Towards Sustainable Land Use

2. Data and trends relevant to sustainable land use

Abstract

This chapter highlights some of the important data and trends in areas relevant to the land-use, biodiversity, climate and food nexus across the case study countries (Brazil, France, Indonesia, Ireland, Mexico and New Zealand), regionally and globally. This includes information on trends in land-cover change and ecosystems, greenhouse gas emissions from agriculture and LULUCF (land use, land use change and forestry), the emissions intensity of agricultural production and trends in protected area coverage. The chapter also highlights the economic importance of international trade in agricultural and forestry products, and its impact on land use in the case study countries.

This chapter presents data and trends relevant to land use, land cover change (partially as a proxy for biodiversity), GHG emissions from LULUCF, food loss and waste and international trade for the six case study countries, Brazil, France, Indonesia, Ireland, Mexico and New Zealand, as well as at OECD and global levels, for comparative purposes. Global and regional trends in food security are also highlighted.

Land cover and ecosystems

The contribution of land cover types to total land in 2015, based on recently available data from the Climate Change Initiative-Land Cover (CCI-LC),1 is depicted in Figure 2.1. Land cover types analogous with unmanaged areas are tree-covered area, grassland, wetland and shrubland. Overall, most of the conversion of natural land since 1992 has been to cropland.

Figure 2.2 shows land-cover conversion patterns in Indonesia, Brazil and the OECD between 1992-2015. The extent of land conversion is highly variable between the different countries and the OECD. For example, 2.3% of total land area in Ireland and 8.4% of total land area in Indonesia was converted from one land cover to another. There are also differences in the patterns of land-cover change between countries. The majority of Indonesian land-cover change was from tree-covered areas to cropland, whereas in OECD as a whole, slightly more land was converted to tree-covered areas than from tree-covered areas. In Brazil, more tree-covered areas were converted to cropland between 1992 and 2015 than in all OECD countries combined. However, the data for conversion between some individual IPCC categories in Figure 2.2 must be treated with caution, as some are difficult to distinguish reliably via remote sensing.2

Figure 2.1. Contribution of primary land cover types to total, 2015

Note: The definitions used to categorise these data differ from those used by the FAO to characterise land use classes. For example, the changes in ‘Tree Covered Areas’ reported in this figure are not analogous to changes in forest area as a share of land area reported in Figure 2.3.

Source: (Haščič and Mackie, 2018[1]), Land cover change and conversions: Methodology and results for OECD and G20 countries, https://doi.org/10.1787/22260935.

Figure 2.2. Land conversion in Brazil, Indonesia and OECD countries

Note: Six IPCC-like top-level categories. “Other” consists of water, bare land, shrubland, and sparse vegetation. Flow sizes are indicated numerically for flows of more than one thousand km2. Due to the different order of magnitude of land area converted in France, Ireland, Mexico and New Zealand, land conversion diagrams for these countries are reported in the Annex.

Source: (Haščič and Mackie, 2018[1]), Land cover change and conversions: Methodology and results for OECD and G20 countries, https://doi.org/10.1787/22260935.

Trends in forest area – to be distinguished from ‘tree-covered areas’ referred to above due to the different definitions used by the various data sources – since 1990 for the six countries, the OECD and the world are depicted in Figure 2.3. While at the global scale, the total forest area has remained stable (from 31.7% of total land area in 1990, to 30.6% of total land area in 2014), this can obscure large variations between different countries. In particular, there was significant forest loss in tropical areas (FAO, 2016[2]), including large declines in forest area in Brazil and Indonesia (FAOSTAT, 2017[3]). In contrast, afforestation, most notably in China, but also in several OECD countries, has increased forest area in temperate regions. The area of primary forest stayed relatively constant in OECD countries (from 10% in 1990 to 9.9% in 2014). There were, however, reductions in primary forest area in Indonesia (from 27.3% of land area in 1990 to 25.5% in 2014) and Brazil (from 26.1% of land area in 1990 to 24.3% of land area in 2014).

Figure 2.3. Trends in forest area (as a share of total land area)

Source: FAOSTAT (2017[3]), Food and Agriculture data, http://www.fao.org/faostat/en/#data

Greenhouse gas emissions

Turning to GHG emissions from the agricultural and forestry sectors, Figure 2.4 illustrates that there is significant variation between the countries. In Indonesia, where deforestation levels have been high and the carbon content of forest areas is high (particularly in peatlands), forestry-related emissions accounted for approximately half of national GHG emissions in 2014. Contrastingly, in many OECD countries (with the notable exception of New Zealand, where agriculture accounted for almost half of national GHG emissions, albeit with significant emissions removals from the forestry sector), GHG emissions from land use, land-use change and forestry contributed a relatively low proportion of total GHG emissions in 2014.

Figure 2.4 also shows the contribution of the agriculture and forestry sectors to GDP in the case study countries. While this is not true for OECD in general, the contribution of agriculture to GDP is higher than forestry sector in all case study countries. The latter remains below 2% for all the countries bar New Zealand where it represents 2.4%, whereas the agricultural share of GDP diverges more substantially across countries. The contribution of agriculture to GDP ranges between 1.9% and 3.2% for France, Ireland and Mexico and between 6 and 13.3% for Brazil, New Zealand and Indonesia, suggesting this sector is significantly more economically important to this latter group of countries.

Figure 2.4. Emissions from land use, land-use change and forestry (LULUCF) and agriculture, and proportion of GDP from forests and agriculture (2014)

Note: To separate out the different trends for GHG emissions from the agriculture and forestry sectors, the y axis is expressed as a percentage of national emissions excluding LULUCF. Data on LULUCF emissions, in accordance with the 1996 IPCC Gudelines for National Greenhouse Gas Inventories and the IPCC’s Good Practice Guidance on Land Use, Land-Use Change and Forestry comes from OECD.stat and is used for comparability between data reported by Annex I- and non-Annex I-Parties to the UNFCCC. LULUCF emissions largely correspond to emissions from forestry and other land use (FOLU) reported by Annex I-Parties following the 2006 IPCC Guidelines.

Sources: Authors based on OECD.stat (2017[4]) National Inventory Submissions 2017 to the United Nations Framework Convention on Climate Change (UNFCCC, CRF tables), and replies to the OECD State of the Environment Questionnaire, http://dotstat.oecd.org/?lang=en, FAOSTAT (2017[3]), Food and Agriculture data, http://www.fao.org/faostat/en/#data, and FAO (2014[5]) Contribution of the Forestry Sector to National Economies, 1990-2011, http://www.fao.org/3/a-i4248e.pdf.

Global GHG emissions from agriculture increased by 11.0% between 1990 and 2010, while those from OECD countries decreased by 9.5% over the same period (Figure 2.5). The gross production value derived from global agriculture increased by a greater proportion than global emissions, demonstrating a relative decoupling. Among OECD countries the data suggest an absolute decoupling of emissions and value from agriculture over the period, as value increases were realised while absolute levels of GHG emissions declined. However, rising food prices between 2000 and 2012 probably explain at least some of the growth in value both inside and outside OECD countries. Figure 2.6 shows GHG emissions intensities from agriculture by country measured in gigagrams of CO₂ equivalent per million US dollars of agricultural revenue. This suggests substantial convergence in direct GHG emissions intensities from the agricultural sector across countries over time. However, it should be noted that indirect emissions due to land-use change are not included, which are substantial in some countries.

Figure 2.5. Direct agricultural GHG emissions and gross production value from agriculture

Note: Data for agricultural value in the OECD are incomplete prior to 2000.

Source: FAOSTAT (2017[3]), Food and Agriculture data, http://www.fao.org/faostat/en/#data

The sources of agricultural GHG emissions differed markedly between countries over time (Figure 2.7). As for OECD as a whole, agricultural emissions declined in France over time. This occurred largely through reductions in emissions from enteric fermentation and synthetic fertilisers. By contrast, GHG emissions from enteric fermentation and synthetic fertilisers have driven substantial increases in agricultural emissions in Brazil. Emissions from synthetic fertilisers have similarly increased in Indonesia, and have added to substantial increases in methane emission from rice cultivation in the country. New Zealand showed little change in agricultural emissions over the time period, as increases in the contribution from synthetic fertilisers were similar in size to decreases in emissions from enteric fermentation. GHG emissions intensities by product, namely for beef and dairy production are depicted in Figure 2.8 and Figure 2.9 respectively, in most cases indicating a decline in emissions intensities, albeit at differing rates.

Figure 2.6. Direct GHG emissions intensities from agriculture

Note: Increases in emissions intensities between 1998 and 2002, most notably in Indonesia, Brazil and New Zealand, reflect reductions in gross agricultural revenue over the period. These reductions may reflect the impacts of the 1997/98 El Niño which have been linked to severe drought and wildfires in Indonesia, drought and severe flooding in Brazil, and substantial reductions in crop and livestock revenue in New Zealand (National Drought Mitigation Center, n.d.[6]). They may also reflect decreases in the value of traded agricultural products resulting from dips in the Indonesian Rupiah in 1998 and the Brazilian Real in 2002 (Trading Economics, 2018[7]). This graph does not include indirect emissions from agriculture, e.g. those associated with land-clearing.

Source: Authors based on FAOSTAT (2017[3]),Food and Agriculture data, http://www.fao.org/faostat/en/#data and OECD.stat (2017[4]), National Inventory Submissions 2017 to the United Nations Framework Convention on Climate Change (UNFCCC, CRF tables), and replies to the OECD State of the Environment Questionnaire, http://dotstat.oecd.org/?lang=en.

Box 2.1. Livestock production systems and trade

Livestock production systems, in particular cattle-rearing dairy and beef production systems, are economically important to almost all case study countries. At the same time, cattle rearing accounts for a significant share of nexus impacts, through land-use change for pastureland or feed production, direct GHG emissions from manure, and pollution. Dairy and beef production systems range from extensive, pasture-based production systems, to intensive feedlot-based systems. These systems differ in both their economic performance and environmental impacts.

For example, Gerber et al. (2013[8]) argue that milk and beef from intensive, high-yielding production systems are less emissions-intensive due to scale effects and productivity gains from emissions-reducing practices, such as high-quality feed and herd management. Figure 2.8 and Figure 2.9 confirm that per-unit emissions are indeed lower in countries such as New Zealand, Ireland and France when compared to the more extensive production systems in Indonesia and Brazil.

However, important differences exist between intensive cattle-rearing systems, which are not discernible from the figures showing only emissions from within the farm gate. Life-cycle analysis which accounts for upstream (and downstream) emissions (see chapter 5)3, shows the predominantly grass-based dairy systems in New Zealand and Ireland to be less emissions-intensive than France, for example, due to their lower reliance on imported feed (Weiss and Leip, 2012[9]). Consequently, indirect land-use change from cattle feed production – which accounts for up to one third of EU GHG emissions in the study by Weiss and Leip (2012[9]) – can severely constrain the scope for intensification to lower the livestock sector’s GHG footprint (Bowles, Alexander and Hadjikakou, 2019[10]; Styles et al., 2018[11]).

Livestock product trade

If a higher proportion of global beef and dairy production originates from the most efficient production systems, international trade can in theory contribute to global production efficiency, reducing GHG emissions and other environmental impacts. Rising (domestic and) international demand is boosting the beef and dairy export industry in the case study countries particularly in New Zealand, Ireland, Brazil, and Mexico. Ireland, for instance, exported 85% of its beef production and 90% of its dairy production in 2016 (Agriculture and Food Development Authority, 2017[12]; Fitzgerald, 2019[13]). While Ireland and New Zealand are relatively efficient producers from a GHG emissions perspective, the increases in production are undermining their ability to meet domestic emissions pledges. Further, whether the wider environmental impacts associated with export-oriented production can be sustained in the long term, is questionable. In New Zealand, for instance, where dairy products accounted for a quarter of total goods exports in 2018 (Stats NZ, 2019[14]), dairy farming has now replaced lower impact sheep farming leading to significant increases in water pollution (Ministry for the Environment, 2019[15]). Despite being an economic boon, the societal costs from environmental impacts of the dairy industry have, however, been found to outweigh the export revenues (Foote, Joy and Death, 2015[16])

Figure 2.7. Agricultural emissions by source

Note: Y-axes on different scales

Source: FAOSTAT (2017[3]), Food and Agriculture data, http://www.fao.org/faostat/en/#data

Figure 2.8. GHG emissions generated within farm gate per kilogram of beef production

Source: FAOSTAT (2017[3]), Food and Agriculture data, http://www.fao.org/faostat/en/#data

Figure 2.9. GHG emissions generated within farm gate per kilogram of milk production

Source: FAOSTAT (2017[3]), Food and Agriculture data, http://www.fao.org/faostat/en/#data

Food systems

Other agri-environment indicators also exist, such as the farmland bird index, pesticide sales and nutrient balance. These indicators provide additional information on the sustainability of agriculture.4 The OECD tracks this information for OECD and some other countries, though complete data is not always available for all countries. For example, time-series data on the farmland bird index is available for France (showing a decline over time). Data on pesticide sales is available for France, Mexico, and New Zealand.

Food security is a growing concern globally, according to FAO et al. (2017[17]), the estimated number of undernourished people – having been on the decline for the past decade – increased from 777 million in 2015 to 815 million in 2016 (i.e. to 11% of global population). Food security issues are in part caused by inefficiencies in global food systems and the recent increases can be traced to the greater number of conflicts. These issues are often exacerbated by climate-related shocks, which are expected to increase in frequency and severity as a result of human GHG emissions, and the reduced resilience of ecosystems from degradation and biodiversity loss. Contrastingly, the same inefficiencies combined with unhealthy consumption patterns also produce the opposite issues in different regions. The global prevalence of obesity is rapidly on the rise, with 13% of the world adult population classified as obese in 2014. The problem is most severe in Northern America, Europe and Oceania, where 28% of adults are classified as obese, compared with 7 % in Asia and 11 % in Africa. In Latin America and the Caribbean, roughly one-quarter of the adult population is currently considered obese (FAO et al., 2017[17]). Addressing the inefficiencies in global food systems, reducing anthropogenic GHG emissions and tackling ecosystem degradation is a key addressing food security issues, especially under environmental change in the future.

The OECD-FAO Agricultural Outlook 2018-2027 (2018[18]) projects that across most commodities, the growth in total demand (including non-food uses) will slow considerably compared to the previous decade (Figure 2.10). Future growth in crop production is expected to come mostly from increasing yields. Yield growth is projected to decrease slightly, but output could be raised by closing large yield gaps, especially in Sub-Saharan Africa. Nevertheless, the Outlook indicates that food insecurity will remain a critical global concern. Further, because the areas of projected food demand growth differ from the areas where supply can be increased sustainably, international trade will become increasingly import for adapting to and mitigating climate change and achieving the SDGs.

Food waste and food loss (FLW) also has major implications for land use, biodiversity, climate change and water. An estimated one third of all food produced for human consumption is either lost or wasted (FAO, 2013[19]). This equates to approximately 1.3 billion tonnes a year, worth an estimated USD 936 billion and generating around 4.4 GtCO₂. Compared to countries FLW is the third biggest emitter globally (behind only the USA and China) (FAO, 2013[19]; FAO, 2015[20]; FAO, 2013[21]). Importantly for the land-use nexus, the production from approximately 30% of global agricultural land is wasted every year, which equates to 1.4 billion hectares (FAO, 2013[21]), an area larger than the total land area of all the case study countries combined.

Figure 2.10. Annual growth in consumption for key commodity groups, 2008-17 and 2018-27

Note: The population growth component is calculated assuming per capita demand remains constant at the level of the year preceding the decade. Growth rates refer to total demand (for food, feed and other uses).

Source: OECD/FAO (2018[18]), OECD-FAO Agricultural Outlook, http://dx.doi.org/10.1787/agr-outl-data-en.

Figure 2.11. Kilograms of food waste per capita per year

Note: The estimates of food waste in this graph should not be directly compared between countries as different methodologies were used in the estimation.

Source: Brazil and Indonesia: EIU (2018[22]) Food Sustainability Index, Country Index and Data, http://foodsustainability.eiu.com/

France and Ireland: OECD (2018[23]) Food Waste database, https://stats.oecd.org/viewhtml.aspx?datasetcode=FOOD_WASTE&lang=en;;

Mexico: Aguilar Gutiérrez (2016[24]) Food Losses and food Waste in Mexico, http://www.cec.org/sites/default/files/pdf/fww/wb-presentations/6-genaro-aguilar.pdf; New Zealand: Yates (2013[25]), Summary of existing information on domestic food waste in New Zealand Document quality control, http://www.wasteminz.org.nz/wp-content/uploads/Report-on-Food-Waste-in-NZ-2013-Final-1.1.pdf and Reynolds et al. (2016[26]), New Zealand’s Food Waste: Estimating the Tonnes, Value, Calories and Resources Wasted, http://dx.doi.org/10.3390/agriculture6010009.

All countries: FAOSTAT (2018[27]), Food Balance Sheets, http://www.fao.org/faostat/en/#data/FBS

Estimates for per capita food waste in the six case study countries vary substantially. The volumes presented in Figure 2.11 are not all directly comparable (due to the differing methodologies used to derive them), and should be treated with caution. Generally, household level FLW is comprised of edible food and is potentially avoidable, whereas feasibility and desirability of avoiding post-production, pre-consumption FLW varies with both the type of food and the position the loss occurs in the food supply chain. There is also considerable uncertainty within the data. Indonesia for example has, according to some sources, both the second highest per capita level of household food waste globally (315kg/capita/year) (behind Saudi Arabia) (EIU, 2018[22]) and over 30% of children under 5 are malnourished (WFP, 2018[28]). However, other estimates of household food waste in Indonesia show large variation from 6kg/capita/year (EIU, 2019[29])5 to 253kg/capita/year (Meidiana and Gamse, 2010[30]) highlighting the considerable uncertainty associated with these data. The high level of post-production pre-consumption waste seen in Brazil (much higher than other countries) is probably a result of viewing the data on a per capita basis which tends to lead to inflated figures in countries which export large volumes of agricultural produce. When viewed as a proportion of production, post-production pre-consumption food waste in Brazil is similar to Indonesia and only marginally more than Mexico and New Zealand. In general, household level food waste is higher in developed than developing countries.

International Trade

The consumption of goods traded internationally accounts for an estimated 25% of agricultural and forestry impacts on bird extinctions, and for 21% of impacts on terrestrial carbon sequestration (Marques et al., 2019[31]). Globally, 20% of wheat, 12% of maize and more than 60% of global soy production is exported (Fischer, Byerlee and Esmeades, 2014[32]). Indeed, Brazil exports two-thirds (41 Mt of 62Mt) of its soybean production (ibid). Indonesia is the world’s largest producer of palm oil, with 76.5% of palm oil production exported in 2013 (FAOSTAT, 2017[3]). Production of palm oil in Indonesia has grown at 4.9% per year between 1991-2010 (Fischer, Byerlee and Esmeades, 2014[32]). Beyond these examples, international trade in goods whose production significantly impacts the nexus is important for all case study countries. When comparing trade in forest products to trade in agricultural products, the latter emerges as the more important category both (i) economically and, in most cases, (ii) in terms of land-use nexus impacts. 6 However, when forest products originate from primary forests, nexus impacts per traded unit, in particular biodiversity impacts, are high and often irreversible.

In economic terms, export revenues from agriculture exceeded those from forest products by a factor of between four (Indonesia) and 52 (Mexico) in 2016 (FAO, 2018[33]). Figure 2.12 illustrates the economic importance of international trade in agricultural products in terms of the case study countries’ exports share of agricultural GDP for 10 years from 2005. The importance of agricultural exports varies significantly between the countries, and has grown in all case study countries except Indonesia (where it slightly declined) in this time period.

Figure 2.12. Share of exports in agricultural GDP

Note: The low Indonesian ratio is partly related to the fact that a large share of the value addition of main Indonesian export products such as palm oil and rubber occurs during their processing and manufacturing, which in accordance with the ISIC4 Classification of Economic Activities is not captured here. Because values for fisheries are reported in conjunction with forestry and agriculture under this classification, fisheries are included here although they are not the focus of this report. Since the economic contribution of fisheries to the variables considered here is marginal for all case study countries, the reported trends are however valid for land-based components of agricultural GDP, too.

Source: OECD (2018[34]), Trade in Value Added (database), oe.cd/tiva

Figure 2.13 shows the economic importance of international trade in agricultural products by another measure, their share in total goods trade. While the share of agricultural products in goods imports lies between 7% (Brazil) and 12.3% (Indonesia) for all case study countries, the share of agricultural exports in total goods exports varies more substantially, ranging from 7.5% (Mexico) to 57.1% (New Zealand). Figure 2.13 also facilitates the interpretation of trends in Figure 2.12 above. Those countries with the highest share of exports in agricultural GDP (Ireland, New Zealand, France) tend to export more higher-value products (such as animal products, processed foods and beverages), the other case study countries tend to export primary commodities such as soy and rubber.

Figure 2.13. Share of different agricultural products in total goods trade (by value), 2016

Note: The total value of all agricultural exports (EX) and imports (IM), indicated in the bar labels, is given in billion (bn) USD.

Source: Authors based on FAOSTAT (2018[33]), Food and Agriculture data (database), http://www.fao.org/faostat/en/#data and (UN Comtrade, n.d.[35]), International Trade Statistics (database), http://comtrade.un.org.

One way to compare the land-use nexus impacts embodied in international trade across countries is the land footprint of national production and consumption. Figure 2.14 illustrates results of the estimation approach adopted by the Global Footprint Network (2018[36]) disaggregated by land type, and shows that land requirements embodied in international flows of goods are substantial for some countries.7 In fact, countries whose ecological footprint of consumption exceeds their export production footprint (Ireland, Mexico) are in effect net importers of land. The other case study countries are net “exporters” of biocapacity, meaning more biologically productive land is embodied in their exports than in their imports. While for Brazil, cropland contributes more than twice as much as forest or grazing land to net biocapacity exports, for both New Zealand and Indonesia forest land constitutes the largest source of net biocapacity exports.8 Lastly, given the important export share of its livestock sector, Ireland is a net “importer” of cropland but a net “exporter” of grazing land. While Figure 2.14 displays the net trade balance of land flows embodied in production and consumption (“virtual land flows”), it must be noted that gross virtual land flows exceed net flows substantially due to circular loops in increasingly integrated markets for agricultural products (Harchaoui and Chatzimpiros, 2017[37]).

Figure 2.14. Net biocapacity exports, 2014

Note: *Note that data displayed for New Zealand refers to 2013. Percentage values indicate the share of net biocapacity exports in a country’s total biocapacity from cropland, grazing land and forest land. Biocapacity is the capacity of a country’s ecosystems to regenerate what people demand from those surfaces including space for growing food, fibre production, and timber regeneration. Cropland is defined as the area needed to grow 177 agricultural products contained in the FAOstat database. Grazing land is defined as the area required to grow grass feed required for feeding a country’s livestock under national average grazing land productivity and after accounting for other feed sources. Forest land represents the area of forest land needed to supply wood for fuel, construction, and paper. The Ecological Footprint of Production indicates the consumption of biocapacity resulting from production processes within a country, and the Ecological Footprint of Consumption indicates biocapacity actually consumed by a country’s inhabitants. Differences between production and consumption footprints result in biocapacity imports or exports attributable to international trade. Biocapacity is measured in global hectares, a productivity-weighted measurement unit making biologically productive areas of different land types comparable.

Source: Authors based on Global Footprint Network (2018[36]), National Footprint Accounts, http://data.footprintnetwork.org/#/

Patterns of trade illustrated by Figure 2.12, Figure 2.13., and Figure 2.14 have important land-use nexus impacts in the case study countries and beyond. In Brazil, for instance, more than 50% of tree cover loss between 2005 and 2015 have been attributed to commodity-driven deforestation (Global Forest Watch, 2019[38]), primarily associated with beef and soy production, two of Brazil's main export commodities (Henders, Persson and Kastner, 2015[39]). While more recent trends in land use mean that this proportion might have changed, a number of earlier studies estimate that an approximate share of 30% of Brazil’s LUC emissions were historically embodied in exports (Saikku, Soimakallio and Pingoud, 2012[40]; Karstensen, Peters and Andrew, 2013[41]). In Indonesia, quantitative data on the drivers of deforestation are rare, but an estimated 23% to 50% of deforestation after 2000 has been attributed to the expansion of oil palm plantations (Austin et al., 2019[42]; Henders, Persson and Kastner, 2015[39]), used to produce palm oil, a key export commodity. Adverse biodiversity impacts attributable to the production of export commodities have also been estimated. Chaudhary and Kastner (2016[43]), for instance, report that among all countries Indonesia has the highest biodiversity impacts in terms of species loss attributable to food exports, more than twice that of second-ranked Thailand.9

In Mexico, research suggests that while the global environmental impacts of agricultural production have been reduced due to trade liberalisation under the North American Free Trade Agreement (NAFTA), more severe environmental impacts have shifted from the US to Mexico (Martinez-Melendez and Bennett, 2016[44]). In fact, agricultural intensification in the wake of NAFTA is likely to have led to significant biodiversity impacts through the wide-scale replacement of traditional cropping systems by input-intensive modern production systems (Orozco-Ramírez et al., 2017[45]; UNCTAD, 2013[46]). In the case of France, studies have focused on the extent to which imported goods embody deforestation and biodiversity threats abroad. Envol Vert (2018[47]), for instance, reports that the deforestation footprint of the average French consumer is 352 m² per year, 59% of which is attributable to soy (mainly from Brazil) embodied in animal products.

Domestic impacts on the land-use nexus attributable to international trade also occur in developed countries. New Zealand and Ireland both export large quantities of, dairy and beef, which are emission-intensive to produce. While exports of these products continue to grow, this growth model is starting “to show its environmental limits” in New Zealand (OECD, 2017, p. 15[48]). Beyond domestic land-use and emissions impacts, animal feed imports imply impacts on land-use outcomes abroad, too. In Ireland, for instance, among agricultural products, animal feed imports constitute the most important import category (in volumetric terms)10 reported by (Department of Agriculture; Food and the Marine, 2018[49]).

References

[12] Agriculture and Food Development Authority (2017), Agriculture in Ireland - Teagasc | Agriculture and Food Development Authority, https://www.teagasc.ie/rural-economy/rural-economy/agri-food-business/agriculture-in-ireland/ (accessed on 22 May 2019).

[24] Aguilar Gutiérrez, G. (2016), Food Losses and Food Waste in Mexico: Quantification and Some Proposals for Public Policy, International Workshop on Food Loss and Food Waste, November 7–9.

[42] Austin, K. et al. (2019), “What causes deforestation in Indonesia?”, Environmental Research Letters, Vol. 14/2, p. 024007, http://dx.doi.org/10.1088/1748-9326/aaf6db.

[10] Bowles, N., S. Alexander and M. Hadjikakou (2019), “The livestock sector and planetary boundaries: A ‘limits to growth’ perspective with dietary implications”, Ecological Economics, Vol. 160, pp. 128-136, http://dx.doi.org/10.1016/J.ECOLECON.2019.01.033.

[43] Chaudhary, A. and T. Kastner (2016), “Land use biodiversity impacts embodied in international food trade”, Global Environmental Change, Vol. 38, pp. 195-204, http://dx.doi.org/10.1016/J.GLOENVCHA.2016.03.013.

[49] Department of Agriculture; Food and the Marine (2018), Annual Review and Outlook for Agriculture, Food and the Marine 2018, Government of Ireland, Dublin, https://www.agriculture.gov.ie/media/migration/publications/2018/AnnualReviewandOutlook2018310818.pdf (accessed on 4 February 2019).

[29] EIU (2019), Food Sustainability Index 2018: Data, http://foodsustainability.eiu.com/wp-content/uploads/sites/34/2018/11/Food-Sustainability-Index-2018_EIU_BCFN_23-November.xlsm.

[22] EIU (2018), Food Sustainability Index, Country Index and Data, http://foodsustainability.eiu.com/ (accessed on 18 October 2018).

[47] Envol Vert (2018), L’Empreinte Foret des Francais, Author, Paris, http://envol-vert.org/wp-content/uploads/2018/11/Envol_Vert_Rapport_Empreinte_Foret_Final.pdf (accessed on 6 February 2019).

[33] FAO (2018), FAOSTAT, Food and Agriculture Organization of the United Nations, http://www.fao.org/faostat/en/#home (accessed on 7 February 2019).

[2] FAO (2016), State of the World’s Forests, Food and Agriculture Organization, Rome, http://www.fao.org/3/a-i5588e.pdf.

[20] FAO (2015), Food wastage footprint and Climate Change.

[5] FAO (2014), Contribution of the Forestry Sector to National Economies 1990-2011, Food and Agriculture Organization.

[19] FAO (2013), Food Wastage Footprint: Full-cost Accounting, Food and Agriculture Organisation of the United Nations, Rome.

[21] FAO (2013), Food wastage footprint: Impacts on natural resources - Summary report, Food and Agriculture Organisation, http://www.fao.org/docrep/018/i3347e/i3347e.pdf (accessed on 29 November 2018).

[17] FAO et al. (2017), The State of Food Security and Nutrition in the World 2017. Building resilience for peace and food security., FAO, Rome, http://www.fao.org/3/a-I7695e.pdf (accessed on 17 September 2019).

[27] FAOSTAT (2018), Food Balance Sheets, http://www.fao.org/faostat/en/#data/FBS (accessed on 18 October 2018).

[3] FAOSTAT (2017), Food and Agriculture data, Food and Agriculture Organization, http://www.fao.org/ag/save-and-Grow/en/1/index.html.

[32] Fischer, T., D. Byerlee and G. Esmeades (2014), Crop yields and global food security: will yield increase continue to feed the world?, Australian Centre for International Agricultural Research, Canberra, https://www.aciar.gov.au/node/12101.

[13] Fitzgerald, C. (2019), Dairy in the Irish Economy!, Teagasc, https://www.teagasc.ie/media/website/publications/2019/Dairy-in-the-Irish-economy!.pdf.

[16] Foote, K., M. Joy and R. Death (2015), “New Zealand Dairy Farming: Milking Our Environment for All Its Worth”, Environmental Management, Vol. 56/3, pp. 709-720, http://dx.doi.org/10.1007/s00267-015-0517-x.

[51] Galli, A. et al. (2016), “Questioning the Ecological Footprint”, Ecological Indicators, Vol. 69, pp. 224-232, http://dx.doi.org/10.1016/J.ECOLIND.2016.04.014.

[8] Gerber, P. et al. (2013), Tackling Climate Change through Livestock: A Global Assessment of Emissions and Mitigation Opportunities, Food and Agriculture Organization of the United Nations, Rome, Italy, http://www.fao.org/publications (accessed on 13 May 2019).

[36] Global Footprint Network (2018), National Footprint Accounts, http://data.footprintnetwork.org/#/ (accessed on 5 June 2019).

[38] Global Forest Watch (2019), Brazil Country Profile, https://www.globalforestwatch.org/dashboards/country/BRA?treeLossTsc=eyJoaWdobGlnaHRlZCI6ZmFsc2V9 (accessed on 18 March 2019).

[37] Harchaoui, S. and P. Chatzimpiros (2017), “Reconstructing production efficiency, land use and trade for livestock systems in historical perspective. The case of France, 1961–2010”, Land Use Policy, Vol. 67, pp. 378-386, http://dx.doi.org/10.1016/J.LANDUSEPOL.2017.05.028.

[1] Haščič, I. and A. Mackie (2018), “Land Cover Change and Conversions: Methodology and Results for OECD and G20 Countries”, OECD Green Growth Papers, No. 2018/04, OECD Publishing, Paris, https://dx.doi.org/10.1787/72a9e331-en.

[39] Henders, S., U. Persson and T. Kastner (2015), “Trading forests: land-use change and carbon emissions embodied in production and exports of forest-risk commodities”, Environmental Research Letters, Vol. 10/12, p. 125012, http://dx.doi.org/10.1088/1748-9326/10/12/125012.

[41] Karstensen, J., G. Peters and R. Andrew (2013), “Attribution of CO2 emissions from Brazilian deforestation to consumers between 1990 and 2010”, Environmental Research Letters, Vol. 8/2, p. 024005, http://dx.doi.org/10.1088/1748-9326/8/2/024005.

[31] Marques, A. et al. (2019), “Increasing impacts of land use on biodiversity and carbon sequestration driven by population and economic growth”, Nature Ecology & Evolution, Vol. 3/4, pp. 628-637, http://dx.doi.org/10.1038/s41559-019-0824-3.

[44] Martinez-Melendez, L. and E. Bennett (2016), “Trade in the US and Mexico helps reduce environmental costs of agriculture”, Environmental Research Letters, http://dx.doi.org/10.1088/1748-9326/11/5/055004.

[30] Meidiana, C. and T. Gamse (2010), Development of Waste Management Practices in Indonesia, http://www.eurojournals.com/ejsr.htm (accessed on 10 January 2019).

[15] Ministry for the Environment (2019), Environment Aotearoa 2019, Government of New Zealand, Wellington, https://www.mfe.govt.nz/sites/default/files/media/Environmental%20reporting/environment-aotearoa-2019.pdf.

[6] National Drought Mitigation Center (n.d.), Reported Effects of the 1997-98 El Niño, http://jrscience.wcp.muohio.edu/studentresearch/climatechange03/elnino/elninoanalysis.pdf.

[23] OECD (2018), Food Waste Database, https://stats.oecd.org/viewhtml.aspx?datasetcode=FOOD_WASTE&lang=en (accessed on 28 November 2018).

[34] OECD (2018), Trade in Value Added, http://www.oecd.org/sti/ind/measuring-trade-in-value-added.htm (accessed on 5 June 2019).

[48] OECD (2017), OECD Environmental Performance Reviews: New Zealand 2017, https://www.oecd-ilibrary.org/docserver/9789264268203-en.pdf?expires=1548772643&id=id&accname=ocid84004878&checksum=34BA8F8BD5F9B3671D35BB832DFAA3C1 (accessed on 29 January 2019).

[50] OECD (2013), “Biodiversity: Farmland bird populations and agricultural land cover”, in OECD Compendium of Agri-environmental Indicators, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264186217-15-en.

[18] OECD/FAO (2018), OECD-FAO Agricultural Outlook 2018-2027, OECD Publishing, Paris/Food and Agriculture Organization of the United Nations, Rome, https://dx.doi.org/10.1787/agr_outlook-2018-en.

[4] OECDSTAT (2017), National Inventory Submissions 2017 to the United Nations Framework Convention on Climate Change (UNFCCC, CRF tables), and replies to the OECD State of the Environment Questionnaire, OECD Publishing, http://dotstat.oecd.org/?lang=en.

[45] Orozco-Ramírez, Q. et al. (2017), “Agricultural Land Use Change after NAFTA in Central West Mexico”, Land, Vol. 6/4, p. 66, http://dx.doi.org/10.3390/land6040066.

[26] Reynolds, C., M. Mirosa and B. Clothier (2016), “New Zealand’s Food Waste: Estimating the Tonnes, Value, Calories and Resources Wasted”, Agriculture, http://dx.doi.org/10.3390/agriculture6010009.

[40] Saikku, L., S. Soimakallio and K. Pingoud (2012), “Attributing land-use change carbon emissions to exported biomass”, Environmental Impact Assessment Review, Vol. 37, pp. 47-54, http://dx.doi.org/10.1016/J.EIAR.2012.03.006.

[14] Stats NZ (2019), Overseas merchandise trade: December 2018 | Stats NZ, https://www.stats.govt.nz/information-releases/overseas-merchandise-trade-december-2018 (accessed on 15 May 2019).

[11] Styles, D. et al. (2018), “Climate mitigation by dairy intensification depends on intensive use of spared grassland”, Global Change Biology, Vol. 24, pp. 681-693, http://dx.doi.org/10.1111/gcb.13868.

[7] Trading Economics (2018), Trading Economics Currency Data, https://tradingeconomics.com/indonesia/currency.

[35] UN Comtrade (n.d.), International Trade Statistics, 2019, https://comtrade.un.org/data/ (accessed on 5 June 2019).

[46] UNCTAD (2013), Mexico’s Agriculture Development: Perspectives and outlook, Author, Geneva, Switzerland, https://unctad.org/en/PublicationsLibrary/ditctncd2012d2_en.pdf (accessed on 7 February 2019).

[9] Weiss, F. and A. Leip (2012), “Greenhouse gas emissions from the EU livestock sector: A life cycle assessment carried out with the CAPRI model”, Agriculture, Ecosystems & Environment, Vol. 149, pp. 124-134, http://dx.doi.org/10.1016/J.AGEE.2011.12.015.

[28] WFP (2018), Indonesia Country Brief, World Food Programme, http://www.wfp.org/countries/Indonesia (accessed on 29 November 2018).

[25] Yates, S. (2013), Summary of existing information on domestic food waste in New Zealand Document quality control, Waste Not Consulting.

Annex 2.A. Land-cover change in France, Ireland, Mexico and New Zealand

Annex Figure 2.A.1. Land Conversion in France, Ireland, Mexico and New Zealand

Note: Six IPCC-like top-level categories. “Other” consists of water, bare land, shrubland, and sparse vegetation. Flow sizes are indicated numerically for flows of more than one thousand km2.

Source: OECD calculations based on CCI-LC database in “Land cover change and conversions: Methodology and results for OECD and G20 countries” (Haščič and Mackie, 2018[1]).

Notes

← 1.  The CCI-LC datasets are currently the only available global datasets that can provide some acceptably harmonised indication of the type and intensity of change between different land cover types. See (Haščič and Mackie, 2018[1]) for a description of the dataset.

← 2.  Misclassification is more likely between different vegetated land types as these classes are often similar and are more difficult to reliably distinguish. For example, the observed conversions from wetlands to tree-cover seen in Figure 2.1 is partly an ambiguous classification issue: the observable biophysical difference between the wetland definition (Shrub or herbaceous cover, flooded, fresh-saline or brackish water) and the flooded forest classes (Tree cover, flooded, fresh or brackish water, Tree cover, flooded, saline water) is small and difficult to distinguish reliably via remote sensing (Haščič and Mackie, 2018[1]).

← 3.  The methodology used to assess the impacts of particular land uses has important implications for decision making as it directly influences the perceived size of any trade-offs and synergies between areas in the land use nexus.

← 4.  For example, the farmland bird index is an average trend in a group of species suited to track trends in the condition of farmland habitats. In general, a decrease in the index means that the balance of bird species population trends are negative, representing biodiversity loss (OECD, 2013[50]).

← 5.  Methodological changes between the 2017 and the 2018 food sustainability index caused the household food waste estimates for Indonesia to change from 315kg/capita/year to 6kg/capita/year, for full details see http://foodsustainability.eiu.com/wp-content/uploads/sites/34/2018/11/FSI-2018-Methodology-Paper_full_December-2018.pdf

← 6.  Clearly, subsistence agriculture and other forms of agricultural production for domestic consumption are important in some contexts, accounting for an important share of both economic value-creation and land-use impacts of total agricultural production. This, however, is not captured in the trade statistics presented in this chapter.

← 7.  The debate about the robustness of this approach is ongoing for reasons including the accurateness of the proxy adopted (land use) for the variable of interest (environmental impact), and the difficulty of comparing heterogeneous environmental impacts across locations (see e.g. (Galli et al., 2016[51]) for a discussion). In Figure 2.14, for instance, land-use nexus impacts associated with virtual exports of forest land are likely to substantially differ between the tropical forests of Indonesia and Brazil and the temperate forests of other case study countries.

← 8.  From an environmental perspective, net biocapacity exports of forest land will not have negative nexus impacts as long as they are produced from sustainably managed forests.

← 9.  Species loss attributable to food exports is estimated using the countryside species area relationship (SAR) model. For full details, see (Chaudhary and Kastner, 2016[43]).

← 10.  Since land-use nexus impacts often represent externalised costs imperfectly reflected in prices, volumetric measures of trade can usefully complement monetary accounts and enhance a quantitative understanding of land-use impacts associated with imports and exports.