Enhancing Climate Change Mitigation through Agriculture

5. Global mitigation potential of biofuels in the transport sector

Abstract

Based on the OECD-FAO Agricultural Outlook 2018-2027 baseline, this chapter examines the potential contribution of biofuels to climate change mitigation in the transport sector

Introduction

The International Energy Agency’s (IEA) Energy Technology Perspectives (IEA, 2017[1]) foresees that bioenergy will play an important role in climate change mitigation. The IEA has defined a mitigation scenario, the 2-degree scenario (2DS), that is consistent with a 50% chance of limiting future global average temperature to an increase of 2°C by 2100. This scenario is developed on assumptions of the future evolution of crude oil prices, the macroeconomic environment, and policies that concern the whole energy system (including the transport sector), including associated intended impacts on transportation fuel demand. In the 2DS, carbon taxes are applied according to the well-to-wheel (WTW) greenhouse gas (GHG) emission profiles of the different types of fuel which increase over time.

The IEA 2DS defines a target for energy-related GHG emissions reduction; which involves lower transport demand and more transport fuel coming from lower GHG emitting fuels. The IEA 2DS does not assess the capacity of agriculture to deliver the ambitious volume of biofuels foreseen in this pathway over the coming decade and is not able to estimate implications for agricultural markets.

This chapter presents the results of the AGLINK-COSIMO 2-degree scenario (AC-2DS) simulated with the AGLINK-COSIMO model. The AC-2DS uses assumptions consistent with the IEA 2DS assumptions. In particular, the IEA 2DS increasing path for carbon taxes is replicated. The AC-2DS can assess the potential impacts of a mitigation scenario on both biofuel and agricultural markets across the globe up to 2030.

The analysis complements and assesses the path developed by the IEA by including an agricultural sector perspective. Indeed, the AGLINK-COSIMO model is a partial equilibrium model that is able to take into account the interconnection between agricultural and biofuel markets1 and the ability of agricultural activities to supply the amount of bioenergy the IEA has determined is needed to meet climate change targets.2 This chapter also proposes ways to enrich and develop the analysis of those linkages.3

Biofuels and greenhouse gas emission savings in the transport sector

Greenhouse gas emissions associated with biofuels

Biofuels are fuels produced by the transformation of biomass. They can be blended with or replace conventional fossil fuels.4 This chapter focuses on two kinds of biofuels: ethanol and biodiesel.5 Ethanol and biodiesel are used, respectively, as gasoline and diesel substitutes or complements. Similar to conventional fossil fuels, biofuels are composed of hydrocarbon chains. Emissions of greenhouse gases (GHGs) occur at each step of the biofuel value chain: during plant growth,6 crop harvest, transportation of the feedstock to the processing plant, conversion process, distribution to the fuel terminal, WTW GHGs emissions for biofuels7. This chapter does not consider specific GHG individually, but looks at the aggregate of all GHGs identified, expressed in CO₂eq.

The WTW carbon intensity of biofuels can be represented by an “emission factor”,8 expressed in kgCO₂eq per unit of energy content. A literature review (Annex 5.A) was conducted to gather WTW emission factors for different biofuel pathways (Box 5.1). The WTW emission factors collected in this literature review were then averaged to derive a set of WTW emission factors for each biofuel type to be used in the present analysis. Most studies reviewed take into account the by-product use of biofuel food and feed feedstock when allocating WTW GHG emissions to the different biofuel feedstock. The literature review does not cover the growing discussion concerning carbon sinks, i.e. carbon capture by oceans and soils when any types of fuel is combusted.

Biofuel-related greenhouse gas emission savings at the 2030 horizon in the transport sector

Table 5.1 provides an overview of WTW emission savings using the baseline of the 2018 OECD-FAO Agricultural Outlook for the period 2015-2017 and by 2030 for major biofuel consuming countries. Box 5.2 describes major trends in biofuel baseline projections.

Table 5.1. Biofuel blending in transportation fuels and associated WTW emission savings

	Ethanol volume share in gasoline type fuels (%)		Ethanol-related WTW emission savings in gasoline type fuels (%) (1)		Biodiesel volume share in diesel type fuels (%)		Biodiesel-related WTW emission savings in diesel-type fuels (%) (2)
	Average 2015-17	2030	Average 2015-17	2030	Average 2015-17	2030	Average 2015-17	2030
World	7.7	8.2	2.0	2.3	3.5	3.8	1.6	1.7
North America
Canada	5.6	5.6	0.9	0.9	1.2	2.0	0.4	0.5
United States	9.6	11.8	1.6	2.1	4.0	4.4	2.1	2.3
Latin America and Caribbean
Argentina	10.0	12.1	3.1	4.0	9.0	12.0	3.3	4.4
Brazil	46.6	51.1	33.6	38.3	7.4	10.0	3.4	4.6
Colombia	7.4	7.1	3.3	3.1	6.4	6.2	2.1	2.0
Paraguay	21.8	26.6	9.8	12.4	0.9	1.0	0.3	0.2
Europe
European Union	4.8	4.7	1.0	1.0	6.1	6.1	2.8	2.9
Asia
China (3)	2.0	2.0	0.3	0.4	0.4	0.7	0.3	0.6
India	2.6	2.4	1.0	0.9	0.1	0.1	0.1	0.1
Indonesia	0.1	0.1	0.1	0.1	5.9	8.9	1.9	3.0
Japan	1.4	1.7	0.7	0.8	0	0	0	0
Malaysia	..	..	..	..	2.1	3.2	0.7	1.0
Philippines	9.6	9.5	3.4	3.7	2.1	1.6	0.9	0.7
Thailand	12.8	15.0	6.2	7.5	5.5	6.0	1.8	2.0
Oceania
Australia	1.1	1.2	0.2	0.2	1.7	1.1	0.6	0.8

Note: Not available

1. The WTW ethanol percentage savings were calculated in a given country as the ratio of the difference between the WTW reference emissions that would have been associated with gasoline if ethanol was not used to replace gasoline, and the WTW emissions associated with the mix of ethanol and gasoline use over the total emissions associated with gasoline use in the country.

2. The WTW biodiesel percentage savings were calculated in a given country as the ratio of the difference between the WTW reference emissions that would have been associated with diesel if biodiesel was not used to replace diesel and the WTW emissions associated with the mix of biodiesel and diesel use and over the total emissions associated with diesel use in the country.

3. Refers to mainland only. The economies of Chinese Taipei, Hong Kong, China (China) and Macau, China (China) are included in the Other Asia Pacific aggregate.

Box 5.1. The pathways of biofuels production

Terms commonly used to classify biofuels relate to feedstock used, the production process, and their capacity to reduce GHGs emissions or policies in place. Biofuels derived from food and feed crops are commonly classified as “conventional biofuels” or “first-generation biofuels”. Another classification is “advanced biofuels”. Identical biofuels may be considered advanced in some countries and not in others. The IEA (IEA, 2017_[3]) defines “advanced biofuels” as sustainable fuels produced from non-food crop feedstocks which are capable of delivering significant life-cycle emissions savings compared with fossil fuel alternatives, and which do not directly compete with food and feed crops for agricultural land or cause adverse sustainability impacts. The terms “second-” and “third-generation” biofuels are used without any agreement on their definitions.

It is not intend here to classify biofuels into categories, but to use the IEA framework to focus on biofuel pathways which combine a source of biomass, a production process, and an end product. Figure 5.1 presents the major liquid biofuel pathways in the transport sector and provides an overall view of the sources of the relevant biomass. Depending on the biofuel production process, certain molecules such as lipid or sugar solutions are extracted from the feedstock. In the most recent conversion processes, ligno-cellulosic material is used directly.

Figure 5.1. Major biofuel pathways in the transport sector

Note: A biofuel pathway is the combination of a source of biomass, a conversion process and an end product. 1. Biofuel pathways not modelled endogenously in AGLINK-COSIMO appear in grey.

1. MSW corresponds to Municipal solid Waste.

2. Some companies are also planning to produce biodiesel from recycled plastics.

WTW GHGs emissions savings from the use of biofuels were calculated as the difference between the amounts of GHGs actually emitted by the use of a mix of biofuels and fossil fuels and the emissions that would have arisen from an equivalent amount of energy being supplied by fossil fuels only9. Annex 4.B presents how WTW GHGs emissions are taken into account in the AGLINK-COSIMO modelling framework for the Agricultural Outlook (OECD/FAO, 2018_[1]).

Box 5.2. Biofuel market prospects towards 2030

Main findings from the 2018 OECD-FAO Agricultural Outlook baseline

The biofuel industry is relatively recent, with the volumes consumed becoming significant only in the 1990s. Biofuel policies continue to play a major role on biofuel markets. In 2017, ethanol accounted for 8.1% in volume of global gasoline-type fuels consumed in the road transport sector while biodiesel accounted for 3% in volume of global diesel-type fuels (OECD/FAO, 2018_[1]).1

Consumption is highly concentrated among several key players with 12 countries representing 97% of the biodiesel and ethanol fuel use. The United States and Brazil dominate the ethanol market, representing respectively 50% and 27% of global ethanol production. The European Union and the United States, representing 39% and 19%, respectively, of global volumes, lead world biofuel production.

The OECD-FAO Agricultural Outlook 2018-2027 describes in detail the expected developments on biofuels markets. The Outlook is based on projections established at the 2030 horizon with the AGLINK-COSIMO model (see Annex 5.B for the main features of the biofuel component of the AGLINK-COSIMO model).

The 2018 Outlook assumes a continuation of current policies, although some general policy targets especially in developing countries would not be met due to the absence of the necessary policy instruments to achieve them. The announcement concerning the Chinese E10 program2 and the Brazilian RenovaBio program3 have not been included in the Outlook.

At the global level, ethanol production (including ethanol used for industrial purposes and beverage) is projected to expand to 133 Bln L by 2030 (compared to 120 Bln L in 2017), while biodiesel production is projected to increase to 40.3 Bln L by 2030 (compared to 36 Bln L in 2017). By 2030, 55% of global ethanol production is expected to be based on maize and 27% on sugarcane. About 21% of global biodiesel production is projected to be based on waste vegetable oils and animal fat. Lignocellulosic biofuels are not expected to take off over the projection period as production costs are likely to remain high.

Biofuel trade is likely to remain limited. Potential ethanol exporters are the United States, as constraints associated with vehicle suitability for high blends and fuel distribution infrastructure will likely limit a further increase in domestic demand, and Brazil. On the biodiesel side, Argentina will likely be the major player, but with limited import demand.

1. The OECD-FAO Agricultural Outlook provides ten-year projections for world agricultural markets. However, the model used for projections in the 2018 report has been extended to 2030 for purposes of scenario analysis.

2. In September 2017, the Chinese government proposed a new nationwide ethanol mandate expanding the mandatory use of E10 fuel from 11 trial provinces to the entire country by 2020. The underlying rationale has not been clearly stated but could be related to abundant grains stocks and to environmental concerns. The mechanisms for the implementation and enforcement have not been announced. If fully implemented these policies could have important impacts on biofuel and agricultural markets.

3. The RenovaBio program was officially signed in January 2018 as a follow up to the Brazilian commitment to reduce greenhouse gas emissions by 37% in 2025 and 43% in 2030 compared to 2005. Its implementation plan is not yet defined. The program defines a minimum blending target for anhydrous fuel ethanol that should reach 30% by 2022 and 40% by 2030 as expressed in volume terms. The fuel ethanol share in the fuels matrix should reach 55% by 2030.

At the global level, while the share of ethanol volume in gasoline type fuels is projected to be 8.2% in 2030, ethanol-related WTW emission savings in gasoline-type fuels are estimated at around 2.3%. For biodiesel, the volume share in diesel type fuels is projected to be 3.8% at the global level in 2030 and biodiesel-related WTW emission savings in diesel-type fuels at around 1.7%. The disparities across countries are significant and depend on the blending mandates or targets in place, as well as the type of biofuels used.

Emissions related to land use change

Concerns about the increasing pressure placed on natural resources and land use changes (LUC) effects created by biofuel production and associated GHG emissions arose in the late 2000s, along with other concerns on the sustainability and potential negative impact of biofuels.

Multiple studies since 2009 examine the extent and consequences of LUC (LCFS, 2009[3]) (Laborde, 2011[4]) (De Cara, 2012[5]) (ECOFYS, 2015[6]), (Overmars et al., 2015[7]), (European Parliament and European Council, 2015[8]). The majority use economic models to estimate LUC impacts: they compare land uses in the baseline situation with a scenario that assumes a different path for biofuel demand. Most evaluations find that the development of biofuels leads to changes in land use10 which result in substantial GHGs emission impacts.

In this context, the overall carbon intensity of biofuels would have to take into account two components: the CO2 emitted along the value chain (WTW emissions) and the CO2 emitted because of LUC changes associated with biofuel use (LUC emissions).

Figure 5.2 provides an overview of potential WTW and LUC carbon intensity of the most important biofuel pathways compared to the fossil fuels they replace. Estimates for LUC emission factors were derived from the study commissioned by the European Commission and conducted by the Ecofys, International Institute for Applied Systems Analysis (IIASA), and E4tech consortium in 2015 based on the GLOBIOM economic model (ECOFYS, 2015[6]). It is the most comprehensive study undertaken to date.

With these estimates, the results in terms of GHGs emission savings from the blending of biofuels in conventional transportation fuels up to 2030 based on the Agricultural Outlook baseline are very different from those presented in the previous section.

Figure 5.2. Carbon intensity (WTW and LUC) of different categories of biofuels in kgCO₂e/GJ

Note: LUC emission factors include direct land use changes and indirect land use changes

Source: Literature review undertaken by the OECD (Annex B) for WTW emission factors and Ecofys (2015) for LUC emission factors.

At the global level, calculations based on the Agricultural Outlook baseline quantity projections establish total ethanol-related emission savings in gasoline-type fuels at 0.7% by 2030, compared to 2.3% when only WTW emissions are taken into account. For biodiesel, the picture even shows negative savings (-3.4% in 2030 compared to 1.7% when only WTW emissions are taken into account). This means that the blending of biodiesel in diesel type fuels could lead to an increase in cumulative carbon dioxide emissions throughout the projection period when emissions related to land use changes are taken into account based on the ECOFYS study.

The 2018 version of the AGLINK-COSIMO model includes a GHG component for the agricultural sector,11 which is used in the analysis of agriculture’s potential contribution to climate change mitigation. To date, however, emissions arising from LUC are not included in the AGLINK-COSIMO biofuel component. In a future version of the GHG component developed in collaboration with the International Institute for Applied Systems Analysis (IIASA), it will be possible to assess GHGs emissions associated with LUC.

The present chapter adopts an approach similar to the IEA by focusing on WTW emissions in the transport sector and uses the GHG component of the agricultural part of AGLINK-COSIMO to measure GHGs emissions related to the agricultural sector. In the scenario analysis the assumptions of IEA 2DS regarding crude oil prices, the macroeconomic environment, taxes applied to fuels according to their GHGs profiles, and future demand of gasoline-type and diesel-type fuels are inputted into AGLINK-COSIMO. The resulting emissions are compared to those projected by the IEA and the baseline presented in the Agricultural Outlook 2018-2027, with the latter projections extended to 2030.

Assessing the potential contribution of biofuels in the decarbonisation of the transport sector: Scenario definition

The IEA perspective

The Energy Technology Perspective (ETP) (IEA, 2017[1]) and the Technology Roadmap (IEA, 2017[9]) focus on the opportunities and challenges of scaling up and accelerating the deployment of clean energy technologies in different sectors. The IEA 2DS sets a path for the energy sector at the 2060 horizon that is consistent with a 50% chance of limiting future global average temperature increases to 2°C by 2100. In the 2DS, carbon taxes increase over time, which partly offsets lower fossil fuel prices occurring due to lower demand.12

The 2DS assumes progressive improvements in the following areas: vehicle technical efficiencies, “avoid-shift measures” for passenger cars,13 systemic and logistic efficiency gains in road-freight, and electrification. Both IEA reports see an important role for bioenergy. They emphasise that the future role of bioenergy will need to be contingent on unambiguous and significant carbon savings (and hence rely on a rapid transition to advanced biofuels), and will need to be consistent with improvements in environmental and social sustainability.

In the 2DS, 17% of the energy consumed in 2060 is derived from bioenergy compared to 4.5% in 2015 and bioenergy is responsible for 17% of the cumulative reductions in emissions to 2060. In the transport sector, fossil fuel consumption is sharply reduced and bioenergy would provide 29% of the total transport final energy demand by 2060.

Towards 2030, the IEA sees gasoline demand retracting more than diesel demand; gasoline is mostly used by passenger road vehicles and thus more easily offset.

By 2030, ethanol and biodiesel use in the 2DS is projected to be respectively 40% and 110% higher than in the OECD-FAO baseline projections. The use of conventional biodiesel derived from vegetable oil is set to be phased out in favour of waste-based biofuels for the diesel pool which offer stronger GHGs emission saving. For ethanol, conventional ethanol mainly derived from sugarcane (that has a stronger GHGs reduction profile than ethanol produced from starch feedstocks) is produced at the expense of other food crops. Advanced ethanol, based on agricultural residues or energy crops, is expected to become widely available as of 2025.

Definition of an AGLINK-COSIMO scenario

The IEA 2DS is based on simulations undertaken with the MoMo model,14 a simulation model that uses detailed projections of transport activity and vehicle activity, energy demand, and WTW GHGs and pollutant emissions to 2060 under alternative policy scenarios.

The IEA sees a role for biofuels in transport sector mitigation within a 2060 horizon. As part of its 2DS, it describes a ten-year transition period, where the use of currently available biofuels would increase before it is replaced by more sustainable biofuels with lower carbon emission profiles. However, the IEA did not take into account the interconnection between agricultural and biofuel markets, and the ability of agricultural activities to supply the amount of bioenergy foreseen in its 2DS to meet climate change targets.

The present AGLINK-COSIMO scenario attempts to fill that need and is a further step in an enhanced collaboration between the OECD and IEA to better capture the potential role of biofuels in climate-change mitigation.15

AGLINK-COSIMO has a medium-term horizon with projections until 2030 and a detailed production, use and trade modelling framework (presented in Annex 5.B) for most categories of biofuels currently available on the market with a direct connection to agricultural markets. It is capable of taking into account alternative assumptions than those used to produce the Agricultural Outlook 2018-2027 baseline.

In particular, the AGLINK-COSIMO 2-degree scenario (AC-2DS) is defined using assumptions consistent with the IEA 2DS assumptions. Assumptions are summarised in Table 5.1. They differ from the Outlook baseline assumptions regarding the future evolution of crude oil prices, the macroeconomic environment, carbon taxes applied to fuels according to their GHGs profiles, as well as the future demand of gasoline-type and diesel-type fuels.

A key difference between the AC-2DS and the 2018 Agricultural Outlook assumptions is that the former projects lower transportation fuel demand. This is crucial given that the AGLINK-COSIMO models biofuel demand as a share of transportation fuel demand. This share is defined as the maximum value between a market-driven share and a mandate-driven share. The market-driven share reacts to the price difference between the biofuel and the conventional transportation fuel it replaces. When the relative consumer price of conventional transportation fuel increases compared to that of biofuel, i.e. when biofuel becomes more competitive, the market-driven share increases. Under AC-2DS, carbon taxes are applied according to the WTW GHGs profiles of the different fuels and encourage or discourage the use of specific biofuels in the transportation mix.16

An additional assumption is made to increase the ethanol blend wall to 15% across all countries to allow additional ethanol to be blended with gasoline17 within the framework of the climate change mitigation scenario, similar to what is included in the IEA 2DS.

To date ethanol production from agricultural and forest residues and specific energy crops is not included endogenously in the AGLINK-COSIMO modelling framework. This is due to the fact that current production levels are low as there is a limited number of commercial scale plants (IEA, 2017[9]) and the fact that information on production costs is not widely available. The AC-2DS scenario setup therefore shows the impact of a mitigation scenario in the transport sector where the availability of biofuels based on ligno-cellulosic material remains limited at the level expected in the 2018 OECD-FAO Agricultural Outlook. In future analysis, additional assumptions concerning the development of advanced ethanol production costs could be included.

Table 5.2. AC-2DS main assumptions

		AC-2DS	AC-2DS	AC-2DS	Baseline	% difference of scenarios vs baseline
		2020	2025	2030	2030	2030
Crude oil prices	USD/barrel	87.6	111.9	140.1	79.7	76%
Additional carbon taxes applied to fuels
Expressed in terms of carbon tax equivalent
Gasoline-type fuels	USD/tCo2eq	24.4	54.6	91.7	..	..
Diesel-type fuels	USD/tCo2eq	18.5	41.5	69.8	..	..
Expressed as a WTW emission based tax
Gasoline	USD/hl	6.5	14.6	24.5	..	..
Diesel	USD/hl	6.6	14.8	24.9	..	..
Sugarcane based ethanol	USD/hl	1.3	2.9	4.9	..	..
Maize based ethanol	USD/hl	3.3	7.5	12.5	..	..
Agriculture residues-based ethanol	USD/hl	0.7	1.5	2.5	..	..
Palm oil based biodiesel	USD/hl	4.0	8.9	14.9	..	..
Soybean oil based biodiesel	USD/hl	3.8	8.5	14.2	..	..
Rapeseed oil based biodiesel	USD/hl	3.4	7.5	12.7	..	..
Waste oil based biodiesel	USD/hl	1.2	2.6	4.3	..	..
Demand for transportation fuels in key countries
Gasoline-type fuels
World	Bln l	1 268	1 140	998	1 318	-24%
United States	Bln l	542	455	367	454	-19%
European Union	Bln l	117	94	76	103	-26%
Brazil	Bln l	47	47	45	51	-11%
China	Bln l	173	176	164	229	-28%
India	Bln l	37	42	48	87	-45%
Diesel-type fuels
World	Bln l	996	1 021	1 024	1 047	-2%
United States	Bln l	217	204	186	199	-7%
European Union	Bln l	229	209	186	204	-9%
Brazil	Bln l	52	53	54	56	-3%
China	Bln l	125	132	134	126	6%
Indonesia	Bln l	38	46	52	52	0%
Argentina	Bln l	13	14	14	16	-13%

Note: It is assumed that WTW emission factors associated with biofuels are constant over the period leading to 2030. If technologies associated with conventional biofuels were to change over the medium-tem – due to new conversion processes, better use of co-products, technical innovation – this could well modify downward the WTW emission factors and change the level of carbon taxes assumed in the scenario

Scenario results

All scenario results, unless otherwise specified, are compared to the 2018 OECD-FAO Agricultural Outlook baseline projections, henceforth referred to as the “baseline”.

Biofuel markets

Under AC-2DS, global WTW GHGs emissions in the transport sector are 15% lower by 2030 than under the baseline (Figure 5.3). The most important factor behind this decrease is the reduction in transportation fuel use due to “avoid and shift measures” and vehicle efficiency gains, while biofuels contribute only marginally.18 Figure 5.3 shows a divergence between the evolution of WTW emissions in the transport sector in the IEA 2DS and in the AC-2DS. This is related to a more limited response of biofuel markets to the policy stimuli in the AC-2DS as compared to IEA 2DS. This is explained in detail below.

Figure 5.3. Evolution of WTW emissions in the transport sector in AC-2DS compared to the baseline and to IEA 2DS

Note: WTW emissions in the transport sector are indexed at 1 in 2015.

Source: OECD for the baseline and AC-2DS, IEA for 2DS, https://www.iea.org/topics/transport/.

Whereas in 2030 gasoline and diesel use are projected to be considerably lower under the AC-2DS compared to the baseline, ethanol fuel and biodiesel use would be 0.9% and 1.2% stronger, respectively, under the AC-2DS than under the baseline. Ethanol-related savings of WTW GHG emissions of gasoline-type fuels would reach 3.1% by 2030 (versus 2.3% in the baseline) and biodiesel related savings in diesel-type fuels would reached 1.8% (versus 1.7% in the baseline) (Figure 5.4).

Expressed in terms of the blending of biofuels in conventional fuels, this means that the volume share of ethanol in gasoline-type fuels at the global level would reach 11% by 2030 in the AC-2DS (versus 8.2% in the baseline) and that the volume share of biodiesel in diesel-type fuels would reach 4% in the AC-2DS (versus 3.8% in the baseline).

Figure 5.5 compares biofuel blending shares between the baseline, the AC-2DS and the IEA 2DS. The development of biofuel blending in transportation fuels is less pronounced in the AC-2DS. The IEA 2DS expects a strong development of sugar cane based ethanol production in the period leading up to 2025 (+130% at the expense of maize based ethanol production) and then an uptake of advanced ethanol production. For biodiesel, the IEA 2DS sees an important increase of vegetable oil based biodiesel in the period leading to 2025 (+39%) and then a take-off of waste-oil and animal-fat based biodiesel and other types of biofuels used for the diesel pool (such as synthetic fuels or animal fats). The AC-2DS foresees much lower biofuel use growth over the period leading to 2030.

Under IEA 2DS and AC-2DS, biofuel use is promoted by taxes applied to fuels according to their GHGs emission profiles. However, in AC-2DS, the production of conventional biofuels (such as sugarcane-based ethanol or vegetable-oil based biodiesel) is constrained by the availability of agricultural feedstock and the potential of agricultural markets to supply more feedstock for biofuels while meeting demand for food and feed. A doubling of sugarcane-based ethanol use as forecast in IEA-2DS would produce a very strong shock on sugar markets and most probably important land use impacts.19

Figure 5.4. Biofuel use and GHGWTW savings, 2030

* Percentage change with respect to the baseline.

Source: OECD, based on Aglink-Cosimo simulations.

Figure 5.5. Comparison of biofuel blending shares in volume, 2015 and 2030

Source: OECD, based on AGLINK-COSIMO simulations and IEA (2017).

Figure 5.6 provides an overview of major AC-2DS results in terms of biofuel blending shares and biofuel use at the country level. In all countries, the ethanol and biodiesel blending shares are higher under AC-2DS than under the baseline, as the taxes applied to the different fuels according to their WTW GHGs profiles and the assumed developments of crude oil prices decrease the price ratio between biofuels and conventional fuels, thus encouraging the market-driven use of biofuels.20

Figure 5.6. Changes in biofuel blending share and fuel use by 2030 for major countries

Note: The percentage change in biofuel fuel use is calculated as the change between the AC-2DS results and the baseline value.

Source: OECD based on Aglink-Cosimo simulations.

Biofuel use does not exceed mandates in the United States; those mandates, expressed in volume terms, are kept at the same volumes as the baseline case (with lower gasoline and diesel use). The assumption of a blend wall gradually increasing to 15% implies that some of the advanced mandate can be met with sugarcane-based ethanol.

In the absence of strong nation-wide ethanol mandate, the market-driven effect is particularly strong in the People’s Republic of China (hereafter “China”) where the ethanol share in gasoline-type fuels doubles to 4.1% by 2030.21 Further increases in ethanol blending is constrained by this country’s ethanol production capacity and the strong domestic demand in major ethanol-producing countries. In Brazil, the use of hydrous ethanol (pure ethanol that can be used by flex-fuel cars) strongly increases in response to the carbon tax stimulus.

The shares of ethanol produced from maize and sugarcane remains relatively stable when compared to the baseline at 51% and 30% respectively. As described above, the current model does not allow the take-off of advanced ethanol based on agricultural residues or energy crops production in the medium term. The picture differs for biodiesel where waste oil-based biodiesel production is 24% stronger than in the baseline, at the expense of vegetable oil based biodiesel.

There continues to be little trade of biofuels compared to global production levels as the biofuel policies in place22 and the taxes applied to fuels based on WTW emissions mostly encourage the consumption of domestically produced biofuels. However, the trade share of biofuels with lower GHGs emission profiles in total biofuel trade increases strongly (+25% for sugarcane based ethanol and +75% for waste oil and animal fat-based biodiesel when compared to the baseline).

Agricultural markets

The OECD-FAO Agricultural Outlook 2018-2027 (OECD/FAO, 2018[2]) reports an alternative scenario to the baseline where crude oil prices and macro-economic assumptions would follow a similar path to what was included in the AC-2DS (see Chapter 1). It highlights that higher oil prices increase agricultural production costs through higher prices for fuel and fertiliser, as well as through general cost increases induced by higher inflation. They can also affect demand for agricultural commodities through biofuels markets. This is also the case in the AC-2DS.

In addition, in the AC-2DS, global demand for agricultural commodities used as biofuel feedstock is affected by the taxes applied to the different fuels according to their WTW GHGs profiles and by the different crude oil price and macroeconomic assumptions. Global maize and sugar cane use for biofuels are supposed to increase by 1% and 0.3%, respectively, by 2030 when compared to the baseline while the demand for vegetable oil to be used for biofuels would be 4% lower than in the baseline (due to the development of waste-based biodiesel).

The overall effects on emissions and food security indicators are presented in Figure 5.7, using the same three indicators as in Chapter 3.

• The Calorie Availability Index represents the average amount of calories available per capita in each country for the subset of the food basket represented in the model.

• The Consumer Food Price Index is calculated as a fixed weight index of the national consumer prices in real terms. The food consumption quantities of 2015 are used as weights. Higher consumer prices are assumed to lead to lower access to food for parts of the population.

• The Agricultural Income Index is calculated as a fixed weight index of a combination of producer prices, subsidies and a cost index. As weights, the production quantities in 2015 are used. This indicator can be used in countries where the agricultural sector is a large contributor to the GDP.

The AC-2DS would imply by 2030 a stronger consumer food price index by about 1.4% when compared to the baseline, while the calorie availability index would remain stable. The agricultural income index would decrease by 1.9%, reflecting higher production costs. Agricultural-related emissions would decrease by 0.15%.

Overall effects of the AC-2DS on agricultural markets are relatively moderate as the increase in biofuel use when expressed, as a share of conventional fuel use does not lead to a strong increase in demand for agricultural feedstock. This is due to the lower demand for gasoline and diesel in the medium term.

Figure 5.7. Impact of the AC-2DS scenario on agricultural markets

Source: OECD based on AGLINK-COSIMO simulations.

Summary of main findings

The main results of the AC-2DS are as follows.

• WTW GHGs emissions in the transport sector would be 16% lower by 2030 than compared to the baseline (Figure 5.3). This is mostly related to the assumption of slower growth of gasoline and diesel use up to 2030.

• The volume share of ethanol in gasoline-type fuels would reach 11% by 2030 and that of biodiesel in diesel-type fuels would reach 4% (versus 8.1% and 3.8% respectively in the Outlook baseline) (Figure 5.5).

• The assumptions regarding the future evolution of crude oil prices coupled with the taxes applied to the different fuels according to their WTW GHGs profiles decrease the price ratio between biofuels and conventional fuels. Market-driven biofuel use is thus encouraged and mandates are not binding in most countries, as is the case in the baseline (Figure 5.6).

• Waste oil-based biodiesel production is set to develop strongly (+24%) by 2030 when compared to the baseline due to its lower WTW GHGs profile as opposed to vegetable oil-based biodiesel.

• The trade share of biofuels with lower GHGs emission profiles, such as sugarcane-based ethanol and waste oil-based biodiesel, in total biofuel trade will increase by 25% and 75%, respectively, by 2030 when compared to the baseline.

• The AC-2DS foresees only moderate increases in the volumes of agricultural feedstock used to produce biofuels despite stronger blending shares of biofuels in volume terms.

• Developments in terms of volume shares are lower than those foreseen in the IEA 2DS. In the AC-2DS, the production of biofuels based on agricultural feedstock (such as sugarcane-based ethanol or vegetable oil-based biodiesel) is constrained by the availability of feedstock and rising agricultural production costs.

• The impact on agricultural markets will be relatively small with a stronger food consumer price index of about 1.4% and a lower agricultural income index by about 1.9% when compared to the baseline. The higher oil prices assumed in the AC-2DS increase agricultural production costs (Figure 5.7).

• The current AGLINK-COSIMO model cannot analyse the effects in terms of global land use, but this should be possible in future work.

Conclusions

This chapter presents the results of a quantitative analysis based on the AGLINK-COSIMO model of a climate change mitigation scenario for the transport sector, namely the AC-2DS; the decarbonisation of the transport sector being an often-stated argument behind biofuel policies.23 The analysis shows that the role of biofuels in the transport sector on climate change mitigation depends, in part, on the ability of the agricultural sector to provide in the medium-term agricultural feedstock to produce biofuels and on the set of policy incentives.

Gasoline substitutes classified as “advanced” according to the IEA definition (Boxes 5.1 and 5.2) are likely to be increasingly produced on an industrial scale. To date, however, the biofuel with the lowest WTW GHGs emission profile able to replace gasoline is sugarcane-based ethanol. The same potential is not seen for the further development in global use of sugarcane-based ethanol in the medium term, as opposed to the IEA. This is due to the position of Brazil as a major supplier of sugar, its own strong use of ethanol, as well as constraints related to the expansion of sugarcane production in the country. In addition, the AC-2DS carbon tax differential between sugarcane-based ethanol and maize-based ethanol is not high enough to promote a massive shift towards the use of sugarcane-based ethanol in the United States, a strong user of ethanol.

On the biodiesel side, waste-based biodiesel offers important WTW GHGs savings when compared to vegetable oil-based biodiesel. The use of waste oil-and animal fat based biodiesel is widespread in the United States and the European Union, constrained mostly by the ability to collect and recycle vegetable oil and animal fats. Production costs are similar to those of vegetable oil-based biodiesel and the policy incentives implemented in AC-2DS lead to an expansion of its production by almost 30%. Further growth is possible, but it could be supported more widely across the globe with other policy incentives, such as targets for the recycling industry and the implementation of traceability measures.

Simulations using the AGLINK-COSIMO model suggest the policy incentives described in the IEA 2DS (mostly carbon taxes according to WTW emissions) may not be sufficient to elicit the expected response in terms of the production of biofuels. Given such policies and constraints on feedstock supplies, biofuels derived from food and feed feedstock have no more than a minor role to play in delivering climate change mitigation from the agricultural sector. A substantially increased role of biofuels in the decarbonisation of the transport sector would require a different set of policy incentives that would need to be cost-effective and take into account effects on food security and the sustainable use of resources.

At present, the AGLINK-COSIMO model cannot evaluate GHGs emissions associated with LUC in the agricultural sector, but this should be possible soon. This would allow for a new scenario where carbon taxes would be applied to the different fuels according to their total emissions (WTW + LUC).

References

[10] Allwood, J. et al. (2014), “I ANNEX Glossary, Acronyms and Chemical Symbols Glossary Editors: Glossary Contributors”, https://www.ipcc.ch/pdf/assessment-report/ar5/wg3/ipcc_wg3_ar5_annex-i.pdf (accessed on 15 March 2018).

[5] De Cara, E. (2012), Revue critique des études évaluant l’effet des changements d’affectation des ... – ADEME, http://www.ademe.fr/revue-critique-etudes-evaluant-leffet-changements-daffectation-sols-bilans-environnementaux-biocarburants (accessed on 14 March 2018).

[6] ECOFYS (2015), The land use change impact of biofuels consumed in the EU, https://ec.europa.eu/energy/sites/ener/files/documents/Final%20Report_GLOBIOM_publication.pdf (accessed on 15 March 2018).

[8] European Parliament and European Council (2015), Directive (EU) 2015/1513 of the European Parliament and of the Council of 9 September 2015 amending Directive 98/70/EC relating to the quality of petrol and diesel fuels and amending Directive 2009/28/EC on the promotion of the use of energy from renewable sources (Text with EEA relevance) - EU Law, https://publications.europa.eu/en/publication-detail/-/publication/8671e480-5b6a-11e5-afbf-01aa75ed71a1/language-en (accessed on 8 February 2018).

[9] IEA (2017), Delivering Sustainable Bioenergy, IEA Technology Roadmaps, IEA, Paris, https://dx.doi.org/10.1787/9789264287600-en.

[11] IEA (2017), Delivering Sustainable Bioenergy, IEA Technology Roadmaps, IEA, Paris, http://dx.doi.org/10.1787/9789264287600-en.

[1] IEA (2017), Energy Technology Perspectives 2017: Catalysing Energy Technology Transformations, IEA, Paris, https://dx.doi.org/10.1787/energy_tech-2017-en.

[4] Laborde, D. (2011), Assessing the Land Use Change Consequences of European Biofuel Policies, http://trade.ec.europa.eu/doclib/docs/2011/october/tradoc_148289.pdf (accessed on 15 March 2018).

[3] LCFS (2009), CALIFORNIA’S LOW CARBON FUEL STANDARD : FINAL STATEMENT OF REASONS, https://www.arb.ca.gov/regact/2009/lcfs09/lcfsfsor.pdf (accessed on 15 March 2018).

[14] OECD (2018), “Taxing Energy Use 2018 COMPANION TO THE TAXING ENERGY USE DATABASE”, https://one.oecd.org/document/23201804/en/pdf (accessed on 26 February 2018).

[12] OECD (2008), Biofuel support policies : an economic assessment., OECD.

[2] OECD/FAO (2018), OECD-FAO Agricultural Outlook 2018-2027, OECD Publishing, Paris, http://dx.doi.org/10.1787/agr_outlook-2018-en.

[7] Overmars, K. et al. (2015), Estimates of indirect land use change from biofuels based on historical data., Publications Office, https://ec.europa.eu/jrc/en/publication/eur-scientific-and-technical-research-reports/estimates-indirect-land-use-change-biofuels-based-historical-data (accessed on 15 March 2018).

[13] Renewable Energy Agency, I. (2018), Biogas for Road Vehicles: Technology brief, https://www.irena.org/-/media/Files/IRENA/Agency/Publication/2017/Mar/IRENA_Biogas_for_Road_Vehicles_2017.pdf?la=en&hash=9261CA2381C7847A515E230D03C9487AE4392B88 (accessed on 18 December 2018).

Annex 5.A. Literature review on WTW emissions

Annex Table 5.A.1. Summary of the literature review

	CARB, 2009	EU, RED 2009	JEC, 2014	Havlik, 2010	Koga et al, 2010	Nguyen et al, 2007	Hoefnagels et al, 2010
Presentation and context
Title	California's Low Carbon Fuel Standard - Final Statement Of Reasons		WTT report, version 4a and Appendix 4	Global land-use implications of first and second generation biofuel targets	Assessing energy efficiencies and greenhouse gas emissions under bioethanol-oriented paddy rice production in northern Japan	Energy balance and GHG-abatement cost of cassava utilization for fuel ethanol in Thailand	GHG footprint biofuels
Publication date	December 2009	2009	2014	2010	2010	2007	2010
Links	CARB, California's LCFS Final Statement Of Reasons, 2009.pdf	http://eur-lex.europa.eu/eli/dir/2009/28/oj	http://iet.jrc.ec.europa.eu/about-jec/sites/iet.jrc.ec.europa.eu.about-jec/files/documents/report_2014/wtt_report_v4a.pdf ; http://iet.jrc.ec.europa.eu/about-jec/sites/iet.jrc.ec.europa.eu.about-jec/files/documents/report_2013/wtt_appendix_4_v4_july_2013_final.pdf	http://users.auth.gr/~akontses/%CE%A4%CE%A0%CE%A0_biofuels/Global%20land-use%20implications%20of%20first%20and%20second%20generation%20biofuel%20targets.pdf	http://www.pubfacts.com/detail/21126818/Assessing-energy-efficiencies-and-greenhouse-gas-emissions-under-bioethanol-oriented-paddy-rice-prod
Focus/scope		Default values for biofuels and bioliquids. Total for cultivation, processing, transport and distribution. "el" (annualised emissions from carbon stock changes caused by land-use change) are excluded. /!\ these are "default" values", whereas "actual" values must be used (actual values include notably emissions from land use)	EU crops (except Soya and palm oil which are imported)	Lifecycle GHG savings from substitution of fuels by biofuels, without LUC related emissions.	Ethanol production from rice (Japan)	Ethanol production from cassava (Thailand)	GHG emissions from biofuel production, No LUC. Allocation of co-products by EU default (energy allocation for co-products, subtraction for co-generation of electricity and heat), energy, mass and market value. JRC DNDC model for N2O emissions from sugar cane, wheat, sugar beet, maize and rapeseed. IPCC model for miscanthus, palm fruit, soy beans, switchgrass, eucalyptus and jatropha.
Remarks	California averages			Reference: - for ET from woody biomass: EF of gasoline = 85.9 - for other biofuels: EF of gasoline = 85, diesel = 86			- Includes co-products. To account for co-products, both allocation (by market value) and system expansion approach are applied. Results with allocation by energy content of mass are also available. - Includes emissions from transport to the EU
		Annex V			Figure 2 and Table 1	Table 4	Figure 2 & Appendix B

Source: OECD literature review.

Annex Table 5.A.2. Comparison of WTW emission factors

	CARB, 2009	EU, RED (2009)	JEC, 2014	Havlik, 2010	Koga et al, 2010	Nguyen et al, 2007	Hoefnagels et al, 2010	Average
WTW emissions (gCO2eq/MJ)
Conventional ethanol
Wheat ethanol		70	69.4					69.7
Maize ethanol	65.66		80.3	49.42				65.1
Barley ethanol			76					76
Rye ethanol			76					76
Sugar beet ethanol		40	40.3					40.2
Sugar cane ethanol	27.4	24	24.8 (excess bagasse used for electricity production)	25.01				25.3
Sweet sorghum ethanol
Rice ethanol					66.3			66.3
Cassava ethanol						45.9		45.9
Conventional biodiesel
Sunflower oil biodiesel		41	45.9					43.5
Palm oil biodiesel		68	50.8				52.5	59.4
Rapeseed oil biodiesel		52	53.9	44.82			45.7	50.2
Soybean oil biodiesel		58	55.1	47.21			52.4 (Brazil), 59.2 (US)	54
Canola oil biodiesel							51	51
Jatropha oil biodiesel							43	43
Biodiesel from waste vegetable oil (UCO)		14	13.8					13.9
Biodiesel from animal fats		14	26.3					20.2
HVO
HVO from rapeseed		44	50.2					47.1
HVO from sunflower		32	44.8					38.4
HVO from palm oil		62	48.6					55.3
HVO from soybean			55.1 (imported soy to the EU)					55.1
HVO from UCO			8.1					8.1
HVO from animal fat			24.5					24.5
ETBE
The part from renewable sources of ETBE		Equal to that of the ethanol production pathway used
Advanced biofuels
Cereal straw ethanol			9.2					9.2
Wheat straw ethanol		13						13
Woody biomass ethanol				22.8				22.8
Switchgrass ethanol (herbaceous)							24	24
Miscanthus ethanol (herbaceous)							17.8	17.8
Eucalyptus ethanol (woody)							4.9	4.9
Farmed wood ethanol		25	22.8					23.9
Waste wood ethanol		22	19.5					20.8
Switchgrass FT diesel (herbaceous)							13.9	13.9
Miscanthus FT diesel (herbaceous)							10.3	10.3
Eucalyptus FT diesel (woody)							7.7	7.7
SRP FT diesel
Forest residues FT diesel
Farmed wood FT diesel		6	7					6.5
Waste wood FT diesel		4						4
Paper & pulp industry waste FT diesel ("waste wood via black liquor")			2.5					2.5

Source: OECD literature review.

Annex 5.B. An overview of the AGLINK-COSIMO biofuel model

The biofuels component of the AGLINK-COSIMO model is a structural partial equilibrium economic model that analyses the world supply and demand of biofuels. The biofuels module, similar to other components of the AGLINK-COSIMO model, is recursive and dynamic. It simulates annual market balances and prices for the production, consumption and traded quantity of ethanol and biodiesel worldwide.

This biofuel model is completely integrated with the cereals, oilseeds and sugar components of the AGLINK-COSIMO model and produces the baseline presented in the annual OECD-FAO Agricultural Outlook. The production of biofuels drives the additional demand for agricultural commodities, in particular for coarse grains, vegetable oil, and sugar.

The AGLINK-COSIMO model has been adapted to explore the environmental impacts of biofuels use in the medium-term. An add-in module to the biofuel model was developed to assess well-to-wheels (WTW) emissions associated with biofuels whereas land-use-changes (LUC) emissions are assessed with the GHG add-in to the global AGLINK-COSIMO model. It is thus possible to compare total GHGs emissions associated with the baseline and different alternative scenarios.

A major update of the AGLINK-COSIMO biofuels (BFL) module was undertaken between 2016 and 2018. In particular, it included:

• The full revision of the model with the introduction of a template for the following countries: Argentina, Australia, Brazil, Canada, China, European Union, Japan, Korea, Mexico, New Zealand, Norway, Russian Federation, Switzerland, United States, Colombia, Chile, India, Indonesia, Malaysia, Paraguay, Philippines and Thailand). Ethanol and biodiesel are modelled separately for each country. The new biofuels module also includes a separate demand and supply for ethanol and biodiesel for the Rest of the World (ROW).

• The introduction of separate fossil fuel (gasoline and diesel) demand equations for each of these countries.

• The modelling of high blend substitute use of ethanol to gasoline and biodiesel use to diesel.

• A revisit of the linkage between the BFL module and other component of the AGLINK-COSIMO model.

• The development of an add-in that allows to calculate WTW GHGs emission and savings derived directly from the use of the biofuels.

• The development of endogenous production functions for biodiesel based on used cooking oil and tallow. This implies to establish price linkages between used cooking oil and vegetable oil.

The documentation on the AGLINK-COSIMO biofuel model is available on www.agri-outlook.org. This annex provides an overview of the main features of the BFL module. In addition to generating baseline outlook reports, the BFL module is used to stimulate policies in the context of global climate change mitigation scenarios.

Significant improvement has been made to the supply side of the biofuels module. The most important was reinforcing the link between the BFL module and other components of the AGLINK-COSIMO model.

Annex Figure 5.B.1. Linkage between the energy, biofuels and agricultural

The production of biofuels (ethanol and biodiesel) derived from each type of feedstock (FS^BF_q,i,c,t) are modelled separately, using the following equation:

logFS^BF_q,i,c,t = ν₀ + ν₁*logRM^BF_q,i,c,t + ν₂* logRM^BF_q,i,c,t-1 + ν₃* logRM^BF_q,i,c,t-2 + ν₄* logRM^BF_q,i,c,t-3 + ν₅*logFS^ET_q,i,c,t-1 + ε^BF_q,i,c,t (4)

where

RM^BF_q,i,c,t = the profit derived from utilising i^th feedstocks for the production of biofuels (ethanol or biodiesel) to be blend in corresponding q type of fossil fuel (petroleum or diesel) in country c and year t.

The RM^ET_q,i,c,t is derived based on the following identity:

RM^BF_q,i,c,t = PP^BF_q,i,c,t + DP^BF_q,i,c,t + VL^BF_q,i,c,t (5)

PI^BF_q,i,c,t

where

PP^BF_q,i,c,t = the biofuels (ethanol or biodiesel) producer price in country c in year t

DP^BF_q,i,c,t = direct government support for biofuels (ethanol or biodiesel) production tied with the use of feedstock i in country c and year t

VL^BF_q,i,c,t = the value of by-products derived from the use of feedstock i in biofuels (ethanol or biodiesel) production in country c and year t

PI^BF_q,i,c,t = production cost index associated with the use of feedstock i in the production of biofuels (ethanol or biodiesel) in country c and in year t

ε^BF_q,i,c,t = the corresponding error term.

PI^BF_q,i,c,t depends on the i^th feedstocks producers price (PPⁱ_c,t) as shown below:

PI^BF_q,i,c,t = f(PPⁱ_c,t) (6)

The purpose of including separate production functions specific to feedstocks is to track changes in greenhouse gas emission, and the direct and indirect consequences of changes in land use due to the use of a variety of food and non-food-based feedstocks in the production of biofuels.

An additional add-in to the biofuel component was developed to calculate GHGs emissions associated with the consumption of biofuels. A review of WTW biofuel emission factors was undertaken in the scope of this project. These emission factors are applied to biofuel use to obtain an estimate of WTW emissions associated with biofuels.

To be able to make that calculation, however, it is necessary to assess the consumption volumes of biofuel by each feedstock types used to produce biofuels on a country basis. This can be done residually if biofuel imports and exports are allocated to the different types of biofuel feedstocks that were used to produce them; this information is known for production as described above. This allocation is made on the assumption that a country’s biofuel export can be divided in exports produced from different feedstocks in the same proportion as its domestic production of biofuels. It is then possible to calculate world exports and also imports by feedstock.

For imports at the country level, the add-in assumes that import shares for major importers (i.e. the European Union and the United States for biodiesel and Canada, and Japan and the United States for ethanol) are fixed and use industry information or US GAINS report to quantify them. For less important countries, it is assumed that national biofuel imports can be split in the same proportion as the world biofuel imports minus the import from the three major importers. It is important to note that biofuel trade remains limited anyway in the baseline projection.

For a given country, WTW emissions associated with ethanol and biodiesel can then be quantified. It is possible to calculate the WTW GHGs emission savings induced by the replacement of conventional fuels by biofuels by comparing the total level of WTW emissions associated by the mix of conventional fuels and biofuel (for example the mix of gasoline and ethanol) in a given country with the emissions that would have occurred if only conventional fossil-based fuels (for the same total energy content) were used.

This can be modelled for a specific country in a given year as:

${\stackrel{-}{GHG\_WTW}}_{BF}=\sum _f{GHG\_WTW}_{{BF}_f}\times V_{{BF}_f}$ (7)

${\stackrel{-}{GHG}}_{transfuel,sub}={GHG}_{transfuel}\times V_{transfuel, sub}$ (8)

${\stackrel{-}{GHG}}_{transfuel,cons}={GHG}_{transfuel}\times V_{transfuel, cons}$ (9)

It is thus possible to derive the emissions savings expressed in percentage as:

$WTW\mathrm{ }GHG\mathrm{ }savings=\frac{{\stackrel{-}{GHG\_WTW}}_{BF}-{\stackrel{-}{GHG}}_{transfuel,\mathrm{ }sub}}{{\stackrel{-}{GHG}}_{transfuel,cons}}$ (10)

Where

$V_{{BF}_f}$ is the volume of biofuels (ethanol or biodiesel) consumed which was produced from feedstock of type f

${GHG}_{transfuel}$ is the emission factor of transportation fuels (gasoline or diesel)

$V_{transfuel,sub}$ is the volume of transportation fuels (gasoline or diesel) substituted by biofuels (ethanol or biodiesel)

$V_{transfuel,sub}=V_{BF}\times {\epsilon }_{transfuel,BF}$ (11)

$V_{transfuel,cons}$ is the volume of transportation fuels of gasoline- or diesel- type consumed

${\epsilon }_{transfuel,BF}$is the energy content ratio between biofuels (ethanol or biodiesel) and transportation fuels (gasoline or diesel)

${\epsilon }_{transfuel,BF}=\frac{\left[GJ/m^{3}BF\right]}{\left[GJ/m^{3}transfuel\right]}$

Notes

← 1.  The main features of the AGLINK-COSIMO model are presented in Annex 5.B.

← 2.  Over the coming decade, the level of food- and feed-based biofuel use in the IEA 2DS is related to their comparative costs (including carbon taxes) compared to the cost of conventional fuels.

← 3.  This chapter does not deal with climate change mitigation in the agricultural sector and does not assume any change in agricultural production costs related to climate change in the medium term compared to the baseline scenario presented in the 2018 Agricultural Outlook.

← 4.  This analysis and more generally all previous analysis undertaken by OECD have focused on liquid fuels produced from biomass (OECD/FAO, 2018[2]), (OECD, 2008[12])). Biogas – gaseous fuels produced from biomass – are not included in the presented analysis principally due to the lack of information on their current development; however, their potential for transport vehicles is clear (Renewable Energy Agency, 2018[13]).

← 5.  Other types of biofuels exist on the market such as jet fuels, a drop-in fuel with the same characteristic as kerosene.

← 6.  Plant growth may lead to GHG emissions despite GHG capture by the plant itself due, for example, to the use of fertilisers.

← 7.  The development of Carbon Capture and Storage (CCS) or Carbon Capture and Utilisation (CCU) could be associated to biofuel production processes to lower CO₂ emissions. CCS and CCU are presently at the prototype or demonstration stages (IEA, 2017[9]).

← 8.  The Fifth Assessment Report of the IPCC defines an emission factor as "the emissions released per unit of activity” (Allwood et al., 2014[10]). The units of activity considered in this study correspond to the consumption of a certain amount of biofuel, expressed in volume or in energy content.

← 9.  Based on the literature, it is calculated to be 2674 gCO₂/l for gasoline and 3208 gCO₂/l for diesel.

← 10.  For example, the conversion of existing cropland to the cultivation of biofuel feedstocks could lead to the expansion of cropland on natural land elsewhere for food production.

← 11.  This component covers GHGs emissions from the following activities: enteric fermentation (linked to ruminant production systems), manure management, rice cultivation, synthetic fertilisers, manure applied to soils, and manure left on pasture. It does not include emissions arising from the burning of crop residues and savannah, from the cultivation of organic soils, nor from crop residues.

← 12.  Carbon taxes applied to transportation fuels according to their GHG profiles are in use in several countries. Nevertheless, most of the taxes applied to transportation fuels take the form of excise taxes, with often the lower rates applied to biofuels (OECD, 2018[14]).

← 13.  “Avoid-shift” measures for passenger cars correspond to measures that promote a decrease in the demand for passenger-car transport and the length of trips, and transport modes that induce fewer emissions.

← 14.  https://www.iea.org/etp/etpmodel/transport/.

← 15.  In 2017, the OECD collaborated with the IEA to review biofuel production costs to be implemented in their respective models (AGLINK-COSIMO and MoMo models).

← 16.  The scenario would yield different results if taxes were applied according to a GHG profile that included LUC-associated emissions.

← 17.  This 15% constraint applies to all countries except Brazil, Paraguay and Thailand, where the use of high-blend ethanol has been developed.

← 18.  In 2030, biofuels use in AC-2DS are behind about one eighth of the WTW GHGs emissions decrease when compared to the baseline. This share is calculated by comparing the WTW GHGs emissions in AC-2DS and in a scenario equivalent in terms of energy demand for the transport sector where only conventional fossil fuels would be used with baseline WTW GHGs emissions.

← 19.  The impact of a shock of this magnitude has not been evaluated with the AGLINK-COSIMO model. The AC-2DS foresees developments on the Brazilian ethanol market that are in line with the RenovaBio program announcements (a fuel ethanol share of 55% by 2030).

← 20.  The market-driven use of biofuel develops when the relative consumer price of conventional transportation fuel increases more than that of biofuel. It occurs either through an increase in high-blend biofuel use or an increase in the share of biofuels being blended in the transportation fuel mix.

← 21.  The baseline does not take into account the 2017 Chinese announcement concerning an E10 mandate (see the biofuel chapter of the OECD-FAO Agricultural Outlook 2018-2027).

← 22.  The AC-2DS assumes a continuation of biofuel policies that are currently in place.

← 23.  There are different types of policy motivations behind the implementation of biofuel policies including climate change mitigation, energy security, rural development, and agricultural market support.