How uncoordinated biofuel policy fuels resource use and GHG emissions

25 July 2013

The year 2010 introduced a new phenomenon into the global biofuels economy: the bilateral trade of bioethanol between Brazil and the United States of America; the most important producers, consumers, and traders of ethanol. Brazilian ethanol is produced primarily from sugarcane, while the US produces ethanol primarily from maize, but the resulting products are physically indistinguishable. Such ethanol intra-industry trade remained small in volume until the end of 2010 and went unnoticed in the global context. However, 2011 saw large increases in ethanol intra-industry trade between the two countries. Here we pose the question regarding its underlying causes and the associated economic and environmental costs, and we assert that under current policies, intra-industry trade is likely to increase to unsustainable levels, increasing costs to consumers and greenhouse gas emissions.
Intra-industry trade in food and agriculture
Trade literature offers some explanations for the less common exchange of undifferentiated products. In this section we examine these reasons and provide evidence that eliminates them as drivers for US-Brazil trade in ethanol since 2010.
Aggregation or classification issues in trade data - In some instances, trade flow classifications may not be fine enough to differentiate between different products. Denatured and un-denatured alcohol (ethanol) are measured at the HS-4 level (HS 2207), which may include other non-fuel alcohol products.
Seasonality - Annual trade statistics may mask the common phenomenon that countries sometimes exchange large quantities of an otherwise homogenous product to accommodate off-season consumer needs and match deficits through an intra-year exchange of goods. Observed ethanol trade flows in recent quarters, however, suggests that intra-industry trade flows were rising and falling simultaneously or pro-cyclically, rather than intermittent or counter-cyclically, particularly if Brazilian exports to the US via Caribbean countries are included. Data for 2012 shows a different pattern but this is related to supply constraints in the US due to the drought.
Border trade - This phenomenon relates to large countries sharing a long physical border, or lacking efficient internal transport channels between supply and demand regions, that may find it profitable to exchange homogenous products across borders rather than within their own borders due to lower transportation costs. The costs of shipment of ethanol between the two countries is greater than the shipment costs between ethanol production and consumption centres within the United States, although potentially high internal ethanol shipping costs in Brazil should be examined.
Discounting these reasons for ethanol intra-industry trade, we pursue the idea that ethanol intra-industry trade is due to policy induced attributes of ethanol, which differ in US and Brazilian policy.
A policy-induced phenomenon of process differentiation
The stated objectives of US biofuel policy have been wide-ranging. While numerous reasons are stated for the expansion of biofuel policy, the primary instrument that is currently applied in the US (mandated usage) contains elements of environmental legislation and aims at fostering environmentally friendly, carbon-saving production processes. Essentially all biofuel classifications are determined by feedstocks and production process rather than the final product.
This policy differentiation creates the potential for differential wholesale pricing of biofuels based on their classification and creates the opportunity for arbitrage with countries which may have different classification schemes or no classification schemes at all. In this process, transportation fuel is wasted in the name of resource-saving policies and transportation costs increase final costs to consumers, thereby suppressing renewable fuel demand. Given the complexity of the policy framework, a review of existing biofuel policy and how it supports ethanol intra-industry trade is in order.
Potential for US-Brazil intra-industry trade
The Renewable Fuel Standard 2 (RFS2) established in the US Energy Independence and Security act of 2007, further segmented biofuels  and mandated volumes were greatly expanded. The four classes of mandates are not individual compartmentalized mandates but quantitative minimums nested within the overall renewable fuel mandate. Over production in each category can be used to meet the larger, less restrictive mandate. This creates a hierarchy among the fuels based on the mandate classification while the physical product, in this case ethanol, is indistinguishable (Thompson et al., 2009).
As different biofuels are indistinguishable for consumers, they cannot be priced differently at retail, and thus the additional benefits are translated into price incentives via an electronic tracking system of traceable mandate obligations. The electronic classification instrument used for tracking mandate compliance, the Renewable Identification Number (RIN), is what differentiates the renewable fuels in the wholesale market. The RIN must accompany the fuel, and can only be separated from the fuel when blended.

Potential for rapid growth in biofuels trade
The US Environmental Protection Agency (EPA) was forced to reduce the 2010-2013 cellulosic biofuel mandate significantly while choosing to leave the total and advanced mandate in place. In reality,  a shortfall in cellulosic ethanol coupled with the EPA decision to maintain the other mandates means that the size of the implied undefined advanced gap has grown and even created the need for undefined advanced fuels, such as sugarcane based ethanol, in 2011. This prompted US ethanol imports from Brazil, and plentiful supplies of maize starch ethanol in the US prompted increased ethanol exports, much of this to Brazil where domestic production of sugarcane ethanol lagged behind domestic demand.
Ethanol demand and the blend wall
While the US could be driven to import as much as 15 billion gallons of ethanol from Brazil, the ability of the US market to absorb such large volumes of ethanol on top of a presumed 15 billion gallons of maize ethanol is uncertain. Until recently, ethanol blending rates in the US for conventional vehicles were capped at 10 percent blends (E10). With motor fuel demand of around 140 billion gallons annually and ethanol disappearance in the US of around 14 billion gallons in 2011 with minimal sales of E85, the 10 percent blend market is approaching saturation (Thompson et al., 2012). There are also a limited number of dispensing options on consumer pumps. These obstacles have worked to constrain E-15 dispensing and use, limiting the outward movement of the blend wall (Wisner, 2012). Further declines in motor fuel use through increased Corporate Average Fuel Economy (CAFE) standards, only exacerbate the problem by shrinking the fuel market and increasing the needed blending rates, as the mandates are in fixed volumes.
The presence of the blend wall will also shape the competition to fill the implied advanced gap. Imported ethanol will need to be absorbed into the motorfuel supply, driving up compliance costs and pushing down the value of ethanol in the retail market. This will push excess US ethanol production out into the export market as opposed to being consumed domestically. The blend wall may also allow for excess biodiesel  to compete more effectively with imported sugar-cane ethanol in filling the undefined advanced gap or overcoming the blendwall as the constraints on the consumption chain in the diesel market are less binding. Biodiesel prices and their associated RIN price may then play a role in the speed and extent of intra-industry trade in ethanol between the United States and Brazil.
The ratio of trade between Brazil and the US is likely to fluctuate from year to year based on relative feedstock prices (sugar and maize), the blend wall, transportation costs, and oil prices. Shipping costs are ultimately borne by motorfuel consumers in both countries. The relative elasticities under policy effects, or how consumer demand responds to changes in the increase in price from transportation costs, will determine who pays for the transport and, ultimately, the number of ethanol ships passing on the high seas. The more restrictive the policies are in Brazil, in terms of blending requirements, the greater the intra-industry trade is likely to be.

Brazil policy and response
The 1973 oil embargo and the associated jump in oil prices came at a time when Brazil was importing over 80 percent of its domestic fuel consumption and low international sugar prices were putting significant economic pressure on producers. Ethanol was promoted through heavy market intervention. Minimum blends were established for ethanol-gasoline blending; the ongoing market intervention spurred the sale of neat vehicles throughout the 1980s, which run on pure ethanol. Increasing sugar prices, lower petroleum prices and an increase in the fixed sales price of ethanol significantly stressed the industry; By the end of the 1990s, ethanol prices had been liberalised along with gasoline and sugar markets, although ethanol still maintained a tax advantage relative to gasoline.
In 2004, the sale of flex fuel vehicles (FFVs) took off in Brazil as the government provided the same tax breaks for the purchase of FFVs as it did for neat vehicles. Currently, ethanol entering the motor fuel market in Brazil is consumed in two ways: first as pure ethanol (E100) by the ageing fleet of neat fuel vehicles, whose sales numbers have plummeted; second, blended with gasoline in the FFV fleet that dominates current vehicle sales. The blending rate minimums support ethanol consumption but do not discriminate between feedstock and process in blending. The blending minimums then provide a mechanism to drive intra-industry trade.

Intersection with Europe's policies
The discussion thus far has been limited to the main elements of US and Brazilian policies. However, recent developments in EU policies and transportation fuel market factors also have the potential to boost ethanol demand in the future, increasing biofuel trade flows between the EU, US and Brazil.
The EU policy framework to promote renewable energy sources for transportation dates from 2003 and to the Directive 2003/30/EC, which foresaw a non-binding target of 5.75 percent market penetration for renewable transportation energy by 2010. The renewable energy in transport targets (RED) were extended to 2020, increased to 10 percent and made binding as part of the climate change and energy package of 2009.
The climate change package also established minimum targets for GHG emission reductions, and included sustainability criteria which would require verification schemes for the achievement of the GHG reduction targets.
The need to take GHG-saving requirements into account opens two possible cases: one in which potential savings include the effects of indirect land-use change (ILUC) and the other where savings are defined through their direct impacts only.  Including ILUC would essentially eliminate all traditional biodiesel feedstocks under the climate change package of the RED. Other things being equal, such a scenario suggests the EU would have to undergo a massive shift from biodiesel to ethanol and, the EU would have to source its needs from feedstocks that provide enough GHG-saving potential. In practice, only cane-based ethanol would qualify and only Brazil has the potential to supply these extra quantities. Such increases in the EU would require an increase in blend rates for ethanol that the EU car fleet is not prepared for technically. As such, a massive increase in imports of cane-based ethanol by the EU may also increase the need for Brazil to cover its own mandatory blending requirements, spurring Brazil's own import needs. These, in turn, could only be covered by US maize ethanol exports to Brazil.

Domestic biofuel trade
Independent policies between government entities which may result in little net gain in biofuel use are not restricted to national governments. Under California Executive Order S-1-07, the California Air Resources Board (CARB) has implemented the Low Carbon Fuel Standard (LCFS) which rates individual fuels based on their GHG reduction score and sets a target for the reduction of GHG emissions.19 The policy requires the fuel to be consumed within California, but the RINs associated with the fuel can still be used to comply with the nation-wide RFS2.
Brazilian imports may be diverted to California ports to comply with the LCFS, with the RINs generated then traded to other states. The consumer cost of renewable fuels could rise due to the need to transport the renewable fuel to California.

Policy-driven intra-industry trade effects and solutions
While the intersection of Brazilian and US biofuel policy provides the necessary condition for intra-industry trade of physically identical but policy differentiated biofuels, other factors will determine if and to what extent this will happen.

Sugarcane and maize yields play a critical role in determining the ratio of intra-industry trade but their yields are likely to have opposing effects on the volume of ethanol trade. Low maize yields in the US are likely to lead to a binding total mandate in the US and reduce "excess" supplies of maize starch ethanol which could be shipped to Brazil in exchange for mandate-driven imports. This describes the market situation with the onset of drought in the United States in the summer of 2012.
Relative demand and supply elasticities in the two markets (as influenced by policies and the blend wall which will make such demand highly non-linear) and the market context (oil and feedstock prices) will ultimately determine the volume of ethanol exchanged. The size of the advanced mandate market in the US is set to expand rapidly in the next decade, making intra-industry trade much more likely and at substantially larger volumes.

Consumer costs and GHG emissions
While an important motivation of biofuel policies is to reduce GHG emissions associated with the use of motor fuel, there is potentially significant efficiency loss in meeting that objective. The transport of ethanol between Brazil and the US generates additional GHG emissions, and those flows identified as policy driven intra-industry trade work against this stated biofuel policy objective. Ultimately the cost of transportation, both the US imports from Brazil and any fuel returned to Brazil, must be borne by motorfuel consumers in both countries. A system of tradable obligations for both the United States and Brazil (and, according to the same arguments, the EU) could avoid the transportation costs and reduce GHG emissions beyond those generated by the uncoordinated policies of each country.
In order to avoid double counting in Brazil, obligations toward its blending minimum would be converted to a RIN system. In this system, the Brazilian sugarcane ethanol, which had RINs sent to the US for its mandate compliance could not be counted toward Brazilian blending minimums, is now based on holding sufficient RIN volumes to equal 20 percent of the volume of fuel sold.

Current uncoordinated policies in the United States, Brazil, and the European Union encourage intra-industry trade of physically homogeneous biofuels which is in contradiction with policy objectives of reducing greenhouse gas emissions. The intra-industry trade we have seen to date remains the "tip of the iceberg" as policy mandated quantities expand. Existing policies may appear unsustainable, through high costs or politically sensitive volumes of imports. We suggest that if blending or consumption mandates for biofuels are going to be an enduring part of energy policies there is an opportunity for a more efficient system that respects differing national objectives in biofuel use through a "book and claim" system, whereby RINs can be separated and used for compliance or sold to other blenders across countries to meet their obligation in lieu of their own physical blending.
This article has been adapted from a longer study, which can be accessed at
Seth Meyer
Economist, Global Perspectives Studies Team (ESA), Food and Agriculture Organization of the United Nations (FAO)

Josef Schmidhuber
Deputy Director, Statistics Division (ESS), Food and Agriculture Organization of the United Nations (FAO)

Jesús Barreiro-Hurlé
Economist, Monitoring African Food and Agricultural Policies (MAFAP) team (ESA), Food and Agriculture Organization of the United Nations (FAO)

This article is published under
25 July 2013
European Parliament has signed off on a proposal that would allow for delaying the auctions of millions of carbon permits, in an effort aimed at propping up prices in the EU's struggling carbon...
25 July 2013
The Nagoya Protocol on Access to Genetic Resources and the Fair and Equitable Sharing of Benefits Arising from their Utilization to the Convention on Biological Diversity (Nagoya Protocol) was...