Download Consensus and Contention in the Food-Versus

EG39CH10-Msangi
ARI
ANNUAL
REVIEWS
27 September 2014
11:57
Further
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
Click here for quick links to
Annual Reviews content online,
including:
• Other articles in this volume
• Top cited articles
• Top downloaded articles
• Our comprehensive search
Consensus and Contention in
the Food-Versus-Fuel Debate
Mark W. Rosegrant and Siwa Msangi
International Food Policy Research Institute, Washington, DC 20006;
email: [email protected], [email protected]
Annu. Rev. Environ. Resour. 2014. 39:271–94
Keywords
First published online as a Review in Advance on
September 17, 2014
biofuel, agriculture, food security, United States, European Union
The Annual Review of Environment and Resources is
online at environ.annualreviews.org
Abstract
This article’s doi:
10.1146/annurev-environ-031813-132233
c 2014 by Annual Reviews.
Copyright All rights reserved
This review discusses research on linkages between biofuels, agriculture,
and food security. The literature indicates that biofuel expansion affects
land use, puts pressure on food and feed markets, and modestly reduces
greenhouse gas emissions. Researchers readily identify these outcomes, as
well as the multitude of factors besides biofuels that have driven up food
prices in recent years. However, precision in quantifying the extent of the
impacts and in attributing effects to various drivers is elusive, resulting in a
wide range of estimates. Nevertheless, the central tendency is that a foodversus-fuel trade-off is created through biofuel production from food crops,
and the continued expansion of biofuel production increases food commodity
prices, reduces the availability of calories, and increases malnourishment
in developing countries. Higher food prices particularly reduce the poor’s
access to food, which has possible long-term, irreversible consequences for
health, productivity, and well-being.
271
EG39CH10-Msangi
ARI
27 September 2014
11:57
Contents
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2. BIOFUEL OBJECTIVES AND POLICIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1. Biofuel Policies: Energy Security, Environmental, and Economic Objectives . . .
2.2. Biofuel Policies Vary by Country . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3. ECONOMIC FACTORS AFFECTING FOOD PRICES
DURING THE BIOFUEL ERA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1. Low Stocks Relative to Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2. Demand Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3. Supply Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4. Other Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.5. Relative Importance of Factors Affecting Food Commodity Prices . . . . . . . . . . . .
4. SOCIOECONOMIC IMPACTS OF BIOFUEL POLICIES . . . . . . . . . . . . . . . . . . . . .
4.1. Biofuel Impacts on Food and Commodity Prices . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2. Poverty and Hunger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3. Feed and Livestock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5. IMPACTS ON LAND USE AND GREENHOUSE GAS EMISSIONS . . . . . . . . . .
5.1. Individual Model Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2. Multistudy Reviews . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6. POLICY IMPLICATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
272
273
273
274
275
275
275
276
276
277
277
279
284
286
286
287
288
289
1. INTRODUCTION
The food-versus-fuel debate has engaged policy makers, policy interest groups, and researchers
since the 2000s, when fuel production from feedstocks grown on agricultural land began increasing
significantly. Global biofuel production expanded from 20 billion L/year in the 1990s to more than
40 billion L/year in 2006 after government policies such as mandated usage and renewable energy
targets encouraged use of biofuels. Output topped 100 billion L/year beginning in 2010 (1). At
recent levels of biofuel production, area used for energy production is approximately 30 million
hectares (ha), or less than 1% of global cropland (2). The share of cropland can be considerably
higher on a country or local level.
The literature on the impacts of biofuel policies for ethanol and biodiesel has produced both
consensus and contention. In general, the evidence thus far has highlighted important trade-offs
between biofuel production, food security, and the environment. The extent that biofuels alone can
be linked to changes in food prices, food security, and land use is debated, however, with differing
research models and policy scenarios yielding different results. In this review, we characterize
potential trade-offs, as well as any evidence for synergies between biofuels and welfare or between
biofuels and the environment.
Although the issue has many dimensions, two broad topics are of keen interest to policy makers
as they consider—partly in response to political reaction engendered by this debate—changes to
policy. The first is tensions between biofuel and welfare. Evidence in the literature indicates
that biofuel production in the United States affects international agricultural commodity prices,
which can be transmitted to net food importers around the world, decreasing calorie availability
in the long run. The second topic involves the interaction between biofuel and the environment.
Increased biofuel production in the European Union and the United States has been shown to
272
Rosegrant
·
Msangi
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
EG39CH10-Msangi
ARI
27 September 2014
11:57
induce land use change (both domestically and internationally), which can potentially offset the
carbon savings from avoided fossil-fuel combustion and can have implications for the land used
for food supply, especially when agricultural land expansion and/or intensification is limited. The
assumptions on productivity growth are particularly critical in this regard.
Our review of the research reveals areas of consensus and of contention regarding the welfare
and environmental impacts of biofuel. Summarizing results from this research provides clarity on
the trade-offs, particularly given the ongoing consideration of biofuel policies in the European
Union, Brazil, and other countries. Much of the literature uses sophisticated modeling techniques
that require multiple assumptions; defined economic linkages between biofuels, food, and energy
markets; and estimates of economic behavior. As a result, research results vary widely. This review
summarizes conclusions from the literature but does not provide a technical review of modeling
approaches provided elsewhere (e.g., 3–5).
Shock: an unexpected
or unpredictable event
that affects an
economic variable or
economic system
positively or negatively
2. BIOFUEL OBJECTIVES AND POLICIES
For several decades, governments have pursued biofuel policy for environmental benefits and to
improve energy security. In this section, we focus particularly on the objectives of and variations
between different national biofuel priorities.
2.1. Biofuel Policies: Energy Security, Environmental, and Economic Objectives
Countries that are net energy importers typically want to reduce their vulnerability to supply
disruptions and price shocks (6). Development of a domestic biofuel industry and other alternative
energy sources reduces such vulnerability to some degree. Following major oil price spikes in the
1970s, Brazil and the United States created ethanol production sectors, and several other countries
also proposed alternative fuel policies. Taking advantage of existing farm production capacities
and addressing a desire to find new demand for agricultural products, Brazil focused on converting
sugarcane, and the United States used corn as a major feedstock.
But for some nations, energy security is not the only, or even the primary, goal of biofuel
policy. Some also have environmental objectives. Using biofuels can reduce the rate of atmospheric
buildup of greenhouse gases (GHGs) (primarily CO2 ). The burning of both biofuels and fossil
fuels results in CO2 emissions, but net emissions are lower for biofuels because atmospheric carbon
is used to grow plant material used as biofuel feedstock. The net gain, though, is reduced by several
other required steps in producing feedstocks and ultimately biofuel. Additional carbon is burned
in the production and application of inputs to grow plants (e.g., fertilizer and tractor fuel) and
when fuel is consumed to process the feedstock into biofuel.
In the European Union, the initial drivers (in addition to energy diversification and farm
sector diversification) included the goal of reducing GHG emissions. Given that diesel engines
dominate the EU motor vehicle fleet, the primary biofuel is biodiesel, which uses oilseeds as
feedstocks. Unlike in Brazil and the United States, the EU’s biofuel targets and its tradition of oil
crop imports led to a dependence on imports of either biofuels (ethanol from Brazil or the United
States) or feedstocks from Latin America, Africa, Asia, or Central and Eastern Europe.
Alongside the combined goals of reduced reliance on fossil-based fuel imports and of reduced
GHG emissions, the policies surrounding biofuels have also aimed to support those sectors that
produce abundant feedstock and that might benefit from the price enhancement provided by a
continuous demand for them. This latter goal is not often explicitly stated, but it can be a powerful
motivation at the political level. Some authors have argued against biofuel policies that try to “pick
winners” in terms of feedstock choice, suggesting that they will not do as well as policies with an
www.annualreviews.org • The Food-Versus-Fuel Debate
273
EG39CH10-Msangi
ARI
27 September 2014
11:57
explicit environmental goal in mind, such as the reduction of carbon. Sperling & Yeh (7, 8) use the
low-carbon fuel standard in California to illustrate that point, supporting its adoption at a more
global level to either complement or replace the existing Renewable Fuel Standard (RFS), which
mandates certain levels of ethanol use in the United States.
2.2. Biofuel Policies Vary by Country
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
In Brazil, federal policies during 1975–1979 expanded the sugarcane distilleries sector with subsidies and increased the ethanol content of gasoline (up to 22%). In 1980, ethanol-powered cars
were introduced that used technology developed in public research centers in the 1970s. In the
2000s, global demand for Brazilian ethanol began to increase, and domestic demand expanded
owing to additional sales of flex-fuel vehicles (1). The flex-fuel vehicles allowed consumers to
choose between gasoline and ethanol at the pump based on relative prices. High prices in global
energy markets helped propel domestic ethanol sales.
In the United States, as in Brazil, the energy crisis of the 1970s piqued policy makers’ interests.
Policies adopted in 1978 and 1980 introduced various instruments to encourage ethanol use. These
included a subsidy for blending ethanol into gasoline, insured loans for small ethanol producers,
federal purchase agreements, and a tariff on imported ethanol. In the late 1990s, fuel economy standards encouraged the production of flex-fuel vehicles powered by E-85 (85% ethanol), although
it increased ethanol use only marginally because of limited adoption. In the early to mid-2000s,
ethanol became the primary octane enhancer when MTBE (methyl tertiary butyl ether), a common
gasoline additive at the time, was identified as a contaminant of water sources and subsequently
banned from use (1). Around the time of the MTBE phaseout, the US government mandated
the use of biofuel at specific volumetric levels: The 2005 mandate required the use of 7.5 billion
gallons (28.4 billion L) of ethanol in transport fuels by 2012. In 2007, the mandate was increased
to 15 billion gallons (56.8 billion L) of corn ethanol by 2015, and a total biofuel target of 36 billion
gallons (136 billion L) by 2022 (including 21 billion L from feedstocks other than corn such as
sugarcane).
Biofuel policy in the European Union was discussed extensively in the 1990s, eventually leading
to an initial goal of replacing 20% of conventional fuels with substitute fuels by 2020. In 2003,
the first EU-level policy on biofuels established a target of 2% for renewables by 2005 and 5.75%
in 2010. In 2009, the European Council established a 10% target for renewable transport fuels
by 2020 (1). The target does not differentiate by feedstock and is based on energy content, unlike
the US mandate (9).
Many other countries have implemented some form of biofuel policy, with approximately
60 countries establishing either mandates or biofuel targets (1, 2). India, for example, uses minimum pricing for feedstocks and tax incentives for ethanol or biodiesel. China employs production
subsidies on ethanol and biodiesel. The motivations are similar to the major players described
above, including bolstering energy security, reducing energy import bills, improving balance of
payments (currency flow into and out of the country), supporting agricultural and rural development and employment, and reducing GHG emissions.
In Africa, several countries have initiated national biofuel programs in response to high fuel
prices, a desire to avoid costly fossil-fuel imports, and a goal of attracting commercial investments
to less-developed rural regions (10). Many of these countries, which have yet to achieve high levels
of productivity in grains or staple crops, have targeted nonfood crops as their intended sources of
feedstocks—such as Jatropha for biodiesel. These may seem to avoid the problem of food versus
fuels, but feedstocks such as Jatropha have proven to be relatively low-yielding, and farmers have
little experience in cultivating them commercially (11). Although many of these African countries
274
Rosegrant
·
Msangi
EG39CH10-Msangi
ARI
27 September 2014
11:57
certainly need additional and more reliable sources of energy, current policies may not be well
targeted; a long-term strategy of research and development to boost the productivity of these
feedstock sectors is necessary to position them for cost-competitive production of biofuels (12).
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
3. ECONOMIC FACTORS AFFECTING FOOD PRICES
DURING THE BIOFUEL ERA
The causes of the food price spikes of 2007–2012 were numerous and wide-ranging. Most research
on this topic concludes that a variety of factors—and not a single, dominant cause—converged to
drive up food prices during this period. Many of these were long-run demand and supply trends
that helped push up prices, with their effects magnified by unfortunate weather, a sharp increase
in use of corn for biofuel production, and linkages between the corn and oil markets. Increased
market volatility encouraged buyers to purchase commodities ahead of normal while offering
potential profit opportunities to financial traders. To limit the adverse impact of rising prices on
consumers, some producing countries banned exports, which put additional upward pressure on
world commodity prices. In this section, we describe these factors and their relative impacts.
3.1. Low Stocks Relative to Use
When analyzing commodity and food price changes during 2007–2012, many researchers have
identified low stocks relative to use as a primary contributor (13, 14). The stock-to-use (s/u)
ratio captures supply and demand in a single indicator to reflect either surplus or “tightness” in
a particular market. The economic concept of the s/u ratio is simple: As food commodity stocks
(measured at the end of the marketing year) decline relative to use for that year, the commodity
becomes more valuable, with buyers willing to pay more and sellers reluctant to sell. For example,
during 2000–2008, global stocks of grains and oilseeds relative to use dropped sharply, while prices
more than doubled (13, 14). This inverse relationship between s/u and price is particularly strong
for corn, a leading agricultural commodity.
For the decade immediately prior to the 2008 food price spike, the general consensus among
researchers is that demand growth outstripped supply growth, resulting in a declining s/u ratio
and stronger prices. This contrasts with the period between 1985 and 1990, when supply mostly
kept pace with demand. After the sharp run-up in 2008, prices retreated as demand softened, but
the market raced ahead again in 2010 and 2011 when crops in the United States and elsewhere
did not meet expectations for stock rebuilding (15, 16).
3.2. Demand Factors
Numerous demand factors contributed to the price spikes and have helped maintain commodity
prices above historical levels since the mid- to late 2000s (17). These factors include income and
population growth, biofuel expansion, and the declining value of the US dollar.
Prior to the price spike in 2008, the primary demand drivers for commodity markets were
income and population growth, particularly in developing countries. Higher incomes led to increased meat and dairy product consumption, which drove up demand for feedstuffs (feed grains
and oilseeds) (14). However, expansion of crop plantings for biofuel production is also cited as a
major demand driver (13, 18, 19). Policies to promote biofuel use, along with increased market
demand for biofuels as a substitute for petroleum-based motor fuels due to higher crude oil prices,
have diverted about one-third of the US corn crop and a similar amount of EU rapeseed. A major
shifting of acreage from feed to fuel production pressured global agricultural markets. Most other
www.annualreviews.org • The Food-Versus-Fuel Debate
275
EG39CH10-Msangi
ARI
27 September 2014
11:57
crop prices rose to maintain planting incentives for commodities other than those used in biofuel
production.
During the 2000s, the declining value of the US dollar contributed to an uptrend in commodity
prices (20). Beginning in 2002, the US dollar began to lose value against other currencies, making
US agricultural commodities less expensive for importers. Buyers began purchasing more grain
and oilseeds from the United States, a major global supplier. This demand put upward pressure on
US commodity prices, as well as on global prices of commodities that are traded widely in world
markets and under contracts with pricing in US dollars.
3.3. Supply Factors
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
The long-term trend in agricultural commodity production has been relatively flat compared
with demand. Annual growth in global harvest area has been near zero, and the global annual
average yield growth for grains and oilseeds slowed to 1.1% between 1990 and 2007 compared
with 2.0% during 1970–1990 (14). Low investment in agricultural research is also often cited as a
reason for slow yield growth and for creating a mismatch with long-term growth in demand (13).
Large commodity supplies in the 1980s, resulting in part from agricultural policies in developed
countries, had a twofold effect. First, the ensuing stable or low world prices reduced incentives for
developing countries to produce crops and develop their agricultural sectors. Second, investments
in agricultural research seemed unattractive if they were to make agricultural surpluses grow
even larger. As a result, agriculture sectors in both developing and developed countries were not
prepared for future expansion in demand and periodic shortages.
The rising cost of energy was another supply factor that adversely affected output and contributed to an uptrend in overall agricultural commodity prices (21). Beginning in 2002 and
continuing until 2008, a rise in crude oil prices drove up prices for inputs, including fuel and
fertilizer. Higher farm production costs thus reduced incentives to expand acreage and crop more
intensively, at least until crop prices increased enough to offset the higher input costs. On balance,
the higher input costs may have slowed what would have otherwise been higher growth in crop
output during a time of rising crop prices.
3.4. Other Factors
Once the market fundamentals were in place in 2008 (i.e., historically low s/u ratios for many
primary commodities such as wheat and corn), several additional factors emerged as market uncertainty grew. Volatility in commodity markets attracted money from other investment sectors
not faring as well, including equity markets and housing. Some have concluded that the “artificial demand created by investors’ speculation in commodities futures put tremendous upward
price pressure on food and energy commodities” (22, p. 35). Although financial flows can affect
market trading momentum, speculators alone have difficulty maintaining particular price levels
indefinitely, and they pursue lower prices with equal vigor when market fundamentals shift. Given
these findings, it is difficult to conclude that market speculation was a large factor in the overall
sustained price level between 2008 and 2012.
For some commodities, such as rice, for which trading accounts for only 7% of global output, the
market suffers from “thinness” that can result in highly volatile markets. Further exacerbating the
explosive rice market in 2008 was the activation of export controls by several exporters, including
India (November 2007), Vietnam and Egypt ( January 2008), China ( January 2008), and Cambodia
(March 2008) (23).
276
Rosegrant
·
Msangi
EG39CH10-Msangi
ARI
27 September 2014
11:57
Bad
weather
Market
driven
Hoarding
Financial
speculation
Export
restrictions
Policy
oriented
Low
stocks
relative
to use
Declining
value of
US $
Biofuels
expansion
Low
investments
in agricultural
research
Preemptive
buying
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
Income and
population
growth
Rising
energy
prices
Short term
Long term
Figure 1
Relative impact of factors behind food price spikes, 2007–2012.
3.5. Relative Importance of Factors Affecting Food Commodity Prices
The consensus among researchers is that the food price spikes during 2007–2012 were caused by
many interconnected factors that have become better understood with time. Most succinctly, a
summary analysis argued: “[T]he more one assesses this crisis, the more one concludes that it is
the result of a complex set of interacting factors rather than any single factor” (24, p. xiii).
While acknowledging the importance of supply and demand factors, others have concluded
that at least half of the 2008 price spike was the result of a speculative bubble in the case of wheat,
corn, and soybeans and a market panic in the case of rice (13). Speculation was also identified
as a key factor in the price spike, along with hoarding and the lack of a commodity reserve that
could have dampened volatile prices (25). Others have identified biofuels and other demand/supply
factors as major drivers of the food price spike (18, 19, 21). Still others have been reluctant to rank
the various factors because the relative impact of these drivers is not yet clear (26).
The factors behind food price spikes during 2007–2012 can be grouped by time period (short
term versus long term) as well as by origin (either market driven or policy related), as shown
in Figure 1. Many of the key papers in the literature touch on one or more of these factors to
explain the observed rise in food prices. The primary factors were low stocks relative to use, with
demand strengthening considerably from income and population growth and biofuels expansion
due to energy policies. Also key were rising energy prices and their subsequent impact on the cost
of production and on demand for biofuel as a substitute for petroleum-based motor fuel. Low
investment in research slowed growth in yields and dampened supply, and bad weather created
short-term supply shocks. Several other factors also contributed to upward price pressure at various
times, such as export restrictions and preemptive buying in the short term and a decline in the
US dollar over the longer term. Lastly, hoarding and financial speculation likely added to price
volatility as market uncertainty reached its peak.
4. SOCIOECONOMIC IMPACTS OF BIOFUEL POLICIES
Determining the impact of biofuel policies on global markets and land use is complicated by the
complex nature of global economic systems and human behavior. Generally, economies function
www.annualreviews.org • The Food-Versus-Fuel Debate
277
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
EG39CH10-Msangi
ARI
27 September 2014
Partial equilibrium
(PE) model:
economic model in
which the market
equilibrium of a
specific good or sector
of interest is obtained
assuming that prices
and quantities in other
markets and sectors of
the macro-economy
remain constant
Computable general
equilibrium (CGE)
model: economic
model that considers
linkages among all
sectors in the economy
by employing feedback
mechanisms between
biofuels and other
markets so as to ensure
that all relevant
transactions between
key economic agents
(including the
government) are
accounted for
278
11:57
in a series of feedback loops, each of which can have positive or negative price impacts that in turn
affect potential outcomes. Price impacts can be short or long term, can overlap and reinforce each
other, and can simply cancel each other out. Importantly, the effects of price spikes on human
welfare are greatest for populations that cannot smooth temporary increases in food expenditures
when prices rise. The poor often already spend a significant part of their income for food and have
little or no savings. Consequently, higher food prices can have the greatest consequence for those
whose food budget is already strained.
Economic models can be used to represent the influence of important drivers of socioeconomic
and environmental change on economic systems, based on some key assumptions of behavior, technology, and defined economic relationships that link the drivers to observed outcomes. Markets
form the core of economic models and represent the medium through which prices are transmitted and influence behavior on the production and consumption sides of the economy. As key
socioeconomic changes such as population and income growth evolve over time, economic models will typically translate this into a change in consumer behavior, which in turn influences the
producers of consumer goods within the economy through changing prices. As the production
mix adjusts to accommodate additional consumer demand, there is a change in the mix of inputs
used in production. Depending on the technologies and the relative intensities at which these
various inputs to (or factors of) production are being used, the demands on the resource base of
the economy change. Given that many of these inputs are themselves priced and supplied through
markets, we might observe a change in the prices of those inputs, which have further consequences
on the economy. Some economic models trace downstream economic impacts of changes in the
demand for the input: For example, consumers and households are affected, particularly when
there are changes in demand for labor, which represents a major source of household income.
Other models do not completely close this circle but rather focus on the environmental dimensions of pulling inputs from the natural environment. It is beyond the scope of this review to fully
describe the range of modeling approaches, but biofuels exemplify how a change in the demand
for energy and transportation fuel has profound downstream impacts and is the key economic link
that models are trying to capture in the food-versus-fuel trade-off. A representation of economic
linkages incorporating how biofuels affect an economy is shown in Figure 2.
Many studies have attempted to measure the impact of biofuel policies across several dimensions, including food prices, poverty, feed markets, and developing countries. In general, either a
partial equilibrium (PE) or a computable general equilibrium (CGE) multi-market model is used
to assess the impacts (27). [These can be referred to as general equilibrium, “applied” general
equilibrium, or “computable” general equilibrium models, in the context of this paper, because
even the PE models we refer to are computable models used for applied, empirical economic
analysis.] Both types of market models explain relationships between supply, demand, and prices
using a system of equilibrium equations. In PE models, market equilibrium of a specific good
or sector of interest is obtained assuming that prices and quantities in other markets and sectors
of the macro-economy remain constant. As such, they provide a detailed sector description but
do not account for the impact that sector may have on other sectors of the wider economy. In
contrast, CGE models take into account linkages among all sectors in the economy by employing
feedback mechanisms between biofuels and other markets so as to ensure that all relevant transactions between key economic agents (including the government) are accounted for. Although
CGE models generally have less sector detail, this approach provides a better global assessment,
as well as one that is generally regarded as more representative of medium- and long-term impact
because it allows price signals to shift resources among sectors, which ultimately results in more
modest price impacts than initially observed. In general, as indicated in the literature cited below,
CGE results tend to show modest global effects of biofuels but acknowledge the potential for
Rosegrant
·
Msangi
EG39CH10-Msangi
ARI
27 September 2014
11:57
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
Macroeconomic aspects
Microeconomic aspects
Global trade in
food and fuel
Agricultural and
trade policy
Demand for
fuel/food/other
commodities
Population and
income growth
Interest/exchange
rates
Energy/environmental
policy
Fuel/food/other
demand elasticities
Fuel/food/other
market prices
Demand for
renewable fuels
Production of
fuel/food/
other goods
Fuel market
Petroleum/biofuel substitution
Oil prices
Crop/livestock supply
Agricultural
prices
Crop inventory
changes
Farm returns
Crop yield
Crop mix/rotation
Crop management
Livestock management
Fertilizer and chemical use
Energy use
Cropland expansion
Food prices
Land quality
Land/other input prices
Risk aversion
Biofuel by-products
Environment
Land cover
Food
Biofuels
Pasture
Forestry
Natural forests
and reserves
Soil fertility/carbon
Erosion
Water availability/quality
Weather events
Climate change
Figure 2
A framework for analyzing biofuel–food market interactions. Figure adapted with permission from Oladosu & Msangi (3) with
permission.
adverse effects on more vulnerable populations, in part because these populations spend a high
proportion of income on food.
4.1. Biofuel Impacts on Food and Commodity Prices
Researchers are in widespread agreement that prices increase with biofuel expansion, but there is a
considerable range in the estimated impact on prices. Generally, for the reasons mentioned above,
the PE analyses result in estimates showing double-digit percentage gains in prices, whereas CGE
and longer-horizon modeling tends to indicate percentage increases in single digits. In multistudy
reviews, authors typically found a wide range of estimates of price effects, usually because of
differences such as model assumptions and scenario construction. Some research, which we discuss
below, has attempted to deal with these differences.
In one of the first papers on the growing influence of biofuels, Schmidhuber (28) examined
the impact on agricultural markets and prices of rising demand for bioenergy. His study reviewed
www.annualreviews.org • The Food-Versus-Fuel Debate
279
EG39CH10-Msangi
ARI
27 September 2014
11:57
quantitative assessments of bioenergy potential and concluded that demand for bioenergy is significant enough to change the historical trajectory of global agriculture, which had been one of
continued supply growth (primarily from decades of growth in crop yields), slow demand growth,
and falling real prices for agricultural commodities. In the last 10 years, the rising price of oil
and biofuel policies created demand for agricultural feedstocks for the energy market, resulting in
higher agricultural commodity prices. As the markets become more integrated between energy and
agricultural feedstock commodities, energy prices increasingly determine both the levels and the
variability of agricultural commodity prices, with implications for food prices. A significant outcome is higher real prices, which help revitalize rural areas and raise incomes, and nonagricultural
households benefit from increased employment as the economy expands.
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
4.1.1. Partial equilibrium models. As discussed above, the food commodity price increases that
began in 2001 and culminated in the food crisis of 2007/2008 reflected a combination of several
factors, including economic growth, biofuel expansion, exchange rate fluctuations, and energy
price inflation. Hochman et al. (29) found that estimates of the price impacts of observed levels of
biofuel production differ widely across studies, ranging between 20% and 60%. For example, using
a PE modeling framework to capture the interactions between agricultural commodity supply and
demand, as well as trade at the global level, Rosegrant et al. (19) found that, compared with the
2020 baseline prices, world prices in 2020 under a biofuel expansion scenario would be higher by
26% for maize, 18% for oilseeds, 12% for sugar, 11% for cassava, and 8% for wheat.
Using a different time period and approach, Hochman et al. (29) developed an empirical model
that included crop inventory adjustments and found that biofuels contributed 19.8% to the increase
in corn price in 2007 relative to 2001. Other factors included income shock, which contributed
29.6%; exchange rate shocks, 15.81%; and energy shocks, 10.8%. From a policy standpoint, as
Hochman et al. note, the food crisis emphasizes both the importance of inventory management
policies to help buffer potential price spikes and the need for mechanisms (e.g., adjustable mandates
or other policies) that either compensate the poor when prices rise to abnormally high levels or
more directly mitigate spikes in food prices. Hochman et al. also see a pressing need for expanding
agricultural supply through investment in research and development and promoting policies that
more effectively utilize existing technologies and invest in outreach and infrastructure that will
enhance productivity.
Durham et al. (30) use the AGLINK-COSIMO PE model of the global agricultural economy,
developed and maintained jointly by the OECD and the UN Food and Agricultural Organization
(FAO). Under highly stylized scenarios in which agricultural prices spike, this model shows that
up to 15% of a hypothetical spike in the price of coarse grains (e.g., maize, barley, oats) could be
avoided if the European Union removed its biofuels mandate when prices start to spike. Babcock
(31) conducted a similar analysis and found that removing subsidies for US ethanol production
in 2011 would have led to a 17% reduction in maize prices. He concluded that if market conditions are tight because of poor corn yields, then the mandate has a larger-than-average impact on
US market prices because it forces all the adjustment to tight supplies onto the livestock sector.
Babcock comments that a large expansion of the US ethanol industry would have occurred even
without the subsidies because higher crude oil prices increased the demand for biofuels and created strong market-driven investment incentives. He concludes that market-driven expansion of
ethanol production had a much larger price effect than did ethanol subsidies. As for retail prices,
the impact of lower crop prices from a lack of ethanol expansion would have been modest because
feed costs make up a relatively small share of retail prices.
Chen & Khanna (32) used another model that simulates welfare-maximizing consumers’ and
producers’ decisions in US agricultural and transportation fuel sectors, including international
280
Rosegrant
·
Msangi
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
EG39CH10-Msangi
ARI
27 September 2014
11:57
trade. They found that due to the increase in food prices, the RFS alone leads to a gain in producer
surplus by 21% relative to a scenario with no government involvement. Consumer surplus is
reduced by 5%. Reduced demand for fossil fuels under the RFS alone results in a 14% reduction
in fuel producers’ surplus compared with the baseline scenario, and low fuel costs benefit fuel
consumers by 2%. Mosnier et al. (33) used GLOBIOM, a global, multisectoral economic model
based on a detailed representation of land use. Results for the 2010–2013 period indicate that an
expanded US biofuel policy would substantially increase the portion of agricultural land needed
for biofuel feedstock production, and US agricultural exports would decrease while production
and exports increase in other exporting regions. Importing regions would face rising agricultural
prices, with price impacts depending upon the commodity. Corn prices, for example, could increase
9% if the biofuels target were increased 50% from baseline levels by the year 2020. Improved
productivity and stabilization of biofuel demand, though, would reduce market pressure and result
in lower prices, possibly ending the impact of biofuel policy. As a result, the authors argue for
enhancing agricultural productivity increases in the United States and globally as a necessary
condition for further development of biofuels, including the use of flexible programs favoring
investments in agricultural research.
De Gorter et al. (34) analyzed biofuel policies in developed countries and found that they
drove the unprecedented price spikes in the food-grain/oilseed sectors, including the policy to
phase out MTBE in the United States. At the same time, high oil prices activated the ethanol tax
credit, creating a direct link between oil and corn prices and food-grain commodity prices. The
authors note that both the 2010 and 2012 price run-ups were influenced by biofuel policies because
stocks were low and because any weather shock could therefore have a big impact on prices. More
generally, de Gorter et al. (34) describe biofuel policies as creating irreversible investments, and
eliminating biofuel policy would have very different outcomes compared with not having biofuel
policies in the first place because sunk costs were subsidized. Removing policies today would not
reduce biofuel production as much as monetary transfers from consumers, taxpayers, and investors
initially promoted ethanol production.
Roberts & Schlenker (35) designed their research to avoid complexities of multiple food products by reducing field crops to one core product. Their framework allows them to estimate demand
and supply elasticities of agricultural commodities with yield shocks (caused primarily by weather
fluctuations), using them to evaluate the impact of ethanol policies on consumers, particularly
world food commodity prices and quantities. They found that US biofuel mandates will increase
the price of food by 30%; if by-products from the biofuel production process are recycled for
livestock feed, then price increase is only 20%.
Producer surplus:
the amount of money
that exceeds what
producers are willing
to accept to sell a
product
Consumer surplus:
the amount of money
consumers are willing
to pay in excess of the
market price
4.1.2. Computable general equilibrium models. Al-Riffai et al. (27) estimated the impact of
EU biofuel policy using CGE models. They found that yield response and land elasticities (i.e.,
how new land is converted to agricultural uses when the rental price of land increases) play critical
roles. The simulations showed that the effects of EU biofuel policies on food prices are expected to
remain limited, with a maximum price change on the food bundle of +0.5% in Brazil and +0.14%
in Europe.
The results of CGE model research by Timilsina et al. (36) suggest that planned biofuel targets
would not cause large impacts on food supply at the global level. Agricultural commodities such
as sugar, corn, and oilseeds that serve as the main biofuel feedstock would experience 1–9% price
increases in 2020 compared with baseline owing to the expansion of biofuels to meet the existing
targets. Timilsina et al. also note that the robustness of the results is related to the relatively
small share of farm value in food processing (hence the moderate food price effects). Also, only
small variations in land use are needed to accommodate additional feedstock demand. Timilsina
www.annualreviews.org • The Food-Versus-Fuel Debate
281
EG39CH10-Msangi
ARI
27 September 2014
11:57
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
et al. conclude that putting land into production takes resources and time; therefore, the results
assume that adjustments can take place over time. Although global impacts appear modest, the
authors note that impacts would be significant in developing countries such as India and those in
sub-Saharan Africa, highlighting the concern with food security of vulnerable populations.
Because CGE models tend to adopt a somewhat more standardized approach to modeling
compared with PE models, the variation in results across different CGE model–based studies
tends to be less than that for PE model–based studies. Zhang et al. (37) also make note of this and
point to some of the key assumptions that do account for differences among this class of models,
such as the degree of substitutability that the models allow between fossil-based fuels and biofuels,
the transportation energy mix, and the degree to which the by-products of biofuels are accounted
for in feed costs.
4.1.3. Multistudy reviews. Zilberman et al. (38) conducted a literature review of six studies
using regression analyses of weekly, monthly, and quarterly data and found that ethanol prices
throughout the world are affected by both food and fuel prices, but they conclude that the linkage
between ethanol prices and food prices is weak. On the basis of their own studies, the authors found
that although the introduction of corn ethanol has had a significant impact on food commodity
prices, economic growth was the most important contributor to increases in prices of food (cropbased and meat from livestock), followed by biofuel and increases in the price of fuel and exchange
rates. For crop prices, the US biofuel policy contributed to 20–25% of the increase in the price
of corn between 2001 and 2007 and contributed to 7–8% of the increase in the price of soybeans.
The authors note that the increase in biofuel production may lead to a lower reduction in food
supply than implied from the allocation of corn to ethanol because extra profits generated by corn
ethanol encourages farmers to expand acreage and invest in equipment and induces farm-input
suppliers to develop new products that will increase corn supply.
Oladosu & Msangi’s (3) review found a wide range of estimates of the impacts of biofuel policy
on the production of grains, sugarcane, and oilseeds. They ranged from 1% to 51% for corn, 1%
to 95% for oilseeds, and 1% to 147% for sugarcane, reflecting the different types of models, data,
scenarios, and parameters. The range of estimated changes in agricultural prices is almost as wide
as for production, at less than 1% to 84% for corn, oilseeds, and sugarcane. The authors found
that the wide range of estimated impacts of biofuels on food markets can be partly explained by
differences in modeling approaches, geographical scope, and assumptions about several crucial
factors including (a) domestic/global availability of land for agricultural expansion, (b) response
of oil prices to biofuel policy, (c) crop yields and livestock productivity, and (d ) availability and
management of crop inventories. Importantly, more recent studies tend to conclude that the
impacts of biofuels on food markets are smaller than initially thought, primarily because these
studies allow for the endogenous expansion/contraction of agricultural land in their simulations,
whereas studies that do not allow for agricultural land expansion tend to produce high production
and price effects in food markets.
Aside from the various literature-based meta-studies that have examined these differences, there
are more thorough ongoing efforts to better identify and document the sources of differences across
various models and how they account for the impact of important drivers such as biofuels. One of
these efforts is the Agricultural Model Intercomparison and Improvement Project (AgMIP), which
has brought together a group of PE and CGE modelers to better understand how differences in
data, technological growth assumptions, and land use (and land expansion) are handled—among
several other points of comparison. Nelson et al. (39), a study coming from this comparison effort,
documents the differences in climate change impacts across models. By carefully isolating the
differences that are due either to model structure or to the underlying assumptions and data, these
282
Rosegrant
·
Msangi
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
EG39CH10-Msangi
ARI
27 September 2014
11:57
efforts will help further narrow the range of estimates of impact coming from biofuels or other
important drivers of change in the global agricultural economy. They will also provide researchers
with a better means of harmonizing assumptions and reconciling important differences.
Similarly, Zhang et al. (37), in a review of nine studies, conclude that overall, each of the modeling efforts expect that biofuel development will likely remain a driving force in world agricultural
markets over the medium term, leading to higher prices and production levels. They also observe
that although the consensus within the literature is that biofuel growth is likely to have at least
some impact on future commodity prices, there is a considerable range in the estimates. Some
of the reviewed studies claim strong linkages and others weak linkages. The research points to
key assumptions and structural differences in the modeling approaches that create variation across
studies. One is the scenario design, which appears to be an important factor for the PE models
and has likely contributed to relatively high estimates of price impact from biofuels. In this case,
differences in biofuel production levels between the baseline (reference) scenario and the policy
change (biofuel) scenario determine the magnitude of biofuel production growth that is assumed
to occur within the projection horizon, and thus differences in the projected impacts of biofuel
growth on agricultural production and prices are based solely on scenario design. A second factor
is the crude oil price trend assumption; higher crude oil price assumptions tend to reduce the
estimate of impacts from imposing biofuel policies. A third is the presence or absence of biofuel
trade, the latter of which ignores important interactions between feedstock commodity prices and
biofuel production and export.
Zhang et al. conclude that because differences in assumptions are significant for estimates, policy makers should take into account the underlying assumption-based and structural differences
among models when using model-generated outcomes to evaluate economic and environmental
impacts and when making decisions. Researchers have an obligation not simply to report results
along with assumptions and explanations of technical merits, but also to offer accessible interpretations of the results with options and recommendations for policy makers.
Several authors point to strong agreement among the studies—including the Joint Research
Centre ( JRC) European Commission report (40), the EU Framework Contract Commission report (2), and the High Level Panel of Experts on Food Security and Nutrition (HLPE) report (1)—
regarding the direction of the changes, but the magnitudes of these effects differ. In general, the
results depend on various underlying assumptions such as future trends in fossil-fuel prices, population, and world GDP. Global land use change estimates due to biofuel policies were also quite
sensitive to yield growth assumptions. Most significantly, perhaps, uncertainty about future technological and productivity developments emerged as a key issue in assessing biofuel policy impacts.
To address the problem of varying assumptions and time periods referred to in many studies,
Condon et al. (4) conducted a meta-analysis of 18 studies (with a total of 78 estimates of the impact
of biofuels on food prices) released between 2008 and 2013. They note that, in the absence of
“normalizing” the results (explained below), a review of the literature reveals that the average estimated impact on the absolute price change of corn from biofuel expansion is 19%, with individual
studies ranging from near zero (or even negative for specific scenarios) to as high as 72%. They
found several factors that explain differences in price effects across studies and scenarios, including
projection year and inclusion of ethanol coproducts. The modeling framework is particularly
significant. In general, studies using CGE) models estimate smaller price effects. For example,
CGE estimates of price change per billion gallons of corn ethanol are about 2.5 percentage points
lower than PE model estimates. CGE models typically allow for adjustments in resource allocation
across all markets in the economy, which lessens the impacts in the directly affected sector.
According to Condon et al. (4), another important factor in the wide range of estimates is the
level of biofuel production used in the baselines and the policy scenarios; a smaller baseline and
www.annualreviews.org • The Food-Versus-Fuel Debate
283
EG39CH10-Msangi
ARI
27 September 2014
11:57
larger policy scenario production (i.e., larger increase in volume) are associated with significantly
higher absolute price changes. To more accurately compare across studies with varying assumptions of baseline and scenario production levels, the authors “normalize” the corn price impacts
by converting the absolute results from each study into a common metric: the percent change
in corn price per billion-gallon increase in corn ethanol. By isolating the price effect of biofuel
expansion while holding the level of expansion constant, the range of estimates within each study
shrinks. Across studies, they found that each billion-gallon expansion in corn ethanol production
results in only a 2.9% increase in long-run corn prices on average. The range of estimates was
−0.27% to 8.4%.
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
4.2. Poverty and Hunger
Again, the literature is in general agreement that there are negative impacts on poverty and
hunger with biofuel expansion. Higher food prices particularly reduce the poor’s access to food,
which has possible long-term, irreversible consequences for health, productivity, and well-being—
particularly if higher prices lead to reduced food consumption by infants and preschool children.
Five studies reviewed by Condon et al. (4) consider the implications of biofuel policies for food
security worldwide. They conclude that biofuel expansion will increase the number of people at
risk of hunger or in poverty in developing countries, with higher agricultural commodity prices
due to increased ethanol production of particular concern in developing countries. According
to their analysis, the number of additional people at risk per billion-gallon increase in ethanol
production ranges from 200,000 to 12 million persons. The range is wide, and methodological
and other issues highlighted above apply here as well. Condon et al. point out that food security is
significant because most developing countries are net importers of food, which means they often
face world prices for agricultural commodities. Also, the world’s poor are disproportionately
affected by higher commodity prices because they rely heavily on raw agricultural products and
spend a far greater portion of their income on staple food expenditures relative to consumers in
the developed world.
Schmidhuber (28) examines the impact of rising demand for bioenergy on agricultural markets
and prices, noting that the impact on food security needs to be analyzed not only in the context of
higher food prices and lower availability but also in terms of rising incomes for farmers and rural
areas as well as changing price variability. Countries that are net buyers of both food and energy
would suffer from increases in both food and energy prices. As net buyers of food and energy, the
very poor could be particularly hard hit.
The EU Framework Contract Commission report (2) comments that the development of biofuels offers both opportunities and challenges for developing countries by providing an additional
source of agricultural income and the development of a biofuel industry that could contribute
to improving local infrastructures and rural development. In particular, high crop prices may be
beneficial for rural poor who will receive a better price and offer new export opportunities, but
this may also be met by a corresponding challenge for food security, notably for poor, urban
populations. Condon et al. (4) also note that offsetting positive effects can occur for farmers in
these countries, who are expected to benefit from higher prices by earning additional income if
they are net food producers. Similarly, a key conclusion from Schmidhuber (28) is that many rural
households stand to benefit through both higher prices for their produce and greater output, and
with 70% of the poor living in rural areas, the overall net effect on food security could be positive.
The HLPE report (1) also expects that for many, biofuels provide important new opportunities
for income and employment generation, in addition to bringing much needed capital, technology,
and knowledge to developing countries’ agriculture. Simulations show that large-scale biofuel
284
Rosegrant
·
Msangi
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
EG39CH10-Msangi
ARI
27 September 2014
11:57
projects for export can generate positive economic results, with an increase of 0.65% in overall
GDP, rising to 2.4% in the case of agriculture and 1.5% for industry.
To illustrate biofuel impacts on food security in developing countries, Tokgoz et al. (41)
calculate—across all commodities and for each region—the per capita utilization levels for food
if biofuel production were fixed, and they show how per capita utilization would change from the
baseline case. If biofuel production were held constant, the levels of calories available from food
would increase in all regions on a per capita basis. The largest increase in percentage terms would
be in Latin America, a major producer of biofuels. Growth in biofuels represents an average of
37 kilocalories per capita per day in food availability across developing countries, representing
a 1.4% difference compared to the case of increasing biofuel production. The authors note that
although 1.4% may not seem significant, when considered in terms of malnutrition for the most
vulnerable population in the developing world, young children, this difference in calorie availability
would represent 2 million fewer malnourished children across all regions, with the largest decrease
occurring in Asia (primarily South Asia).
Rosegrant et al. (19) found that the increase in crop prices resulting from expanded biofuel
production is also accompanied by a net decrease in availability and access to food. Rosegrant
et al. estimated that calorie consumption would decrease across regions under the two biofuel
scenarios (“biofuel expansion” and “drastic biofuel expansion”) compared with baseline levels.
For example, biofuel expansion has negative impacts on calorie availability in sub-Saharan Africa,
Latin America and the Caribbean, the Middle East and North Africa, and the rest of the regions.
The adverse effects on calorie consumption are particularly high in Africa, with a reduction of
more than 8% in calorie consumption compared with a scenario in which biofuel production did
not increase. Moreover, sub-Saharan Africa shows lower levels of imports for wheat and sugar
and higher levels of exports for maize and cassava under the two biofuel expansion scenarios.
These cause the numbers of preschool malnourished children in sub-Saharan Africa to increase by
1.5 million and 3.3 million for the “biofuel expansion” and “drastic biofuel expansion” scenarios,
respectively, compared with the projection in the baseline scenario. World total numbers of
preschool malnourished children are projected to increase by 4.4 million and 9.6 million under
the two scenarios, respectively.
Other researchers using CGE models tend to find modest impacts on overall income levels
but greater impacts on poor countries. Al-Riffai et al. (27) conclude that EU biofuel policy has
no significant real income consequences for the European Union, though some countries may
experience a slight decline in real income, including −0.12% for sub-Saharan Africa, due to a rise
in food prices. Similarly, Timilsina et al. (36), using a CGE model, found that global effects seem
small, especially in percentage terms. Importantly, though, low- and middle-income countries are
more negatively affected than high-income countries because about two-thirds of the world food
supply decrease is imputable to developing countries. Results indicate that China, sub-Saharan
Africa, the Middle East and North Africa, and India would suffer the most if biofuel targets are
implemented. For example, food availability would be reduced by $1.4 billion in India under that
country’s announced target scenario for biofuels.
De Jongh & Nielsen (42) summarize lessons learned from the three Jatropha pilot projects
in Mali, Mozambique, and Honduras, from 2007 to 2009. The projects focused on production
of biofuels by poor and subsistence farmers to increase their income and to substitute fossil
fuels used locally with biofuel, in contrast to large-scale commercial plantations focused on
export markets, which have been the goal of other projects. The three projects show that the
profitability is lower than expected and that only under some circumstances is Jatropha an
attractive option, primarily because the yield capabilities of the Jatropha plant have been below
expectations.
www.annualreviews.org • The Food-Versus-Fuel Debate
285
EG39CH10-Msangi
ARI
27 September 2014
11:57
According to the EU Framework Contract Commission report (2), one negative impact of
biofuel development is the risk that smallholders might lose their land due to unclear land tenure
systems and the increased interest in agriculture production and acquisition of large agricultural
areas. These forces can change land property relations that favor (re)concentration of wealth and
power in the hands of the dominant classes.
4.3. Feed and Livestock
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
Only a few studies have addressed the feed and livestock sectors. Those studies indicate negative
impacts for the sectors. In their CGE model, Timilsina et al. (36) consider distillers’ dried grains
with solubles (DDGS) as a feed crop substitute (DDGS are a by-product of ethanol and provide
additional revenues for ethanol plants, while also supplementing livestock feed rations by displacing some of the demand for coarse grains). These authors note that indirect land use and price
effects linked to feed grains are dampened once DDGS are accounted for. The JRC European
Commission report (40) notes that impacts of biofuel policies on total EU livestock production
are small to negligible, although other results show there is likely a shift of production within the
European Union due to higher feed costs.
Oladosu et al. (43) apply a decomposition approach to estimate contributions from several
major processes to the supply of corn for ethanol production in the United States between 2001
and 2009. According to their analysis, the 23% reduction in the share of domestic uses of corn
in total supply occurred mainly in the feed and residual uses category, but it did not lead to large
reductions in livestock production activities because about one-third of the corn used for ethanol
production is returned as by-product feed (DDGS). De Gorter et al. (34) note that biofuel policy
raises feed costs for the livestock sector, essentially operating as a tax on value-added agriculture
by reducing incomes of these farmers and reducing economic growth in rural areas net of the
economic growth due to biofuel production.
Babcock (31) concludes that there was no rationale for the now-expired blender tax credit (a
reimbursement against US federal gasoline taxes for adding ethanol into gasoline fuel) because
it did little to help the biofuel industry while mandates were in place, except in years when high
gasoline prices had already stimulated demand beyond mandated levels. In this situation, the extra
demand stimulus helped biofuel manufacturers but at great cost to the livestock sector because
it pushed world maize prices even higher than either energy prices or mandates would support.
Babcock recommends additional flexibility in US policy by relaxing blending mandates when
feedstock supplies are low because otherwise all adjustment to low feedstock supplies are forced
onto the livestock sector when consumers have gasoline as a ready substitute for lower biofuel
supplies. One option to increase flexibility in US mandates is to increase the limits by which fuel
blenders can bank or borrow blending credits when meeting their blending obligations more than
they are currently able to.
5. IMPACTS ON LAND USE AND GREENHOUSE GAS EMISSIONS
Scholars agree that land use change occurs with biofuel expansion, although some of the early
work may have overstated the degree of land use change. Several rigorous intermodel comparisons
highlight the sources of divergence in estimates that we discuss further below.
Separately, a portion of the literature addresses the dynamic nature of the issue by pointing
out that the impact of high crop productivity can reduce the land use change effects because of
both the initial shock from biofuels and the longer-term effect of increased investment in research
and technology. Notably, the issue of land use change embodies not only the crop production
used for food, feed, and biofuels feedstocks but also the other sectors that compete for land with
286
Rosegrant
·
Msangi
EG39CH10-Msangi
ARI
27 September 2014
11:57
crops, such as livestock, forestry, and urban uses. The degree to which these complex dynamics
of competition are captured quantitatively differs greatly across the literature.
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
5.1. Individual Model Results
Searchinger et al. (44) use a global agricultural market model to estimate emissions from land use
change, finding that corn-based ethanol, instead of producing a 20% savings, nearly doubles GHG
emissions over 30 years and increases GHGs for 167 years. This is in contrast to prior studies
that had found that substituting biofuels for gasoline reduces GHGs because biofuels sequester
carbon through the growth of the feedstock. The authors argue that these analyses failed to count
the carbon emissions that occur as farmers worldwide respond to higher prices and convert forest
and grassland to new cropland to replace the grain (or cropland) diverted to biofuels.
Al-Riffai et al. (27) use a CGE model to estimate the impacts of EU biofuel policy. The main
finding is that indirect land use change (ILUC)—which occurs when additional biofuel feedstock
acreage results in a global conversion of native vegetation to cropland—does indeed have an
important effect on the environmental impacts of biofuels. However, the size of the additional EU
2020 mandate, under current assumptions regarding the future evolution of renewable energy
use in road transport, is sufficiently small (5.6% of road transport fuels in 2020) and does not
threaten the environmental viability of biofuels. Specifically, direct emission savings from biofuels
are estimated at 18 Mt CO2 e,1 whereas additional emissions from ILUC are estimated at 5.3 Mt
CO2 e (mostly in Brazil), resulting in a global net balance of nearly 13 Mt CO2 e savings in a 20-year
horizon. Simulations for EU biofuel consumption above 5.6% of road transport fuels show that
ILUC emissions can rapidly increase.
Also using a CGE model, Timilsina et al. (36) obtain large effects in several countries implementing large biofuel targets with the general tendency to expand land devoted to feedstock crops
(sugar crops, grains, and oilseeds for vegetable oil for biodiesel). In the longer run, the land expansion would recede and productivity gains would reduce land use for feedstock and the trade-off
between food and biofuel.
Havlik et al. (45) provide a detailed analysis of the ILUC effect. Using a PE model of the
global forest, agriculture, and biomass sectors, results indicate that second-generation (advanced)
biofuel production fed by wood from sustainably managed existing forests would lead to a 27%
decline in overall emissions compared with the “No biofuel” scenario by 2030. Given that the
GLOBIOM model that they use in this analysis is uniquely designed to handle the forest sector
and its interaction with agriculture and other land uses, we are inclined to place some degree
of confidence in these results, although we have no immediate points of comparison to other
model results. The ILUC factor of first-generation biofuel global expansion is generally positive,
requiring 25 years to be paid back by the GHG savings from the substitution of biofuels for
conventional fuels.
Tyner et al. (46) estimate land use changes associated with US corn ethanol production up to
the 15 billion-gallon RFS using a special version of the Global Trade Analysis Project (GTAP)
model. On average, 24.4% of the land use change occurs in the United States and 75.6% in the
rest of the world. Forest reduction accounts for 32.5% of the change and pasture for 67.5%.
On average, 0.13 ha of land are needed to produce 1,000 gallons of ethanol. They conclude
that modeling land use change, like all economic modeling, is quite uncertain and that there is a
Indirect land use
change (ILUC):
the unintended
consequences of
releasing more carbon
emissions due to global
land use changes that
are induced by
market-mediated
effects and that happen
outside the region
from which the driver
of change originated
Second-generation
(advanced) biofuel:
fuel manufactured
from lignocellulosic
biomass or woody
crops, agricultural
residues, or waste
First-generation
biofuel: fuel
manufactured from the
sugars and vegetable
oils found in arable
crops and easily
extracted using
conventional
technology
1
CO2 e refers to “equivalent carbon dioxide,” a standard way of reporting GHG impacts in the scientific literature that
denotes the “climate warming potential” that various emissions could have when translated into the equivalent units of CO2
concentrations. Mt CO2 e refers to million metric tons of equivalent CO2 .
www.annualreviews.org • The Food-Versus-Fuel Debate
287
EG39CH10-Msangi
ARI
27 September 2014
11:57
relatively wide range of estimation differences. For example, the land needed to meet the ethanol
mandate ranges between 0.13 and 0.22 ha/1,000 gallons. The land use ethanol CO2 emissions
per gallon range between 1,167 and 1,676 g CO2 e/gallon. Total ethanol CO2 emissions due to
production and consumption of gasoline (including land use) range between 78.1 g CO2 e/MJ and
84.4 g CO2 e/MJ.
Research by Dumortier et al. (47) indicates that the impact of cropland expansion on carbon
emissions is extremely sensitive to model assumptions, especially with respect to the price-induced
yield response. As a result, it is difficult to narrow the range of reasonable parameter values to a
level that would allow robust policy conclusions (47).
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
5.2. Multistudy Reviews
A review of models is provided by Plevin & Kammen (53). Given the differing modeling approaches, data sets, and parameter choices, models of ILUC emissions have produced widely
divergent results. Figure 3 shows the results from several studies modeling the ILUC effects of
expanded corn ethanol production. The ranges reflect various approaches to exploring the sensitivity of ILUC emissions to model assumptions. The magnitude and time profile of emissions
from land-cover change depend on the type of land cover and the mode of clearing.
Edwards et al. (40) compare the ILUC results of five different models for marginal changes
in biofuels demand from different feedstocks in terms of hectares of ILUC. To enable direct
comparison, the results were standardized to kha per Mtoe biofuels (kilohectare per million tonnes
of oil equivalent). All biofuels in all models showed significant increases in land use for crops. In
the EU ethanol scenarios, the total estimated ILUC (in the world) ranges from 223 to 743 kha per
Mtoe. The major factors causing dispersion of model results include how much yields increase
225
200
175
g CO2e/MJ
150
125
100
75
50
Plevin
Dumortier
Searchinger
USEPA
(2012)
USEPA
(2017)
IFPRI
USEPA
(2022)
CARB
Hertel
0
Tyner
25
Figure 3
Emission estimates by various studies: Tyner (46), Hertel (55), CARB (48), USEPA 2022 (49), IFPRI (27),
USEPA 2017 (49), USEPA 2012 (49), Searchinger (44), Dumortier (50), Plevin (51). (Modified with
permission from Reference 53, p. 295.)
288
Rosegrant
·
Msangi
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
EG39CH10-Msangi
ARI
27 September 2014
11:57
with price and how much crop production is shifted to developing countries. For most of the
EU ethanol scenarios the models project that the largest share of ILUC would occur outside the
European Union. In the EU biodiesel scenarios, total ILUC ranges from 242 to 1928 kha per
Mtoe. Furthermore, the models project that the largest share of land use change would occur
outside the European Union. In US ethanol scenarios, total ILUC ranges from 107 to 863 kha
per Mtoe. In the palm oil scenarios, ILUC ranges from 103 to 425 kha per Mtoe.
In a review of 12 research articles, the US Government Accountability Office (52) also found
that there is disagreement about assumptions and assessment methods for estimating global ILUC.
The reviewed studies estimated a wide range of lifecycle GHG emissions of biofuels relative to
fossil fuels: from a 59% reduction to a 93% increase in emissions for conventional cornstarch
ethanol, a 113% reduction to a 50% increase for cellulosic ethanol, and a 41% to 95% reduction
for biodiesel. Large differences in the estimates were explained by differences in assumptions about
the agricultural and energy inputs used in biofuel production and how to allocate the energy used
in biofuel production to coproducts, such as DDGS.
Khanna & Crago (54) report that recent studies found an ILUC effect that is significantly
lower than the initial estimate from Searchinger et al. (44) of 104 g CO2 e per MJ (with a range
of 20–200 g CO2 e per MJ). Searchinger et al. (44) found that diverting 12.8 million ha of corn
cropland to increase US corn ethanol production by 56 billion L would require 10.8 million ha of
additional land to be brought into production globally. Hertel et al. (55) found that an additional
50 million L of US corn ethanol would lead to a global increase in cropland of only 3.8 million
ha; the associated emissions intensity of corn ethanol owing to land use change is 27 g CO2 e
per MJ (with a range of 15–90 g CO2 e per MJ). Compared with Hertel et al. (55), the ILUC
effect found by Tyner et al. (46) decreases by as much as 50% with modifications to the GTAP
model. These modifications include adding new land categories, unused cropland, and cropland
pasture; increasing the level of disaggregation in the model in terms of regions, commodities,
industries, and feedstocks for biofuels; and changing baseline conditions, including improvements
in technology, land productivity, and crop yields.
Lifecycle greenhouse
gas (GHG)
emissions: aggregate
GHG emissions
related to the full fuel
lifecycle, including all
stages of fuel and
feedstock production
and distribution
6. POLICY IMPLICATIONS
The biofuel literature reviewed here uses various economic modeling approaches to analyze the
socioeconomic and land use change impacts due to biofuel policies. The consensus of the studies
is that biofuel polices have resulted in higher commodity prices, but many other factors have also
pushed these prices up. Estimated impacts for biofuels in percentage terms range from single
digits to high double digits. As many researchers have pointed out, the complexity of the models
and the variety of assumptions create variability in estimated outcomes. Among the many factors
that affect the estimates are differences in modeling approaches, geographic scope, response of
oil and other commodity prices to biofuel policy, and assumptions used for the growth in crop
yields. Additional factors include the availability of crop inventories, scenario design, the structural
way that agriculture and energy market linkages are modeled, availability of land for expansion,
projection period, and time period analyzed. When scenarios and results are standardized across
various studies, however, the calculated price impact is reduced. Similarly, studies using CGE
models tend to report smaller impacts than PE models because the CGE models allow for resource
adjustments across all sectors of an economy and represent a longer-term outcome.
Several studies have reported impacts on poverty and hunger, with low- and middle-income
countries affected the most because they absorb a large share of any decline in the world food
supply. Moreover, the biofuel expansion certainly adds to overall commodity demand and raises
the number of people at risk for poverty and hunger due to higher commodity prices. Likewise,
www.annualreviews.org • The Food-Versus-Fuel Debate
289
ARI
27 September 2014
11:57
the feed and livestock sectors worldwide are negatively affected, but some of the loss is offset by
increased amounts of feed by-products generated by the biofuel production process.
There is also widespread agreement that biofuels have increased GHG emissions and land use
change, pressing more land into agricultural and energy production, but estimates vary widely.
Many of the same factors that create variability in price impacts across studies also affect impacts
for land use. Other factors also affect the estimates, including the size of the biofuel policy shock
and the scale of biofuel production.
A key observation across the literature is the dynamic nature of the food-versus-fuel issue.
Resolution of the issue over time is highly dependent on future rates of technological change and
productivity growth. For example, faster rates of yield gains or modifications in the government
policies could relieve much of the economic (and political) pressure associated with biofuel policies.
And as observed by Dumortier et al. (47), there are essentially no robust policy conclusions. The
primary lesson for the policy maker is that results from economic models depend heavily on
assumptions, and significant differences can be present in those assumptions, particularly because
they are used to predict human behavior and economic decisions and responses.
Given the impacts and potential adverse market effects, several researchers recommend more
flexible biofuel policies. Flexibility in mandates and other biofuel policies could allow for adjustments and relief when commodity supplies become tight and adversely affect food and feed
markets. Longer-term, variable requirements for ethanol could reduce market volatility for all
participants by allowing ethanol demand to flex lower (rather that remain fixed) as all other sectors do when prices rise. This in turn could also improve the investment climate for agriculture.
Such flexibility may be entering into US biofuel policy in 2014 as the Environmental Protection Agency considers a downward adjustment in the biofuel mandate due to practical limits on
blending more ethanol with gasoline than the market can absorb and because there is no current
capacity to produce ethanol with cellulosic feedstocks on a commercial scale.
Although the range of estimated impacts in the literature is large, the central tendency from
these estimates is that a food-versus-fuel trade-off is created through the expansion of biofuels
from food crops. Continued rapid expansion of biofuel production, whether mandated through
blending requirements or planned according to self-sufficiency goals, will affect the food sector,
including price increases for food commodities, reductions in the availability of calories, and
increased levels of malnourishment in developing countries, hitting the poor, and poor children,
the hardest.
Given these impacts and the relatively modest reductions in GHG emissions from biofuels
once ILUC is accounted for, subsidies and mandates for biofuels cannot be justified on economic
or environmental grounds and should be phased out. For the longer term, it is critical to focus on
increasing agricultural productivity growth and improving policies and infrastructure related to
the storage, distribution, and marketing of food. These factors will continue to drive the future
health of the agricultural sector and will play the largest role in determining the food security and
well-being of the world’s poorer and more vulnerable populations.
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
EG39CH10-Msangi
SUMMARY POINTS
1. Factors behind food price spikes during 2007–2012 included low stocks relative to use,
with demand strengthening considerably from income, population growth, and biofuels
expansion due to energy policies. Other factors include rising energy prices and their
subsequent impact on the cost of production and on demand for biofuel as a substitute
for petroleum-based motor fuel.
290
Rosegrant
·
Msangi
EG39CH10-Msangi
ARI
27 September 2014
11:57
2. As the markets become more integrated between energy and agricultural feedstock commodities, energy prices increasingly determine both the levels and variability of agricultural commodity prices and, therefore, for food prices as well.
3. The literature on the impacts of biofuels on food and commodity markets is in widespread
agreement that price increases occur with biofuel expansion, but there is a considerable
range in estimated impact due to various factors such as model assumptions and scenario
construction.
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
4. The impact of biofuels on poverty and hunger is negative as higher food prices reduce
the poor’s access to food, which has possible long-term, irreversible consequences for
health, productivity, and well-being.
5. Biofuel expansion has negative effects on the feed and livestock sectors.
6. Land use change occurs with biofuel expansion as well, though some of the early literature
may have overstated the degree of land use change; subsequent research has yielded
smaller but significant impacts on land use.
7. Great care must be taken when drawing policy implications from modeling results based
on available data and limitations of current methods. Moreover, researchers have an
obligation not simply to report results along with assumptions and explanations of technical merits but also to offer accessible interpretations of the results with options and
recommendations for policy makers.
FUTURE ISSUES
1. Research has provided clear directional impacts but uncertain magnitudes. This results in
a lack of robust and detailed policy conclusions and recommendations for biofuel policies
with respect to food and crop markets, food security, land use, and environmental issues.
As a result, there is a critical need for examining how to develop data and methods that
can produce credible and robust policy-relevant analyses.
2. Many analysts argue for improvements or amendments to biofuel technologies, policies,
and strategies, but not necessarily for their wholesale removal. This is a useful line of
inquiry that could be further expanded in the literature so as to better guide policy and
strategic thinking within the sector.
3. Special attention should be given to assessing uncertainties about future technological
and productivity developments when assessing biofuel policy impacts.
DISCLOSURE STATEMENT
The authors are not aware of any affiliations, memberships, funding, or financial holdings that
might be perceived as affecting the objectivity of this review.
ACKNOWLEDGMENT
The authors are grateful to Thomas P. Tomich for useful comments that helped improve the
review.
www.annualreviews.org • The Food-Versus-Fuel Debate
291
EG39CH10-Msangi
ARI
27 September 2014
11:57
LITERATURE CITED
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
1. HLPE. 2013. Biofuels and food security. HLPE Rep. No. 5, High Level Panel Experts Food Secur.
Nutr., Comm. World Food Secur., Rome. http://www.fao.org/fileadmin/user_upload/hlpe/hlpe_
documents/HLPE_Reports/HLPE-Report-5_Biofuels_and_food_security.pdf
2. Diop D, Blanco M, Flammini A, Schlaifer M, Kropiwnicka MA, Markhof MM. 2013. Assessing the impact
of biofuels production on developing countries from the point of view of policy coherence for development. Final Rep.,
AETS Consort., Feb., Eur. Union Framew. Contract Comm., Contract No. 2012/299193, Brussels
3. Oladosu G, Msangi S. 2013. Biofuel-food market interactions: a review of modeling approaches and
findings. Agriculture 3:53–71
4. Condon N, Klemick H, Wolverton A. 2013. Impacts of ethanol policy on corn prices: a review and meta-analysis
of recent evidence. Presented at Jt. Annu. Meet. AAEA & CAES, Aug. 4–6, Washington, DC
5. Chakravorty U, Hubert MH, Nostbakken L. 2009. Fuel versus food. Annu. Rev. Resour. Econ. 1:645–63
6. Babcock BA. 2007. High crop prices, ethanol mandates, and the public good: Do they coexist? Iowa Agric.
Rev. 13(2):1–12
7. Sperling D, Yeh S. 2009. Low carbon fuel standards. Issues Sci. Technol. 25(2):57–66
8. Sperling D, Yeh S. 2010. Toward a global low carbon fuel standard. Transport Policy 17:47–49
9. Tyner WE. 2010. Comparison of the US and EU approaches to stimulating biofuels. Biofuels 1(1):19–21
10. Mitchell D. 2011. Biofuels in Africa: Opportunities, Prospects and Challenges. Washington, DC: World Bank
11. Hazell PBR, Evans M. 2011. Environmental economic and policy aspects of biofuels. In Handbook
of Sustainable Energy, ed. I Galarraga, M González-Eguion, A Markandya, pp. 375–92. Cheltenham,
UK/Northampton, MA: Edward Elgar
12. Msangi S, Evans M. 2013. Biofuels and developing economies: Is the timing right? Agric. Econ. 44:501–10
13. Piesse J, Thirtle C. 2009. Three bubbles and a panic: an explanatory review of recent food commodity
price events. Food Policy 34:119–29
14. Trostle R. 2008. Global agricultural supply and demand: factors contributing to the recent increase in food commodity prices. Outlook Rep. No. WRS-0801, Econ. Res. Serv., US Dep. Agric., Washington, DC. http://www.
ers.usda.gov/publications/wrs-international-agriculture-and-trade-outlook/wrs-0801.aspx#.
U8k365RdWSo
15. Trostle R, Marti D, Rosen S, Westcott P. 2011. Why have food commodity prices risen again? Outlook
Rep. No. WRS-1103, Econ. Res. Serv., US Dep. Agric., Washington, DC. http://www.ers.usda.gov/
publications/wrs-international-agriculture-and-trade-outlook/wrs1103.aspx#.U8k5upRdWSo
16. Off. Chief Econ., Agric. Mark. Serv., Farm Serv. Agency, Econ. Res. Serv., Foreign Agric. Serv. 2011.
World agricultural supply and demand estimates. Mon. Rep., May 11, US Dep. Agric., World Agric.
Outlook Board, Washington, DC. http://usda01.library.cornell.edu/usda/waob/wasde//2010s/2011/
wasde-05-11-2011.pdf
17. Rosegrant MW, Tokgoz S, Bhandary P. 2013. The future of the global food economy. In Food Security
and Sociopolitical Stability, ed. CB Barrett, pp. 35–63. Oxford, UK: Oxford Univ. Press
18. Mitchell D. 2008. A note on rising food prices. Policy Res. Work. Pap. 4682, World Bank, Washington, DC.
http://www-wds.worldbank.org/external/default/WDSContentServer/IW3P/IB/2008/07/28/
000020439_20080728103002/Rendered/PDF/WP4682.pdf
19. Rosegrant MW, Zhu T, Msangi S, Sulser T. 2008. Global scenarios for biofuels: impacts and implications.
Rev. Agric. Econ. 30(3):495–505
20. Abbott P, Hurt C, Tyner W. 2008. What’s driving food prices? Issue Rep., July, Farm Found., Oakbrook, IL. http://www.farmfoundation.org/webcontent/Farm-Foundation-Issue-Report-WhatsDriving-Food-Prices-404.aspx
21. Elliott K. 2008. Biofuels and the food price crisis: a survey of the issues. Work. Pap. No. 151, Cent. Glob. Dev.,
Washington, DC. http://www.cgdev.org/files/16499_file_Elliott_Biofuels_Food_Crisis.pdf
22. Mittal A. 2009. The blame game. In The Global Food Crisis, ed. J Clapp, M Cohen, pp. 13–28. Waterloo,
Can.: Cent. Int. Gov. Innov./Wilfrid Laurier Univ. Press
23. Headey D, Malaiyandi S, Fan S. 2009. Navigating the perfect storm—reflections on the food, energy, and
financial crisis. IFPRI Discuss. Pap. 00889, Int. Food Policy Res. Inst., Washington, DC. http://www.
ifpri.org/sites/default/files/publications/ifpridp00889.pdf
292
Rosegrant
·
Msangi
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
EG39CH10-Msangi
ARI
27 September 2014
11:57
24. Headey D, Fan S. 2010. Reflections on the global food crisis—how did it happen? How has it hurt? And how
can we prevent the next one? IFPRI Res. Monogr. 165, Int. Food Policy Res. Inst., Washington, DC.
http://www.ifpri.org/sites/default/files/publications/rr165.pdf
25. von Braun J, Torero M. 2009. Exploring the price spike. Choices 24(1):16–21. http://www.
choicesmagazine.org/magazine/pdf/article_58.pdf
26. Rapsomanikis G, Sarris A, eds. 2010. Commodity Market Review. Rome: Food Agric. Org. United Nations.
http://www.fao.org/docrep/012/i1545e/i1545e00.pdf
27. Al-Riffai P, Dimaranan B, Laborde D. 2010. Global trade and environmental impact study of
the EU biofuels mandate. Final Rep., ATLASS Consort., Int. Food Policy Res. Inst., Dir. Gen.
Trade Eur. Comm., Washington, DC/Brussels. http://www.ifpri.org/sites/default/files/publications/
biofuelsreportec.pdf
28. Schmidhuber J. 2006. Impact of an increased biomass use on agricultural markets, prices and food security: a
longer-term perspective. Presented at Int. Symp. Notre Eur., Nov. 27–29, Paris
29. Hochman G, Rajagopal D, Timilsina G, Zilberman D. 2011. The role of inventory adjustments in quantifying
factors causing food price inflation. Policy Res. Work. Pap. 5744, World Bank, Washington, DC
30. Durham C, Davies G, Bhattacharyya T. 2012. Can biofuels policy work for food security? An analytical
paper for discussion. Res. Pap., Dep. Environ. Food Rural Aff., London. http://www.gov.uk/government/
publications/can-biofuels-policy-work-for-food-security
31. Babcock BA. 2011. The impact of US biofuel policies on agricultural price levels and volatility. Issue Pap.
No. 35, ICTSD Prog. Agric. Trade Sustain. Dev., Int. Cent. Trade Sustain. Dev., Geneva
32. Chen X, Khanna M. 2013. Food versus fuel: the effect of biofuel policies. Am. J. Agric. Econ. 95(2):289–95
33. Mosnier A, Havlik P, Valin H, Baker JS, Murray BC, et al. 2012. The net global effects of alternative U.S.
biofuel mandates: fossil fuel displacement, indirect land use change, and the role of agricultural productivity growth.
Rep. NI R 12-01, Nicholas Inst. Environ. Policy Solut., Duke Univ., Durham, NC
34. de Gorter H, Drabik D, Just DR. 2013. Biofuel policies and food grain commodity prices 2006–2012: all
boom and no bust? Agbioforum 16(1):1–13
35. Roberts MJ, Schlenker W. 2010. Identifying supply and demand elasticities of agricultural commodities: implications for the US ethanol mandate. NBER Work. Pap. 15921, Natl. Bur. Econ. Res., Cambridge, MA
36. Timilsina GR, Beghin JC, van der Mensbrugghe D, Mevel S. 2012. The impacts of biofuels targets on
land-use change and food supply: a global CGE assessment. Agric. Econ. 43:315–32
37. Zhang W, Yu E, Rozelle S, Yang J, Msangi S. 2013. The impact of biofuel growth on agriculture: Why
is the range of estimates so wide? Food Policy 38:227–39
38. Zilberman D, Hochman G, Rajagopal D, Sexton S, Timilsina G. 2013. The impact of biofuels on commodity food prices: assessment of findings. Am. J. Agric. Econ. 95(2):275–81
39. Nelson GC, Valin H, Sands RD, Havlik P, Ahammad H, et al. 2014. Climate change effects on agriculture:
economic responses to biophysical shocks. Proc. Natl. Acad. Sci. USA 111:3274–79
40. Edwards R, Mulligan D, Marelli L. 2010. Indirect land use change from increased biofuels demand: comparison
of models and results for marginal biofuels production from different feedstocks. JRC Sci. Tech. Rep., EUR 24485
EN, JRC, Eur. Comm., Inst. Energy, Luxembourg
41. Tokgoz S, Zhang W, Msangi S, Bhandary P. 2012. Biofuels and the future of food: competition and
complementarities. Agriculture 2:414–35
42. de Jongh J, Nielsen F. 2011. Lessons learned: Jatropha for local development. Rep., FACT Found., Wageningen, Neth. http://www.jatropha.pro/PDF%20bestanden/2011-12_Jatropha_Lessons_Learned_
final.pdf
43. Oladosu G, Kline K, Uria-Martinez R, Eaton L. 2011. Sources of corn for ethanol production in the
United States: a decomposition analysis of the empirical data. Biofuels Bioprod. Bioref. 5:640–53
44. Searchinger T, Heimlich R, Houghton RA, Dong F, Elobeid A, et al. 2008. Use of U.S. croplands for
biofuels increases greenhouse gases through emissions from land-use change. Science 319:1238–40
45. Havlik P, Schneider UA, Schmid E, Bottcher H, Fritz S, et al. 2011. Global land-use implications of first
and second generation biofuel targets. Energy Policy 39(10):5690–702
46. Tyner WE, Taheripour F, Zhuang Q, Birur D, Baldos U. 2010. Land use changes and consequent
CO2 emissions due to US corn ethanol production: a comprehensive analysis. Final Rep., GTAP Resour.
www.annualreviews.org • The Food-Versus-Fuel Debate
293
EG39CH10-Msangi
ARI
27 September 2014
47.
48.
49.
50.
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
51.
52.
53.
54.
55.
294
11:57
No. 3288, Dep. Agric. Econ., Purdue Univ., West Lafayette, IN. http://www.gtap.agecon.purdue.
edu/resources/download/5200.pdf
Dumortier J, Hayes DJ, Carriquiry M, Dong F, Du X, et al. 2011. Sensitivity of carbon emission estimates
from indirect land-use change. Appl. Econ. Perspect. Policy 33(3):428–48
CARB. 2009. Proposed regulation to implement the low carbon fuel standard, Vol. II. Staff Rep., Initial Statement
of Reasons, Calif. Air Resour. Board, Sacramento, CA
USEPA. 2010. Renewable Fuel Standard Program (RFS2): Regulatory Impact Analysis. Washington, DC: US
Environ. Prot. Agency
Dumortier J, Hayes DJ, Carriquiry M, Dong F, Du X, et al. 2009. Sensitivity of Carbon Emission Estimates
from Indirect Land-Use Change. Ames, IA: Cent. Agric. Rural Dev., Iowa State Univ.
Plevin RJ, O’Hare M, Jones AD, Torn MS, Gibbs HK. 2010. Greenhouse gas emissions from biofuels:
Indirect land use changes are uncertain but may be much greater than previously estimated. Environ. Sci.
Technol. 44(21):8015–21
US Gov. Account. Off. 2009. Biofuels: potential effects and challenges of required increases in production and use.
Rep. GAO-09-446, Washington, DC. 184 pp. http://www.gao.gov/assets/160/157718.pdf
Plevin RJ, Kammen DM. 2013. Indirect land use and greenhouse gas impacts of biofuels. In Encyclopedia
of Biodiversity, Vol. 4, ed. SA Levin, pp. 293–97. Waltham, MA: Academic. 2nd ed.
Khanna M, Crago CL. 2012. Measuring indirect land use change with biofuels: implications for policy.
Annu. Rev. Resour. Econ. 4:161–84
Hertel TW, Golub AA, Jones AD, O’Hare M, Plevin RJ, Kammen DM. 2010. Effects of U.S. maize
ethanol on global land use and greenhouse gas emissions: estimating market-mediated responses. Bioscience
60(3):223–31
Rosegrant
·
Msangi
EG39-FrontMatter
ARI
8 October 2014
23:26
Annual Review of
Environment
and Resources
Contents
Volume 39, 2014
Introduction p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p pv
Who Should Read This Journal? p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p pvii
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
I. Integrative Themes and Emerging Concerns
Environmental Issues in Australia
Alistair J. Hobday and Jan McDonald p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 1
Gender and Sustainability
Ruth Meinzen-Dick, Chiara Kovarik, and Agnes R. Quisumbing p p p p p p p p p p p p p p p p p p p p p p p p p29
II. Earth’s Life Support Systems
Implications of Arctic Sea Ice Decline for the Earth System
Uma S. Bhatt, Donald A. Walker, John E. Walsh, Eddy C. Carmack, Karen E. Frey,
Walter N. Meier, Sue E. Moore, Frans-Jan W. Parmentier, Eric Post,
Vladimir E. Romanovsky, and William R. Simpson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p57
Modeling the Terrestrial Biosphere
Joshua B. Fisher, Deborah N. Huntzinger, Christopher R. Schwalm,
and Stephen Sitch p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p91
Tropical Forests in the Anthropocene
Yadvinder Malhi, Toby A. Gardner, Gregory R. Goldsmith, Miles R. Silman,
and Przemyslaw Zelazowski p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 125
Life’s Bottleneck: Sustaining the World’s Phosphorus for a Food
Secure Future
Dana Cordell and Stuart White p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 161
Tropical Intraseasonal Modes of the Atmosphere
Yolande L. Serra, Xianan Jiang, Baijun Tian, Jorge Amador-Astua,
Eric D. Maloney, and George N. Kiladis p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 189
III. Human Use of the Environment and Resources
Dynamics and Resilience of Rangelands and Pastoral Peoples Around
the Globe
Robin S. Reid, Marı́a E. Fernández-Giménez, and Kathleen A. Galvin p p p p p p p p p p p p p p p p 217
Carbon Dioxide Capture and Storage: Issues and Prospects
Heleen de Coninck and Sally M. Benson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 243
viii
EG39-FrontMatter
ARI
8 October 2014
23:26
Consensus and Contention in the Food-Versus-Fuel Debate
Mark W. Rosegrant and Siwa Msangi p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 271
Energy for Transport
Maria Figueroa, Oliver Lah, Lewis M. Fulton, Alan McKinnon,
and Geetam Tiwari p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 295
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
The Environmental Costs and Benefits of Fracking
Robert B. Jackson, Avner Vengosh, J. William Carey, Richard J. Davies,
Thomas H. Darrah, Francis O’Sullivan, and Gabrielle Pétron p p p p p p p p p p p p p p p p p p p p p p p 327
Human Appropriation of Net Primary Production: Patterns, Trends,
and Planetary Boundaries
Helmut Haberl, Karl-Heinz Erb, and Fridolin Krausmann p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 363
Consumer End-Use Energy Efficiency and Rebound Effects
Inês M.L. Azevedo p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 393
IV. Management and Governance of Resources and Environment
Environmental Ethics
Clare Palmer, Katie McShane, and Ronald Sandler p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 419
The Psychology of Environmental Decisions
Ben R. Newell, Rachel I. McDonald, Marilynn Brewer, and Brett K. Hayes p p p p p p p p p p p p 443
The Business of Water: Market Environmentalism in the Water Sector
Karen Bakker p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 469
V. Methods and Indicators
Advances in Measuring the Environmental and Social Impacts of
Environmental Programs
Paul J. Ferraro and Merlin M. Hanauer p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 495
Concepts and Methodologies for Measuring the Sustainability of Cities
Marı́a Yetano Roche, Stefan Lechtenböhmer, Manfred Fischedick,
Marie-Christine Gröne, Chun Xia, and Carmen Dienst p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 519
Measuring the Co-Benefits of Climate Change Mitigation
Diana Ürge-Vorsatz, Sergio Tirado Herrero, Navroz K. Dubash,
and Franck Lecocq p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 549
Networks and the Challenge of Sustainable Development
Adam Douglas Henry and Björn Vollan p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 583
Water Security and Society: Risks, Metrics, and Pathways
Dustin Garrick and Jim W. Hall p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 611
Contents
ix
EG39-FrontMatter
ARI
8 October 2014
23:26
Citizen Science: A Tool for Integrating Studies of Human and Natural
Systems
Rhiannon Crain, Caren Cooper, and Janis L. Dickinson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 641
Indexes
Cumulative Index of Contributing Authors, Volumes 30–39 p p p p p p p p p p p p p p p p p p p p p p p p p p p 667
Cumulative Index of Article Titles, Volumes 30–39 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 672
Errata
Annu. Rev. Environ. Resourc. 2014.39:271-294. Downloaded from www.annualreviews.org
Access provided by University of Bern on 05/08/15. For personal use only.
An online log of corrections to Annual Review of Environment and Resources articles may
be found at http://www.annualreviews.org/errata/environ
x
Contents