled income support to decoupled income support already resulted in a lower vulnerability of European
573
agricultural production to additional liberalizing policies. However, liberalization will have a major negative impact on real farm income. Nevertheless, our analysis shows that studies on the effects of trade liberalization for Europe cannot be performed properly without taking the global context (economic circumstances, size of population, global food and feed demand) into consideration. Considering the sustainability of environmental services, some threats are identified in this paper, i.e. climate change and nutrient loading. Interestingly, these environmental threats do not seem to impact the European agriculture, especially since biodiversity in Europe is already at a very low level (Bakkenes et al., 2006). Here, the global context of European studies becomes essential. By looking at environmental changes, especially climate change and soil erosion/depletion are the major threats to agriculture in arid and semi-arid tropical regions. Unfortunately, these regions are expected to experience most of the increased pressure on land, even without these environmental feedbacks. This environmental concern is insufficiently reflected in current WTO trade discussions and in scientific studies. We showed that a globalizing world without any environmental focus (GE) experiences severe environmental burden (i.e. climate and nutrients), which may endanger the economic benefits of trade liberalization. Therefore, we conclude that environmental and trade agreements and policies must be sufficiently integrated or coordinated, to assure that they work together to improve the environment and attain the benefits of free trade. In our analysis we did not consider water shortage as another environmental threat to sustained agricultural productivity. This factor needs to be included in further integrated analyses. Especially, the price of possible scarce inputs like water and sustainable agricultural practices are insufficiently included in current models. Another question is whether developing countries and the poor in particular are able to reap the ex ante benefits calculated by the model studies. Imperfect markets and institutions (credit, human capital), instable political climate, a bad infrastructure and enhanced quality standards by Western countries may lead to lower benefits. Therefore complementary investments (e.g. infrastructure, education) are needed to reap the potential benefits. Concluding, our analysis shows that to learn lessons on the future of European land use, not only changes in desired production need to be taken into account, but also changes in land productivity (which may offset the increase in production), changes in trade regimes (and therefore the global context) and changes in land management (intensification versus extensification) as a response to increase or decrease in land demand. The total sum of these mechanisms may lead to a coherent understanding of Europe’s land use. Here, we have tried to deal with these mechanisms in one modeling framework, taking future uncertainties in desired production and trade regimes into account by using four contrasting scenarios. Future work will be focused on improving our current modeling
ARTICLE IN PRESS 574
B. Eickhout et al. / Land Use Policy 24 (2007) 562–575
framework by dealing with other land use options, which may conflict with food production (e.g. biofuels, forestry, urban sprawl, recreation), by improving our understanding of options in land management (further intensification) and by expanding our knowledge on the impacts of environmental circumstances affecting crop yields (e.g. soil depletion, desertification, eutrophication, water impacts).
Acknowledgments The work described in this paper formed part of the EURURALIS project, a Dutch collaboratiung effort of the Netherlands Environmental Assessment Agency (MNP), the Agricultural Economics Research Institute (LEI), the Laboratory of Soil Science and Geology from Wageningen University and Alterra Wageningen. Thanks are due to Lex Bouwman, Tom Kram and Henk Westhoek (MNP) for advising on the set-up of the study and especially to Kasper Kok (WUR) for providing extensive comments, which led to major improvements of this paper. Thanks are also due to two anonymous reviewers for improving this paper.
References Alcamo, J., Leemans, R., Kreileman, G.J.J., 1998. Global Change Scenarios of the 21st Century. Results from the IMAGE 2.1 Model. Pergamon and Elsevier Science, London. Anderson, K., 1999. Agriculture, developing countries and the WTO millennium round, CIES, Discussion paper 99/28. Bakkenes, M., Eickhout, B., Alkemade, J.R.M., 2006. Ecosystem impacts for biodiversity of different stabilization scenarios. Global Environmental Change 16, 19–28. Bollen, J.C., Manders, T., Mulder, M., 2004. Four futures for energy markets and climate change. CPB Special Publication 52, ISBN 905833-171-7, Netherlands Bureau for Economic Policy Analysis, The Hague, and National Institute for Public Health and the Environment, Bilthoven, the Netherlands. See: http://www.mnp.nl/en. Bouwman, A.F., Van Drecht, G., Van der Hoek, K.W., 2006. Nitrogen surface balances in intensive agricultural production systems in different world regions for the period 1970–2030. Pedosphere 15 (2), 137–155. Bruinsma, 2003. World Agriculture: Towards 2015/2030. An FAO Perspective. Food and Agriculture Organization, Rome, Italy. Cassman, K.G., Dobermann, A., Walters, D.T., Yang, H., 2003. Meeting cereal demand while protecting natural resources and improving environmental quality. Annual Review of Environmental Resources 28, 315–358. Castles, I., Henderson, D., 2003a. The IPCC Emission Scenarios: an Economic-Statistical Critique. Energy and Environment 14 (2–3), 159–185. Castles, I., Henderson, D., 2003b. Economics, Emissions Scenarios and the Work of the IPCC. Energy and Environment 14 (4), 415–435. CPB, 2003. Four Futures of Europe. Netherlands Bureau for Economic Policy Analysis, the Hague, the Netherlands. See: http://www.cpb.nl Eickhout, B., Van Meijl, H., Tabeau, A., Van Zeijts, H., 2004. Between liberalization and protection: four long-term scenarios for trade, poverty and the environment. Presented at the seventh annual conference on global economic analysis, June 2004, Washington, USA. Eickhout, B., Bouwman, A.F., Van Zeijts, H., 2006. The role of nitrogen in world food production and environmental sustainability. Agriculture, Ecosystems and Environment 116, 4–14.
Fischer, G., Shah, M., Van Velthuizen, H., 2002. Climate Change and Agricultural Vulnerability. International Institute of Applied Systems Analysis, Vienna, Austria. Francois, J., Van Meijl, H., Van Tongeren, F., 2005. Trade liberalization and developing countries under the Doha round. Economic Policy 20–42, 349–391. GEO3, 2002. Global Environment Outlook 3. United Nations Environment Programme (UNEP), Nairobi, Kenya. Gru¨bler, A., Nakic´enovic´, N., Alcamo, J., Davis, G., Fenhann, J., Hare, B., Mori, S., Pepper, B., Pitcher, H., Riahi, K., Rogner, H., La Rovere, E.L., Sankovski, A., Schlesinger, M., Shukla, R.P., Swart, R., Victor, N., Jung, T.Y., 2004. Emission scenarios: a final response. Energy and Environment 15 (1), 11–24. Hertel, T., 1997. Global Trade Analysis. Modelling and Applications. Cambridge University Press. Hertel, T., Anderson, K., Hoekman, B., Francois, J.F., Martin, W., 1999. Agriculture and nonagricultural liberalization in the millennium round, paper presented at the Agriculture and New Trade Agenda, October 1–2, 1999, Geneva, Switzerland. Huang, H., Van Tongeren, F., Dewbre, F., Van Meijl, H., 2004. A new representation of agricultural production technology in GTAP. Paper presented at the seventh annual conference on global economic analysis, June, Washington, USA. IMAGE Team, 2001. The IMAGE 2.2 Implementation of the SRES Scenarios. A Comprehensive Analysis of Emissions, Climate Change and Impacts in the 21st century. RIVM CD-ROM publication 481508018, National Institute for Public Health and the Environment, Bilthoven, the Netherlands. Klijn, J.A., Vullings, L.A.E., Van den Berg, M., Van Meijl, H., Van Lammeren, R., Van Rheenen, T., Tabeau, A.A., Veldkamp, A., Verburg, P.H., Westhoek, H., Eickhout, B., 2005. The EURURALIS study: technical document. Alterra Report nr. 1196, Alterra Wageningen, the Netherlands, 215pp. Leemans, R., Eickhout, B., 2004. Another reason for concern: regional and global impacts on ecosystems for different levels of climate change. Global Environmental Change 14, 219–228. Leemans, R., Eickhout, B., Strengers, B., Bouwman, L., Schaeffer, M., 2002. The consequences of uncertainties in land use, climate and vegetation responses on the terrestrial carbon. Science in China, Series C 45 (Supp.), 126. MA, 2005. Biodiversity Synthesis Report. Millennium Ecosystem Assessment Report. Island Press, Washington, DC, USA. MNP, 2005. Quality and the future. Sustainability outlook. Summary. Netherlands Environmental Assessment Agency, Bilthoven, the Netherlands. See also: http://www.mnp.nl/en Nakicenovic, N., Alcamo, J., Davis, G., De Vries, B., Fenhann, J., Gaffin, S., Gregory, K., Gru¨bler, A., Jung, T.Y., Kram, T., La Rovere, E.L., Michaelis, L., Mori, S., Morita, T., Pepper, W., Pitcher, H., Price, L., Riahi, K., Roerhl, A., Rogner, H.-H., Sankovski, A., Schlesinger, M., Shukla, P., Smith, S., Swart, R., Van Rooijen, S., Victor, N., Dadi, Z., 2000. Special Report on Emission Scenarios. Intergovernmental Panel on Climate Change (IPCC), Cambridge University Press, Cambridge, UK. Nakicenovic, N., Gru¨bler, A., Gaffin, S., Jung, T.T., Kram, T., Morita, T., Pitcher, H., Riahi, K., Schlesinger, M., Shukla, P.R., Van Vuuren, D., Davis, G., Michaelis, L., Swart, R., Victor, N., 2003. IPCC SRES revisited: a response. Energy and Environment 14 (2–3), 187–214. OECD, 2001. Market EFFECTS of Crop Support Measures. Organization for Economic Co–operation and Development, Paris. OECD, 2003. Agricultural Policies in OECD Countries 2000. Monitoring and Evaluation. Organization for Economic Co–operation and Development, Paris. Parry, M., Arnell, N., McMichael, T., Nicholls, R., Martens, P., Kovats, S., Livermore, M., Rosenzweig, C., Iglesias, A., Fischer, G., 2001. Millions at risk: defining critical climate change threats and targets. Global Environment Change 11 (3), 1–3. Rosegrant, M.W., Cai, X., Cline, S.A., 2002. World Water and Food to 2025: Dealing with Scarcity. International Food Policy Research Institute (IFPRI), Washington, DC, USA (pp. 338).
ARTICLE IN PRESS B. Eickhout et al. / Land Use Policy 24 (2007) 562–575 Rosenzweig, C., Parry, M., Fischer, G., 1995. World Food Supply. In: Strzepek, K.M., Smith, J.B. (Eds.), As Climate Changes: International Impacts and Implications. Cambridge University Press, UK (pp. 27–56). Rounsevell, M.D.A., Reginster, I., Ara´ujo, M.B., Carter, T.R., Ewert, F., House, J.I., Kankaanpa¨a¨, S., Leemans, R., Metzger, M.J., 2006. A coherent set of future land use change scenarios for Europe. Agriculture, Ecosystems and Environment 114, 57–68. Sere´, C., Steinfeld, H., 1996. World Livestock Production Systems. Current Status, Issues and Trends. Animal Production and Health Paper 127. Food and Agriculture Organization of the United Nations, Rome. Strengers, B.J., 2001. The Agricultural Economy Model in IMAGE 2.2. RIVM Report no. 481508015. National Institute for Public Health and the Environment, Bilthoven, the Netherlands. See: http://www.mnp.nl/ image. Strengers, B., Leemans, R., Eickhout, B., De Vries, B., Bouwman, A.F., 2004. The land use projections in the IPCC SRES scenarios as simulated by the IMAGE 2.2 model. Geojournal 61, 381–393.
575
UN, 2002. World Population Prospects: 2002. Department for Economic and Social Information and Policy Analysis, United Nations. Van Meijl, H., Van Tongeren, F.W., 2002. The Agenda 2000 CAP reform, world prices and GATT-WTO export constraints. European Review of Agricultural Economics 29-4, 445–470. Van Meijl, H., Van Rheenen, T., Tabeau, A., Eickhout, B., 2006. The impact of different policy environments on land use in Europe. Agriculture, Ecosystems and Environment 114, 21–38. Verburg, P.H., Schulp, C.J.E., Witte, N., Veldkamp, A., 2006. Downscaling of land use change scenarios to assess the dynamics of European landscapes. Agriculture, Ecosystems and Environment 114, 39–56. Westhoek, H.J., Van den Berg, M., Bakkes, J.A., 2006. Scenario development to explore the future of Europe’s rural areas. Agriculture, Ecosystems and Environment 114, 7–20. World Bank, 2003. Global Economic Prospects 2004. The World Bank, Washington DC. WTO, 2003. Understanding the WTO. World Trade Organization, Geneva, Switzerland. Downloaded from http://www.wto.org/english/ thewto_e/whatis_e/tif_e/understanding_e.pdf.
