South African Biofuels - Policy Brief

Foreword Due to many countries having signed the Kyoto protocol, the production of biofuels, bioethanol and biodiesel in particular, is on the rise. South Africa has in the past experienced huge maize surplusses as well as low producer prices and this together with the SA government, being a signatory to the protocol, has been partially responsible for the recent publishing of the draft strategy for the development of a biofuels industry within the country. The successful establishment of the biofuels industry involves the dynamic interaction between the field crops, livestock and fuel industries with all the inherent risks and uncertainties of each industry. The purpose of this policy brief is to explore some of the impacts of the biofuels draft strategy that has recently been published by government and to provide guidelines for the formulation of an alternative biofuel strategy that will assist government to reach its targets of renewable energy, growth in the agricultural sector, and the creation of job opportunities. The Bureau for Food and Agricultural Policy (BFAP) acknowledges the financial assistance of the NAMC, ABSA, Maize Trust, Eskort and THRIP. In the past few years BFAP has undertaken extensive research on this topic and the purpose of this policy brief is to clearly indicate what the possible effects of implementing such a draft strategy might be on specific sectors within the agricultural sector. BFAP further analyses the overall impact on the gross value of agriculture and the possible effect on the labour market. A set of policy recommendations are made and the effects of an alternative scenario, where the policies have a more attractive spin-off effect on the sector, are simulated and the results illustrated. The scenarios are compared to the latest BFAP baseline projections, which are generated under a set of basic assumptions regarding certain macro-economic variables, policies and weather patterns. Therefore, the results that are presented in this brief serve as a benchmark. The modelling results of the scenarios for the draft and alternative strategy are constructed and presented in such a way that the decision-maker can form a picture of possible future changes and what their likely effects could be. The results of this document accentuate the importance of government policies in order to facilitate the successful establishment of the biofuel industry. It is therefore hoped that this document will contribute towards assisting the various government agencies in designing policies that will ensure the viability of a biofuels industry and at the same time create employment opportunities as well as viable alternative markets for agricultural products without compromising food security. Bureau for Food and Agricultural Policy University of Pretoria.

1

Executive summary The purpose of this policy brief is to explore some of the impacts of the biofuels draft strategy that has recently been published by government and to provide guidelines for the formulation of an alternative biofuel strategy that will assist government to reach its targets of renewable energy, growth in the agricultural sector, and the creation of job opportunities. This could have major positive spin-off effects for the rural economies and improve the livelihood of the poorest of poor. The BFAP team was able to draw on the resources of the BFAP modelling framework that has recently been updated with a complete system that has the ability to model the impact of a biofuel industry on the agricultural sector. Important to note is that the modelling system is based on a partial equilibrium framework where an equilibrium is dynamically simulated over time. The first section of the report provides a brief overview of the draft strategy document and presents the series of assumptions about the general economy, agricultural policies, the weather, and technological change, which are necessary to simulate a baseline with the BFAP model. The baseline can also be regarded as a “reference scenario” that is used to benchmark/compare any shocks that a biofuel industry could have on the agricultural sector under a certain combination of policies and support measures. In the second section the draft strategy is summarized in the form of a possible scenario. This scenario is presented on page 9 of the report. The scenario is simulated in the BFAP model and the deviations from the baseline are presented. In the third section an alternative scenario is presented (page 15). The alternative scenario is also simulated in the BFAP model and compared to the baseline. The purpose of this scenario is to highlight some of the critical drivers that need to be taken into consideration if a successful biofuel industry is to be established. The combination of policies and support measures that are used in the alternative scenario are based on international experiences and basic local requirements to make the industry feasible. The fourth section presents the main findings and recommendations to guide the process of designing optimal policy solutions that will ensure the viability of a biofuels industry and at the same time create employment opportunities as well as viable alternative markets for agricultural products without compromising food security especially that of the poor. The modelling results and main findings for each of the sections in the report are summarized as follows: Baseline • Under baseline assumptions and market conditions, it becomes clear that no commercial crop will yield positive plant profits from the production of biofuels without government support. The potential plant profits are presented on page 7. • The main reason for negative plant profits under the baseline conditions is the unfavourable ratio of feedstock prices relative to fuel prices. Scenario 1: Draft scenario • Only very little biofuel production will take place in SA and the majority of biofuel will be imported. • Importing biofuels on large scale in order to achieve the renewable energy goals will have a negative impact on the balance of payments. • The overall impact of the biofuel industry under the draft scenario on the contribution of the agricultural sector to the economy (gross value added) is expected to be modest. The percentage change from the baseline remains fairly constant until 2011 and increases thereafter on average by 1.6% until 2015.

2

Executive summary • Approximately 5000 additional jobs will be generated in the primary agricultural sector. • Bioethanol from maize will account for most of the local production. Bioethanol from sugar only plays a small role later on as sugarcane prices become more competitive compared to maize. Approximately 200 million litres of bioethanol will be produced locally and 80% of domestic consumption will have to be imported. • The production of biodiesel from soybeans and sunflower is marginal and only some on-farm production for own use will take place. The majority (90%) of the biodiesel will be imported. • The draft strategy presents a set of policies that are not optimal to reach the targets with respect to renewable energy, growth in the agricultural sector, job creation and uplifting of the rural economy. Scenario 2: Alternative scenario • Most of the bioethanol is produced locally to make up the E8 blend and less than 10% of local requirements are imported. The local production of biodiesel accounts for approximately 45% of local requirements. • Under the alternative scenario the biofuel industry boost the agricultural sector’s contribution to the economy (gross value added) by 4.3%. • More than 10 000 new jobs are generated in the primary agricultural industry. • The annual area planted under field crops increases by 4.65% (229 000 ha). • The average increases in feed costs range from approximately 10% in the cattle feed market to approximately 6.5% in the pork and chicken feed market. This negatively impacts on the production of meats and dairy products in SA. However, due to the inelastic demand, the percentage increase in price is larger than the percentage decrease in the quantity produced and, therefore, the real gross value of animal production increases. • Approximately 870 million litres of bioethanol are produced locally, with 440 million litres coming from maize and 430 million litres coming from sugar. • Biodiesel production from soybeans accounts for 60 million litres of local production, while biodiesel from sunflowers accounts for 15 million litres of local production. • Under the alternative scenario, SA is competitive in the production of bioethanol from maize and sugar since both commodities are frequently trading at export parity levels. The production of biodiesel from vegetable oil is less favourable due to the high value of the vegetable in the human market and the commodity markets frequently trading at import parity levels. Only major improvements in the efficiency of oilseed production in SA or a sharp decrease in vegetable oil prices will influence the economic feasibility of biodiesel production in SA. Recommendations • Without any government policies, it is highly unlikely that any production of biofuels can be economically viable and, therefore, government needs an accurate estimate of the costs of establishing a successful biofuel industry in South Africa and weigh that up against the benefits.

3

Executive summary • No country in the world uses a staple to produce biofuels. Therefore, policy alternatives have to be carefully considered to understand the impact on food security. With approximately 1 million tons of maize required for the production of ethanol under the alternative scenario, the projected increase in maize prices is moderate, but the probability of the maize market trading at import parity levels, if weather conditions are below normal, is very high, thus putting pressure on food security to be maintained. • The biofuel industry around the world is highly subsidized and protected from competition. The least cost and most efficient producer of bioethanol, Brazil, also safeguards the industry by applying a 20% import levy and maintaining a tax differential. Taking into consideration that Brazil has undergone a learning curve in the successful establishment of a biofuel industry over the past three decades, it is apparent that the South African biofuel industry will struggle to remain competitive without due trade protection and support. • It appears as if trade policies are one of the critical drivers in determining the successful establishment of the biofuel industry in SA. However, there exists a disjoint between trade policies and domestic policies. Following the process of aggressive market reform, South African agriculture finds itself in a position where only a few mechanisms are left through which the industry can be supported to achieve sector-level development goals. Government has only two options of intervention and support; either through trade policies or domestic support mechanisms. • The positive effects of an established biofuel industry are apparent in the modelling results of the alternative scenario. Apart from boosting agriculture’s contribution to GDP, the positive upstream and downstream effects also need to be taken into consideration. Future research should focus on the economy-wide effects of a biofuel industry and estimate the positive impact of rural economies. • The cost of information in the biofuel industry is very high because of the risk and uncertainty inherent in the industry. It is an infant industry and more time is needed to analyse various policy alternatives and to get a sense of the tradeoffs that accompany different policy packages. Some role players are already paying the price of uninformed decision making. The government needs to allow for more time to research the initiative and for policy makers to find the optimal policy by means of the harmonization of various policy targets and developmental goals. • The following factors that will definitely influence the nature of the ultimate SA biofuel industry, are not taken into consideration in this study: the locations and size of biofuel plants, the improvement in farming practices to reduce the costs of production, the drastic increase in the yield of commercial field crops, climate change, the development of alternative crops, improved efficiency of the production of biofuels and the development of an alternative market for by-products.

4

Background and Assumptions In the draft strategy on the biofuels industry recommendations are made regarding the mandatory blending levels, the prices at which biofuels should be sold, the structure of tariffs, the depreciation allowance and the possibility of implementing an equalisation fund. The draft strategy proposes a number of policy options which the government can follow and describes a number of goals which the government has set out to reach in a specified time period. The government has set out to achieve its goals of producing 10 000 GWh of renewable energy by the year 2013. The draft strategy highlights the biofuels initiative as being one of the major drivers to achieving the 2013 target. A 4.5% blend, for example, will translate into 75% of the 10 000 GWh target being reached. The strategy proposes that “incentives and regulations be developed that enable biofuels producers to supply the refineries, which then blend up to E10 and B5 at market penetration levels.” The task team that wrote the draft strategy does however recommend that an E8 and a B2 blend should rather be implemented for bioethanol and biodiesel respectively. The team also recommended that this implementation be done on a regional basis such that the targets are reached in 5 years’ time. The Basic Fuel Price, also referred to as the BFP, is the price it would cost a South African importer of petrol to buy petrol from an international refinery, transport that petrol from that refinery, insure the petrol against losses at sea and land the product on South African shores. The biofuels strategy proposes that bioethanol, or fuel ethanol be sold at 95% and biodiesel is to be sold at 100% of the basic fuel price. The reason why bioethanol be sold at 95% of the BFP is that the 5% margin, which oil depots retain, helps to cover ethanol handling costs. The same principle does not apply to the biodiesel price since biodiesel is proposed to sell at 100% of the BFP. However, the ultimate price discovery of the biofuel prices can largely depend on the location of the plants. Plant locations have not been factored into the model. A possible formulation for the basic fuel price can include elements of the basic fuel price, tax exemption and location differentials. Thus, a typical calculation of the biofuel price can be as follows: BFP plus location differentials plus tax exemption less transport costs to the depots. The same principle can be applied for the costs of feedstock. Depending on the location of the plant, feedstock prices can vary considerably. In the draft strategy, import tariffs on energy crops are not advised as it is argued by the national biofuels task team that tariffs normally degenerate into artificially shaped economic structures that may spill over to other agricultural sub-sectors. The main argument is that such tariffs may discriminate against biofuel producers compared to oil refiners, as crude oil, for example, carries no import tariff and produces a directly sustainable product. The team further recommended that fuel tax reductions and hedge mechanisms should only be applied to biofuels produced locally from South African feedstock. There is very little mention of import tariffs being applied directly to biofuels. Specific sections of the task team document seem to suggest that no tariffs will be applied to any biofuel product as this creates an unfair advantage in relation to the oil industry. Currently a fuel levy tax exemption of 40% for biodiesel is regulatory. It is proposed that bioethanol receives 70% of the 40% of the fuel levy tax exemption as it only contains 70% of the energy of biodiesel. The draft strategy proposes that the fuel levy tax exemption of biodiesel be increased to 50%. It is mentioned that this 50% appears justified based on various job related criteria. The draft makes provision that this fuel levy tax exemption be raised to 75% if production needs to be stimulated. Over the past year, BFAP has developed the capacity to model the biofuels industry in a system of equations which interacts directly with the crops and livestock industries. The new BFAP sector model now has the ability to simulate the impact of various policy scenarios and macro-economic factors on the potential biofuel industry in South Africa. 5

Background and Assumptions In order to compare various scenarios and policy options a “reference scenario” needs to be simulated that can serve as a benchmark or the most likely outcome if no major shocks occur. This reference scenario is called a baseline. A baseline is a simulation of the sector model under agreed policy and certain assumptions with respect to macro-economics, the weather and technological change. The baseline projections, which the draft scenario and the alternative scenario are compared to in this policy brief, are grounded on a series of assumptions about the general economy, agricultural policies, weather and technological change. Macro-economic assumptions are based on forecasts prepared by a number of institutions, such as Global Insight, the Food and Agricultural Policy Research Institute (FAPRI) at the University of Missouri, ABSA bank and the Actuarial Society of South Africa (for projections on population). Tables 1 and 2 present the baseline projections for key economic indicators and world commodity prices in the model. Table 1: Economic indicators – Baseline projections. Item Crude Oil Persian Gulf: fob Population Exchange Rate Real per capita GDP CPIF (Inflation)

$/barrel Millions SA c/US$ R/capita Index (‘00)

2007

2008

2009

2010

2011

2012

63.22

60.79

57.01

53.44

51.41

50.33

47.68

47.65

47.54

47.39

47.22

47.04

715.96 17600.81 217.55

754.65 18390.69 227.28

785.38 19233.4 237.37

811.13 20120.7 247.18

828.40 21042.24 257.23

844.66 22017.87 267.53

Source: Global Insight, FAPRI, Actuarial Society, ABSA, as quoted in the 2006 BFAP baseline. Table 2: World prices – Baseline projections. Item Yellow maize, US No.2, fob, Gulf Wheat US No2 HRW fob Gulf Sorghum, US No.2, fob, Gulf Sunflower Seed, EU CIF Lower Rhine Sunflower cake (pell 37/38%) , Arg CIF Rott Sunflower oil, EU FOB NW Europe Soya Beans seed: Arg. CIF Rott Soya Bean Cake(pell 44/45%): Arg CIF Rott Soya Bean Oil: Arg. FOB

US$/t US$/t US$/t US$/t US$/t US$/t US$/t US$/t US$/t

2007

2008

2009

2010

2011

2012

168.00

143.01

145.66

145.87

146.33

145.44

188.82 177.55

187.64 177.66

191.03 180.93

192.71 182.09

195.00 183.61

196.68 183.74

338.24

343.53

348.44

343.27

335.93

331.01

122.32

120.86

120.28

119.49

118.89

117.51

726.01

729.93

732.82

737.66

743.40

747.45

328.30

343.84

343.82

340.34

335.77

333.66

245.49

249.36

244.55

236.84

231.76

228.98

648.43

684.69

698.42

712.77

714.35

716.18

Source: FAPRI, 2006, BFAP

The figures above have been sourced from the mentioned organisations. Each set of figures is based on a number of assumptions which the various institutions have published. BFAP has made some adjustments to the 2007 world prices of selected commodities based on certain assumptions. Apart from the series of assumptions necessary to generate the baseline projections, a number of assumptions are made on the production processes of biofuels. These assumptions are based on the inputs by industry. The technical factors with respect to extraction rates of ethanol, vegetable oil and by-products have been benchmarked by using data and norms received from industry role players and international experience. An ethanol from sugar cane extraction rate of 76 litres, an ethanol from maize extraction rate of 402 litres per ton, a biodiesel from soybean extraction rate of 194 litres and a biodiesel from sunflower seed extraction rate of 398 litres are applied in the model. DDGs from maize has an extraction rate of 304 kilograms per ton, soy cake an extraction rate of 800 kilograms per ton of soybeans and sunflower cake, a rate of 420 kilograms per ton of sunflower seed. An average quality of DDGs is assumed for this model. The BFAP model only takes commercial agricultural commodities into account but it is acknowledged that there are other commodities that could also contribute to the production of biofuels. Apart from the technical factors, a set of prices and costs are required to calculate plant profits. 6

Background and Assumptions Table 3 presents an example of the calculation of possible plant profits in 2006. These plant profits change over time as projected prices and costs change according to the baseline projections. Under baseline simulations ethanol is selling at 95% of the basic fuel price and biodiesel selling at 100% of the basic fuel price. This is, however, not the case for the scenario analyses where factors of demand, supply and world prices determine the local biofuel prices. Table 3: Plant profit calculations, 2006 average prices. Commodity* Sugar cane (Eth) Yellow maize (Eth) Soybeans (BIOD) Sunflowers (BIOD)

Cost of feedstock (R/ton) 193 1310 1959 2338

Income from byproduct (R/ton)

Income from sales (c/litre)

1092 2076 1505

312.91 312.91 336.45 336.45

* Abbreviations of bio fuels: Eth – Ethanol; and BIOD – Biodiesel.

Total costs of production (c/litre) 165.59 171.54 212.67 170.47

Profit (c/litre) -89.87 -101.80 -29.87 -262.27

Due to the sensitive nature of the costs of production, only the total costs of production are provided in cents per litre, which include variable costs and capital costs. These costs were collected from technology providers, financial institutions and refineries. Another important assumption is that these costs are representative for an “average sized plant” and the researchers acknowledge that the costs structures for different sized plants will differ from the values that are presented table 3. From table 3 it becomes clear that under 2006-market conditions, no commercial crop would have yielded a positive plant profit producing biofuels. The main reason for this being the unfavourable ratio of feedstock prices relative to fuel prices. Important to note is that plant profits are dynamically simulated over the baseline projection period. Thus, the projected loss of biofuel plants is even higher under the current high levels of commodity prices. It is projected that with the current policies in place, biofuel production plants will consistently make a loss over the baseline projection period. It is crucial to note that the production of vegetable oil for the human market is and remains profitable under the baseline projection period. This is due to very high prices of vegetable oil in the human market. Thus, incentives in the biodiesel industry will have to be large enough to divert vegetable oil from the human market to the biofuels market. The figure below, illustrates the dynamic interaction between the fuel, field crops and livestock industry in the model. The figure also indicates at which levels the various tariffs and proposed incentives under the draft strategy and the alternative scenario will influence the local industries. Alternative Scenario: 35% tariff on Ethanol and Biodiesel

Draft Scenario: NO Import tariffs!

Draft Scenario: E8 and B2

Biofuel imports / exports

Domestic biofuel use Custom & Excise

SA GRAIN, LIVESTOCK and BIOFUELS SECTOR MODEL DDGS / oilcake consumption

Mandatory blending

Oil price

Road accident fund

Fuel tax: 116 c/l on 7 Feb 2007 for Petrol and 100 c/l for Diesel.

Domestic biofuel production

Alternative Scenario: E 8 and B 2

Yellow maize: Variable levy White maize: Variable levy Wheat: 2% Sunflower oil: 10% Sunflower seed: 9.4% Sunflower cake: 6.6% Soybean oil: 10% Soybeans: 8% Soybean cake: 6.6%

BIOFUELS

CROPS

Fuel tax Taxes

Draft Scenario: 75% fuel levy reduction for biodiesel and 70% thereof for ethanol, i.o.w 52.3%. Alternative Scenario: 40% fuel levy tax reduction ~ 60% of the fuel levy still applies.

DDGS / oilcake production

7

LIVESTOCK

Acronyms Acronyms BFAP – Bureau for Food and Agricultural Policy FAPRI – Food and Agricultural Policy Research Institute BFP – Basic Fuel Price E8 – 8% bioethanol blend in petrol B2 – 2% biodiesel blend in diesel Eth – Ethanol BIOD – Biodiesel Sce- Scenario Bsl – Baseline

8

Draft strategy: Scenario The following section presents a scenario on what the possible impact of the current draft strategy could be given the policy options as set out in the document. The "background environment" in which this analysis is done is presented by means of a scenario, after which the simulated results are presented on pages 5 to 8. The scenario reads as follows: The government is set on achieving its 2013 goals on renewable energy. Mandatory blending incentives of 8% bioethanol and 2% biodiesel have been imposed and implementation of these levels will be phased in gradually from 2009 onwards. The government understands that the infant ethanol industry will need some time to get production capacity established and therefore it announces the blending levels at an early stage but only makes it mandatory later on so to avoid too many foreign firms taking up most of the market share before the South African industry is established. The government does not impose import tariffs on energy crops which are to be used for the production of biofuels. Import tariffs are also not imposed on direct imports of both refined bioethanol and biodiesel. Government decides to increase the fuel levy tax exemption to 50% for biodiesel and the exemption of bioethanol to 70% of the 50% of biodiesel. This exemption is increased to 75% in 2010 to ensure that the 2013 renewable energy targets will be met.

9

Draft strategy: Total agriculture

Real Gross Value of Fieldcrops The proposed draft strategy will augment the real gross value of field crops modestly. It increases the value by 2.88 percent above the baseline between 2011-2015.

Biofuel balance, 2012-2015 averages

m en ts

po rt s im

re qu ire E8

Et ha no l

-s

ug ar

ai ze -m

Et ha no l

en ts

Et ha no l

po rts B

2

im

re qu ire m

un fl -s B

io di es el

io di es el B

B

io di es el

-s

oy b

million litres

1000 900 800 700 600 500 400 300 200 100 0

Real Gross Value of Field Crops 120 100 80 Index

Biofuel Composition It is projected that with a E8 and B2 mandatory blending policy, approximately 950 million litres of bioethanol and 180 million litres of biodiesel will be required to meet the local blending requirements for the period 2012 - 2015. Given the draft scenario assumptions, it seems like bioethanol from maize will account for most of the local production. Bioethanol from sugar only plays a small role later on as sugarcane prices become more competitive compared to maize. Very small quantities of biodiesel are produced from both sunflowers and soybeans. 80% of bioethanol and 90% of biodiesel will be imported.

60 40 20 0 2006

2007

2008

2009

2010

Baseline

Real Gross Value Added The overall impact of the biofuel industry on the contribution of the agricultural sector to the economy (gross value added) under the draft strategy is expected to be modest. It increases the value by 1.36% above the baseline from 2011-2015.

2012

2013

2014

2015

2014

2015

2014

2015

Scenario

Index

Real Gross Value of Animal Production 160 140 120 100 80 60 40 20 0 2006

2007

2008

2009

2010

Baseline

2011

2012

2013

Scenario

Real Gross Value Added of Agricultural Sector 140 120 100 Index

Real Gross Value of Animals The impact of the draft strategy on real gross value of animal production is relatively small. Production in the livestock sector decreases due to higher feed costs. However, the percentage increase in prices is larger than the percentage decrease in production due to inelastic demand. Thus, the real gross value increases by 0.64 percent above the baseline during 2011 to 2015.

2011

80 60 40 20 0 2006

2007

2008

2009

2010

Baseline

10

2011

2012

Scenario

2013

Draft strategy: Ethanol White and yellow maize producer prices 1700 1500 R/ton

Maize prices Given the policy proposals made in the draft strategy it can be expected that the maize price is likely to increase due to the increased demand for maize. Taking 2012 to 2015 averages, the white maize price is expected to increase by 4.19% and the yellow maize price by 6.67%.

1300 1100 900 700 500 2006

2008

2009

2010

2011

2012

2013

2014

White maize prod price (sce)

Yellow maize prod price (sce)

White maize prod price (bsl)

Yellow maize prod price (bsl)

Yellow maize, DDGS prices and DDGS production 1800

800

1600

700

1400

600

R/ton

1200

500

1000

400

800

300

600 400

200

200

100

0

0 2006

2007

2008

2009

2010

2011

Yellow maize prod price (sce) R/ton DDG prod (sce) 1000 t

2013

2014

2015

DDG price (sce) R/ton

Sugar exports and average cane prices 1000 tons

1300

300 250

1100

200

900

150 100

700

50

500

0 2006

2007

2008

2009

2010

2011

2012

2013

2014

Sugar exports (sce) 1000 tons

Sugar exports (bsl) 1000 tons

Sugar cane avg pr (sce) R/ton

Sugar cane avg pr (bsl) R/ton

2015

Plant profits, 2012-2015 averages 40 30 c/litre

Plant Profits: 2012 - 2015 Although there are some years where the production of ethanol from sugar will be profitable, the average plant profit for the period 2012-2015 is negative. The average plant profit for ethanol production from maize is projected to be 30c/litre. Positive plant profits are induced by the higher ethanol prices that are simulatd in the model. Bioethanol prices rise because mandatory blending requirements shift the demand for ethanol.

2012

20 10 0 -10 -20

Ethanol-sugar 11

Ethanol-maize

R/ton

Sugar Only a very small quantity of ethanol is produced from sugar. Ethanol production from sugar will be sourced from sugar that is being exported. The sugarcane production area is limited in SA and the area under production will increase marginally. The local sugar cane price in the scenario remains relatively stable compared to the baseline price.

2015

1000 tons

DDGS The DDGS prices are expected to increase with the yellow maize price, given the feed ration composition and nutritional value of DDGS and its potential use in the different livestock industries. DDGS production will start up relatively slowly and then reach a level of approximately 150 000 tons.

2007

Draft strategy: Biodiesel

1000 tons

Soybean production and net imports 500 450 400 350 300 250 200 150 100 50 0

200 150 100 50 0 -50 2006 2007

2010 2011

2012 2013

2014 2015

Soybean production (bsl) 1000 tons

Soybean production (sce) 1000 tons

Soybean net imports (bsl) 1000 tons

Soybean net imports (sce) 1000 tons

Soycake prices and production

250 Prod (1000 tons)

2008 2009

700 600

200

500

150

400

100

300 200

50

100

0

0 2006

2008

2009

2010

2012

2013

2014

2015

Soybean Cake Production (bsl)

Soybean Cake Production (sce)

Soybean Cake Imports (bsl)

Soybean Cake Imports (sce)

Sunflower cake production and net imports 300 200 100 0 2006

2007

2008

2009

2010

Sunflower Cake Production (scenario)

2011

2012

2013

2014

0 -50 -100 -150 -200 -250 12

Biodiesel-soybeans

2015

Sunflower Oil Net Imports

Plant profits, 2012-2015 averages

c/litre

Plant profit Despite higher biodiesel prices due to mandatory blending requirements that shift the local demand for biodiesel, production of biodiesel is not commercially viable and it is projected that only some on-farm production for own use will take place.

2011

400 1000 tons

Sunflower cake Under the baseline assumptions, sunflower cake production increases by 12% from 2006 to 2012. As a result sunoil imports decline slightly over time. The production of biodiesel from sunflowers is marginal and has a very some effect on the sunflower industry.

2007

Biodiesel - sunflower

Imports (1000 tons)

Soybean cake Overtime there is a slight decrease in the imports of soybean cake due to the marginal increase in the local production of soybean cakes.

1000 tons

Soybean Seed Soybean seed production is not stimulated significantly by the incentives in the draft strategy, hence the increase in the area harvested is insignificant. Soybean imports are projected to grow as the domestic demand for soybeans in the feed market grows.

Draft Strategy: Farm Labour The introduction of a biofuels industry will effect prices and profitability within the markets for grains and livestock. As a result of these changes, production patterns, especially of feedstocks used for the production of biofuels, are likely to change, which ultimately has consequences for employment patterns. The table 4 presents the possible impacts on employment given the simulated changes in production as a result of the policies as assumed in the draft strategy. The proposed changes in the labour market for the various grains are based on two coefficients. The first coefficient, namely 0,1 is taken from the draft biofuels strategy document. The 0,1 implies that for every hectare expansion in grain production 0,1 permament employment opportunities are created. The second coefficient was taken from the literature (Kirsten & Vink, 2001) on employment in agriculture. The coefficient (0,03) is much lower than that used in the draft strategy. Hence, the idea of using both coefficients is to give an indication of the possible upper and lower limit in terms of creation or destruction of employment opportunities in the various grain industries as a result of the policies being followed. In using these coefficients, no assumption is made whatsoever on whether feedstocks for biofuel plants will be sourced from emerging farmers or commercial farmers or both, hence just a simple calculation is done to give some idea on potential employment impacts. The feedstock procurement models that will eventually be used by the biofuel industry, are likely to change the employment impacts significantly from what is presented below. It must be noted that these potential employment impacts are only calculated at farm level, and therefore exclude direct employment opportunities at plant level, other levels of the economy, as well is indirect employment opportunities.

Table 4: Potential Employment impacts of draft biofuel strategy Industry White maize Yellow maize Wheat Barley Canola Sorghum Soybeans Sunflower Sugar Potential Net effect

Coeffient 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.08

Change in Employment opportunities -789 2812 -299 -4 -1 -7 65 -237 5 1543

White maize Yellow maize Wheat Barley Canola Sorghum Soybeans Sunflower Sugar Potential Net effect

0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1

-2656 9459 -1007 -14 -3 -24 217 -797 6 5181

Attempting to understand the employment effect, one needs to look at the potential net effect, since employment opportunities are not necessarily "destroyed" in practise, since labour is used to produce alternative products that are more profitable on the farm due to its uptake in the biofuels market.

13

Draft Strategy: Conclusion The simulation results, presented in pages 8 to 11, indicate that the likely impacts of a potential biofuel industry, established under the policies and incentives of the draft stratgy are likely to be small in terms of production, prices, contribution to economic growth, and employment. Only very little biofuel production will take place in SA and the majority of biofuel will be imported. This will have a negative impact on the balance of payments. 5000 additional employment opportunities are created on primary (farm) level and the agriculture' s contribution to the economy increases on average by 1.6% above baseline projections. To conclude, the draft strategy presents a set of policies that are not optimal to reach the targets with respect to renewable energy, growth in the agricultural sector, job creation and uplifting of the rural economy.The question, therefore, arises of what is required to make the production of biofuels in SA economically viable? In the following section an alternative scenario is presented.

14

Alternative scenario This section illustrates the positive impacts that a successful biofuel industry could have on the agricultural sectors if the necessary measures of support and protection are put in place. In order to illustrate this point, the approach used in this document is to scetch an alternative scenario where the policies differ from the ones proposed in the draft strategy document. The purpose of this excercise is to inform policymakers what combination of policies and suport is likely to support the viability of biofuels production in SA. The scenario reads as follows: In the government’s paper on renewable energy, the government has set standards of achieving 10 000 GWh of renewable energy by 2013. The aim is to have biofuels account for 40% of this quota. The government implements an E8, 8% bioethanol blend, and a B2, 2% biodiesel blend, mandatory blending policy in 2008. It is keen on achieving its 2013 renewable energy goals and expects that its policy on renewable energy will uplift the emerging and small scale farmers, if managed correctly. The mandatory blending policies for bioethanol and biodiesel differ considerably in that the quantities of petrol and diesel consumed in South Africa are so different. In 2005 petrol consumption in South Africa amounted to approximately 11.1 billion litres, while diesel consumption in 2005 amounted to approxminately 8.1 billion litres (SAPIA, 2006). The bioethanol policy is gradually phased in, changing with 2% blending every year until a 8% blend is achieved. The biodiesel mandate is also systematically phased in from 2008 onwards with a 1% blending mandate at first and this is then upgraded to a 2% mandate from 2010 onwards. The government is very conscious as to what is happening within the global biofuel industry and understands that it is nearly impossible for such an infant industry to survive without any government support. Currently a fuel levy tax exemption of 40% for biodiesel is regulatory. It is proposed that bioethanol receives 70% of the 40% of the fuel levy tax exemption as it only contains 70% of the energy of biodiesel. The draft strategy proposes that this fuel levy tax exemption of biodiesel be increased to 50%. It is mentioned that this 50% appears justified based on various job related criteria. The draft makes provision that this fuel levy tax exemption be raised to 75% if production needs to be stimulated. On the biodiesel side, supply of feedstock seems to be the biggest concern for the South African biofuels industry. The past 5 years’ averages indicate that South Africa has produced 686 thousand tons of sunflower seed and 245 thousand tons of soybeans, but consumed an average of 717 thousand tons of sunflower seed and 258 thousand tons of soybeans. This means that South Africa is on average a net importer of both of these commodities. SA is also a net imported of soybean cake with 80% - 90% of soybean cake being imported. World soybean prices could also be expected to follow an upward trend as the European Union and countries like the US start to increase their biodiesel capacity, which in turn goes hand in hand with a reduction in exports and a potential supply squeeze on the international market. On the bioethanol side, ethanol prices rise as the demand has to be satisfied and there are just not enough plants producing ethanol. Internationally, the developed countries are moving towards a biofuel blend in their local transport fuels. This increase in the world demand for ethanol leads to the steady increase in the price of ethanol on the world market and as a result of this an increase in the South African ethanol import parity price. On top of the increase in the import parity prices, the financial survivability of the local industry is further supported by the introduction of import tariffs (35% on bioethanol and 35% on biodiesel) in 2010. The local industry only starts to make a real contribution to the mandatory requirements of ethanol once the import tariffs are in place. The local biodiesel industry struggles to remain competitive compared to the local vegetable oil market. Biodiesel will only be produced from vegetable oil where full tax concessions are applicable (like on-farm use). The bulk volume of vegetable oil produced locally is still consumed in the human market.

15

Alternative scenario: Total agriculture Bio fue l b a la nce , 2012 - 2015 a ve ra g e s 1000 800

million litres

Biofuel Composition In the alternative scenario, maize and sugar supply almost the whole market with equal amounts of bioethanol. On the biodiesel side, soyoil produces the largest quantity, approximately 60 million liters, while sunflower oil produces only 17 million liters. Local production is induced by positive plant profits.

600 400 200 0

Real Gross Value of Field Crops 120 100 Index

Real Gross Value of Field Crops The alternative scenario expands the real gross value of field crops by 9 percent above the baseline from 2011 to 2015; a comparable difference to the 2.88% increment in the case of the draft strategy.

80 60 40 20 0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Baseline

Real Gross Value of Animal Production 200 150 Index

Real Gross Value of Animals The impact of the alternative scenario on the livestock industry is also larger than the impact under the draft strategy. This scenario augments the real gross value of animal production by 2.88 percent above the baseline between 2011-2015. This is caused by a larger increase in the prices and a smaller decrease in production due to inelastic demand.

100 50 0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Baseline

Scenario

Real Gross Value Added of Agricultural Sector

Index

Real Gross Value Added Under the alternative scenario, the biofuel industry will boost the real gross value added of the agricultural sector by 4.3 percent above the baseline from 2011 to 2015.

Scenario

140 120 100 80 60 40 20 0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Baseline

16

Scenario

Alternative scenario: Ethanol

2015

2014

2013

Yellow maize prod price (sce)

White maize prod price (bsl)

Yellow maize prod price (bsl)

DDGS prices, Yellow maize prices and DDGS production 1800

700

1600

600

1400

500

1200 1000

400

800

300

600

200

400

100

200 0

0 2006

2007

2008

2009

2010

2011

Yellow maize producer price (sce) DDG production (sce)

2012

2013

2014

2015

DDG price (sce)

1000 tons

Sugar exports and average cane prices 1400

300

1200

250

1000

200

800

150

600

100

400

50

200 0

0 2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

Sugar exports (sce) 1000 tons

Sugar exports (bsl) 1000 tons

Sugar cane avg pr (sce) R/ton

Sugar cane avg pr (bsl) R/ton

Ethanol, Petrol plant and retail prices 1000 SA cents/ litre

Ethanol vs Petrol prices The bioethanol prices flow in tandem to the petrol price until a mandatory blending policy is enforced. The inclusion of import tariffs gives the prices another boost and from then onwards biofuels trade at a premium to the petrol price. The higher bioethanol prices bring on positive plant profits.

R/ton

Sugar As in the previous scenario, exported sugar is diverted to ethanol production. Interestingly enough this only occurs once the import tariff is implemented in 2011. The sudden diversion from exports forces the local sugar cane price up and it increases on average by 4.9%.

2012

2011

2010

2009

2008

2007

R/ton

2006

White maize prod price (sce)

1000 ton

DDGS The total production of DDGS is projected to reach approximately 330 000 tons. The higher levels of production will cause DDGS prices to trade at lower levels than the yellow maize prices, as the critical up-take level in the feedmarket is reached. This critical level of uptake is determined by an optimization model which shows how prices of DDGS will have to decrease to ensure uptake of the product in feed rations.

White and yellow maize producer prices 1800 1600 1400 1200 1000 800 600 400 200 0

R/ton

Maize prices Due to the higher demand for yellow maize, the price of yellow maize is above that of white, especially after the import tariff on biofuels is applied in 2011. This induces an increase in the area planted under yellow maize and a decrease in the area planted to white maize. On average yellow maize prices increase by 15.57% and white maize prices increase by 8.71% above baseline projections. Long-run equilibrium in reached in 2014 and 2015 when white maize and yellow maize prices are trading at comparable levels.

800 600 400 200 0 2006

2007

2008

2009

2010

2011

2012

2013

2014

Ethanol price, plant SA cents/l

Ethanol price, retail SA cents/l

Petrol price - plant (bfp) SA cents/l

Petrol, retail price SA cents/l

17

2015

Alternative scenario: Biodiesel Prod (1000 tons)

350 300 250 200 150 100 50 0 2006 2007 2008 2009 2010 Soybean production (bsl) 1000 tons Soybean net imports (bsl) 1000 tons

2011

-50 2012 2013 2014 2015 Soybean production (sce) 1000 tons Soybean net imports (sce) 1000 tons

1000 tons

Soybean production and imports

350 300 250 200 150 100 50 0

700 600 500 400 300 200 100 0

1000 tons

Soybean cake Local soybean cake production increases dramatically from the baseline as the imported soybeans are crushed and biodiesel, soybean cake and oil are the by-products. Soybean cake imports also decrease as domestic production increases.

Soybean production and net imports 500 450 400 350 300 250 200 150 100 50 0

Imports (1000 tons)

Soybean seed In the alternative scenario soybean production increases slightly due to the more protective tariffs implemented on biofuel imports. Soybean imports also increase from the level of the baseline. These imports even out at around 300 000 tons compared to the 170 000 tons which would have entered the country had biodiesel policy incentives not been taken into consideration.

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Soybean Cake Production (bsl) Soybean Cake Imports (bsl)

Sunflower cake production and net imports 400 350 1000 tons

Oilcake production and net imports Sunflower oil imports increase as very little sunflower oil is used for biodiesel production. The bulk of the sunflower oil is still used for consumption in the food industry.

Soybean Cake Production (sce) Soybean Cake Imports (sce)

300 250 200 150 100 50 0 2006

2007

2008

2009

2010

2011

Sunflower Cake Production (sce)

2013

2014

2015

Sunflower Oil Net Imp

Biodiesel prices vs Diesel price (retail and plant) 1200 SA cents / litre

Biodiesel versus Diesel prices As there is a relative shortage of biodiesel feedstock in SA, one can expect the price of biodiesel to trade well above the wholesale price of fossil diesel. If government were to fix the biodiesel price to the fossil diesel price this would create a disequilibrium associated with welfare losses which would have to be born by either producers or consumers. Positive plant profits for biodiesel production from soybeans are projected due to the high biodiesel prices.

2012

1000 800 600 400 200 0 2006

2007

2008

2009

2010

2011

2012

2013

2014

Biodiesel price, plant (sce)

Biodiesel price, retail (sce)

Diesel price - plant (bfp) (sce)

Diesel, retail price (sce)

18

2015

Alternative scenario: Farm Labour The potential employment impacts of the alternative scenario is presented in Table 5. The same coeffients were used as in Table 5. Again it must be noted that these potential employment impacts are only calculated at farm level, and therefore exclude direct employment opportunities at plant level, other levels of the economy, as well as indirect employment opportunities.

Table 5: Potential Employment Impacts under Alternative Scenario Industry White maize Yellow maize Wheat Barley Canola Sorghum Soybeans Sunflower Sugar Potential Net effect

Coeffient 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.08

Change in Employment opportunities -1574 5585 -606 -8 -2 9 146 -446 256 3360

White maize Yellow maize Wheat Barley Canola Sorghum Soybeans Sunflower Sugar Potential Net effect

0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1

-5295 18789 -2040 -26 -7 31 490 -1501 320 10761

The potential employment impacts of the alternative scenario differ significantly from the results of the draft strategic document' s simulation results. The reason for the significant increase in employment opportunities is an increase in the area of field crops production. The increase in area usage is due to higher profitability as a result of higher feedstock prices received by the farmers. Feedstock prices are higher because of higher demand for feedstocks to be used for production of biofuels. The downside in terms of employment opportunities is the "destruction" of employment opportunities in some of the grain industries as a result of substitution of products being produced on the farm. However, one can argue that since substitution takes place on the farm, the employment opportunities are not destroyed in practice since labour is just used to produce the alternative product on the farm. Therefore, the most basic figure to look at when attempting to understand the employment effect is the potential net effect.

19

Alternative scenario: Aggregate agriculture Plant profit: 2012-2015 averages

60 cents / litre

Average plant profits Under the alternative scenario, plant profits increase to economically viable levels for bioethanol from maize, bioethanol from sugar and biodiesel from soybeans. Initial plant profits for biodiesel from soybeans are larger, but as the prices of soybeans increase due to increased demand and the prices of soybean cake decrease due to an increase in local production, plant profits dwindle over time.

40 20 0

-20 -40

Ethanol-sugar

Ethanol-maize

-60

Biodieselsoybeans

Biodiesel sunflower

-80

Total area under production

5400 5200 Thousand ha

Total land use in SA On average, the total area planted to commercial field crops that have been included in the model, increases by 4.65% under the alternative scenario compared to 2.89% under the draft scenario.

5000 4800 4600 4400 4200 4000 2010

2011 Baseline

2013

2014

2015

Scenario

Percentage rise in feed costs

12.00 10.00 8.00 %

Costs of feed rations The average increases in feed costs range from approximately 10% in the cattle feed market to approximately 6.5% in the pork and chicken feed market. This negatively impacts on the production of meats and dairy products in SA.

2012

6.00 4.00 2.00 0.00 Cattle feed cost

20

Pork feed cost Chicken feed cost

Dairy feed costs

Conclusion: Alternative Scenario The combination of policies and support measures that are used in the alternative scenario are based on international experiences and basic local requirements to make the industry feasible. Under this combination of policies and support measures most of the bioethanol is produced locally to make up the E8 blend with less than 10% of local requirements being imported. The local production of biodiesel accounts for approximately 45% of local requirements. Under the alternative scenario, the biofuel industry expands the real gross value of field crops by 9 percent above the baseline from 2011 to 2015 and boost the real gross value added of the agricultural sector by 4.3 percent above the baseline from 2011 to 2015. The positive impact of a successfull biofuel industry can also be related to more than 10 000 new jobs on farm-level. The alternative scenario represents only one possible combination of policies that can be implemented to support the infant industry. There are many more options and support measures that can be looked at. However, following the process of aggressive market reform, South African agriculture finds itself in a position where only a few mechanisms are left through which the industry can be supported to achieve sector-level development goals.The policy gap that exists to support the local biofuel industry needs to be established. Government can support the local industry on two levels, namely by means of trade policies and domestic policies. Domestic support can be justified under article 6.2 of the WTO regulations. Trade policies can also be justified when the import tariffs on bioethanol that are imposed by major bioethanol producers are taken into consideration (table 6). The 35% import tariff on bioethanol and biodiesel was discovered by testing the model to find a breakeven point for the production of biofuels. A 35% tariff on bioethanol implies that plants make a small profit, and a 35% tariff on biodiesel implies that biodiesel plants just break even. Table 6: Import tariffs of selected countries on ethanol. Country Ethanol Import Tariff USA Brazil Argentina Thailand India Canada European Union China

2.5% + 54 cents per gallon (approx. 45%) 20% 20% 30% 186% 4.92 cents per liter = 19 cents per gallon 19.2 cents per liter = 87 cents per gallon 30%

Source: RFA (2006), Latner et.al. (2006) and Elabeid and Tokgoz (2006).

21

Main findings and recommendations The simulation results presented, clearly underline various key factors and policies that are likely to support the viable production of biofuels in South Africa. Firstly, it needs to be considered when it comes to the viable production of biofuels in South Africa that nowhere in the world has biofuel production been stimulated and developed without some form of government support, be it in a large or small degree. The degree of the fuel tax levy reduction, the level of import tariff protection and the efficiency and power of regulations governing these policies have a huge impact on how local biofuels production is stimulated and developed. What needs to be emphasized is that an early level of government support is vital as this will determine the sustainability of the industry until it reaches the required economies of scale. Similar levels of government support have been implemented in other countries with the result that biofuel production in those countries is feasible. Examples of countries where these policies have created a viable biofuels industry are the USA and Brazil. The selection of crops is also crucial. Multi feedstock plant technology already exists. The question is in what crops SA will be competitive if support measures are put in place. The modelling results under the alternative scenario show that SA is competitive in the production of ethanol from maize and sugar since both commodities are frequently trading at export parity levels. The production of biodiesel from vegetable oil is less favourable due to the high value of the vegetable in the human market and the commodity markets frequently trading at import parity levels. Major improvements in the efficiency of oilseed production in SA or the sharp decrease in vegetable oil prices will influence the economic feasibility of biodiesel production. No country in the world uses a staple to produce biofuels. Therefore, policy alternatives have to be carefully considered to understand the impact on food security. With approximately 1 million tons of maize required for the production of ethanol under the alternative scenario, the projected increase in maize prices is moderate, but the probability of the maize market trading at import parity levels, if weather conditions are below normal, is very high. Even though the policy of minimal tariff protection is in place, the emphasis of the entire draft strategy document is still very much on locally produced commodities and fuels. The fuel levy tax exemption, for example, only applies to the locally produced products and importers cannot benefit from these. There is a problem with this policy in that biofuel producers, especially biodiesel producers who cannot source the product locally since local production is too small, will not qualify for the fuel levy exemption and will as a result struggle to produce biofuels viably. The equalization fund proposed in the draft strategy was not taken into account. The proposed equalization fund is based on the movement of the oil price. This will, however not be suffcient to cover the exposure to price risk as the swing in feedstock prices also will have a major impact on the profitability of the biofuel plants. The location of biofuel plants can have a substantial impact on profitability. This model presents a general picture and does not take specific plant locations into account. Thus, the location of the plant needs to be carefully considered. The proposed equalization fund in the draft strategy is based on the movement of oil prices only. It is recommended that the equalization fund is not only based on the movement of oil prices, but on the ratio of feedstock and crude oil prices.

22

Main findings and recommendations It appears as if trade policies are one of the critical drivers in determining the successful establishment of the biofuel industry in SA. However, there exists a disjoint between trade policies and the domestic policies. Following the process of aggressive market reform, South African agriculture finds itself in a position where only a few mechanisms are left through which the industry can be supported to achieve sector-level development goals.The policy gap that exists to support the local biofuel industry needs to be established. Government can support that local industry on two levels, namely by means of trade policies and domestic policies. Domestic support can be justified under article 6.2 of the WTO regulations. Lastly, it needs to be mentioned that the results presented in this study take the information and technology that is currently available and applied in the market into account. Thus, it has to be regarded as a benchmark under current conditions, rather than a forecast. It does not take into consideration factors like the improvement in farming practices to reduce the costs of production, the drastic increase in the yield of commercial field crops, the development of alternative crops that could be grown specifically for the production of biofuels, improved efficiency in the production of biofuels and the development of an alternative market for bioproducts.

23