Bio-based and Applied Economics 6(1): 37-56, 2017 ISSN 2280-6180 (print) © Firenze University Press ISSN 2280-6172 (online) www.fupress.com/bae Full Research Article DOI: 10.13128/BAE-16372 What if meat consumption would decrease more than expected in the high-income countries? Fabien Santini*, tevecia Ronzon, ignacio PeRez Dominguez, SeRgio Rene aRaujo enciSo, ilaRia PRoietti European Commission, Joint Research Centre, IPTS, Seville, Spain Date of submission: 2015 30th, June; accepted 2017 2nd, January Abstract. Changes in meat consumption patterns could induce significant adjust- ments in agricultural markets. In this paper alternative scenarios envisaging lower meat consumption over the coming decade in high income countries and some select- ed emerging economies have been tested, with or without compensation by other sources of proteins. From a European perspective, results show a livestock farming sector having to deal with contradictory market signals. On the one side, the reduc- tion in feedstuffs prices is an incentive to produce more, with lower output prices affecting positively the trade balance with developing countries, where demand keeps increasing. However, on the other side, the lower domestic demand for meat would affect profitability of meat production in the EU. Overall, the European beef meat sec- tor would be the most affected, with some higher demand for dairy products. This possible evolution of European diets is a challenge for European livestock farmers, which will be required to adapt their production mix and rely on the portfolio of poli- cies the CAP offers. Keywords. Meat consumption, agricultural markets, agro-economic modelling, agri- cultural commodities JEL codes. Q13, Q17, Q18 1. Introduction The nutrition transition worldwide has been extensively described since Popkin’s sem- inal article (Popkin, 1994). Associated with income growth and higher food availability, predominantly starchy diets firstly diversify with the integration of fruits and vegetables as well as animal proteins. It then ensures a new stage of the nutrition transition character- ised by increased per capita consumption of fat (particularly animal fat), refined carbohy- drates and sugar. This stage of the nutrition transition has been first observed in western countries in the first half of the 20th century. It then occurred in low- and middle-income *Corresponding author: fabien.santini@ec.europa.eu 38 F. Santini et al. countries from the early 1980’s onwards, and by then it is often referred to as a “west- ernisation of diets”. Here, the adoption of a “western” dietary pattern is mainly an urban phenomenon associated with the adoption of sedentary lifestyles, although it is increasing also in rural areas. As a consequence, many regions and countries, both in the developing and developed world, are experiencing a lower dietary intake of legumes, vegetables and coarse grains and a higher intake of refined carbohydrates, added sugars, fats, processed foods and animal-source foods (Popkin et al., 2012). Sign of economic development and opening markets, it has also been subject to early warnings with regards to its corollary health effects: development of overweight, obesity, diabetes and coronary heart diseases (Popkin, 1994; Popkin, 1999; Tsolekile, 2007; Schmidhuber, 2004). In response, “health conscious behavioural changes” emerged, especially among better educated groups of population who tend to substitute part of their animal-based consump- tion by plant-based products (Popkin, 1994). Ethical concerns in relation to animal wel- fare in farming/marketing practices or the growing uptake of philosophical and/or religious conceptions recognising a status to animals were other reasons for a growing number of people to exclude meat in their diet. Moreover, especially with the release of the FAO live- stock‘s long shadow report (Steinfeld et al., 2006) and with the spread of life-cycle assess- ments (LCA) in the early 21st century, the adverse effects of intensive meat production systems - in terms of land use, greenhouse gas emissions and biodiversity losses – were pointed out in environmental assessments, determining the scaling up of the vegetarianism linked to concerns with the environmental and ecological impact (Ruby, 2012). Concerning climate change, livestock are a major source of greenhouse gas (GHG) emissions within the agricultural sector, mainly carbon dioxide, nitrous oxide and in particular methane. Meth- ane is 25 times more powerful than carbon dioxide and is estimated to be responsible for approximately one-fifth of man-made global warming (Reay et al., 2010). On a global scale, livestock contributes about 15% of the total global anthropogenic greenhouse gas emissions (Steinfeld et al., 2006) and ruminants, in particular, represent more than 80% of total GHG emissions related to livestock (Herrero et al., 2013; Hristov et al., 2013). In the wake of such concerns, medias and NGOs conveyed the health, environmen- tal and ethical benefits of lowering individual meat consumption including ‘Meatless day campaigns’, pro-vegetarianism and pro-flexitarianism communications (Laestadius et al., 2013). The rationale of the meatless day campaigns was to rely on responsible citizen to achieve mass effects from small individual changes. In other words, cutting out meat only one day a week equals to a 14% reduction in meat consumption, which can be highly sig- nificant in terms of health, environmental and animal welfare impacts. These campaigns also claimed for governmental support to educate and inform the targeted “responsible citizens” (Gold, 2004). But, this strategy has been later judged at risks in the sense that “a simple ‘eat less meat’ message alienates some people and could have unintended conse- quences, not least on farmers’ livelihoods” (Sutton and Dibb, 2013). That led to the emer- gence of ‘less-but-better’ recommendations still oriented towards a reduction of individ- ual meat consumption but jointly with purchases favouring meat proceeding from more extensive farming systems and/or of higher quality (de Boer et al., 2014; Sutton and Dibb, 2013; National Food Administration, 2009). The impact of current western diets also moved up in the policy agenda. This was motivated by a context of implementation of sustainability principles, of fixing green- 39Meat markets and a stronger consumption decrease in high-income countries house gas reduction targets, and of growing costs for public health systems. In 2009, the government of Sweden was the first government to officially recommend to lower meat consumption as one ‘environmentally effective food choice’ among others and to notify such a proposal to the European Union (EU)1 (National Food Administration, 2009). In this case, the health perspective argument was not put forward but reducing meat con- sumption appeared as a win-win option for both human health and the environment in the recommendations to government of the Sustainable Development Commission of the United Kingdom (2009) as well as the ones of the Health Council of the Netherlands (2011) and more recently the recommendations of the Dietary Guidelines Advisory Group (GDAP) of USA (2015) (U.S. Department of Agriculture and U.S. Department of Health and Human Services, 2015; Sustainable Development Commission, 2009; Health Council of the Netherlands, 2011). All these activities resulted in the rise of people’s awareness and in the growth of veg- etarian population over time. Nevertheless, data on the development of vegetarianism and / or flexitarianism behaviour is scarce. First because the definition of vegetarianism is not stabilised, giving space to a co-existence of various typologies of vegetarians in the sci- entific literature (Phillips, 2005; de Bakker and Dagevos, 2012; Ruby, 2012). Vegetarian types range from flexitarians (still eating meat but reducing their meat consumption) to vegans (not eating any product of animal origin). In between, a large diversity of vegeta- ble-based diets can be found: lacto-, lacto-ovo-vegetarianism, macrobiotic, and pescatar- ians (or pesco-vegetarians) according to the variety of personal motivations for adopting a lower meat or meat-free diet. Few surveys have been conducted in the United States in particular by the Vegetarian Research Group, which commissions a yearly poll, showing that 4 to 5% of the adult population can be defined as vegetarian and that up to 15% of the population do not eat meat at more than half of the meals (Stahler, 2011; Casalena, 2011). Similarly, the European Vegetarian Union compiled in 2008 estimations on the number of vegetarians in developed countries. The share of vegetarians among European countries ranged between less than 1% in Poland and Portugal and close to 10% in Ger- many (Pichler and Blackwell, 2008). The same source estimated 3% vegetarians in Aus- tralia, 3.2% in USA and 4% in Canada. Finally, empirical observations like the increased numbers of vegetarians in western societies (Gossard and York, 2003) and in particular growing flexitarianism (Dagevos, 2014; de Boer et al., 2014; Friends of the earth, 2014) motivated the analysis of two simplified alternative protein consumption scenarios, mainly resulting in a stronger reduction of per capita meat consumption in developed countries, compared to the baseline projections of the European Commission (European Commis- sion, 2014). In the first section, the quantitative framework and the market projections resulting are described and in the second section the alternative scenarios implemented in order to capture the nutritional pattern changes anticipated are analysed. The model ouputs for the main agricultural markets concerned by the alternative scenarios are discussed in the third section and finally, in the fourth section, a reflection on policy implications and on the limitations of the present work is presented 1 Proposal notified to the EU 15.05.2009. 40 F. Santini et al. 2. Methodology and modelling approach Annually, the OECD and the FAO jointly release a ten-year horizon assessment of medium-term projections of national, regional and global agriculture commodity markets (OECD/FAO, 2015a). The baseline scenario is taken from the European Commission con- tribution to the latter published as “prospects for EU agricultural markets and income” (European Commission, 2014) which is produced within the Aglink-Cosimo modelling framework. The modelling framework ensures that the overall set of equations balances with plausible outcomes. Aglink-Cosimo is a global economic recursive-dynamic, partial equilibrium, supply demand modelling framework which covers the main agricultural commodities (Arau- jo Enciso e. al., 2015; OECD, 2015). The model is a collaborative work integrating the OECD’s Aglink and FAO’s Cosimo sub-modules. It is used to simulate the developments of annual supply, demand and prices for the main agricultural commodities produced, consumed and traded worldwide. The Aglink-Cosimo model covers 44 individual coun- tries and 12 regions, 93 commodities and 40 world market clearing prices with a total of around 36000 equations. Most behavioural equations in Aglink-Cosimo can be linearised in logarythms (i.e. “double-log” functions), including those for estimating production and demand functions, where the underlying relationship between y and x resembles a logarithmic function (e.g. y experiences diminishing marginal returns with respect to increases in x): ln yi( )=ai +ξij i ln xij( )+γ i it+ ln ei( ) where i and j correspond to the agricultural commodities covered in the model. The rela- tionship between x and y is parameterized through the introduction of a constant term αi a slope term ξij which corresponds to the elasticity between y and x (Araujo Enciso et al., 2015) and a term trend (γi). The residual is captured by the error term (ei) which is fre- quently referred to as ‘calibration term’. For the purpose of this paper, we focus our attention on the demand side of the mod- el. Total consumption is modelled as an aggregate of different uses including food, feed, biofuels and industrial uses. For meat, consumption only covers food use (FO), and is modelled as a function of the relative ratio consumer prices (CPc,r,t) for each commodity and the total consumer price index (CPIr,t) the GDP index (GDPIr,t) and the population (POPr,t) with a constant, a trend and an error term. For any region and time period, the food demand equation can be expressed as: ln FOc( ) =α + c ∑ξFOc ,CPci i ln CPc CPI ⎛ ⎝ ⎜⎜ ⎞ ⎠ ⎟⎟+ξFOc ,DGPI i ln GDPI POP ⎛ ⎝ ⎜⎜ ⎞ ⎠ ⎟⎟+ ln POP( )+γ c i t + ln ec( ) (1) The trend serves to depict changes in the consumption patterns; the population serves to model changes due to demographic changes and the GDPI the income effect on food consumption. 41Meat markets and a stronger consumption decrease in high-income countries 2.1 Reference scenario The reference scenario (REF) is taken from the European Commission “prospects for EU agricultural markets and income” (European Commission, 2014). Such a scenario is the so-called baseline and it is the result from an interactive process between market analysts, policy analysts and modellers. The reference scenario reflects the expert knowl- edge on the possible developments of each agricultural commodity market in view of the recent trends and anticipated developments. Noteworthy, it is built upon a certain num- ber of exogenous assumptions deemed most plausible at the time of the analysis concern- ing macro-economic and energy conditions, agricultural and trade policy arrangements in force, as well as yield trends under ‘normal’ climatic conditions. For the meat markets, the assumptions of REF reflect a stabilisation or lower growth of per capita meat consumption in the richest countries (a limited growth in meat con- sumption is exclusively attributable to a continuing increase of consumption of poultry meat) and, on the other hand, a further development of the nutrition transition in devel- oping countries (i.e. an upward trend in sweeteners and animal-based products con- sumption). These assumptions are reflected by the trends in the model for different meat consumption and selected countries. Accordingly, in terms of total meat consumption, the ‘baseline’ projections foresee some relative stability in developed countries, while the total consumption per capita is projected to keep on increasing in the ten coming years in developing countries (Figure 1). As a result, domestic EU meat demand is projected to slightly decrease while exports are expected to increase, mainly due to the population and economic growth in low- income countries and driven by a sustained growth in poultry meat. The medium term outlook of EU and world prices is accordingly rather positive, in particular for EU poultry and pig meat. The gap between EU and world prices is projected to significantly reduce over the projection period, as shown in Figure 2 comparing the level of prices for three periods 2005-07, 2012-14 and 2022-24. The use of such partial equilibrium model for a ten years ahead projections of a change in consumption patterns does not take into account possible changes in price and income elasticities over the period. To this respect, such a tool represents a simplification of what is likely to happen in the event of lower meat consumption. In addition, such a model consid- ers each meat product as one single commodity, while the reality is likely to be character- ised by segmentation into different types of meat products, each of them reacting different- ly to the shock. Organic or ethically claimed products are going to follow another path of development and this could nuance the results. In addition, each country is considered as a whole, while consumption preferences will depend on regional and/or population (gender, demography) characteristics. Lastly, the scenario by essence does not take into account the possibility of external unforeseen events (e.g. an animal health related problem). 2.2 Low meat consumption scenarios In order to assess the impact of lower meat consumption on agricultural commodity markets, two “lower meat consumption scenarios” are designed: LOWMEAT1 and LOW- MEAT2 (see Table 1). 42 F. Santini et al. Figure 1. Baseline total meat consumption (kg per capita). 0 10 20 30 40 50 60 70 80 90 100 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 20 20 20 21 20 22 20 23 20 24 North America Oceania Developed European Union-28 Latin America and Caribbean China Japan Asia Africa Source: OECD-FAO world outlook, 2015a. Figure 2. World and EU producer meat price development indices (base 100 = 2005-2007) in the base- line (EU in EUR, World in USD). 100 110 120 130 140 150 160 170 180 190 EU World EU World EU World 2012-14 2022-24 Beef Pig meat Poultry Source: own elaboration from European Commission (2014). 43Meat markets and a stronger consumption decrease in high-income countries Table 1. Scenario characteristics. Scenarios LOWMEAT1 LOWMEAT2 Common elements Linear decrease of meat consumption for : - all meat types (i.e. beef and veal, sheep, pig, poultry) meat consumption (food use) in the EU, USA, Canada, Australia and New Zealand by -11% over ten years and, - certain (beef and veal, pig) meat consumption (food use) in Argentina, Brazil, Mexico and Uruguay by -5% over ten years Distinctive elements No compensation with non-meat proteins Linear increase of non-meat protein sources human consumption (food use) in the EU, USA, Canada, Australia and New Zealand - Wheat, other cereals, fresh dairy products and butter, cheese, eggs: +5% over the whole period - Oilseeds and pulses: +2% over the whole period Reduced (by half) linear increase of the same products in Argentina, Brazil, Mexico and Uruguay. In line with the nutrition transition theory, meat consumption projections are based on steady path, which is depicted with the trend. The alternative scenarios pre- sented in this study are precisely based on a change of consumer preferences towards lower meat consumption. These changes are assumed to happen in high income coun- tries, and in developing countries where the protein intake is higher than the “safe lev- el of protein intake”2 defined by WHO/FAO/UNU (2007). The high income countries considered are the United States, Canada, Australia, New Zealand and the European Union, altogether representing 29.1% of the total world meat consumption (2011-13). Given the high levels of protein intake in certain Latin American countries and the rising concerns on health in particular and other issues, the scenarios are also imple- mented to three MERCOSUR partners where meat consumption is the highest per cap- ita (Argentina, Brazil, Uruguay) and Mexico as North American OECD country only for specific types of meat (bovine and pig meat), poultry being considered to keep on following the reference scenario. Such countries represent an additional 10.5% of the total meat consumption. Various dietary guidelines for high-income countries confirm that in general the average per capita level of meat consumption is high enough to rec- ommend lower meat consumption without putting people at risks of nutrient deficien- cies (National Food Administration, 2009; Health Council of the Netherlands, 2011). In the rest of the world, it is assumed that these trends are not yet likely to express themselves within the time horizon chosen (ten years ahead) and therefore the baseline trends are not changed in the scenarios. Per capita meat consumption, especially poul- try, would continue to increase in these countries as shown in Figure 1. 2 The safe level for a population is defined as the average protein requirement of the individuals in the popula- tion, plus twice the standard deviation (SD) (WHO/FAO/UNU 2007). 44 F. Santini et al. Concerning the magnitude of the meat consumption decrease in the two scenarios, the basic assumption reflects a possible doubling of vegetarians and flexitarians over the 10 coming years in high income societies. Concretely, estimating that around 3% of the population of high-income countries is vegetarian in 2014 and does not consume any meat products, we assume that this share would double (6%) by 2024 in our alternative scenarios. Concerning flexitarians, we assume the conservative assumption that 15% of the population of these countries eat 50% of the average per capita meat consumption and that this share would double by 2024. Under these assumptions, the total meat consump- tion per capita decreases by 11% by 2024 relative to the baseline in the EU, United States, Canada, Europe and Oceania. Concerning the three MERCOSUR countries and Mexico, information on the devel- opment of vegetarianism and flexitarianism is even scarcer than for high income coun- tries, which tends to confirm that these trends are less developed for a series of social, economic, ethical and political reasons. Therefore, we assume a lower total meat con- sumption decline in these countries than in the other countries included in the scenario. Per capita meat consumption is reduced by only 5% by 2024 relative to the baseline in these countries, and in addition, the decrease is only applied to beef and veal meat as well as pig meat (see Table 1). In both scenarios, meat consumption reduction is uneven among types of meat. As the baseline foresees a stronger poultry meat consumption trend than for other types of meat (+5% above the average 2005-07 in the EU, see Figure 3), these assumptions corre- spond to a stability / slight increase of poultry meat consumption in developed countries over the projected period. On the contrary, the baseline scenario implies strong reduction in pig meat and beef meat per capita consumption. Therefore the two scenarios result in an even stronger decrease of consumption of those types of meat. In the case of the EU, the 2024 level of consumption is close to 20% below the average 2005-07 for pig meat and to 25% for beef and veal meat. Over the period 2002-11, 32% of the protein intake comes from meat and derived products in the high income countries (see Figure 4). Population turning to vegetarian- ism or reducing its consumption of meat would see their intake in protein reduced, unless different sources of protein intake are envisaged. This significant source of proteins is like- ly to be compensated by an increase of consumption of other sources of proteins. In the LOWMEAT2 scenario, a partial compensation of protein intake losses by other sources is envisaged, corresponding to an increased food consumption of cereals, eggs, dairy prod- ucts and oilseeds-pulses. In total, these four groups of products represent 53% of the total protein intake in the countries concerned, therefore the increased protein intake from these products should be of 6.7% to fully compensate the 11% decrease of meat consump- tion. Fish, fruit and vegetables intake, although they represent around 15% of the total protein intake of high-income countries, were not considered because not covered by the modelling tool used. Given the fact that some commodities are richer in weight (oilseeds and pulses in particular) than others in protein and that the scenario only aims to repre- sent a partial compensation (being reasonable to assume that in countries where the daily protein intake is well above recommendations, a full compensation will not occur), the assumption tested is of a 5% increase in cereals, dairy and eggs food use over the period and of 2% for oilseeds and pulses food use (see Table 1). 45Meat markets and a stronger consumption decrease in high-income countries Figure 3. Scenarios LOWMEAT1 and LOWMEAT2 (in bold) and baseline (in thin) – Meat consumption per capita index (EU-28), Base 100 = 2005-2007. 70 80 90 100 110 120 130 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19 20 20 20 21 20 22 20 23 20 24 Poultry Pig meat Beef and veal Thin lines are for the baseline and bold lines for the LOWMEAT1 and LOWMEAT2 scenarios. As the two scenarios are equal in terms of meat consumption assumptions (see table 1), both scenarios are repre- sented by the same bold line. Source: own calculations from Aglink-Cosimo database. Figure 4. Average protein intake in high income countries (% of total protein intake, 2002-11). 0% 5% 10% 15% 20% 25% 30% 35% Meat Milk Cereals Oilseeds and pulses Eggs Fish Other Source : Author’s calculations from FAO Food Balance Sheets. 46 F. Santini et al. Both scenarios LOWMEAT1 and LOWMEAT2 do not claim to fully reflect plausible pathways in their complexity but were designed to mainly capture simplified elements of what could imply for commodity markets the decrease of meat consumption and their partial compensation by consumption of other protein sources, as well as the contribution of each aspect of both scenarios (meat / other protein sources). 3. Results In scenario LOWMEAT1, meat consumption reduction in selected countries results on a moderate meat demand reduction at global level. In terms of world prices, the differ- ence to the baseline trend ranges between 4% and 12% by 2024, depending on the types of meat (Figure 5). Lower world prices for meat provoke meat consumption increase in the rest of the world, mitigating overall the impact on world prices. The subsequent analysis of results focuses on the situation on the EU markets. At EU level, the meat price difference to the baseline is wider than at global level. Indeed, with a decrease in the domestic consumption and domestic meat prices (significantly high- er in general than world prices), the EU faces difficulties to fully compensate losses of domestic demand only by additional exports. As a consequence the price gap between the EU and world meat markets is closing down, depending on the type of meat consid- ered. Poultry, pig and sheep meat domestic prices are below the baseline level in 2024, but they remain 10 to 25% above the average price 2005-2007. EU domestic price of beef and veal meat results to be more affected, but remains around 10% below the average 2005-2007 domestic price level (Figure 6). At EU level, the decrease of domestic demand is expected to induce an adjustment either through a drop in meat production and/or a drop in EU meat imports and/or a Figure 5. Impact of the LOWMEAT1 scenario on the world price for different meat (% difference to the baseline in 2024). -12% -10% -8% -6% -4% -2% 0% Poultry Pig Meat Beef and veal Sheep meat Source: own calculations from scenarios results. 47Meat markets and a stronger consumption decrease in high-income countries boost of extra-EU exports. However, in scenario LOWMEAT1, meat markets respond dif- ferently. For poultry and pig meat, the decrease of consumption is compensated in the EU by both a decrease of domestic production and increased exports in similar proportions (Figure 7). This market response is favoured by domestic prices not too far away from the world prices and a reasonable degree of competitiveness of the EU industry in these sectors. On the contrary, concerning beef and veal meat, the price wedge between the EU and world markets remains too important and the adjustment to reduced consumption is mostly achieved through a reduced EU production. Finally, sheep meat consumption reduction results almost entirely in an import reduction. These impacts result in an improved self-sufficiency ratio and trade-balance for meat in the EU (see Table 2). The EU would become self-sufficient for beef meat and would be very close for sheep meat. The positive trade balance would also amplify significantly for poultry and pig meat. A reduction in meat consumption also implies a contraction in feed demand, resulting in price decrease for these commodities. Prices are particularly affected for coarse grains and protein meals (i.e. commodities mainly used for feed) but less for wheat and oilseeds (i.e. commodities used for both animal and human consumption as well as for the industry). Overall the feed cost index3 calculated in the Aglink - Cosi- 3 Defined as the average price of all types of commodities potentially used as feed weighed by the actual quanti- ties of each commodity used for feed ; only commercial feed bulks considered. Figure 6. Evolution of EU meat domestic price index in the baseline compared to the LOWMEAT1 sce- nario (Base 100= 2005-07). 70 80 90 100 110 120 130 140 150 Poultry Pig meat Beef and veal Sheep Thin lines are for the baseline and bold lines for the LOWMEAT1 and LOWMEAT2 scenarios. Source: own calculations from scenarios results. 48 F. Santini et al. Figure 7. EU market balances for meat markets in the LOWMEAT1 scenario (change in 1000T relative to the baseline). -2000 -1000 0 1000 Poultry -3000 -2000 -1000 0 1000 2000 Pig meat -1000 -500 0 500 Beef and veal -200 -100 0 100 Sheep meat Source: own calculations from scenarios results Table 2. EU Self-sufficiency ratio and net exports (in % of production). Self-sufficiency ratio Net Exports Baseline Scenario LOWMEAT1 Baseline Scenario LOWMEAT1 Poultry 113 121 +11% +17% Pig meat 104 110 +4% +9% Beef and veal 99 101 -1% +1% Sheep meat 88 97 -13% -3% Source: own calculations from scenario results. 49Meat markets and a stronger consumption decrease in high-income countries mo for the EU would decrease by nearly 8% over the period relative to the baseline (Figure 8). Concerning the meat market, the LOWMEAT2 scenario (with partial protein com- pensation) shows similar results as the scenario LOWMEAT1, although most prices (except for beef meat) are slightly higher than in the scenario LOWMEAT1. The impacts in crop markets are significantly contrasted in the two scenarios. As in LOWMEAT2 meat proteins are partially substituted in the diet with other products - crop or animal - the demand in cereals and oilseeds is less affected than in the first scenario: the increased dairy and eggs production is associated with a higher demand in feed, and food demand for cereals and oilseeds also increases. All in all, EU domestic and world prices for these commodities tend to decrease less than in the scenario LOWMEAT1 (without protein compensation). The feed cost index in the EU would only be 4% below the baseline in this second scenario. Figure 8. EU domestic and world prices for crops and EU feed cost index (Difference to baseline in % change, 2024). scenario LOWMEAT1 scenario LOWMEAT2 -10% -8% -6% -4% -2% 0% World EU -8% -6% -4% -2% 0% World EU Source: own calculations from scenarios results It is to be noted that the LOWMEAT2 scenario includes higher dairy products con- sumption in compensation for the meat consumption reduction, and it would therefore translate into higher EU milk and cheese prices above baseline trends, already quite posi- tive. This scenario with protein compensation leads to particular market stress: on the one hand the domestic demand for beef meat is reduced, but on the other hand beef meat 50 F. Santini et al. production is stimulated by the boost in milk domestic demand as an output of dairy farming. This explains why the beef meat price is the only one to be more affected in sce- nario LOWMEAT2 than in scenario LOWMEAT1. This is also the reason why in the EU in scenario LOWMEAT2, the herd itself would evolve towards less suckler cows (-3%) and more dairy cows (+1%), which would put under pressure those livestock systems special- ised in beef meat production. 4. Discussion on policy implications These two scenarios depict a ten-year transition during which individual changes in food consumption patterns could induce market adjustment and changes in relative pric- es. The European livestock farming sector would have to cope with contradictory market signals: the reduction in feedstuffs prices is an incentive to produce more at lower costs of production, but at the same time domestic demand for meat is losing momentum and lower domestic price would affect profitability of meat production in the EU. This is particularly the case for cattle breeders specialised in meat, who cannot take advantage of the higher demand for milk and dairy products. Some elements of the new CAP might be relevant to help producers facing such challenges, for instance in terms of support to livestock farming systems and in particular for extensive grazing livestock farming systems or through its rural development. First, the CAP 2014-2020 provides recoupling possibilities to EU Member States for livestock (regulation (EU) No 1307/20134). Considering that 24 out of 28 Member States5 opted for the voluntary coupled support for Beef and Veal (Agra Europe, 2015), it is likely that the majority of EU cattle farmers will receive direct coupled support in the future. Beef and veal meat coupled support is expected to represent 10.2 billion euros over the period 2015-20, which represents over 40% of the total voluntary coupled support granted by Member States to farmers. Sheep and goat meat coupled support (around 12% of the total) would add to this. Other meat sectors would not represent a significant share (Euro- pean Commission DG-AGRI, 2015). Thus, recoupling support could play a role of par- tial safety net for livestock farmers and especially for suckler-cow based farming systems. However, such recoupling possibilities might delay the structural adaptations . Second, as highlighted in the introductory section, civil society organizations are con- veying the message to eat “less-but-better” meat. Thus, extensive grazing system and/or quality cuts might benefit from niche markets, based on specific voluntary labelling, quali- ty schemes or simply further market segmentation, for example on the base of breeds spe- cific to one or the other areas of production. Extensive grazing systems are also acknowl- edged to deliver substantial environmental positive externalities. Some instruments of the new CAP other than recoupled support can then play a role in the reorientation of the EU meat sector. Grazing livestock systems should first mechanically be the prime beneficiar- ies of the redistribution of direct payments for internal convergence envisaged in the new 4 Regulation (EU) No 1307/2013 of the European Parliament and the Council of 17 December 2013 establishing rules for direct payments to farmers under support schemes within the framework of the common agricultural policy and repealing Council Regulation (EC) No 637/2008 and Council Regulation (EC) No 73/2009. 5 Only Ireland, Germany, Cyprus and Luxembourg did not recouple support for beef and veal. 51Meat markets and a stronger consumption decrease in high-income countries CAP 2014-2020. Also, grassland-based farms are over-represented in less-favoured areas (LFAs) and areas of natural constraints (ANCs) eligible for specific CAP supports, which may therefore indirectly contribute to foster the extensive livestock systems. In addition, the rural development component of the current CAP contemplates a list of measures relevant for the development of “high quality” livestock systems (Regulation (EU) No 1305/20136), among others support to quality or agri-environment-climate measures (e.g. on the maintenance or introduction of extensive livestock management). The promotion of extensive livestock systems within or without the CAP would also have some benefits for the farming sector. In addition to the environmental adverse effects, intensive meat production systems might face biological risks (such as infectious diseases) and high costs if practices aimed at protecting animals’ wellbeing and safety are not put in place (Hinchliffe et al., 2013; Tilbrook and Hemsworth, 2015). It should be also mentioned that most policy measures presented above will be implemented differently in the Member States. The ability to target financial resources to different categories of breeders according to specific policy or local objectives is the strength of RDPs as a policy tool. However the impact assessment CAP 2020 showed that RDPs sometimes suffer from path dependence (authorities tend to favour past suc- cessful measures over new ones) and unbalanced ability of areas/groups to weight in the process of defining RDPs (European Commission, 2011). From these observations, we can anticipate that in our scenarios livestock producers will receive uneven support across Member States. Noteworthy, most of the policy tools mentioned above are either not designed for the benefit of non-ruminant meat sectors (pig and poultry) or not targeted by Member States to such sectors. EU farmers active in these sectors therefore find less policy tools in the CAP than the ones involved in ruminant livestock breeding. 5. Conclusion This paper analyses the possible impacts on agricultural markets of a significant change in meat consumption in high income countries and selected Latin American coun- tries. The scenarios described here have to be considered as two single and subjective alternative pathways among many other possible and plausible ones. The level of uncer- tainty concerning the pace and features of future development of meat demand and of the development of alternative sources of protein intake in the EU, in other developed coun- tries or in developing countries remains high. Also, the baseline and the two scenarios presented consider a certain set of macroeconomic and yield conditions and the absence of any ‘black swan’ events, like the emergence of new zoonoses or food scares. Furthermore, the scenarios have been elaborated with the same modelling tool as the baseline. Since the alternative scenarios encompass nutritional issues, they could jus- tify that the model is improved by including relevant missing commodities (fish products, fruit and vegetables) for this purpose. 6 Regulation (EU) No 1305/2013 of the European Parliament and the Council of 17 December 2013 on support for rural development by the European Agricultural Fund for Rural Development (EAFRD) and repealing Coun- cil Regulation (EC) No 1698/2005. 52 F. Santini et al. Actually, including the fish commodity would allow for a deeper analysis of protein substitution in lower meat consumption scenarios. But fish inclusion is not straightfor- ward. It would also require further investigation to correctly integrate fish linkages with feed in aquaculture. Another model improvement could be the conversion of quantity of commodities produced and traded into quantities of calories or nutrient equivalent. This would allow to better assess the nutritional value in each scenario: are we modelling healthy purchase behaviours/diets or not? are we correctly simulating protein substitution options? Linking the different food uses of the different commodities by specific equations (developing nutritional module) could be a way forward. Finally, as one of the drivers of a lower meat consumption choice is the individuals’ concerns of their food impact on the environment, an interesting field of research would be to couple the agro-economic model with environmental indicators like the commodi- ties’ virtual content in water or land (like for example in (Wirsenius et al., 2010). In the same perspective, agricultural production could be linked with greenhouse gas emis- sions (Garnett, 2011; Tukker et al., 2011) or biodiversity losses and other environmental impact indicators (as in (Tukker et al., 2011)). Indeed, reducing animal production and/ or improving its efficiency may be key in the mitigation and adaptation to climate change. The reduction in numbers and enhancement of productivity of animals would represent an important mitigation strategy that might have side effects on animal welfare and other environmental issues (Gerber et al., 2013). Further assessment of the environmental and climate impact of nutritional changes as explored by Wolf et al. (2011) would also comple- ment the present analysis. As a conclusion, it is clear that a shift in the meat consumption trends in developed countries would impact on global meat markets, with stronger consequences at EU lev- el. The impact appears to be stronger for beef and veal meat than for non-ruminant live- stock sectors, in particular because of the linkage of this sector with dairy products. There are also significant spill-overs to the crops markets through the feed use (and to a lesser extent possible increased human consumption). A broad range of policy tools could be of importance for farmers affected at EU level by such evolutions. Some tools might allow a swifter transition and/or facilitate a redistribution of the production at EU level between geographical areas and/or farming systems; others might be used for reinforcing competi- tiveness of the EU meat sectors and/or manage diversification where needed. The views expressed are purely those of the author and may not in any circumstances be regarded as stating an official position of the European Commission. References Agra Europe (2015). Market Support - Meat & Livestock. CAP Monitor, 7(1): 7-24. Araujo Enciso, S. R., Perez Dominguez, I., Santini, F. and Hélaine, S. (2015). Documen- tation of the European Commission’s EU Module of the Aglink-Cosimo Modelling System. JRC Scientific and Policy Report EUR 27138 EN. Luxembourg: Publications Office of the European Union. Casalena, N. (2011). How Often Do Americans Eat Vegetarian Meals? And How Many Adults in the US Are Vegan? The VRG Blog. 53Meat markets and a stronger consumption decrease in high-income countries Dagevos, H. (2014). Flexibility in the Frequency of Meat Consumption – Empirical Evi- dence from The Netherlands. EuroChoices 13(2): 40-45. de Bakker, E. and Dagevos, H. (2012). Reducing Meat Consumption in Today’s Consumer Society: Questioning the Citizen-Consumer Gap. Journal of Agricultural and Envi- ronmental Ethics 25(6): 877-894. de Boer, J., Schösler, H. and Aiking, H. (2014). “Meatless Days” or “Less but Better”? Exploring Strategies to Adapt Western Meat Consumption to Health and Sustain- ability Challenges. Appetite 76: 120-128. European Commission (2011). Impact Assessment Common Agricultural Policy towards 2020. Annex 4: Rural Development. European Commission (2014). Prospects for EU Agricultural Markets and Income 2014-2024. European Commission DG-AGRI (2015). Direct Payments Schemes. Friends of the earth (2014). Flexitarianism: The Environmentally Friendly Diet. Briefing. Garnett, T. (2011). Where Are the Best Opportunities for Reducing Greenhouse Gas Emissions in the Food System (Including the Food Chain)? Food Policy 36, Supple- ment 1(0): S23-S32. Gerber, P.J., Steinfeld, H., Henderson, B., Mottet, A., Opio, C., Dijkman, J., Falcucci, A. and Tempio, G. (2013). Tackling Climate Change Through Livestock – A Global Assessment of Emissions and Mitigation Opportunities. FAO, Rome. Gold, M. (2004). The Global Benefits of Eating Less Meat. Petersfield, UK: Compassion in World Farming Trust. Gossard, M.H. and York, R. (2003). Social Structural Influences on Meat Consumption. Human Ecology Review 10(1): 1-9. Health Council of the Netherlands (2011). Guidelines for a Healthy Diet: The Ecological Perspective. Vol. publication no. 2011/08E, 84 The Hague: Health Council of the Netherlands. Herrero M, Havlík, P., Valin, H., Notenbaert, A., Rufino, M.C., Thornton, P.K., Blümmel, M., Weiss, F., Grace, D. and Obersteiner, M. (2013). Biomass Use, Production, Feed Efficiencies, and Greenhouse Gas Emissions from Global Livestock Systems.  Pro- ceedings of the National Academy of Sciences USA 110, 20888-20893.  Hinchliffe, S., Allen, J., Lavau, S., Bingham, N. and Carter, S. (2013). Biosecurity and the Topologies of Infected Life: From Borderlines to Borderlands. Transactions of the Institute of British Geographers 38: 531-543. Hristov, A., Oh, J., Lee, C., Meinen, R., Montes, F., Ott, T., Firkins, J., Rotz, C., Dell, C., Adesogan, A., Yang, W., Tricarico, J., Kebreab, E., Waghorn, G., Dijkstra, J. and Oosting, S.  (2013). Mitigation of Greenhouse Gas Emissions in Livestock Produc- tion - A Review of Technical Options for Non-CO2 Emissions. FAO, Rome. Keyzer, M.A., Merbis, M.D., Pavel, I.F.P. and van Wesenbeeck, C.F.A. (2005). Diet Shifts towards Meat and the Effects on Cereal Use: Can We Feed the Animals in 2030? Ecological Economics 55(2): 187-202. Laestadius, L., Neff, R., Barry, C. and Frattaroli, S. (2013). Meat Consumption and Climate Change: The Role of Non-Governmental Organizations. Climatic Change 120(1-2): 25-38. National Food Administration (2009). The National Food Administration’s Environmen- tally Effective Food Choices. Proposal notified to the EU. Stockholm. 54 F. Santini et al. National Health and Medical Research Council (2013). Eat for health. Australian Dietary Guidelines. Providing the Scientific Evidence for Healthier Australian Diets. Can- berra, Australia: National Health and Medical Research Council. OECD/FAO (2015a). OECD-FAO Agricultural Outlook 2015. OECD Publishing. OECD/FAO (2015b). Aglink-Cosimo Model Documentation. A Partial Equilibrium Mod- el of World Agricultural Markets. OECD Publishing Phillips, F. (2005). Vegetarian Nutrition. Nutrition Bulletin 30(2): 132-167. Pichler, R. and Blackwell, G. (2008). How Many Veggies…?: European Vegetarian Union. Popkin, B.M. (1994). The Nutrition Transition in Low-Income Countries: An Emerging Crisis. Nutrition Reviews 52(9): 285-298. Popkin, B.M. (1999). Urbanization, Lifestyle Changes and the Nutrition Transition. World Development 27(11): 1905-1916. Popkin, B.M., Adair, L.S. and Ng, S.W. (2012). Global Nutrition Transition and the Pan- demic of Obesity in Developing Countries.Nutrition Reviews, 70(1): 3-21. Reay, D., Smith, P. and van Amstel, A. (2010). Methane and Climate Change. Earthscan, 261 pp. Rivers Cole, J. and McCoskey, S. (2013). Does global meat consumption follow an envi- ronmental Kuznets curve. Sustainability: Science, Practice and Policy 9: 26-36. Ruby, M.B. (2012). Vegetarianism. A blossoming field of study. Appetite 58(1): 141-150. Schmidhuber, J. (2004).The Growing Global Obesity Problem: Some Policy Options to Address It. In Globalization of Food Systems in Developing Countries: Impact on Food Security and Nutrition, FAO, Rome: 82-97 Stahler, C. (2011). How Often Do Americans Eat Vegetarian Meals? And How Many Adults in the US Are Vegan? Vegetarian Journal. Steinfeld, H., Gerber, P., Wassenaar, T., Castel, V., Rosales, M. and De Haan, C. (2006). Livestock’s Long Shadow. Environmental issues and options. FAO Rome. Sustainable Development Commission (2009). Setting the Table: Advice to Government on Priority Elements of Sustainable Diets. Sutton, C. and Dibb, S. (2013). Prime Cuts: Valuing the Meat We Eat. A Discussion Paper by WWF-UK and the Food Ethics Council. WWF-UK and Food Ethics Council. Tilbrook, A. and Hemsworth, P.H. (2015). The Costs and Benefits of Animal Welfare. Conference ‘ Animal Welfare, from Science to Law’, LFDA-UNESCO, Paris Decem- ber 10-11, 2015. Tsolekile, L. P. (2007). Urbanization and Lifestyle Changes Related to Non-Communicable Diseases: An Exploration of Experiences of Urban Residents Who Have Relocated from the Rural Areas to Khayelitsha, an Urban Township in Cape Town. Minithesis. University of the Wetsern Cape. Tukker, A., Goldbohm, R.A., de Koning, A., Verheijden, M., Kleijn, R., Wolf, O., Pérez- Domínguez, I. and Rueda-Cantuche, J.M. (2011). Environmental Impacts of Chang- es to Healthier Diets in EUROPE. Ecological Economics 70(10): 1776-1788. U.S. Department of Agriculture and U.S. Department of Health and Human Services (2015). Scientific Report of the 2015 Dietary Guidelines Advisory Committee. Advi- sory Report to the Secretary of Health and Human Services and the Secretary of Agriculture, 571. 55Meat markets and a stronger consumption decrease in high-income countries Vranken, L., Avermaete, T., Petalios, D. and Mathijs, E. (2014). Curbing Global Meat Con- sumption: Emerging Evidence of a Second Nutrition Transition. Environmental Sci- ence and Policy 39: 95-106. Westhoek, H., Rood, G., Van den Berg, M., Janse, J., Nijdam, D., Reudink, M. and Ste- hfest, E. (2011). The Protein Puzzle: The Consumption and Production of Meat, Dairy and Fish in the European Union. European Journal of Food Research and Review 1(3): 124-144. Wirsenius, S., Azar, C. and Berndes, G. (2010). How Much Land is Needed for Global Food Production under Scenarios of Dietary Changes and Livestock Productivity Increases in 2030? Agricultural Systems 103(9): 621-638. Wolf, O., Pérez-Domínguez, I., Rueda-Cantuche, J. M., Tukker, A., Kleijn, R., de Koning, A., Bausch-Goldbohm, S. and Verheijden, M. (2011). Do Healthy Diets in Europe Matter to the Environment? A Quantitative Analysis. Journal of Policy Modeling 33(1): 8-28. World Health Organization Europe (2003). Food-Based Dietary Guidelines in the WHO European Region. EUR/03/5045414. Copenhagen, Denmark: WHO Regional Office for Europe. Annex – Summary results for meat markets (2024) Poultry meat REF LOWMEAT1 LOWMEAT2 World price (USD/t) EU producer price (EUR/t) 1600 2076 1498 1800 1516 1827 EU production (1000t) 13977 13298 13244 EU imports (1000t) 1018 1002 1003 EU consumption (1000t) per capita (kg1/cap/yr) 13466 22.8 12052 20.4 12030 20.4 EU exports (1000t) 1529 2249 2217 Pig meat REF LOWMEAT1 LOWMEAT2 World price (USD/t) EU producer price (EUR/t) 1924 1857 1715 1581 1734 1596 EU production (1000t) 22779 21824 21843 EU imports (1000t) 21 18 18 EU consumption (1000t) per capita (kg2/cap/yr) 20211 30.3 18031 27.1 18018 27.0 EU exports (1000t) 2590 3812 3843 56 F. Santini et al. Beef and veal meat REF LOWMEAT1 LOWMEAT2 World price (USD/t) EU producer price (EUR/t) 3022 3618 2895 2649 2894 2641 EU production (1000t) 7443 6827 6843 EU imports (1000t) 334 272 271 EU consumption (1000t) per capita (kg2/cap/yr) 7530 10.1 6759 9.1 6772 9.1 EU exports (1000t) 252 346 347 Sheep and goat meat REF LOWMEAT1 LOWMEAT2 World price (USD/t) EU producer price (EUR/t) 4948 5779 4560 4986 4593 5023 EU production (1000t) 957 939 939 EU imports (1000t) 205 106 107 EU consumption (1000t) per capita (kg2/cap/yr) 1083 1.8 964 1.6 964 1.6 EU exports (1000t) 79 82 82 A systematic approach to understanding and quantifying the EU’s bioeconomy Tévécia Ronzon1*, Stephan Piotrowski2, Robert M’Barek1, Michael Carus2 Cost function and positive mathematical programming Quirino Paris What if meat consumption would decrease more than expected in the high-income countries? Fabien Santini*, Tevecia Ronzon, Ignacio Perez Dominguez, Sergio Rene Araujo Enciso, Ilaria Proietti A stakeholder engagement approach for identifying future research directions in the evaluation of current and emerging applications of GMOs Davide Menozzi1*, Kaloyan Kostov2, Giovanni Sogari1, Salvatore Arpaia3, Daniela Moyankova2, Cristina Mora1 A spatial analysis of terrain features and farming styles in a disadvantaged area of Tuscany (Mugello): implications for the evaluation and the design of CAP payments Laura Fastelli1*, Chiara Landi2, Massimo Rovai1, Maria Andreoli1