“Unbundling” the biofuel promise: Querying the ability of liquid biofuels to deliver on socio-economic policy expectations

Energy Policy xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Energy Policy journal homepage: www.elsevier.com/locate/enpol

“Unbundling” the biofuel promise: Querying the ability of liquid biofuels to deliver on socio-economic policy expectations Carol Hunsbergera, Laura Germanb, Ariane Goetzc a b c

University of Western Ontario, Canada University of Georgia, USA Institute of Advanced Sustainability Studies, Potsdam, Germany

A R T I C L E I N F O

A BS T RAC T

Keywords: Biofuel Agrofuel Environmental governance Energy security Employment Smallholder inclusion

While excitement around biofuels initially focused on finding a clean and secure alternative to fossil fuels, many other expectations have subsequently been attached to the “biofuel boom.” Biofuels are not only expected to mitigate climate change or foster domestic energy security, but also to generate employment, provide opportunities to smallholders and support decentralized energy systems. This paper interrogates the expectations attached to biofuels. We begin by examining how policies in consumer and producer countries articulate a series of expectations for biofuels that are “bundled” with the promise of cleaner energy, using the rationales behind these expectations to derive criteria for success. We then review evidence from the published literature on biofuel outcomes against these criteria to assess whether the most prevalent assumptions have been met. We find that policy expectations for biofuels are often expressed in narrow terms, failing to capture important potential impacts – for example focusing on new jobs in the formal sector rather than job quality or whether employment offsets livelihood costs associated with biofuel investments. Some expectations have proven elusive irrespective of the metrics employed, for example using biofuels to improve energy access in remote rural areas. The paper concludes by discussing implications for policy and practice. This article is part of a Virtual Special Issue entitled 'Scaling Up Biofuels? A Critical Look at Expectations, Performance and Governance’.

1. Introduction While assumptions of climate neutrality were a key discursive driver of the recent “biofuel boom” (Searchinger, this issue), a host of additional expectations were quickly attached to liquid biofuels as different actors sought to capitalize on the opportunity to achieve other social, economic and environmental goals. Positive discourses about biofuels rest on the assumption that producing and using biofuels can achieve several goals simultaneously, such as enhancing energy security, job creation, and rural development. Such “bundled” expectations appear in biofuel policies around the world – policies that have played a critical role in promoting the expansion of biofuels by setting fuel blending targets, creating financial incentives for producers, and funding research and development (Bailis and Baka, 2011; Hall et al., 2009). But can biofuels meet these multiple objectives? Key claims related to the benefits of biofuels have been questioned, with greenhouse gas balance receiving particular scrutiny (Fargione et al., 2008; Searchinger et al., 2008, Searchinger this issue). Other critiques include concerns over perceived competition between food and fuel production (Thompson,

2012; Zilberman et al., 2012), local land rights (Cotula et al., 2008; White and Dasgupta, 2010), and labour issues (Labruto, 2014). Previous work finds that the social protection measures in biofuel policies and sustainability certification schemes are particularly weak (German and Schoneveld, 2012a; Hunsberger et al., 2014), while Europe's “hybrid governance” approach allows producers to choose between standards that vary considerably in their comprehensiveness and rigour (German and Schoneveld, 2012a), enabling a ‘race to the bottom’ that evidence suggests is already underway (Ponte, 2014, de Man and German, this issue). Despite these critiques, the “biofuel promise” – the expectation that biofuels can achieve a suite of social, economic and environmental goals – remains influential in guiding policy. There is thus a need to query the evidence on whether positive aspirations for biofuels are being met on the ground. This paper aims to fill this gap by “unbundling” policy expectations linked to biofuels and reviewing evidence on whether each one has been achieved. We focus on firstgeneration, crop-based liquid fuels – the basis of many biofuel policies – because they have been pursued long enough and are well enough established to assess to what extent and under what conditions the

E-mail address: [email protected] (C. Hunsberger). http://dx.doi.org/10.1016/j.enpol.2017.04.017 Received 16 November 2016; Received in revised form 2 April 2017; Accepted 6 April 2017 0301-4215/ © 2017 Elsevier Ltd. All rights reserved.

Please cite this article as: Hunsberger, C., Energy Policy (2017), http://dx.doi.org/10.1016/j.enpol.2017.04.017

Energy Policy xxx (xxxx) xxx–xxx

C. Hunsberger et al.

1. 2. 3. 4. 5.

aims of these policies have been met. While others have approached the task of reviewing the performance of biofuels (Clancy, 2013; Creutzig et al., 2014; German et al., 2011b; Robledo-Abad et al., 2016), our study extends this work by evaluating the impacts of biofuels in relation to the policies that played such a crucial role in promoting them. The question guiding this analysis is, “To what extent have biofuels delivered on expectations commonly attached to them, aside from climate change mitigation?” This study is designed to complement two other papers in this collection: one that assesses biofuels’ climate change mitigation potential (Searchinger, this issue), and another that investigates the extent to which negative effects on food security, land rights and the environment have been avoided (Goetz et al., this issue). The paper proceeds as follows: Section 2 explains our methods, including the selection criteria for countries and policy documents. Section 3 develops operational definitions for the most consistently mentioned expectations in these documents – drawn as much as possible from the policies. Section 4 presents a targeted literature review that assesses outcomes for four expectations: energy security, employment creation, smallholder inclusion and decentralized energy systems. Section 5 reflects on implications for policy and practice.

Domestic energy security Job creation Smallholder inclusion Decentralized energy systems Rural development

Given their prevalence, we chose the expectations as the focus of our analysis – however, rural development proved too difficult to assess as a coherent category due to widely differing interpretations (discussed further below). Section 3 explains how the policy documents defined and operationalized the core expectations and how we in turn derived criteria for assessing outcomes for each one, drawing as closely as possible on terms used in the selected policies. To evaluate empirical evidence of outcomes we conducted a targeted literature review, focusing on empirical work published within the last 10 years (since 2006). The start date of 2006 reflects a year in which biofuels production and consumption increased rapidly worldwide – both in terms of world production of fuel ethanol, the main form of biofuel, and technical innovations in biofuels (Albers et al., 2016). As far as possible, we matched evidence on outcomes for each expectation with the countries and regions whose policies expressed those expectations. For example, all the focal countries identified energy security goals, so all are included in the discussion of energy security outcomes. On the other hand, expectations related to job creation, smallholder inclusion and decentralized energy systems were expressed by focal countries in the global South (plus the EU, whose 2009 policy acknowledges the need to import biofuels to meet use targets), but not by the UK or US – so our review of evidence on these expectations focuses on studies from the global South, drawing in findings from additional countries to increase the available evidence. While we initially intended to restrict our search to articles sourced through Web of Science and Scopus, for some themes these databases proved too limited, and we expanded our search to Google Scholar or even Google (for example, to identify specialized datasets). Table 3 summarizes the literature search process, including the search terms used. The volume of evidence made it impossible to tackle all policy expectations in depth within a single paper. We focus on four of the most widely mentioned expectations related to benefits that biofuels are supposed to provide. Recent evidence on greenhouse gas emissions is reviewed in a separate paper (Searchinger, this issue), as are policy statements and outcomes related to mitigating anticipated risks of biofuels (Goetz et al., this issue). Further in terms of scope, we focus on crop-based sources of liquid biofuels. Recognizing that most biofuel feedstocks can be considered ‘flex crops’ due to their multiple possible uses (Borras et al., 2015), we include in our analysis crops that can be used for food (corn, sugarcane, cassava, soy, oil palm) as well as some that cannot (jatropha, castor).

2. Methods The research followed a five-step process: 1. Selecting a diverse set of countries with bioenergy policies; 2. Identifying core expectations raised in policy documents from the selected countries to find the most prevalent themes; 3. Deriving assessment criteria for each theme based on rationales expressed in policies, and where necessary, also drawing on international standards or literature; 4. Assessing published evidence of outcomes for each theme using the criteria derived in step 3; 5. Reflecting on whether the criteria derived from policy documents are sufficient to assess outcomes for each expectation and how they could be improved. While many state and non-state actors were involved in formulating and propagating discourses of potential benefits of biofuels as well as the governance instruments that have emerged to govern their effects (Bailis and Baka, 2011), we focus our review on expectations enshrined in government policies in key producer and consumer countries/blocs. We justify this focus on policies on two grounds. The first is a question of accountability: any effort to evaluate actual impacts against expectations should consider what outcomes actors are aiming for, and accountable to. This is best evaluated from the perspective of policy aims officially endorsed by nation-states. The second is a recognition that non-state actors, including industry lobbies and civil society, have heavily influenced existing policies – suggesting these policies reflect the intersection of multiple sets of interests (Bailis and Baka, 2011; German and Schoneveld, 2012b). In selecting policy documents we sought to include countries with diverse characteristics including key producers and consumers of biofuels, countries with earlier biofuel policies and more recent adopters, while aiming for a sample of countries at different levels of industrialization and from diverse world regions (Table 1). A close reading of the selected documents yielded a summary of expectations raised across the documents. The results, compiled in Table 2, highlight the key policy aims linked to biofuels in 11 countries. As the table shows, while emissions reduction is a nearly universal aim, many additional expectations have been “bundled” with climate mitigation expectations. Expectations related to avoiding negative outcomes are explored in a companion paper (Goetz et al. this issue). Five expectations stand out as particularly prominent across the countries reviewed:

3. Bundling the biofuel promise: expectations in national biofuel policies 3.1. Clarifying policy expectations To assess outcomes related to each expectation, we must: (i) understand what is meant by each expectation and the degree to which different countries are aiming for the same things; and (ii) distil indicators linked to these policy expectations against which the evidence will be evaluated. 3.2. How expectations are defined or operationalized within focal countries 3.2.1. Domestic energy security The International Energy Agency defines energy security as “the uninterrupted availability of energy sources at an affordable price,” identifying three components: accessibility (available supply), afford2

Energy Policy xxx (xxxx) xxx–xxx

C. Hunsberger et al.

Table 1 Selected Countries and Policy Documents. Country/Region

Selection Criteria

Sources/Policy Documents

Identified by Bailis and Baka (2011) as having a biofuel mandate AND… Brazil China Colombia EU India Indonesia Mozambique South Africa Thailand UK USA

Pioneer in biofuels production; major producer; major consumer Emerging economy, major producer (top 10) Regional balance, major producer (top 10) Major consumer bloc Emerging economy Regional balance, major producer (top 10) Regional balance Regional balance Regional balance, major producer (top 10) Major consumer Major producer; major consumer

(Brazilian National Congress, 1993, 1997, 2005; de Andrade and Miccolis, 2011; MAPA, 2006; MDA, 2009a, 2009b) (Koizumi, 2008; USDA-FAS, 2006) (Conpes, 2008; MADR, 2010; Minambiente, 2016; MME, 2007) (EU, 2009) (MNRE, 2009) (Caroko et al., 2011; Sardjono, 2014; Suharto, 2012) (Republic of Mozambique, 2009) (Republic of South Africa, 2007) (DEDE, 2012; Tongsopit and Greacen, 2012) (Bailis and Baka, 2011; UK, 2007) ( U.S. Congress, 1990, 2005; US EPA, 2007, 2010, 2016)

laws while accommodating specific demands of sugarcane workers (Government of Brazil, 2009). A key expectation is eliminating middlemen from the recruitment of cane cutters (de Andrade and Miccolis, 2011).

ability (supply at competitive prices) and reliability (uninterruptible supply) (IEA, 2014). Energy insecurity may result from physical unavailability or prices that are uncompetitive or overly volatile (OECD-IEA, 2007). The transport sector is particularly dependent on oil, and the share of oil in the sector has grown steadily since the 1980s (IEA, 2014). As biofuels help diversify the energy mix in the transport sector, they are seen as an option to enhance accessibility through diversification and affordability by buffering against price shocks. While the focal countries vary in the primary concerns driving their interest in energy security, these concerns manifest in much the same way. The United States and China, for example, are motivated by geopolitical and national security concerns related to overreliance on oil imports (U.S. Congress, 2007; USDA-FAS, 2006), while many developing and emerging economies are more concerned about import substitution and balance of payments (e.g. Colombia, Indonesia, Mozambique, Thailand, South Africa). Colombia and Indonesia are also motivated to conserve and diversify away from domestic petroleum supplies (Caroko et al., 2011; MME, 2007). For all countries reviewed, biofuels are expected to enhance energy security by reducing dependence on oil and natural gas imports or by diversifying the energy mix (even if achieved through imports). Developing countries that import fossil fuels are particularly motivated by the associated fiscal burden, given their vulnerability to price increases – which can badly affect their balance of payments (ESMAP, 2005).

3.2.3. Smallholder inclusion Several policies express a desire for biofuel production to benefit farmers who are seen as disadvantaged in some way. Most often a distinction is made based on farm size, with policies encouraging ‘smallholders’ (India, Indonesia and Mozambique) or ‘family farmers’ (Brazil) to participate. In other cases the reviewed policies give special consideration based on poverty, remoteness or economic exclusion: Brazil's biodiesel program encourages farmers from semi-arid and remote areas to enter the biofuel sector, while South Africa's biofuel strategy targets locations “worst hit by poverty and deprivation” and offers to support efforts “in the previous homelands by historically disadvantaged farmers” (Republic of South Africa, 2007). These policies share an emphasis on generating income and linking farmers to previously inaccessible markets. Some also state that smallholders can benefit by receiving technical support and producing energy for local use. Policy measures to encourage smallholder participation include direct funding, low-interest loans, technical support, a guaranteed minimum price for biofuel feedstock, and a requirement that processors purchase part of their feedstock from smallholders. Contract farming involving individuals or cooperatives is the most frequently mentioned form of production.

3.2.2. Job creation Expectations related to job creation are framed vaguely in the reviewed policy documents. Most simply state that investments in biofuel will generate employment, in some cases specifying expectations in numbers (e.g. 7.25 million people employed by biofuels by 2015 in Indonesia). Some specify expectations related to the nature of jobs, with Brazilian, Colombian and EU policies specifying job creation in rural and isolated areas as an expected benefit, and Mozambique specifying employment creation in factories. South Africa's Biofuels Industrial Strategy is one of the few documents to project expectations about the consequences of employment for workers, stating that employment in energy crops and the biofuel value chain should provide a bridge “from the second economy to first economy status” (Republic of South Africa, 2007). Expectations related to job quality are rare, and addressed only indirectly. In the EU, labor implications of renewable energy policies are limited to “reporting on” whether countries that are significant suppliers of feedstock have ratified relevant ILO treaties. Brazil goes the furthest among the focal countries in responding to concerns over labor conditions. The National Commitment for the Improvement of Labor Conditions in Sugarcane Production, a voluntary code of conduct, incentivizes business practices which respect existing labor

3.2.4. Decentralized energy systems Several policies endorse small-scale biofuel production for local use based on the expectation that it can increase rural energy access, generate income or create employment. For example, Thailand promotes community collaboration in establishing and managing local bioenergy stations while the Indonesian biofuels taskforce promotes energy self-sufficient villages. Brazilian policy states the importance of enabling “isolated and traditional communities” and farmers to produce their own energy (CIFOR 2011: 10). Similarly, Mozambique envisions a national program of support for small-scale biofuel production, helping farmers to generate income and “respond to their energy needs” (Mataveia, 2009). The EU associates decentralized renewable energy with income generation, employment, local energy security and lower transmission losses than large-scale energy generation (EU, 2009). While these policies promote local biofuel production and use, they do not specify concrete indicators of success. This is especially true for goals related to energy access and local energy security (we evaluate income and employment under smallholder inclusion and job creation). In the absence of other criteria, we return to the IEA (2014) 3

√ — — — — √ — — √ (carbon) — —

Tax Revenues

— — — — — √ — — — — —

Reduced Fossil Fuel Subsidies

√ — — √

√ √ √ — — √

Alternative /Expanded Market for Farmers — — — — — — √ — — — —

Value Addition to Agricultural Products

√ — √ √ — — √ — √* — —

Technological Innovation/ Agroindustrial Development √ — √* √ √ √ √ √ — — —

Job Creation

√ √ √ √ — √ √ √ √ — —

Rural Development

√ √ — √ — √ √ — √ — —

Decentralized Energy Systems

√ √ — — √ √* √ √ √* — —

Smallholder Inclusion / Income / Productivity

4

Decentralized energy systems

Smallholder inclusion

Job creation

-Biofuel* AND/OR -Agrofuel*

Domestic energy security

- Energy security OR energy independence - country name AND (energy security OR enery independence) – job quality OR employment intensity OR job creation (refined as needed to) – specific feedstock AND (employment intensity OR job quality) – smallholder inclusion OR social inclusion – smallholder farm* OR small-scale farm* OR small farm* OR family farm* – upgrading AND value chain – outgrower OR outgrowing – farmer income* – contract farming OR inclusive business model – pro-poor – energy access (then refine to) rural energy access – energy poverty – decentralized energy production OR local energy production OR local energy sufficiency – energy transition AND (traditional OR biomass OR modern)

Search Terms (all possible combinations of the two columns have been applied for each row of policy expectations)

Policy Expectation

Web of Science, Scopus, Google Scholar

Official data or reports published by globally recognized authorities (the International Energy Agency, U.S. Energy Information Agency, and country reports published by the USDA-Foreign Agricultural Service) or energy data clearinghouses (Enerdata, REN21) Peer-reviewed articles with case studies on outcomes of biofuels production Excluded from the findings: modelling articles, scenarios, grey literature

Google Scholar, Google

√ — √* — — — √ — √ — —

Increase Exports

Targeted Data or Article Type

— √ — — — — — — — — —

Reduce Air Pollution

Search Engines

This is motivated by different factors: citizen and policy concerns about climate change (e.g. the EU), and international commitments and/or global leadership ambitions (e.g. China) (USDA-FAS, 2006).

— — — √ — — — —

√ —

Reduce Energy Costs

Table 3 Literature Search Strategy for Review of Evidence.

a

√* √* √* √* √* √* √* √ √* √* √*

√ √ √ √* √

Brazil China Colombia EU India Indonesia Mozambique S. Africa Thailand UK USA

√ √ √ √* √

Domestic Energy Security

GHG Emissions Reductiona

Country

Table 2 Policy Expectations that Biofuels will Produce Benefits (√ = vaguely worded assumption; √* = assumption backed by measures to promote benefits). Sources: Brazil: (Brazilian National Congress, 1993, 1997, 2005; de Andrade and Miccolis, 2011; MAPA, 2006; MDA, 2009a, 2009b); China: (Koizumi, 2008; USDA-FAS, 2006); Colombia: (Conpes, 2008; MADR, 2010; Minambiente, 2016; MME, 2007); EU: (EU, 2009); India: (MNRE, 2009); Indonesia: (Caroko et al., 2011; Sardjono, 2014; Suharto, 2012); Mozambique: (Republic of Mozambique, 2009); South Africa: (Republic of South Africa, 2007); Thailand: (DEDE, 2012; Tongsopit and Greacen, 2012); UK: (Bailis and Baka, 2011; UK, 2007); USA: (U.S. Congress, 1990, 2005; US EPA, 2007, 2010, 2016)

C. Hunsberger et al.

Energy Policy xxx (xxxx) xxx–xxx

Energy Policy xxx (xxxx) xxx–xxx

C. Hunsberger et al.

tribution of biofuels to global net energy consumption; total consumption in the road transport sector; and the transport sector energy mix at country level. Global biofuel production reached 127 billion liters in 2014 (IEA 2014), representing a 0.8% of share of global final energy consumption. This figure rose slightly in 2015, with global ethanol production reaching 98.3 billion liters and biodiesel reaching 30.1 billion liters (REN21, 2016). Yet with the core usage of biofuels and their primary contribution to energy security being in the road transport sector, it is more meaningful to focus on this sector. Biofuels contributed 4% of global road transport fuel demand in 2014 (IEA 2014), representing a small but significant contribution to energy diversification for transport. With energy security concerns focused on the national level, it is important to explore how diversification plays out at the country level. Table 5 shows net biofuel production, net biofuel consumption, and the contribution of biofuels to energy diversification in the transport sector for focal countries. For the majority of countries reviewed, biofuels contribute only small percentages to the national fuel mix. Brazil is the outlier, with an ethanol share of over 50% – with clear energy security gains. While no other country comes close to this, the ethanol industry in the United States and Thailand has reduced gasoline demand by roughly 13% and 12%, respectively – producing significant advances in energy diversification. Colombia follows, with an approximately 7.5% biofuel share. To explore whether biofuel production has lowered fossil fuel imports, we look at whether diversification is achieved through domestic production or imports. With few exceptions, the advances come from domestic production – with consumption closely matching production in most countries reviewed. While the EU produces much of what it consumes overall, member states vary considerably in the percentage of biofuels produced domestically – with up to 72% supplied through trade (e.g. ethanol in the UK, biodiesel in Italy). EU ethanol imports have fluctuated widely over time, with imports from the US reaching 900 million liters in 2011. These dropped to under 100 million liters in 2014 following a complaint from the European bioethanol industry, and a resulting duty imposed by the European Commission on bioethanol from the US (USDAFAS, 2014). Importantly, countries in most need of a domestic biofuel industry to offset oil imports and improve the balance of trade (e.g. Mozambique, South Africa) have negligible production levels despite policy directives. Colombia is the exception, where an E8 blending mandate and subsidized prices have boosed a domestic ethanol industry (USDA-FAS, 2015b). Indonesia and India export more biofuels than they consume, yet this production capacity is a potential source of energy security for the future.

definition of energy security as a combination of accessibility, affordability and reliability. 3.2.5. Rural development Rural development is mentioned in nearly all of the focal policies though the term refers to widely varying aspirations. One set of goals relates to agricultural development: policies interpret biofuel production as a way to foster economic growth, promote modernization, or adopt new technologies in the agriculture sector. Another theme is international engagement: some policies see biofuels as a driver of rural economic activity that can attract foreign investment and help achieve international leadership through trade. Still another theme includes national ambitions for alleviating poverty, improving rural infrastructure and services, and promoting the advancement of particular disadvantaged groups (e.g. South Africa's Black Economic Empowerment plan). Thus, while rural development is consistently mentioned across policies, understandings of the term are too diverse to develop common criteria for success. For this reason we do not assess outcomes for rural development as a coherent category. 3.3. Indicators of effectiveness Table 4 distils a set of indicators linked to the first four policy expectations reviewed above. These will be used to assess whether or not each policy expectation as defined in the reviewed policy documents has been met. In many cases these performance criteria drawn from policy documents reflect narrower interpretations of the expectations (energy security, employment creation etc.) than those used by international agencies or scholars. Section 5 discusses the implications of using such narrow criteria and offers suggestions on how policies could and should expand their criteria for effectiveness based on the review of evidence. 4. Reviewing the evidence: have expectations been met? 4.1. Energy security While over 50 countries have blending mandates in place, many struggle to reach their targets (Lane, 2016; OECD-IEA, 2013). To explore whether biofuels have diversified the energy mix, we must look at production, consumption and trade patterns. Recognizing that the relationship between a country's biofuel consumption and total fossil fuel use depends on a complex set of interactions that we cannot fully investigate using available empirical data, here we explore the conTable 4 Criteria for success linked to each policy expectation. Expectation

Criteria for Success

Basis on which criteria are defined

1. Domestic energy security

(a) Has biofuel production and/or consumption led to a diversification of the energy mix? (b) Has biofuel production led to a reduction in fossil fuel imports? (a) Has the biofuel industry generated employment? (b) Do the distributional outcomes of biofuel-related employment meet policy expectations (jobs in rural areas, in factories)? (c) Are companies producing biofuel adopting good labor standards / “best practices”? (a) Are small-scale, poor, or remote farmers participating in biofuel production? (b) Are smallholders gaining income from growing biofuel crops – enough to compensate for foregone activities?

These criteria focus only on expectations raised by policy documents. Developing country concerns associated with the effects of the fuel import bill on the balance of trade are captured by the second criterion (reduced fossil fuel imports).

2. Job creation

3. Smallholder inclusion

4. Decentralized energy systems

(a) Are biofuels used as an energy source in rural areas where they are produced? (b) Do locally produced biofuels extend energy access to those who previously lacked sufficient energy?

5

The expectation that biofuel production will generate employment is almost a given, provided the investment is economically viable – setting expectations low. We also consider employment intensity, both in absolute terms (jobs created per area of investment) and relative to the jobs or economic values displaced (e.g. jobs created on plantations relative to labor intensity on smallholder farms). These criteria focus only on expectations raised by policy documents. Literature on the social impacts of biofuels identifies additional issues, e.g. whether smallholders ultimately benefit from closer integration with international markets, and whether benefits are equitably distributed within households and communities (BlaberWegg et al., 2015; Creutzig et al., 2013). Additional criteria addressed in other sections: (c) Does small-scale biofuel production for local use generate income and/or employment? (see sections on smallholder inclusion and job creation)

Energy Policy xxx (xxxx) xxx–xxx

C. Hunsberger et al.

Table 5 Biofuel production and consumption trends for major producers, 2014. Sources: USDA-FAS (2015, 2016); US-EIA (2016); Enerdata (2016). Country

USA Brazil EU 1. France 2. Germany 3. Italy 4. UK China Thailand Indonesia India Colombia S. Africa (2012) Mozambique (2012) a

Biofuel Production (million liters)

Biofuel Consumption (million liters)

Biofuels as % of Fuel Consumption

Ethanol

Biodiesel

Ethanol

Biodiesel

Ethanol

Biodiesel

Average

54,181 22,555 5250 1180 850 n/a 230

4842 3500 10,890 1930 3180 340 340 1133 1170 3300 130 606 0 0

50,891 23,200 5635 800 1520 360 820 2951 1053 0 350 423 6 0

5364 3410 12,280 2500 2390 1220 680 1,233 1180 1600 80 606 0 0

13 53.3 3.7 n/a n/a n/a n/a 2.3 12.3 0 1.4 7.6 ~0.01 0

1.4 6.2 5.6 n/a n/a n/a n/a 0.2 5.6 6.3 0.08 7.4 0 0

5.5 15.7 n/a 6.7 5.9 4.2 2.3 n/a n/a n/a n/a 7.5 ~0.01 0

2951 1058 4 2002 406 6 0

Items in italics designate estimates.

it typically becomes – producing trade-offs between productivity and poverty alleviation objectives (Binswanger, 1986). From the perspective of smallholders, employment intensity on one's own farm and on a commercial farm mean very different things (with higher employment intensity representing a burden on the former, and having value on the latter). A more meaningful metric is the relative returns to smallholders growing feedstock on their own farm vs. working as wage laborers. Data compiled by Deininger et al. (2011) across different crops suggest a trend of greater returns to the former (Table 7). With limited employment opportunities on plantations restricting the availability of wage labor to most smallholders, it becomes hard to justify plantations on the basis of their employment benefits alone. The above tables do not capture significant differences in labor intensity during different phases of the production cycle. The decline in employment intensity on jatropha plantations following the labor-intensive land clearing and planting phase has been estimated at 61% to 84% (Leonardo et al., 2015; Schoneveld et al., 2011). Labor intensity also tends to fall over longer time periods as industries mature and intensify. In Brazil, the number of workers employed in sugarcane production has decreased by 62 percent nationally as a result of mechanization (Dufey, 2008; Ortiz and Rodrigues, 2006), and the employment intensity of unskilled workers in south-central Brazil fell from 57 to 40 workers per 1000 ha from 2007 to 2009 (Baccarin et al., 2011). Similar trends characterize the soy sector, in which mechanized cultivation generates on average only 5 to 6 jobs per 1000 ha (Bickel and Dros, 2003). Land concentration has reduced jobs by 44 percent since

4.2. Employment Employment outcomes of biofuels are explored through three lenses: employment intensity (jobs created per unit area or unit volume of biofuel); distributional effects (who secures employment benefits); and job quality.

4.2.1. Employment intensity With job creation argued to be one of the main benefits of biofuels, it is important to explore how many jobs are created in the sector. Most direct employment creation is in feedstock production, with fewer jobs in processing. The number of jobs created in feedstock cultivation, and thus “the scope for agricultural growth to reduce poverty” (Deininger et al., 2011), is often understood to depend on the feedstock and the level of mechanization (Deininger et al., 2011; Diop et al., 2013; cited by Pradhan and Mbohwa, 2014). Table 6, which summarizes employment intensity across feedstocks and business models, largely supports this view. Oil palm, cassava and jatropha have far higher employment intensities than maize and soybean, largely because the harvest is difficult to mechanize. Employment intensities among commercial scale operations also vary widely with levels of mechanization. The four-fold difference in employment intensities of oil palm plantations between Indonesia and Malaysia, for example, reflects differences of scale and sophistication (Barlow et al., 2003; cited by World Bank, 2010). The more capital intensive an enterprise, the less labor intensive

Table 6 Employment intensity for different feedstocks on smallholder farms and commercial plantations. Feedstock

Employment Intensity (jobs /1000 ha) Smallholder

Commercial Plantations

Cassava Jatropha Maize Oil palm

546–792 119–412b 154 500

110–370a 60–420c 8–10 9–380

Soybean Sugarcane

296–373d

2.7–18 110–700e

a b c d e

Source

Diop et al. (2013); Tshiunza (1996); van Eijck et al. (2012) Deininger et al. (2011); Loos (2009); Pradhan and Mbohwa (2014); Schoneveld et al. (2011); van Eijck et al. (2012) Deininger et al. (2011) Barlow et al. (2003); Deininger et al. (2011); Li (2011); Pradhan and Mbohwa (2014); PT SMART (2008); World Bank (2010) Bickel and Dros (2003); Deininger et al. (2011); Lima et al. (2011) Deininger et al. (2011); Pradhan and Mbohwa (2014); Schuenemann et al. (2016)

Thailand and Mozambique. Varies by degree of intensification and cultivation year (with higher labor inputs for plantation establishment). Romijn et al. (2014) cite a wider range of 30 to 1030 for Mozambique, but the highly standardized nature of data acquired suggests it is official data – which tend to be inflated. Rainfed and irrigated outgrowers, respectively (Malawi). Lower figures are for irrigated, mechanized harvest (Brazil, Mozambique) and higher figures are for irrigated, manual harvest (Tanzania).

6

Energy Policy xxx (xxxx) xxx–xxx

C. Hunsberger et al.

Colombia, Indonesia and Malaysia, and soya-based biodiesel in Brazil (van Gelder and German, 2011). Other focal countries have had negligible investments in biofuel production.

Table 7 Relative returns to labor for smallholders growing feedstock on their own land vs. employed on biofuel estates (adapted from Deininger et al., 2011). Feedstock

Farm Income–to–Wage Ratio

Maize Oil palm Sugarcane

3.13 to 9.91 0.88 to 3.05 6.09

4.2.2. Distributional outcomes (who is employed) While it is often argued that employment can offset the costs of land loss, consistent evidence reveals employment biases that favor migrant workers over local residents – both for biofuels and in plantation agriculture generally (Colchester, 2011; Deininger et al., 2011; German et al., 2011c; Li, 2011; Obidzinski et al., 2012; Smalley, 2013; Thondhlana, 2015; World Bank, 2010). One study found employment benefits on a jatropha plantation in Ghana accrued to as few as 4% of land losing households (Schoneveld et al., 2011). Another found the majority of jobs associated with a sugarcane investment in Zimbabwe (66% in the plantation, and 80% in the factory) went to people outside the district – despite concerns voiced by displaced farmers (Thondhlana, 2015). Women, indigenous people and other vulnerable groups are also less likely to obtain employment from investors (Deininger et al., 2011; Dihr, 2015). This is of particular concern in cases where communities gave up productive land with the expectation of employment benefits (Deininger et al., 2011). While local residents are often seen to resist plantation employment because of poor employment conditions and competition with traditional livelihood activities (e.g.Dayang Norwana et al., 2011; German and Parker, 2015), mounting evidence suggests that new employment prospects are desirable but inaccessible for local residents. Reasons may include greater familiarity with sedentary agriculture among migrants; a perceived work ethic among certain ethnic groups; absence of qualified skilled workers; and difficulty extracting consistent, cheap labor from those with alternative economic activities in the vicinity of plantations (Colchester, 2011; German and Parker, 2015; Li, 2011; Thondhlana, 2015; World Bank, 2010). While some oil palm companies in Indonesia have implemented affirmative action policies to recruit local residents, hopes of employment continue to attract spontaneous migrants – sometimes producing conflict over land and jobs (Feintrenie et al., 2010). Research from Brazil's sugarcane industry illustrates the draw of migrant labor for plantation operators. Manual laborers in the sugarcane industry in Sao Paolo state tend to come from the northeast, where there is limited opportunity for employment (Silva, 2007). These workers are perceived as “socialized in the arduous and difficult work of agriculture in their region of origin”, “more dedicated to work and grateful to employers”, and having a “pressing need to earn money to ensure the livelihood of the distant family”, and are therefore considered effective workers who are more likely to tolerate breaches of labor laws (Novaes and Alves, 2007). Companies also reportedly prefer young men for their “good conduct,” relative “subordination” and ability to maintain good health and attendance, and therefore productivity, throughout the harvest (Rosa, 2009).

the 1980s despite the explosion of soy demand and acreage (Wilkinson and Herrera, 2010). It is also important to look at factors shaping the magnitude of benefits from employment, such as the success of the investment and the number of jobs created relative to those displaced (German and Parker, 2015). While disappointing employment benefits have been a general trend of the recent land rush (Deininger et al., 2011), biofuel investments have been particularly prone to technical or economic failure – and thus some of the worst performers from the perspective of job creation. The confluence of new market opportunities associated with policy commitments to renewable energy and the hype surrounding untested biofuel crops like jatropha contributed to a boom in risky investments, resulting in the widespread failure of many investments. Comparing jobs created with jobs displaced, evidence is patchy. An evaluation of a jatropha plantation in Ghana found displaced land uses to generate greater value per unit area than employment (Schoneveld et al., 2011). Evidence of the opportunity costs associated with oil palm is contradictory. Li (2011) found employment intensities across 15 oil palm plantations in West Kalimantan, Indonesia to be less than a hundredth of those of non-oil palm smallholdings displaced by these plantations. On the other hand, a comparison of returns to labor from smallholder oil palm and other crops in Bungo District, Indonesia (Feintrenie et al., 2010) found oil palm to out-compete rubber and rice by a factor of 1.7 to 2.1 and 21.2, respectively, and rice by a factor of 9.4 even during a slump in oil palm prices. The fact that these cases differ not only in the business model analyzed (plantation vs. outgrower) but also in the metrics employed (returns to land vs. labor) makes comparison difficult. A study of household income and poverty among households integrated and not integrated into a sugarcane investment in Malawi found net annual income to be equivalent for plantation workers and independent smallholders, and no difference in poverty indices (Herrmann and Grote, 2015). Reasons were related to trade-offs associated with new labor allocations, with average losses in farm income of 48% offsetting gains in off-farm income. These studies demonstrate that net employment effects depend not only on the nature of the investment (crop, business model) and the context, but also on the evaluation metrics employed. Additional employment may be generated in processing, with estimates of potential employment creation from different feedstocks provided in Table 8. In reality, much of the capital invested in biofuel production from 2000–2009 was concentrated in a few countries and sectors: sugar-based ethanol in Brazil, palm-based biodiesel in

4.2.3. Job quality Job quality in the biofuel sector varies considerably by job type (plantation vs. factory vs. small-scale farms), and by skill level and seniority. Factory work tends to be more permanent and higher paying, even for low-level positions. Brazilian workers in ethanol refining receive about 30 percent higher salaries than sugarcane harvesters (UNEP, 2008). Feintrenie et al. (2010) found that work in a palm oil refinery in Sumatra, Indonesia was locally considered a ‘good job’ – representing an opportunity to move out of farming and improve one's social status (if not always one's livelihood), and providing access to housing and social services. While factory work is more permanent, formal and high-paying than plantation work, far fewer jobs are created in processing than on plantations. While job quality varies considerably, plantation employment tends to be unskilled and temporary – undermining poverty alleviation benefits (Macedo, 2005; Marti, 2008; World Bank, 2010). A

Table 8 Employment intensity in biofuel processing. Biofuel Type

Employment (jobs/million liters/yr)

Source

Ethanol

1.04–39

Cassava Corn Sugarcane Biodiesel

117 1.1 5.1–121 6.0–83.3

Oil palm Soy

73.3–128 3.5

Jupesta (2010); Pradhan and Mbohwa (2014) Pradhan and Mbohwa (2014) Pradhan and Mbohwa (2014) Pradhan and Mbohwa (2014) Jupesta (2010); Pradhan and Mbohwa (2014) Pradhan and Mbohwa (2014) Pradhan and Mbohwa (2014)

7

Energy Policy xxx (xxxx) xxx–xxx

C. Hunsberger et al.

cutting on their health. An organization serving migrant workers registered 23 worker deaths during working hours from 2004 to 2009 (Facioli and Peres, 2007, cited in Galiano et al., 2012). Epidemiological studies further demonstrate the health impacts of cane cutting. In Nicaragua, an epidemic of chronic kidney disease has been linked to employment in sugarcane fields, likely due to strenuous labor and poor hydration (Laws et al., 2015). These conditions extend to other world regions: a study from Zimbabwe (Thondhlana, 2015) reports worker concerns over low wages, late salary payment, ad hoc retrenchments and grossly exploitative working conditions. A study from Zambia (German and Parker, 2015) reports poor living conditions and a high risk of HIV/ Aids from employment policies favoring workers from distant provinces and prohibiting families from visiting. Evidence on employment conditions on soy plantations also comes from Brazil. While some reports complain that the local population can only find temporary, unskilled work (Bickel and Dros, 2003), national statistics suggest the proportion of temporary positions ranged from 18% to 24% in the early 2000s – significantly less than for other crops. On the other hand, labor conditions during land clearing are poor and are generally below Brazilian and ILO standards. The Ministry of Labor and ILO document widespread use of slave labor for forest clearing and land preparation for soybean and other crops in Brazil (723 cases) (Bickel and Dros, 2003; ILO, 2007). This is practised by falsifying dates on employment contracts, failing to pay or under-paying relative to agreed wages, and sub-contracting intermediaries for labor-intensive tasks. With the cost of opening new areas for soy varying by a factor of 10 (Bickel and Dros, 2003), such practices are likely to be highly variable across companies. Most evidence on employment conditions on jatropha plantations comes from Africa. Wages are reportedly between US $33/month to $333/month, with the majority in Ghana earning $53/month (Acheampong and Campion, 2014). The US $50/month reported in one study was found to be only 51% of the district average for household income (Schoneveld et al., 2011), yet even then, 67% of respondents considered plantation employment to have had a net positive impact on their livelihoods. This was attributed to the regularity of income flows rather than income per se. Farming remained an important component of household livelihoods for employees, contributing to income and food security. In a rare study of jatropha from Latin America, Skutsch et al. (2011) found women's employment in nurseries to be poorly remunerated and short-term – only 50 days in each location to produce seedlings for distribution to farmers. In Mozambique, Romijn et al. (2014) found companies provided on-the-job training, education, meals, health care and transport to employees; however, most of these benefits were channeled to permanent workers. Perhaps most importantly, most employment benefits in the jatropha industry evaporated quickly due to project failure.

review of biofuel impacts on developing countries found documented wages both above and below the official minimum wage and the World Bank-defined poverty line of US $2 a day (Diop et al., 2013). The literature suggests a host of factors shape the quality of plantation jobs, including legislation, crop characteristics, commodity prices, labor scarcity, and specific labor arrangements (German et al., 2011a; Loewenson, 1992; Tiffen and Mortimore, 1990). Civil society organizations and some academics are unwaveringly negative about employment conditions on oil palm plantations. Some focus on poor working conditions and wages too low to reduce poverty (Marti, 2008), others on captive labor created by failed transmigration programs (Li, 2011) or exploitative relationships in migrant worker schemes (Colchester and Chao, 2011). While these situations undoubtedly favor employers and unfair contractual arrangements, the views of workers themselves provide a more nuanced picture. In Malaysia, where most plantation workers are Indonesian migrants, 77% of respondents on two oil palm estates reported that employment improved their incomes and livelihood conditions (Dayang Norwana et al., 2011). These perceptions are undoubtedly shaped by the facilities provided to employees: housing, potable water, electricity, medical services, schooling and sports facilities. In Indonesia, a study across three sites (Obidzinski et al., 2012) found mixed worker perceptions: employees in one site overwhelmingly reported positive net livelihood impacts, while nearly half of respondents in another site (predominantly indigenous Papuans) perceived negative impacts. Most positive impacts were attributed to higher income and more regular income flows. Yet in terms of livelihood consequences of employment, only 46% of respondents reported positive effects. In West Kalimantan, 50 to 80% of plantation workers had temporary employment and no job security,1 yet their ability to maintain complementary income from farming meant that 75% reported improved living conditions due to employment. In the other sites in Papua, less than one-third of respondents reported livelihood improvements. Experiences with plantation employment also depend on cultural differences, including prior experience with waged labor and the magnitude of the shift away from customary ways of life. Indigenous communities in Indonesia and Malaysia have had the hardest time adapting to wage labor on oil palm plantations, and are more likely to regret land loss to oil palm (93% of respondents in one site) and to prefer customary ways of life (Dayang Norwana et al., 2011; Obidzinski et al., 2012). Poor working conditions linked to performance-based pay and long working hours were also observed (Obidzinski et al., 2012). Working conditions in Brazil's sugarcane sector—where 200,000 people work as harvesters—are notoriously poor, marked by crowding, poor hygiene and nutrition, violence by company security guards, and temporary employment (Balsadi, 2007; UNEP, 2008). Many find themselves in a form of debt peonage resulting from exorbitant charges by employers for transportation, accommodation, and food (UNEP, 2008). Exploitative working conditions are particularly striking in the Northeast, due to a centuries-long history of sugarcane cultivation that entrenched unequal employee-worker relations (Lehtonen, 2011). Poor working conditions are also linked to performance-based pay (Galiano et al., 2012). In some companies, the amount of cane cut must reach a minimum equivalent salary for a worker to be re-hired, reportedly to comply with labor standards (World Bank, 2008). Daily cutting quotas can often only be achieved with the help of family members (Lehtonen, 2011). This system has increased the amount of cane cut per worker from 3 to 14 t per day during the 1990s and 18 t by the mid-2000s (Alessi and Scopinho, 1994; IEA, 2008, cited in Galiano et al., 2012). Eighty-six percent of workers interviewed by Galiano et al. (2012) reported physical exhaustion and concerns over the impacts of cane

4.3. Smallholder inclusion 4.3.1. Smallholder participation While providing opportunities to smallholder farmers was an explicit goal for most focal countries, many of the same policies provide incentives for large-scale biofuel production. In Brazil, ethanol policies since the 1970s have favoured large-scale sugarcane producers in the Central and Southeast regions (Hall et al., 2009). Lehtonen (2011) finds that efforts to boost sugarcane production in the poorer Northeast have benefited regional elites while “the benefits have failed to trickle down to the poorest.” Despite social inclusion measures in its biodiesel policy, Brazil has fallen short of its goals to increase participation by family farmers in poorer regions (Padula et al., 2012; Wilkinson and Herrera, 2010). Biodiesel production remains concentrated in the Central and Southeast regions where large-scale soy production dominates (Florin et al., 2013). In 2010, only 246 of more than

1 Annual reports from oil palm companies themselves suggest use of casual labor may be commonplace (PT Smart, 2015).

8

Energy Policy xxx (xxxx) xxx–xxx

C. Hunsberger et al.

outcomes and ‘exclusion’ with negative ones, ‘adverse incorporation’ arises as a more complicated possibility – one that can cause smallholders to sink deeper into poverty. Similarly, one can ask whether opportunities to grow jatropha have been a blessing or a curse for precarious farmers. Smallholders who rely more on income from crops have been found to take larger risks when planting jatropha than farmers with more land and diverse income sources: smallholders are more vulnerable to losses if a market fails to develop, and to incurring debts if their crop fails (Rittenburg et al., 2011).

100,000 family farmers participating in the biodiesel program (PNPB) were located in the north of the country (Cesar and Batalha, 2013), while in 2014 only 1% of biodiesel produced in Brazil came from crops considered suitable for smallholder production, such as castor and sunflower (USDA Foreign Agricultural Service, 2015). Lima et al. (2011) report that while PNPB has compelled large soy companies to support some small-scale farmers, there has also been a trend toward consolidating smaller soy plots into larger ones. By contrast, smallholders account for 30–40% of the area planted with oil palm in Indonesia (McCarthy, 2010; Rist et al., 2010) – a relatively high level of smallholder participation. The Government of Indonesia supported this outcome during the New Order period by facilitating smallholder access to credit, technical assistance, inputs and markets, while requiring that at least 80% of land in nucleus estate-smallholder schemes be allocated to smallholders (McCarthy 2010). However, outcomes for smallholders have been highly uneven – even more so with the increasing withdrawal of the state. Some consider oil palm to be “a rich farmer's crop” because it requires significant up-front investment, labour and time before it bears fruit. Successful migrants and local elites in Indonesia have reportedly been better able to benefit from the spread of oil palm than poor smallholders (Feintrenie et al., 2010; McCarthy, 2010). Evidence is mixed on whether jatropha – touted as a “pro-poor” crop (Brittaine and Lutaladio, 2010) – is accessible to the poorest farmers. In Tamil Nadu, India, Ariza-Montobbio and Lele (2010) found large landowners disproportionately represented among 45 jatropha adopters, with farmers who owned a well (and could thus irrigate their plantations) having greater success than those who did not. However, jatropha-growing households in Andhra Pradesh had smaller landholdings and were more dependent on agriculture than those in Tamil Nadu (Axelsson et al., 2012). In Zambia, a survey of 249 farmers found that those who were better-off were most likely to benefit from jatropha activities (Kuntashula et al., 2014), though a study in Kenya concluded that financial assets did not seem to influence jatropha adoption (Mogaka et al., 2014). In Malawi, poorer farmers without off-farm income were found to be more likely to plant jatropha than richer farmers with livestock and off-farm income (Mponela et al., 2011). Overall, evidence suggests that incentives to participate in biofuel production have been preferentially awarded to farmers with more land and resources. Smallholders face barriers including lack of land tenure (de Andrade and Miccolis, 2011), lack of control over water resources (Ariza-Montobbio and Lele, 2010), and lack of capital needed to make up-front investments and wait for slow-growing plants to mature (Ariza-Montobbio and Lele, 2010). The literature provides examples of companies avoiding farmers who they consider riskier or less desirable (Lima et al., 2011), for example gravitating toward the largest farmers who qualify as ‘smallholders’ to reduce transaction costs. There is also evidence of government subsidies being directed to better-off households in better-off communities (Soto et al., 2015). Research in the Philippines finds that indigenous people face specific barriers to benefiting from contract farming: they tend to be either excluded entirely or compelled to give up their traditional agricultural practices of polyculture and shifting cultivation (Montefrio and Sonnenfeld, 2013). Smallholders’ decisions about whether to engage with biofuel crops often appear to be influenced by economic and demographic factors. Multiple studies warn that smallholders can be enticed by short-term incentives into signing contracts disadvantageous to them over the longer term (German et al., 2011a; Rist et al., 2010; Skutsch et al., 2011). Farmers’ decisions are also shaped by their perceptions about the environmental, livelihood and cultural consequences of biofuel crops – an area where research is emerging (Montefrio et al., 2015). While policies assume that participating in biofuel production benefits smallholders, the literature reviewed presents a mixed picture. McCarthy (2010) shows that much depends on the terms of engagement with oil palm: rather than equating ‘inclusion’ with positive

4.3.2. Smallholder income This section investigates how much income smallholders gain from growing biofuel crops and whether it is enough to compensate for foregone activities. Despite isolated successes, income opportunities for many smallholders remain disappointing. The most notable beneficiaries of smallholder energy crop production have been oil palm producers due to the relatively high returns to land and labor, and the infrastructure and services provided by Indonesia's Transmigration program (Feintrenie et al., 2010; McCarthy, 2010; Obidzinski et al., 2012; Rist et al., 2010). Yet benefits are uneven, for reasons McCarthy (2010) documents in four villages in Sumatra, Indonesia. The most successful farmers earned over US $22,000 per year, using their surplus to buy more land and hire labour. Moderately successful smallholders invested smaller profits into converting “unproductive rubber lands” into oil palm production (McCarthy, 2010). However, poorer farmers lost land because they were unable to repay debts, or became trapped in a cycle of low productivity because they could not afford fertilizer or high-yield seedlings. Factors affecting these outcomes included differences in smallholder development programs, land tenure systems, community influence over local institutions, spatial characteristics such as proximity to a major road, and ways in which oil palm cultivation interacted with existing livelihood strategies (McCarthy, 2010). In Brazil, low income generation, especially in poorer regions, and lack of value addition opportunities for small farmers are among the reasons why a literature review concluded that the country's social inclusion measures for biodiesel were not working (de Andrade and Miccolis, 2011). Based on a quantitative model drawn from farmer surveys and field experiments, Dal Belo Leite et al. (2015) conclude that castor and sunflower offer only “small or absent” economic benefits for family farms and consider the possibility of expanding these small-scale alternatives to soy to be limited (22). Not only the amount of income matters, but also how it is distributed. Cash crops, including oil palm, tend to produce income in lump sums rather than steady increments, making it difficult for households to manage expenses throughout the year (Anderman et al., 2014). Long delays before slow-maturing crops such as oil palm and jatropha produce a harvest can mean smallholders must wait 3–5 years before receiving a return on their investments of money, land and labour. Many jatropha farmers in India could not wait that long and abandoned their plantations before their first harvest (Axelsson et al., 2012). Numerous studies find that jatropha has produced disappointing levels of income for smallholders – if any at all. In Tanzania, Grimsby et al. (2012) found that a jatropha economy depended on people being willing to work for less than US$1/day (0.90); prices were so low that farmers allowed children and poor people to harvest their seeds for free. In Zambia, 69 out of 100 smallholders reported that their economic situation had declined since planting jatropha (Kalinda et al., 2015), while German et al. (2011a) calculated that the economic return on labour for growing jatropha was as low as US$0.06/day. A livelihood analysis of 314 jatropha growers in Tanzania found that better-off farmers earned more from jatropha agroforestry than poorer farmers, though this income was proportionally more important to the poorer farmers (Fasse and Grote, 2015). In Mali, interviews with 30 smallholders who had grown jatropha for at least three years found that 9

Energy Policy xxx (xxxx) xxx–xxx

C. Hunsberger et al.

biofuels being used in the same areas where the feedstock was grown. It is not clear whether this discrepancy occurs because communitylevel projects are less likely to be written up in peer-reviewed publications or because the number of working decentralized energy projects is actually small. Certainly many jatropha projects that aimed to provide energy for local use failed to achieve their goals and/or changed their focus to non-energy products because of low yields, fluctuating prices and other obstacles (see for example Favretto et al., 2014; Grimsby et al., 2012; Mogaka et al., 2014). A community palm oil project in Tanzania experienced a similar shift, initially planning to sell palm oil for transportation but finding it more economical to sell their product for edible oil, cosmetics or pharmaceuticals instead (FAO and PISCES, 2009). Several decentralized energy projects focus on providing electricity. In India, a project to produce domestic electricity from pongamia oil was reportedly being expanded to 100 villages (Ewing and Msangi, 2009) and a rural electrification project using jatropha oil had enabled over 100 households to access 3 h of electricity per day (FAO and PISCES, 2009). Other local electricity projects include supplementing a community-level diesel generator with plant oil, as in Kipini, Kenya (Archer, 2012). Other projects focus on providing cooking fuel. In Ethiopia, cookstoves using ethanol derived from surplus molasses from the sugar industry have been distributed in refugee camps, reducing time and health burdens on women who would otherwise need to collect firewood while relieving pressure on nearby forests (Rogers et al., 2013). While this project uses ethanol that was produced non-locally in a centralized way, the organizers plan to introduce the same stove technology in Brazil together with ethanol micro-distilleries (FAO and PISCES, 2009). Nigeria's “cassakero” project involves 10,000 cassava micro-distilleries that similarly aim to substitute ethanol for wood as a cooking fuel; however, running the micro-distilleries in areas that lack reliable electricity requires burning 6.4 kg of wood per litre of ethanol produced, undermining the project's goal of reducing deforestation (Ohimain, 2015). Still other projects aim to supply fuel for local transportation or farm equipment. A sugarcane micro-distillery in Brazil (600 L/day, 70 farmers) produces ethanol to be used first and foremost by local cooperatives for equipment and vehicles. A study calculated that after 5 years spent paying back the initial loan, farmers would earn the equivalent of almost 4 minimum salaries per hectare of sugarcane and be able to use byproducts of sugarcane processing as supplementary animal feed and fertilizer (Maroun and La Rouere, 2014).

selling seeds into an energy value chain was not profitable; farmers gained more income by making jatropha soap (Favretto et al., 2014). These outcomes for jatropha growers cast doubt on the aspirational policy statement that energy crops give smallholders access to new markets. Markets for jatropha seeds failed to develop in Kenya (Hunsberger, 2014; Mogaka et al., 2014) and in Zambia, where 92/ 100 farmers in one survey said there were no markets for selling jatropha seeds, and the few who were able to sell some seeds lacked information about prices that could have helped them bargain (Kalinda et al., 2015). Another study in Zambia reported that a company claiming to have contracts with 25,000 farmers vanished before buying any harvests (German et al., 2011a). In India, Axelsson et al. (2012) report that 21 out of 106 jatropha farmers had harvested their plantations and only three managed to sell their crop. The lack of market development for jatropha reflects its experimental nature; many smallholders ended up bearing the risk of working with this unproven crop. Regarding foregone activities, the literature provides examples of how smallholder biofuel production can mean sacrificing food and feed production as well as other ecological resources. In Tamil Nadu, ArizaMontobbio and Lele (2010) found that 82% of those growing jatropha had converted land previously used for food crops, and half had dedicated 50% of their total land to jatropha. Giving up groundnut cultivation was particularly important as it represented a source of protein, edible oil and a unit of exchange (Ariza-Montobbio and Lele, 2010). Also in Tamil Nadu, Baka (2014) found that the tree species Prosopis juliflora growing on ‘wastelands’ already provided energy services, jobs and economic opportunities that were threatened by plans to replace Prosopis with jatropha for biodiesel. A key aspect of whether smallholders gain income from growing biofuel crops is whether buyers honour the terms of their contracts. Contract violations have been noted with castor growers in Brazil (de Andrade and Miccolis, 2011) and jatropha farmers in India (ArizaMontobbio and Lele, 2010). The largest jatropha outgrower scheme in Zambia involved one-sided contracts signed by farmers but not the company; terms included a 30-year agreement to sell exclusively to the company, while the costs of technical support and inputs were to be deducted from future profits rather than provided for free. In 2008 the company disappeared altogether (German et al., 2011a). Promises to provide smallholders with technical support or other services have also proved less than reliable. While four jatropha pilot projects in Mali all reportedly provided farmers with technical assistance (Favretto et al., 2014), outcomes were less favourable in other settings. Lack of technical assistance has been identified as a shortcoming in achieving Brazil's social inclusion measures for biodiesel (de Andrade and Miccolis, 2011) and as a source of conflict with oil palm growers in Indonesia (Rist et al., 2010). In India, most farmers interviewed in Andhra Pradesh and Tamil Nadu reported that they received free jatropha seedlings as the government promised, but less than 40% had received other forms of support (Axelsson et al., 2012). While some farmers in Zambia who received technical support for jatropha stated that it helped them improve their maize and bean farming, jatropha growers also reported that promised incentives were not always delivered (Kalinda et al., 2015). A review of relevant literature therefore suggests that smallholders are participating in biofuel production to a much lesser extent than envisioned in policy documents, and those who do participate receive uneven benefits in terms of income and connection to new markets.

4.4.2. Do locally produced biofuels extend energy access? While ethanol stoves for refugees in Ethiopia filled a gap for people who lacked access to cooking fuel, in other cases biofuels have done little to improve patterns of energy access. Research from India suggests that growing jatropha as an energy crop threatened to displace existing energy crops: in Tamil Nadu, one-fifth of farmers surveyed had substituted jatropha for crops such as pigeon peas that already produced fuel (Ariza-Montobbio and Lele, 2010), while Baka (2014) found that developing jatropha would disrupt an existing local energy economy involving Prosopis. Some efforts to produce biofuels for local use appear to have supplied energy to users who already had other alternatives. In Mpeketoni, Kenya, a machine for producing jatropha oil and milling grain sat idle as farmers harvested jatropha seeds in 2009, while down the road a diesel generating station built by the Kenya Power and Lighting Company sold electricity at nationally standardized rates (Hunsberger, 2010). The same jatropha project, which also aimed to provide fuel for household lighting, had by 2013 given up on jatropha lamps due to problems with their design and was selling jatropha oil as a skin product instead (Hunsberger, under review). In 2013 a project in Shimba Hills, Kenya was processing jatropha oil and selling it to a local restaurant where it was used in special floating-wick lamps – providing

4.4. Decentralized energy systems 4.4.1. Are biofuels used where they are produced? This section explores the desire to improve rural energy access by producing biofuels for local use. The literature reveals many hypothetical discussions and models describing potential benefits of decentralized biofuel systems, but relatively few documented cases of 10

Energy Policy xxx (xxxx) xxx–xxx

C. Hunsberger et al.

production, notably in Indonesia, government policies and programs have played a crucial role. Smallholders who have benefited from producing oil palm have tended to be those who had ample government support or adequate landholdings and sufficient alternative income to afford the significant investments required to establish successful oil palm plots and wait for the plants to mature (McCarthy, 2010). Including the poorest farmers in oil palm production remains an unsolved challenge as successful oil palm smallholders in Indonesia have tended to be migrants heavily supported by government, or local elites (McCarthy, 2010). Different business models shape opportunities and risks; independent producers face greater obstacles to acquiring high quality germplasm and selling their produce than those who are affiliated with state or private sector schemes (McCarthy, 2010; Vermeulen and Goad, 2006).

income for farmers but providing energy only to those who already had other sources (Hunsberger, under review). Similarly, biodiesel made from jatropha grown by smallholders in Tanzania was sold to nearby safari companies branding themselves as ‘eco-green’ (van Eijck, 2009). These examples suggest that while many biofuel initiatives start out with the aim of enhancing energy access in remote rural areas, economic incentives and changing circumstances may pull these projects in other directions: producing non-energy products such as soap, cosmetics or edible oil; burning substantial amounts of wood to produce a substitute for fuelwood; or supplying energy to users who already have reliable energy sources. 5. Discussion: unbundling biofuels 5.1. Have expectations for biofuels been met?

5.2. Gaps and omissions in the expectations This section summarizes the extent to which biofuels have met expectations – in policy documents’ own terms – and considers the circumstances surrounding more successful outcomes. Based on the evidence reviewed here, three expectations appear more likely to be fully or partially met. First, biofuel production and use targets have been reached in particular countries, substituting some demand for fossil fuels – especially in Brazil, the United States and Thailand. The increased use of biofuels in these countries has diversified energy sources, reduced import dependence and saved foreign exchange. Second, the biofuel sector has indeed created employment in some locations, particularly plantation work for labour-intensive crops – with several caveats discussed below. Third, growing biofuel crops has generated income for some smallholders, for example some oil palm growers in Indonesia. At the other end of the spectrum, three expectations appear less likely to be met. First, biofuel production has not done well at creating stable, long-term, high-quality jobs, especially for local residents of areas where biofuels are produced. Many of the jobs created in the sector are unskilled, strenuous, temporary or seasonal, and tend to be awarded to migrants rather than local residents. Second, while some smallholders have been able to benefit from producing biofuels, the expectation that biofuels would provide opportunities and alleviate poverty for the most disadvantaged farmers remains far from being met. Farmers with the least land and assets face barriers to participating in biofuel production, while most smallholders who do participate fail to gain a higher level of income than they could have earned through alternative livelihood activities. Third, with a few exceptions, there is little evidence to suggest that energy access in remote rural areas has improved due to decentralized biofuel production and use. Where expectations have been met, it is important to consider what circumstances may have helped enable success. In Brazil, often regarded as the world's best success story in terms of biofuel production, strong subsidies and incentives for biofuel production, research and development were sustained over a period of decades. The country's climate is favourable for growing sugarcane and soy, two crops that have high yields per hectare and are suitable for mechanized large-scale cultivation. Low-cost labour has also boosted Brazil's biofuel sector. While Brazil has taken steps to improve labour conditions, diversify its energy crops and encourage smallholder participation from poor regions, results of these efforts have been mixed. It is significant that even with considerable policy intervention (e.g. the Social Seal program for biodiesel and the National Commitment for the Improvement of Labor Conditions in Sugarcane Production), benefits continue to flow to larger-scale operators. Where biofuel production has created jobs, these most often involve crops with well-known agronomy and established markets. Agricultural employment intensity is greater for crops whose cultivation is not easily mechanized. Factory work in biofuel processing tends to be more skilled, high paying and permanent than plantation work. Where significant numbers of smallholders participate in biofuel

This section asks whether biofuel policy expectations capture an appropriate range of possible outcomes, keeping in mind that risks related to food security, land rights and environmental impacts are discussed in a companion paper (Goetz et al. this issue). After identifying gaps and omissions in the expectations as defined in the focal policies we propose an expanded set of criteria for these expectations. Based on the literature reviewed here we conclude that many of the expectations derived from the policies are framed in terms that are far too narrow to capture what might constitute a sustainable industry. For example, policy expectations for employment focus on whether jobs are created, with scant attention to crucial considerations such as employment intensity, job quality, duration, or distribution. A full accounting of the net employment effects of biofuel investments should also take into account the full suite of effects, positive and negative, at the local level – as well as economy-wide effects beyond the local level. Peters and Thielmann (2008), for example, argue that the indirect effects associated with “crowding-out” – in which job losses occur in rival industries competing for land or buyers – should be considered for a full accounting of employment effects. Similar gaps occur in policy statements on the presence of smallholder participation rather than the nature and outcomes of this participation, and on income as the primary measure of outcomes for farmers. A related gap is that policy expectations focus on aggregate outcomes while paying little attention to distributional effects. Criteria for success drawn from policies alone are blind to the household-level distribution of benefits and impacts. With the exception of efforts to promote opportunities for farmers in poor regions (Brazil) or historically disadvantaged black farmers (South Africa) they are also largely blind to systematic biases along lines of class, ethnicity, gender and age. This study reinforces the call for biofuel analysts and decisionmakers to devote much more effort to explicitly targeting equitable benefits, and to performing disaggregated analysis and addressing equity concerns (Blaber-Wegg et al., 2015; Creutzig et al., 2013). Some of the expectations outlined in policy documents are simply too vague to operationalize. The theme of rural development is expressed in very broad terms ranging from alleviating poverty and extending rural services to achieving world leader status through trade. Not only do the interpretations of rural development vary widely between countries; for this theme in particular, objectives and desired outcomes are weakly specified, making it very difficult to assess progress. Based on these observations we propose an expanded set of criteria to assess the expectations reviewed in this paper (Table 9). 5.3. Tensions and trade-offs between the expectations In addition to leaving out important considerations, ‘bundled’ expectations for biofuels contain tensions and trade-offs that most 11

Energy Policy xxx (xxxx) xxx–xxx

C. Hunsberger et al.

Table 9 Expanded criteria linked to selected policy expectations. Expectation

Criteria Drawn from Policies

Additional Criteria

1. Domestic energy security

(a) Has biofuel production and/or consumption led to a diversification of the energy mix? (b) Has biofuel production led to a reduction in fossil fuel imports? (a) Has the biofuel industry generated employment? (b) Do the distributional outcomes of biofuel-related employment meet policy expectations (jobs in rural areas, in factories)? (c) Are companies producing biofuel adopting good labor standards/“best practices”? (a) Are small-scale, poor, or remote farmers participating in biofuel production? (b) Are smallholders gaining income from growing biofuel crops – enough to compensate for foregone activities?

(c) Is energy security at scales other than the national also enhanced by biofuel production and use? (d) Do biofuels advance energy security goals over both the long and short term?

2. Job creation

3. Smallholder inclusion

4. Decentralized energy systems

(a) Are biofuels used as an energy source in rural areas where they are produced? (b) Do locally produced biofuels extend energy access to those who previously lacked sufficient energy?

(d) Are jobs created in the biofuel sector of sufficiently high quality in terms of working conditions, remuneration, duration and other terms of employment? (e) Are employment opportunities preferentially offered to households that are displaced or otherwise negatively impacted by biofuel activities, and are the returns to land and labor greater than for the land uses displaced? (c) Are the benefits and impacts of participating in biofuel production equitably distributed within households and communities? (d) Do the benefits of producing biofuels compensate for impacts not usually measured as income, such as reduced access to land and resources and reduced time for other activities? (c) Do decentralized energy projects use accessible, appropriate technologies that are welcomed by local residents? (d) Are decisions concerning local energy projects made in inclusive, transparent ways?

6. Conclusion and policy implications

policies leave unacknowledged. In some cases the assumptions underlying policy goals push in different directions; in other cases competition between interests can arise when goals are viewed at different spatial scales or time frames. While biofuel policies tend to imply that their multiple expectations are complementary, some combinations of objectives are likely to compete such that advancing one goal undermines others. For example, goals related to national energy security, agricultural modernization and production efficiency all favour a capital-intensive, high quantity production model, while goals related to employment creation, smallholder inclusion and decentralized energy systems favour a labour intensive, geographically distributed production model. Experience from Brazil demonstrates how difficult it can be to encourage both strategies simultaneously, with large-scale production continuing to dominate. Another example of conflicting assumptions involves a trade-off between how much biofuel is desired and ideas about how biofuels should be desirably produced. The very prevalent expectation that biofuels will reduce greenhouse gas emissions relies in part on avoiding high-carbon land, while national fuel blending targets are more easily achieved using land with the highest biomass production potential. Unsuccessful attempts to cultivate jatropha on land seen as ‘marginal’ in terms of its agricultural potential suggest there is no easy fix for this dilemma. Biofuel policy aims can look different when viewed at various scales. National and local interests may compete over the potential benefits of biofuels – for example if national blending targets for road transportation take priority over the goal of extending energy access in rural areas by using biofuels to provide electricity, household lamps or cookstoves. Here trade-offs may arise between pursuing energy security for “the country” and for particular communities or under-served populations. In terms of temporal scale, countries pursuing biofuel production face decisions over whether to cash in short-term economic gains from immediate agricultural investments or to pursue deferred, longer-term gains by slowly and carefully establishing enterprises likely to be viable for a longer period of time. The ‘hype’ of enthusiasm over jatropha followed by widespread project failure provides a cautionary example of short-term actions running ahead of research, knowledge and informed extension work. These examples show how expectations assumed to be synergistic in many cases work against each other. Treating the assumed benefits of biofuels as an undifferentiated bundle obscures competition between them.

The expectations for biofuels shown in Table 2 make an attractive package – but ‘unbundling’ these expectations reveals the difficulty, if not impossibility, of achieving these goals together. By reviewing published evidence, this study has shown that while some policy aspirations linked to biofuels have been partly achieved under particular circumstances, others have remained out of reach. Further, this study has identified problems of omission and competition within bundles of policy expectations. Vague or narrowly defined aspirations exclude important considerations (such as equity issues in how outcomes are distributed), while policies have so far largely failed to acknowledge tensions between goals that pull in different directions (such as incentivizing large-scale production while seeking to benefit small-scale farmers). Similarly, policies remain silent on how to reconcile conflicts between interests spanning multiple spatial and temporal scales. Tilman et al. (2009) anticipated many of these challenges, writing: “Legislation that is vague could allow significant portions of the biofuels industry to develop along counterproductive pathways. Complementary policies must directly target related goals, such as land- and water-efficient food production, reduced agricultural greenhouse-gas emissions, and the prevention of habitat loss from landclearing.” We believe this assessment remains prescient today. Fulfilling a bundle of goals requires clear, specific policies in each area, together with coordination, capacity to monitor outcomes, and mechanisms to enforce regulations. The extent to which this is achievable for biofuels remains an open question. We conclude that it is crucial to ‘unbundle’ biofuel expectations along several dimensions: by theme, by scale, by time frame, and by social characteristics of the people affected. It is also important to identify different ways in which aspirations can be evaluated, and to develop metrics that are meaningful in advancing specific policy aims. Doing so can help anticipate the contradictions and challenges likely to arise when trying to implement a multi-pronged biofuel agenda. While pursuing multiple and integrated benefits is a normal, even admirable policy goal, successfully doing so requires a curiously high level of disaggregated thinking. Acknowledgements This paper grew out of the workshop “Bioenergy: Status quo, trends and sustainability governance,” held in Berlin in October 2015, organized and funded by the Institute of Advanced Sustainability 12

Energy Policy xxx (xxxx) xxx–xxx

C. Hunsberger et al.

Global Interest in Farmland: Can it Yield Sustainable and Equitable Benefits?. World Bank, Washington D.C. Dihr, R.K., 2015. Indigenous Peoples in the World of Work in Asia and the Pacific: A Status Report. International Labor Organization, Geneva. Diop, D., Blanco, M., Flammini, A., Schlaifer, M., Kropiwnicka, M.A., Mautner Markhof, M., 2013. Assessing the impact of biofuels production on developing countries from the point of view of policy coherence for development, The European Union's Framework Contract Commission 2011 Lot1 - Contract N° 2012/299193. AETS, Lons, France. Dufey, A., 2008. Impacts of sugarcane bioethanol towards the Millennium Development Goals. In: Zuurbier, P., Vooren, Jvd (Eds.), Sugarcane Ethanol: Contributions to Climate Change Mitigation and The Environment. Wageningen AcademicPublishers, Wageningen, The Netherlands, 199–225. Enerdata, 2016. Global Energy Statistical Yearbook. ESMAP, 2005. Impact of higher oil prices on low income countries and on the poor. Energy Sector Management Assistance Program, United Nations Development Program and the World Bank, Washington, D.C. EU, 2009. Directive 2009/28/EC of the European Parliament and of the Council on the promotion of the use of energy from renewable sources, L 140/16. Official Journal of the European Union, Brussels. Ewing, M., Msangi, S., 2009. Biofuels production in developing countries: assessing tradeoffs in welfare and food security. Environ. Sci. Policy 12, 520–528. Facioli, I., Peres, G., 2007. Histórico de cortadores de cana mortos no setor canavieiro. Pastoral do Migrante, Guariba. Fargione, J., Hill, J., Tilman, D., Polasky, S., Hawthorne, P., 2008. Land clearing and the biofuel carbon debt. Science 319, 1235–1238. Fasse, A., Grote, U., 2015. The role of Jatropha curcas cultivation in livelihood strategies of small-scale households in rural Tanzania. Reg. Environ. Change 15, 1203–1214. Favretto, N., Stringer, L.C., Dougill, A.J., 2014. Unpacking livelihood challenges and opportunities in energy crop cultivation: perspectives on Jatropha curcas projects in Mali. Geogr. J. 180, 365–376. Feintrenie, L., Chong, W.K., Levang, P., 2010. Why do farmers prefer oil palm? Lessons learnt from Bungo District, Indonesia. Small-scale For. 9, 379–396. Florin, M.J., van Ittersum, M.K., van de Ven, G.W.J., 2013. Family farmers and biodiesel production: systems thinking and multi-level decisions in Northern Minas Gerais, Brazil. Agric. Syst. 121, 81–95. Galiano, Ad.M., Vettorassi, A., Navarro, V.L., 2012. Trabalho, saúde e migração nos canaviais da região de Ribeirão Preto (SP), Brasil: o que percebem e sentem os jovens trabalhadores? Revista Bras. saúde Occup. 37, 51–64. German, L., Schoneveld, G.C., 2012a. A review of social sustainability considerations among EU-approved voluntary schemes for biofuels, with implications for rural livelihoods. Energy Policy 51, 765–778. German, L., Schoneveld, G.C., Gumbo, D., 2011a. The Local Social and Environmental Impacts of Smallholder-Based Biofuel Investments in Zambia. Ecol. Soc., 16. German, L., Schoneveld, G.C., Pacheco, P., 2011b. Local social and environmental impacts of biofuels: global comparative assessment and implications for governance. Ecol. Soc., 16. German, L., Schoneveld, G.C., Pacheco, P., 2011c. The social and environmental impacts of biofuel feedstock cultivation: evidence from multi-site research in the forest frontier. Ecol. Soc., 16. German, L.A., Schoneveld, G., 2012b. Biofuel Investments in Sub‐Saharan Africa: a review of the early legal and institutional framework in Zambia. Rev. Policy Res. 29, 467–491. German, L.A., Parker, L., 2015. Model development? Zambia Sugar and the uneven terrain of social benefit, Presented at the World Bank Conference on Land and Poverty. The World Bank, Washington, D.C. Government of Brazil, 2009. The national commitment to improve labor conditions in the sugarcane activity. General Secretariat of the President of the Republic. Grimsby, L.K., Aune, J.B., Johnsen, F.H., 2012. Human energy requirements in Jatropha oil production for rural electrification in Tanzania. Energy for Sustain. Dev. 16, 297–302. Hall, J., Matos, S., Severino, L., Beltrao, N., 2009. Brazilian biofuels and social exclusion: established and concentrated ethanol versus emerging and dispersed biodiesel. J. Clean. Prod. 17, S77–S85. Herrmann, R., Grote, U., 2015. Large-scale agro-industrial investments and rural poverty: evidence from sugarcane in Malawi. J. Afr. Econ. 24, 645–676. Hunsberger, C., 2010. The politics of Jatropha-based biofuels in Kenya: convergence and divergence among NGOs, donors, government officials and farmers. J. Peasant Stud. 37, 939–962. Hunsberger, C., 2014. Jatropha as a biofuel crop and the economy of appearances: experiences from Kenya. Rev. Afr. Political Econ. 41, 216–231. Hunsberger, C., Bolwig, S., Corbera, E., Creutzig, F., 2014. Livelihood impacts of biofuel crop production: Implications for governance. Geoforum 54, 248–260. IEA, 2014. Energy Supply Security 2014: Emergency Response of IEA Countries. International Energy Agency, Paris. IEA, 2008. Banco de dados. Instituto de Economic Agricola. ILO, 2007. Trabalho Escravo no Brasil do Século XXI. International Labor Organization in Brazil, Brasilia. Jupesta, J., 2010. Impact of the introduction of biofuel in the transportation sector in Indonesia. Sustainability 2, 1831–1848. Kalinda, C., Moses, Z., Lackson, C., Chisala, L.A., Donald, Z., Darius, P., Exildah, C.-K., 2015. Economic impact and challenges of Jatropha curcas L. projects in NorthWestern Province Zambia: a case of Solwezi district. Sustainability 7, 9907–9923. Koizumi, T., 2008. Biofuel Policies in Asia, Presented at FAO Expert Meetings 5 and 6, Rome, Italy. Kuntashula, E., van der Horst, D., Vermeylen, S., 2014. A pro-poor biofuel? Household

Studies. We appreciate the ideas and perspectives of the workshop participants as well as the helpful insights of the anonymous reviewers. References Acheampong, E., Campion, B., 2014. The effects of biofuel feedstock production on farmers’ livelihoods in Ghana: the case of Jatropha curcas. Sustainability 6, 4587–4607. Albers, S.C., Berklund, A.M., Graff, G.D., 2016. The rise and fall of innovation in biofuels. Nat. Biotechnol. 34, 814–821. Alessi, N.P., Scopinho, R.A., 1994. In: Alessi, N.P. (Ed.), A saúde do trabalhador do corte de cana-de-açúcar. Saúde e Trabalho no Sistema Único de Saúde, Hucitec, São Paulo, 121–151. Anderman, T.L., Remans, R., Wood, S.A., DeRosa, K., DeFries, R.S., 2014. Synergies and tradeoffs between cash crop production and food security: a case study in rural Ghana. Food Security 6, 541–554. de Andrade, R.M.T., Miccolis, A., 2011. Policies and Institutional and Legal Frameworks in the Expansion of Brazilian Biofuels. CIFOR, Bogor Barat, Indonesia. Archer, D.-E., 2012. Energy system assessment and modeling of a potential future energy system on Lamu Island, Kenya. Ariza-Montobbio, P., Lele, S., 2010. Jatropha plantations for biodiesel in Tamil Nadu, India: Viability, livelihood trade-offs, and latent conflict. Ecol. Econ. 70, 189–195. Axelsson, L., Franzen, M., Ostwald, M., Berndes, G., Lakshmi, G., Ravindranath, N.H., 2012. Jatropha cultivation in southern India: assessing farmers' experiences. Biofuels Bioprod. Biorefin. 6, 246–256. Baccarin, J.G., Gebara, J.J., Borges Júnior, J.C., 2011. Expansão canavieira e ocupação formal em empresas sucroalcooleiras do Centro-Sul do Brasil, entre 2007 e 2009. Revista de Econ. Soc. Rural 49, 493–506. Bailis, R., Baka, J., 2011. Constructing sustainable biofuels: governance of the emerging biofuel economy. Ann. Assoc. Am. Geogr. 101, 827–838. Baka, J., 2014. What wastelands? A critique of biofuel policy discourse in South India. Geoforum 54, 315–323. Balsadi, O.V., 2007. Qualidade do emprego na agricultura brasileira no período 2001– 2004 e suas diferenciações por culturas. Revista de Economia e Sociologia Rural 45, 409–444. Barlow, C., Zen, Z., Gondowarsito, R., 2003. The Indonesian oil palm industry. Oil Palm, Ind. Econ. J. 3, 8–15. Bickel, U., Dros, J.M., 2003. The Impacts of Soybean Cultivation on Brazilian Ecosystems: Three Case Studies. World Wildlife Fund Forest Conversion Initiative. Binswanger, H., 1986. Agricultural mechanization a comparative historical perspective. World Bank Res. Observer 1, 27–56. Blaber-Wegg, T., Hodbod, J., Tomei, J., 2015. Incorporating equity into sustainability assessments of biofuels. Curr. Opin. Environ. Sustain. 14, 180–186. Borras, S.M., Franco, J.C., Isakson, S.R., Levidow, L., Vervest, P., 2015. The rise of flex crops and commodities: implications for research. J. Peasant Stud. 43, 93–115. Brazilian National Congress, 1993. Lei Federal No. 8723/93, Brasilia. Brazilian National Congress, 1997. Lei No. 9478/97, Brasilia. Brazilian National Congress, 2005. Lei do Biodiesel (Lei nº 11.097/05), Brasilia. Brittaine, R., Lutaladio, N., 2010. Jatropha: a smallholder energy crop: the potential for pro-poor development. Integr. Crop Manag., 114, Food and Agriculture Organization of the United Nations, Rome. Caroko, W., Komarudin, H., Obidzinski, K., Gunarso, P., 2011. Policy and Institutional Frameworks for the Development of Palm Oil–based Biodiesel in Indonesia. CIFOR, Bogor Barat, Indonesia. Cesar, Ad.S., Batalha, M.O., 2013. Brazilian biodiesel: the case of the palm's social projects. Energy Policy 56, 165–174. Clancy, J., 2013. Biofuels and Rural Poverty. Routledge, Abingdon, UK. Colchester, M., 2011. Palm Oil and Indigenous Peoples in South East Asia: Land Acquisition, Human Rights Violations and Indigenous Peoples on the Palm Oil Frontier. Forest Peoples Programme and the International Land Coalition, Moretonin-Marsh, UK and Rome, Italy. Colchester, M., Chao, S., 2011. Oil Palm Expansion in South East Asia: Trends and Implications for Local Communities and Indigenous Peoples. Forest Peoples Programme and Sawit Watch, Moreton-in-Marsh, UK and Bogor, Indonesia. Conpes, 2008. Lineamientos de Política para Promover la Producción Sostenible de Biocombustibles en Colombia. Documento Conpes 3510. Consejo Nacional de Política Económica y Social, República de Colombia, Bogotá. Cotula, L., Dyer, N., Vermeulen, S., 2008. Fuelling Exclusion? The Biofuels Boom and Poor People's Access to Land. IIED, London. Creutzig, F., Corbera, E., Bolwig, S., Hunsberger, C., 2013. Integrating place-specific livelihood and equity outcomes into global assessments of bioenergy deployment. Environ. Res. Lett. 8, 035047. Creutzig, F., Ravindranath, N.H., Berndes, G., Bolwig, S., Bright, R., Cherubini, F., Chum, H., Corbera, E., Delucchi, M., Faaij, A., Fargione, J., Haberl, H., Heath, G., Lucon, O., Plevin, R., Popp, A., Robledo-Abad, C., Rose, S., Smith, P., Stromman, A., Suh, S., Masera, O., 2014. Bioenergy and climate change mitigation: an assessment. GCB Bioenergy. Dayang Norwana, A., Kanjappan, R., Chin, M., Schoneveld, G., Potter, L., Andriani, R., 2011. The local impacts of oil palm expansion in Malaysia; An assessment based on a case study in Sabah State. Center for International Forestry Research (CIFOR) Working Paper. CIFOR, Bogor, Indonesia, 1–17. DEDE, 2012. The Renewable and Alternative Energy Development Plan (AEDP 20122021), English version. Deininger, K., Byerlee, D., Lindsay, J., Norton, A., Selod, H., Stickler, M., 2011. Rising

13

Energy Policy xxx (xxxx) xxx–xxx

C. Hunsberger et al.

395–405. PISCES, F.A.O., 2009. Small-Scale Bioenergy Initiatives: Brief Description and Preliminary Lessons on Livelihood Impacts from Case Studies in Asia, Latin America and Africa. Food and Agriculture Organization of the United Nations (FAO) Policy Innovation Systems for Clean Energy Security (PISCES), Rome, 139. Ponte, S., 2014. ‘Roundtabling’ sustainability: lessons from the biofuel industry. Geoforum 54, 261–271. Pradhan, A., Mbohwa, C., 2014. Development of biofuels in South Africa: challenges and opportunities. Renew. Sustain Energy Rev 39, 1089–1100. PT SMART, 2008. Annual Report. REN21, 2016. Renewables 2016 Global Status Report. Renewable Energy Policy Network for the 21st Century, Paris. Republic of Mozambique, 2009. Política e Estratégia de Biocombustíveis, Resolução n° 22/2009. Boletím da República, Maputo. Republic of South Africa, 2007. Biofuels Industrial Strategy of the Republic of South Africa. Department of Minerals and Energy. Rist, L., Feintrenie, L., Levang, P., 2010. The livelihood impacts of oil palm: smallholders in Indonesia. Biodivers. Conserv. 19, 1009–1024. Rittenburg, R.A., Kummel, M., Perramond, E.P., 2011. The local climate-development nexus: Jatropha and smallholder adaptation in Tamil Nadu, India. Clim. Dev. 3, 328–343. Robledo-Abad, C., Althaus, H.J., Berndes, G., Bolwig, S., Corbera, E., Creutzig, F., GarciaUlloa, J., Geddes, A., Gregg, J.S., Haberl, H., Hanger, S., Harper, R.J., Hunsberger, C., Larsen, R.K., Lauk, C.h., Leitner, S., Lilliestam, J., Lotze-Campen, H., Muys, B., Nordborg, M., Ölund, M., Orlowsky, B., Popp, A., Portugal-Pereira, J., Reinhard, J., Scheiffle, L., Smith, P., 2016. Bioenergy production and sustainable development: science base for policy-making remains limited. GCB Bioenergy. Rogers, C., Sovacool, B.K., Clarke, S., 2013. Sweet nectar of the Gaia: lessons from Ethiopia's “Project Gaia”. Energy Sustain. Dev. 17, 245–251. Romijn, H., Heijnen, S., Rom Colthoff, J., de Jong, B., van Eijck, J., 2014. Economic and Social Sustainability Performance of Jatropha Projects: Results from Field Surveys in Mozambique Tanzania and Mali. Sustainability 6, 6203–6235. Rosa, L.A., 2009. Trabalho e Trabalhadores dos Canaviais: Perfil dos Cortadores de Cana da Região de Ribeirão Preto (SP). Psychology, Faculty of Philosophy, Sciences and Letters, Ribeirão Preto. Sardjono, M., 2014. Indonesian Policy on Sustainable Oil Palm Development. Schoneveld, G.C., German, L.A., Nutakor, E., 2011. Land-based investments for rural development? A grounded analysis of the local impacts of biofuel feedstock plantations in Ghana. Ecology and Society, 16. Schuenemann, F., Thurlow, J., Zeller, M., 2016. Leveling the Field for Biofuels: Comparing the Economic and Environmental Impacts of Biofuel and Other Export Crops in Malawi. International Food Policy Research Institute, Washington, D.C.. Searchinger, T., Heimlich, R., Houghton, R.A., Dong, F., Elobeid, A., Fabiosa, J., Tokgoz, S., Hayes, D., Yu, T., 2008. Use of U.S. croplands for biofuels increases greenhouse gases through emissions from land use change. Science, 319. Silva, M.A.M., 2007. Trabalho e trabalhadores na região do “mar de cana e do rio de álcool”. In: Novaes, R., Alves, F. (Eds.), Migrantes: Trabalho e Trabalhadores no Complex Agroindustrial Canavieiro (Os Heróis do Agronegócio Brasileiro). Editora da Universidade Federal de São Carlos, São Carlos, Brazil, 55–86. Skutsch, M., de los Rios, E., Solis, S., Riegelhaupt, E., Hinojosa, D., Gerfert, S., Gao, Y., Masera, O., 2011. Jatropha in Mexico: environmental and social impacts of an incipient biofuel program. Ecol. Soc. 16. Smalley, R., 2013. Plantations, Contract Farming and Commercial Farming Areas in Africa: A Comparative Review. Land and Agricultural Commercialisation in Africa (LACA) Working Paper 55. Future Agricultures Consortium, Sussex. Soto, I., Achten, W.M.J., Muys, B., Mathijs, E., 2015. Who benefits from energy policy incentives? The case of jatropha adoption by smallholders in Mexico. Energy Policy 79, 37–47. Suharto, R., 2012. ISPO Certification – An Update, Presented at IPOSC 2012, Malaysia. Thompson, P., 2012. The Agricultural Ethics of Biofuels: The Food vs. Fuel DebateAgriculture (Basel) 2, 339–358. Thondhlana, G., 2015. Land acquisition for and local livelihood implications of biofuel development in Zimbabwe. Land Use Policy 49, 11–19. Tiffen, M., Mortimore, M., 1990. Theory and Practice in Plantation Agriculture: An Economic Review. Overseas Development Institute, London. Tilman, D., Socolow, R., Foley, J.A., Hill, J., Larson, E., Lynd, L., Pacala, S., Reilly, J., Searchinger, T., Somerville, C., Williams, R., 2009. Beneficial biofuels-the food, energy, and environment trilemma. Science 325, 270–271. Tongsopit, S., Greacen, C., 2012. Thailand's Renewable Energy Policy: FiTs and Opportunities for International Support. Tshiunza, M., 1996. Agricultural intensification and labor needs in the cassava-producing zones of sub-Saharan Africa, Applied Biological Sciences. Catholic University Leuven, Belgium. U.S. Congress, 1990. The Clean Air Act, Section 211(o), Washington, D.C. U.S. Congress, 2005. Energy Policy Act of 2005. Public Law 109-58, Washington, D.C. U.S. Congress, 2007. Energy security and independence act, United States of America, p. 311. UK, 2007. The Renewable Transport Fuel Obligations Order. Statutory Instrument 3072. UNEP, 2008. Green Jobs: Towards decent work in a sustainable, low-carbon world. UNEP/ILO/IOE/ITUC. US EPA, 2007. Regulation of Fuels and Fuel Additives: Renewable Fuel Standard Program, Final Rule. Federal Register 72(83): 23900–24014. United States Environmental Protection Agency. US EPA, 2010. In: Quality, Oo.TaA. (Ed.), EPA Finalizes Regulations for the National Renewable Fuel Standard Program for 2010 and Beyond. EPA-420-F-10-007.. United States Environmental Protection Agency, Washington, D.C.

wealth and farmer participation in Jatropha curcas seed production and exchange in eastern Zambia. Biomass Bioenergy 63, 187–197. Labruto, N., 2014. Experimental biofuel governance: Historicizing social certification in Brazilian ethanol production. Geoforum 54, 272–281. Lane, J., 2016. Biofuels Mandates Around the World: 2016. Laws, R.L., Brooks, D.R., Amador, J.J., Weiner, D.E., Kaufman, J.S., Ramírez-Rubio, O., Riefkohl, A., Scammell, M.K., López-Pilarte, D., Sánchez, J.M., 2015. Changes in kidney function among Nicaraguan sugarcane workers. International J. Occup. Environ. Health 21, 241–250. Lehtonen, M., 2011. Social sustainability of the Brazilian bioethanol: Power relations in a centre-periphery perspective. Biomass Bioenergy 35, 2425–2434. Leite, Dal Belo, Justino, J.G., Silva, F.B., Florin, J.V., van Ittersum, M.K., M.J., 2015. Socioeconomic and environmental assessment of biodiesel crops on family farming systems in Brazil. Agric. Syst. 133, 22–34. Leonardo, W.J., Florin, M.J., van de Ven, G.W., Udo, H., Giller, K.E., 2015. Which smallholder farmers benefit most from biomass production for food and biofuel? The case of Gondola district, central Mozambique. BiomassBioenergy 83, 257–268. Li, T.M., 2011. Centering labor in the land grab debate. J. Peasant Stud. 38, 281–298. Lima, M., Skutsch, M., de Medeiros Costa, G., 2011. Deforestation and the Social Impacts of Soy for Biodiesel: Perspectives of Farmers in the South Brazilian Amazon. Ecol. Soc., 16. Loewenson, R., 1992. Modern Plantation Agriculture: Corporate Wealth and Labour Squalor. Zed Books, London. Loos, T.K., 2009. Socio-economic impact of a Jatropha-project on smallholder farmers in Mpanda, Tanzania: case study of a public-private-partnership project in Tanzania, Institute for Agricultural Economics and Social Sciences in the Tropics and Subtropics. University of Hohenheim, Hohenheim, The Netherlands, 157. Macedo, Id.C., 2005. Sugar Cane's Energy: Twelve Studies on Brazilian Sugar Cane Agribusiness and Its Sustainability. União da Agroindústria Canavieira de São Paulo, São Paulo, Brazil. MADR, 2010. Política Nacional de Biocombustibles en Colombia, in: Rural, M.d.A.y.D. (Ed.). Presented at the V Seminario Latinoamericano y del Caribe de Biocombustibles, 17–18 August 2010, Santiago de Chile. MAPA, 2006. Diretrizes De Política De Agroenergia, 2006 − 2011. Ministério da Agricultura, Pecuária e Abastecimento, Government of Brazil, Brasília. Maroun, M.R., La Rouere, E.L., 2014. Ethanol and food production by family smallholdings in rural Brazil: Economic and socio-environmental analysis of micro distilleries in the State of Rio Grande do Sul. Biomass Bioenergy 63, 140–155. Marti, S., 2008. Losing Ground: The Human Rights Impacts of Oil Palm Plantation Expansion in Indonesia. Friends of the Earth, Life Mosaic and Sawit Watch, London, UK. Mataveia, M., 2009. Biofuel policy and strategy for Mozambique. Presentation slides by representative of the Ministry of Energy, Government of Mozambique, delivered at UN CSD-17 side event, May 14 2009. Global Bioenergy Partnership, New York. McCarthy, J.F., 2010. Processes of inclusion and adverse incorporation: oil palm and agrarian change in Sumatra, Indonesia. J. Peasant Stud. 37, 821–850. MDA, 2009a. Biodiesel Combustível Social. Ministry for Agrarian Development, Government of Brazil. MDA, 2009b. PRONAF portal. Ministry of Agrarian Development, Government of Brazil. Minambiente, 2016. Biofuel. Ministry of Environment, Housing and Territorial Development of the Republic of Colombia. MME, 2007. Política General en Materia de Biocombustibles. Ministerio de Minas e Energia. MNRE, 2009. National Policy on Biofuels. Ministry of New and Renewable Energy, Government of India, New Delhi. Mogaka, V., Ehrensperger, A., Iiyama, M., Birtel, M., Heim, E., Gmuender, S., 2014. Understanding the underlying mechanisms of recent Jatropha curcas L. adoption by smallholders in Kenya: a rural livelihood assessment in Bondo, Kibwezi, and Kwale districts. Energy Sustain. Dev. 18, 9–15. Montefrio, M.J.F., Sonnenfeld, D.A., 2013. Global-local tensions in contract farming of biofuel crops involving indigenous communities in the Philippines. Soc. Nat. Resour. 26, 239–253. Montefrio, M.J.F., Sonnenfeld, D.A., Luzadis, V.A., 2015. Social construction of the environment and smallholder farmers' participation in 'low-carbon', agro-industrial crop production contracts in the Philippines. Ecol. Econ. 116, 70–77. Mponela, P., Jumbe, C.B.L., Mwase, W.F., 2011. Determinants and extent of land allocation for Jatropha curcas L. cultivation among smallholder farmers in Malawi. Biomass Bioenergy 35, 2499–2505. Migrantes: Trabalho e Trabalhadores no Complex Agroindustrial Canavieiro (Os Heróis do Agronegócio Brasileiro). In: Novaes, J.R.P., Alves, F. (Eds.), Editora da Universidade Federal de São Carlos, São Carlos, Brazil.. Obidzinski, K., Andriani, R., Komarudin, H., Andrianto, A., 2012. Environmental and social impacts of oil palm plantations and their implications for biofuel production in Indonesia. Ecol. Soc., 17. OECD-IEA, 2007. Contribution of Renewables to Energy Security, IEA Information Paper. Organization for Economic Cooperation and Development and the International Energy Agency, Paris. OECD-IEA, 2013. Renewable Energy in Transport. OECD-IEA, Paris. Ohimain, E., 2015. Smallholder Bioethanol Production from Cassava Feedstock Under Rural Nigerian Settings. Energy Sources Part B—Economics Planning and Policy 10, 233–240. Ortiz, L., Rodrigues, D., 2006. Case Study Sugar Cane Ethanol from Brazil. CREM, Núcleo Amigos da Terra (NAT). Vitea Civilis Institute, São Lourenço da Serra, Brazil, 49. Padula, A.D., Santos, M.S., Ferreira, L., Borenstein, D., 2012. The emergence of the biodiesel industry in Brazil: current figures and future prospects. Energy Policy 44,

14

Energy Policy xxx (xxxx) xxx–xxx

C. Hunsberger et al.

van Gelder, J.W., German, L., 2011. Biofuel Finance: Global Trends in Biofuel Finance in Forest-rich Countries of Asia, Africa and Latin America and Implications for Governance. CIFOR, Bogor, Indonesia. Vermeulen, S., Goad, N., 2006. Towards Better Practice in Smallholder Palm Oil Production. IIED, London. White, B., Dasgupta, A., 2010. Agrofuels capitalism: a view from political economy. J. Peasant Stud. 37, 593–607. Wilkinson, J., Herrera, S., 2010. Biofuels in Brazil: debates and impacts. J. Peasant Stud. 37, 749–768. World Bank, 2008. World Development Report: Agriculture for Development. World Bank, Washington, DC.. World Bank, 2010. Environmental, Economic and Social Impacts of Oil Palm in Indonesia: A Synthesis of Opportunities and Challenges. World Bank, Washington, D.C.. Zilberman, D., Hochman, G., Rajagopal, D., Sexton, S., Timilsina, G., 2012. The impact of biofuels on commodity food prices: assessment of findings. Am. J. Agric. Econ..

US EPA, 2016. Renewable Fuel Standard Program. United States Environmental Protection Agency, Washington, D.C.. USDA Foreign Agricultural Service, 2015. Brazil Biofuels Annual - Ethanol and Biodiesel. GAIN Report Number BR15006. USDA-FAS, 2006. China, Peoples Republic of: Bio-Fuels, An Alternative Future for Agriculture. USDA Foreign Agricultural Service, Washington, D.C.. USDA-FAS, 2014. EU Biofuels Annual. USDA Foreign Agricultural Service, Washington, D.C. USDA-FAS, 2015. Biofuels Annual Reports. USDA-FAS, 2016. Biofuels Annual Reports. US-EIA, 2016. International Energy Statistics. U.S. Energy Information Agency, Washington, D.C. van Eijck, J., 2009. Case Study: The Smallholder Model of Biofuel Production in Tanzania. Commissioned by GTZ and ProBEC for the SADC Secretariat, Arusha. van Eijck, J., Smeets, E., Faaij, A., 2012. The economic performance of Jatropha, Cassava and Eucalyptus production systems for energy in an East African smallholder setting. GCB Bioenergy 4, 828–845.

15