The Climate Change and Economic Impacts of Food Waste in the United States

Abstract

This study analyzes the climate change and economic impacts of food waste in the United States. Using loss-adjusted national food availability data for 134 food commodities, it calculates the greenhouse gas emissions due to wasted food using life cycle assessment and the economic cost of the waste using retail prices. The analysis shows that avoidable food waste in the US exceeds 55 million metric tonnes per year, nearly 29% of annual production. This waste produces life-cycle greenhouse gas emissions of at least 113 million metric tonnes of CO2e annually, equivalent to 2% of national emissions, and costs $198 billion.

Full text

Available online a
t www.fooddynamics.org
Int. J. Food System Dynamics
2 (4), 2011, 431-446
431
The Climate Change and Economic Impacts of Food Waste in
the United States
Kumar Venkat
Clean Metrics Corp., Portland, Oregon, USA
kvenkat@cleanmetrics.com
Received January 2012, accepted April 2012, available online April 2012
ABSTRACT
This study analyzes the climate change and economic impacts of food waste in the United States. Using loss-
adjusted national food availability data for 134 food commodities, it calculates the greenhouse gas emissions due to
wasted food using life cycle assessment and the economic cost of the waste using retail prices. The analysis shows
that avoidable food waste in the US exceeds 55 million metric tonnes per year, nearly 29% of annual production.
This waste produces life-cycle greenhouse gas emissions of at least 113 million metric tonnes of CO2e annually,
equivalent to 2% of national emissions, and costs $198 billion.
Keywords: food waste, climate change, greenhouse gas emissions, life cycle assessment (LCA)
1 Introduction
A recent study by the Food and Agriculture Organization (Gustavsson, et al., 2011) reported that one-third
of all food produced for human consumption is lost or wasted globally, amounting to as much as 1.2
billion metric tonnes annually. Food waste is a global problem of staggering proportions, but the
underlying reasons differ between countries. While food waste in industrialized countries is dominated by
retail and consumer waste, developing countries have high losses at the post-harvest and processing
stages due to spoilage in warm and humid climates resulting from the lack of modern transport and
storage infrastructures. Gustavsson et al. (2011) estimated the magnitude of worldwide food losses, but
did not assess the corresponding climate change or economic impacts.
Food waste is an issue in all of the major economies in the world. Japan’s households and food industry
together discard nearly 17 million metric tonnes of edible food annually, an estimated 30% of production
(Morisaki, 2011; MAFF, 2012; Srinivas, 2010). In India, nearly 30% of the country’s fruits and vegetables
are lost due to lack of cold-storage facilities, and more than 30% of the grain supplied through the public
distribution system is lost as well (Mukherji and Pattanayak, 2011). Food waste in China has increased
rapidly and now accounts for about 70% of household and commercial waste (Xin et al., 2012).
Stenmarck et al. (2011) examined food waste from the retail and wholesale sectors in Nordic countries
(Denmark, Finland, Norway and Sweden) and found that annual retail waste ranged from 40,000 to 83,500
metric tonnes in each of these countries. In comparison, the authors noted that retail and wholesale
waste is orders of magnitude higher in larger economies such as Japan, the United Kingdom (UK) and the
United States (US). This study looked at the causes and prevention of waste, but not the climate change or
economic impacts.

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
432
A report from the Waste & Resources Action Programme (Chapagain and James, 2011) is one of the first
country-level assessments of the environmental and economic impacts of food waste, focusing on
household waste. It found that households in the UK waste 8.3 million metric tonnes of food and drink
each year, with a value of at least $18.6 billion and responsible for about 3% of UK’s domestic greenhouse
gas (GHG) emissions.
In one of the earliest analyses of US food waste, Kantor et al. (1997) estimated that 27% of the available
food was wasted in 1995. Food waste is now receiving increasing attention in the US, with major news
organizations frequently covering it as an issue of interest (Barclay, 2011; Nassauer, 2012). The US
Department of Agriculture regularly publishes data on food losses (USDA ERS, 2009). While US cities such
as Portland are focusing on food waste recycling through composting (Walker, 2011), the food industry
has launched an initiative to help reduce food waste at the source (EL, 2011). However, a comprehensive
evaluation of the environmental and economic impacts at the national level has been lacking. Two recent
studies have taken a first look at the larger environmental impacts of food waste in the US beyond just the
disposal stage.
Hall, et al. (2009) used energy balance to calculate that nearly 40% of the food was wasted in the US as of
2003, accounting for more than one quarter of the total freshwater use and 4% of petroleum oil
consumption. The water and energy estimates were based on the overall freshwater consumption by
agriculture and the fossil energy used by the average farm to produce food containing 1 kcal of energy.
Cuellar and Webber (2010) used food loss data from the US Department of Agriculture for 1995 – which
showed that 27% of edible food was wasted – and estimated that the energy embedded in wasted food
represents about 2% of annual energy consumption in the US. The total energy required for food
production (agriculture, processing, transportation and handling) at the national level was compiled from
various literature sources. Agricultural energy use for 10 broad food categories was derived from the total
energy used by agriculture using relative intensity factors and production mass.
Both of these studies used top-down methods (i.e., starting from an economy-wide estimate of energy
used in agriculture and deriving from it a national average for the farm level or food category level) to
estimate the energy needed to produce food that is ultimately wasted. While Hall, et al. (2009) discussed
GHG emissions from the decomposition of wasted food (but not emissions from production or other
upstream stages), neither of these studies directly addressed climate change, arguably the most pressing
environmental problem of our times.
Garnett (2008) has pointed out that food waste contributes to GHG emissions in two ways: A relatively
minor impact from decomposition of the wasted food after disposal in landfills, and a potentially far more
significant impact from the embedded emissions associated with its production, processing, transport and
retailing. This second impact requires a life-cycle view of the wasted food.
It should also be noted that the climate change impact of food waste – as quantified by life-cycle GHG
emissions – is a more complete measure of environmental impact than embedded energy or barrels of oil:
It includes not only the emissions from the burning of fossil fuels but also significant other GHG emissions
that are not energy-related such as methane (in agriculture and waste disposal) and nitrous oxide (in
agriculture).
Besides environmental impacts, food waste also imposes an economic cost on consumers and retailers. If
quantified correctly, this could provide a unique incentive to simultaneously mitigate emissions and save
money through waste reduction.
The motivation for the present study is to quantify in a comprehensive manner, for the first time, the
annual climate change and economic impacts of the food wasted in the US using the most recent national
data available (as of this writing). This is particularly important given the position of the US as the world’s
largest economy and a major consumer of resources. In conjunction, a secondary goal is to develop and
demonstrate a robust food waste model and methodology – based on the principles of life cycle
assessment (LCA) – that can be used to monitor the future impacts of food waste not only in the US but
also in other parts of the world.
The approach adopted in this study is both bottom-up and life-cycle based: It analyzes 134 distinct food
commodities accounting for most of the food consumed in the US, and then groups them into 16 food
categories. Each of the 134 commodities is modeled using one or more representative production
systems, based on detailed North American production data in most cases. Foods such as beef, chicken,
pork and cheese are placed in their own separate categories because of their unique production
characteristics and significant climate change impacts. Such an approach can provide a degree of precision
and rigor that may not be possible with top-down methods.

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
433
The rest of the paper is organized as follows. We will first describe the methods used in our analysis
including: the life-cycle food waste model; LCA standards, software and database used in the calculation
of GHG emissions; the system boundary; important assumptions; and the how the economic impact is
calculated. Following this, we will describe the food waste data used in this study, including a detailed
summary of annual food production, consumption and waste in the US. We will then present detailed
results for the climate change and economic impacts of US food waste, including a sensitivity analysis to
test a critical subset of the assumptions used in this study. We will then close with concluding remarks
and recommendations for further work on quantifying the full impacts of food waste.
2 Methods
2.1 Life-Cycle Food Waste Model
Figure 1 illustrates the life-cycle model of material flow from production to disposal for each of the food
commodities. This model has been developed specifically to fit the loss-adjusted food availability data
series from the US Department of Agriculture (USDA ERS, 2009).
Figure 1. Life-cycle model of material flow from production to disposal
Equation 1 below defines the basic mass balance in the life cycle of a food commodity. The difference
between production (P) and consumption (C) is the total gross waste made up of waste at the distribution
(, retail ( ) and consumer ( ) levels. All quantities are product weights.
(1)
The food availability data series provides values for each of the terms in Equation 1 for all commodities on
an annual basis from 1970 through 2009. This is described further in the Food Waste Data section below.
is the gross consumer waste, the sum of avoidable and unavoidable consumer waste:
(2)
The avoidable consumer waste ( ) – also referred to as “consumer waste” in this paper – represents
uneaten food that is wasted at the consumer level and is defined in Equation 3. excludes the
unavoidable waste in consumed foods due to non-edible parts (such as skins and shells) as well as fat or
moisture losses in cooking. N is the fraction of a food commodity that is non-edible, and L is the fraction
that is lost as fat or moisture during cooking.
(3)
The non-edible fraction N for each commodity is obtained directly from the food availability data. The fat
or moisture lost in typical cooking is estimated from USDA ERS (1998) based on certain cooking
assumptions as shown below. These estimates apply only to meats, fish, eggs and oils, all of which lose fat
and possibly moisture during cooking. Vegetables may lose moisture in cooking, but we assume that this is
compensated on average by added moisture during cooking. Since cooking methods and cooking losses
can vary considerably, these typical loss estimates are subjected to a sensitivity analysis as described in
the Results and Discussion section.
Distribution waste
(Wd)
Retail
waste (Wr)
Unavoidable waste
from eaten food
(Wu)
Production (P)
Retail
Consumer
Distribution
Avoidable
waste
from
uneaten
food (Wc)

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
434
Table 1.
Estimated fat and moisture losses in typical cooking
Category
Cooking assumption
Fat/moisture loss
fraction
Beef
Steaks
0.25
Pork
Chops
0.25
Chicken
All cooking
0.26
Turkey
All cooking
0.27
Lamb
Chops/Steaks
0.25
Shellfish
Boiled
0.24
Fish
Baked
0.19
Fats & Oils
All cooking
0.10
Eggs
Scrambled
0.08
The total avoidable waste , then, is the sum of distribution waste, retail waste and avoidable
consumer waste.
(4)
The avoidable consumer waste is further adjusted for moisture and fat losses in cooking as follows in
order to estimate the remaining solid waste that is actually landfilled. Although cooking is not explicitly
included in this study, we assume that half of the consumer waste occurs after cooking. This assumption is
necessary for calculating , the quantity of waste sent to landfills after accounting for fat and moisture
losses in the cooking of certain foods.
(5)
2.2 LCA Standards and Methodology
Life-cycle GHG emissions for the food commodities have been modeled and analyzed based on the PAS
2050:2008 standard (BSI Group, 2008), which in turn builds on ISO standards (ISO, 2006) by specifying
additional requirements for the assessment of GHG emissions in the life cycles of products and services.
Within this framework, GHG emissions from agricultural processes and waste disposal are modeled based
on the IPCC tier 1 guidelines (IPCC, 2006). In the context of typical life cycle impact assessment
methodologies used in the food sector (Amani and Schiefer, 2011), this study uses a single impact
category: climate change, as quantified by life-cycle GHG emissions. The characterization step is included,
but normalization and weighting are not. A sensitivity analysis on parametric uncertainties and an
interpretation of the final results are presented in the Results and Discussion section.
2.3 LCA Software and LCI Database
FoodCarbonScopeTM (CleanMetrics, 2011b), a web-based LCA software tool for food and beverage
products, was used to perform the detailed cradle-to-grave GHG emissions modeling and analysis of all
the farming systems. FoodCarbonScope supports all of the LCA and GHG standards on which this study is
based (BSI Group, 2008; ISO, 2006; IPCC, 2006).
FoodCarbonScope includes CarbonScopeDataTM (CleanMetrics, 2011a), which is a life cycle inventory
(LCI) database. CarbonScopeData includes cradle-to-gate and unit process data for over 1100 products
and processes in the food and agriculture sectors, covering a full range of crop and animal production
systems, commercial food processing, commercial cooking appliances, packaging, and waste disposal. The
majority of this data is for US and Canadian production and processing drawn from over a dozen major
agricultural states and provinces. In addition, the database includes: food production data for Europe and
other parts of the world; data for all energy sources including electricity by grid regions; all common
freight transport modes used for food products, including refrigerated transport; and
refrigerators/freezers used for food storage in distribution and retail locations. FoodCarbonScope and
CarbonScopeData have been used previously in major LCA studies of North American food systems

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
435
(Hamerschlag and Venkat, 2011; Venkat, 2012).
Each of the 134 food commodities is mapped to one or more food production systems in the LCI database
for the purposes of this study. Most of the high-volume commodities – including beef, pork, chicken, fish,
some dairy, common nuts and legumes, and several fruits and vegetables – are modeled as the average of
two or more representative production systems in North America. Most of the other commodities are
modeled using the closest available North American production system in the database, with the
exception of tropical fruits and tuna which are modeled using overseas production systems.
2.4 LCA Goal and Scope Definition
The goal of the LCA portion of this study is to assess the climate change impact of the food wasted in the
US on an annual basis. The functional unit for the LCA of each food commodity is the actual annual
quantity consumed in the United States as calculated from USDA ERS (2009). This in turn requires a higher
quantity of production, and the difference between the two quantities determines the wasted food.
The spatial boundary for the LCAs of the food commodities is cradle to grave. This starts with extraction of
raw resources from the ground and ends with the disposal of uneaten food. The system boundary includes
the production, processing and packaging of food products, transport and storage through typical
distribution networks, storage at retail locations, and landfilling of waste. Disposal methods are discussed
further under Other Assumptions. Food production and processing are generally assumed to occur within
the United States, except for specialty items such as tropical fruits and tuna which are imported. Food
waste is considered at the distribution, retail and consumer levels for which data exist, but not at the farm
or processing level (USDA ERS, 2009). Certain food processing steps are excluded where the data are in
terms of the primary ingredients only, as explained further under Other Assumptions.
All energy used at the consumer level – including shopping trips, refrigeration and cooking – is excluded
from this analysis because of uncertainties and lack of adequate data. Therefore, the total climate change
and economic impacts of food waste as calculated in this study represent conservative lower bounds on
the actual impacts.
The temporal boundary consists of one year of production and consumption based on 2009 data. The
assessment period for the LCAs is 100 years, meaning that the climate change impact of one year’s food
waste is evaluated over the standard 100-year time horizon in this study. This is particularly important for
the calculation of long-term emissions from landfills due to food waste deposited in any one year (BSI
Group, 2008; IPCC, 2006).
2.5 Other Assumptions
The food waste analysis undertaken in this study considers the entire US food system, which necessitates
a number of reasonable assumptions. These assumptions are listed below, and a critical subset of the
assumptions is subjected to a sensitivity analysis as described in the Results and Discussion section.
• The vast majority of the food consumed in the US is assumed to be produced in North America, except
as indicated below. This is justified by the fact that nearly 99% of GHG emissions from the provision of
food in the US are due to domestic production and value chain activities (Stolaroff, 2009).
• All meat product weights are boneless-equivalent (edible) weights as specified in the food availability
data (USDA ERS, 2009).
• All fish and shellfish are assumed to be produced in North America through aquaculture. Tuna is wild
caught in Europe and imported to the US.
• Typical food processing is included for all commodities that are listed in their processed forms in the
food availability data (USDA ERS, 2009). Examples of such processed foods include fruit juices, canned
and frozen vegetables and fruits, canned tuna, various meats and fish, milled flour, etc. About half of
the 134 commodities in this study are processed in some way before entering the distribution stage.
On the other hand, some commodities are listed in the data only in terms of the primary ingredients –
additional processing steps are excluded from our analysis in such cases. These commodities mostly
include processed grain products such as breakfast cereals, pasta and bread, which are listed in terms
of the primary grains and flour.
• All fresh foods are stored (in refrigerators, freezers or otherwise) in distribution centers and retail
stores for an average of 7 days before purchase.
• All food commodities are assumed to be transported an average of 2400 km within North America
from production or processing locations to typical retail locations. Out of this, 2240 km are through
semi-trailer trucks and 160 km through single-unit trucks. Tropical fruits are transported an additional
5000 km by ocean, and canned tuna is transported an additional 10,000 km by ocean. All transport
modes include refrigerator or freezer compartments as needed.

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
436
• Two-thirds of all meat and fish is distributed frozen and the rest is distributed fresh.
• Fresh meat, fresh fish, dairy, fruits and vegetables are assumed to require refrigeration throughout the
distribution and retail stages. (While some fruits and vegetables may not be refrigerated for certain
periods at the retail stage, almost all are refrigerated during distribution.)
• Food packaging materials and configurations for meat and fish products are based on commercial
packaging information from Sealed Air (2011). All other commodities – including dairy, vegetables,
fruits, nuts, grains, oils, and juices – are assumed to be packaged in typical materials and
configurations as found in retail stores. In the case of grains, the food availability data (USDA ERS,
2009) are in terms of the primary grains and flour produced and consumed, but not in terms of final
processed products such as breakfast cereals or pasta. In such cases, the packaging assumptions apply
to the forms of the food commodities found in the data.
• All solid waste from wasted food is assumed to be landfilled under typical US conditions in anaerobic
landfills, with 21% of the landfill methane flared and 23.25% of the methane recovered for electricity
(EPA, 2006). The landfills are assumed to be distributed equally in Boreal temperate wet and dry
climate zones as defined by the IPCC (2006). Long-term carbon storage in the waste matter present in
landfills offsets a small portion of the final emissions. (Based on recent data from the US
Environmental Protection Agency (EPA, 2010), it appears that about 87% of the US food waste is
landfilled. As of 2009, only 2.5% of food waste was composted. While 10.7% of general municipal
waste was incinerated in 2009 (EPA, 2010), there is no explicit data available on the portion of food
waste incinerated. Given this lack of data and the dominant role of landfilling, we use landfilling to
model all food waste disposals.)
• Fluid milk and juice products are assumed to be disposed through waste water which is then treated in
an anaerobic reactor. Energy used in waste water processing comes from US average grid electricity.
2.6 Calculating the Economic Impact of Wasted Food
The economic impact of avoidable food waste is calculated in this study using current US retail prices for
all the food commodities. The retail price of a commodity reflects all the value added throughout the
value chain – including agriculture, processing, packaging, distribution and retail – and provides a very
good measure of the total economic value embedded in the commodity as delivered to consumers.
Therefore, retail prices are used to uniformly calculate the economic impact of all avoidable food waste
occurring after the production/processing stages – specifically waste at the distribution, retail and
consumer levels.
The US Department of Agriculture provides current national retail prices for most meats, eggs, vegetables
and fruits (USDA AMS, 2011; USDA ERS, 2011). Prices for the other commodities are based on current
advertised prices at a major online food retailer (Safeway, 2011). While the food waste data is for the year
2009, all retail prices used in this study are as of December 2011 because a complete set of 2009 prices is
not readily available.
Most of the food waste is generally landfilled, as assumed in this study. The typical cost structure for
municipal solid waste collection and disposal in North America is a flat rate for a fixed volume of waste
(Rosenberg, 1996), which makes it difficult to quantify the real disposal cost of a marginal increase or
decrease in the quantity disposed. Therefore, disposal cost is excluded from our calculation of the
economic impact of food waste. It should also be noted that disposal costs are likely to be negligible
compared to the retail prices of the wasted quantities.
3 Food Waste Data
The loss-adjusted food availability data series from the US Department of Agriculture (USDA ERS, 2009) is
the basis for the food waste analysis in this study. The USDA maintains the sole national database of food
availability and food loss data in the US. The data series provides annual per-capita food production,
waste and availability data for a full spectrum of food commodities in the United States, adjusted for food
spoilage and other losses to closely approximate per-capita intake. Food waste is further broken down
into waste at the distribution, retail and consumer levels. The US population estimate for 2009 (US Census
Bureau, 2011) is used to convert the annual per-capita data for all commodities into national aggregate
data.
This study uses the most recent year in the food availability data series, which is 2009 as of this writing,
and analyzes a total of 134 commonly consumed food commodities accounting for most of the food
consumed in the US. These commodities include common meats, fish, shellfish, dairy products, oils and
fats, eggs, sweeteners, nuts, legumes, grains, vegetables, fruits, and fruit juices. Note that this data is for
a very recent year compared to the 1995 data used by Cuellar and Webber (2010).

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
437
Table 2 summarizes the annual aggregate food production, consumption and avoidable waste data for the
major food commodity categories as derived from the food availability data (USDA ERS, 2009). Waste at
the consumer level has been adjusted to remove the unavoidable waste in consumed foods as per the
food waste model defined in the Methods section. The result is the avoidable consumer waste (defined by
Equation 3) and includes both the edible and non-edible portions of foods available at the retail level that
are not consumed.
All quantities in Table 2 are in millions of metric tonnes (MMT) per year. Using this data, our total
estimate of avoidable food waste in the US is 55.41 MMT/year for 2009, which amounts to 28.7% of total
annual production by weight. This translates to 180 kg/year of total avoidable waste on a per-capita basis
– this is less than the 280-300 kg/year reported for Europe and North America by Gustavsson et al. (2011)
because it excludes both production losses and the unavoidable consumer waste. Consumer waste
dominates the total waste, accounting for just over 60% of the total avoidable waste. Per-capita consumer
waste is 110 kg/year, which is within the 95-115 kg/year range estimated for Europe and North America
by Gustavsson et al. (2011). Retail waste – including waste in institutional food service – amounts to 34%
of the total. Figure 2 illustrates this in terms of absolute quantities (MMT), and Figure 3 depicts the same
data as percentage of food wasted in each category.
Table 2.
US annual food production, consumption and avoidable waste in 2009 (MMT/year)
Category Production
P
Consumption
C
Distribution
Waste
Retail
Waste
Avoidable
Consumer
Waste
Total
Avoidable
Waste
Beef
8.09
5.26
0.00
0.35
0.72
1.07
Pork 6.48 3.78 0.00 0.28 1.16 1.44
Chicken
7.80
4.50
0.00
0.31
1.42
1.73
Other Meats 1.95 1.27 0.00 0.08 0.14 0.21
Fish &
Shellfish
1.98 1.29 0.00 0.17 0.25 0.42
Cheese 4.90 3.93 0.00 0.33 0.64 0.97
Milk & Yogurt
26.47
18.63
0.00
3.18
4.66
7.84
Other Dairy 5.70 4.18 0.00 0.62 0.90 1.52
Butter, Fats &
Oils
10.88 7.23 0.00 2.08 0.87 2.95
Eggs 4.48 2.93 0.07 0.40 0.40 0.86
Sweeteners
18.00
12.82
0.00
1.98
3.20
5.18
Nuts 1.28 1.08 0.00 0.08 0.12 0.20
Legumes
0.96
0.81
0.00
0.06
0.09
0.15
Grains
27.02
18.89
0.00
3.24
4.89
8.13
Vegetables 37.60 21.91 1.95 2.96 8.72 13.63
Fruits & Juices
29.48
17.59
0.90
2.65
5.56
9.11
Total 193.10 126.13 2.92 18.76 33.73 55.41

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
438
Figure 2. US annual avoidable food waste in 2009 (MMT/year)
Figure 3. US annual avoidable food waste in 2009 as percentage of production

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
439
4 Results and Discussion
4.1 Climate Change Impact of US Food Waste
Table 3 summarizes the results of the LCA portion of this study and shows the GHG emissions due to
avoidable waste throughout the life cycles of food commodities. The total emissions from all stages are
presented as both aggregate national emissions and per-capita emissions. Figure 4 illustrates the national
emissions graphically. The emissions are reported in carbon dioxide equivalents (CO2e).
Table 3.
GHG emissions from avoidable US food waste in 2009 (MMT CO2e/year for all emissions, except per-capita emissions in Kg
CO2e/year)
Category
Production
+ Processing
Emissions
Packaging
Emissions
Distribution
+
Retail Emissions
Disposal
Emissions
Total
National Emissions
Total
Per capita
Emissions
Beef 17.27 0.10 0.32 0.34 18.03 58.74
Pork
7.12
0.13
0.43
0.45
8.13
26.49
Chicken 6.17 0.16 0.52 0.54 7.38 24.05
Other Meats
1.33
0.02
0.06
0.07
1.48
4.82
Fish & Shellfish 2.37 0.05 0.12 0.14 2.68 8.72
Cheese
8.60
0.23
0.24
0.34
9.40
30.63
Milk & Yogurt
6.89
1.72
1.89
0.20
10.70
34.84
Other Dairy
2.04
0.35
0.45
0.53
3.37
10.98
Butter, Fats & Oils
5.11
0.49
0.65
1.02
7.26
23.66
Eggs 1.82 0.14 0.21 0.29 2.47 8.03
Sweeteners 2.15 1.04 1.13 1.81 6.12 19.94
Nuts
0.20
0.01
0.04
0.07
0.33
1.06
Legumes 0.11 0.01 0.03 0.05 0.20 0.67
Grains
5.82
0.57
1.68
2.83
10.91
35.53
Vegetables 5.67 0.72 3.23 4.75 14.37 46.81
Fruits & Juices 4.79 0.50 2.11 2.68 10.08 32.84
Total
77.46
6.23
13.12
16.11
112.92
367.82

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
440
Figure 4. US national GHG emissions from avoidable food waste in 2009 (MMT CO2e/year)
Beef – accounting for 16% of the total emissions – is the single largest contributor to the emissions from
wasted food, even though the quantity of beef wasted amounts to less than 2% of the total waste by
weight. This is because of the high emissions intensity of beef (Hamerschlag and Venkat, 2011). Animal
products have a disproportionate climate change impact because of their relatively high emission footprints. They
make up about 30% of all wasted food by weight, but account for nearly 57% of the emissions. On the other hand,
grains, vegetables and fruits make up 56% of the waste, but contribute just 31% of the emissions due to their
relatively low emission footprints.
Figure 5. Components of US national GHG emissions from avoidable food waste in 2009
Figure 5 shows that the wasted GHG emissions are dominated by the production and processing emissions
which account for 68.6% of the wasted emissions. The total emissions from the production, processing,
packaging, distribution, retail and disposal of the avoidable food waste in the US amounts to 112.9 MMT
CO2e per year. These emissions are equivalent to 2% of net US GHG emissions for 2009 based on the
national emissions inventory published by the US Environmental Protection Agency (EPA, 2011). Note that

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
441
these emissions represent a conservative lower bound on the actual emissions attributable to food waste,
since all energy used at the consumer level – for example, in shopping trips, refrigeration and cooking –
has been excluded from this analysis.
4.2 Economic Impact of US Food Waste
The economic impact of the wasted food is considerable. Using 2011 retail prices, the avoidable food
waste (for the year 2009) has a total retail value of $197.7 billion, as shown in Table 4 and Figure 6. Out of
this, the consumer waste alone amounts to $124.1 billion, which is nearly 63% of the total retail value of
wasted food. Using the 2009 US population estimate of 307 million (US Census Bureau, 2011), the per-
capita retail value of total avoidable waste is $643.95 per year. The avoidable consumer waste portion of
this works out to about $1600 per year for a family of four. This suggests a promising opportunity to
motivate consumers to reduce waste, which would yield additional dividends by way of lower emissions.
Retail waste – including waste in institutional food service – is valued at $64.6 billion, which shows that
businesses and organizations also have much to gain by reducing waste. The economic value reported
here is a conservative lower bound because the cost of consumer-level energy use and the cost of waste
disposal are not included.
Table 4.
Retail values of US avoidable food waste in 2009 using 2011 prices (billions of dollars per year for all waste, except per-
capita waste in dollars per year)
Category
Value of
Distribution
Waste
Value of Retail
Waste
Value of
Avoidable
Consumer
Waste
Total
National
Value of
Avoidable
Waste
Total
Per-capita
Value of
Avoidable Waste
Beef
0.00
3.45
7.09
10.54
34.35
Pork
0.00
2.57
10.53
13.10
42.66
Chicken
0.00
1.86
8.52
10.38
33.81
Other Meats
0.00
0.94
1.60
2.54
8.28
Fish & Shellfish
0.00
2.94
4.42
7.37
23.99
Cheese
0.00
3.44
6.64
10.08
32.82
Milk & Yogurt
0.00
3.46
5.07
8.54
27.80
Other Dairy
0.00
3.26
4.76
8.02
26.14
Butter, Fats & Oils
0.00
9.26
3.93
13.19
42.95
Eggs
0.06
0.36
0.36
0.78
2.53
Sweeteners
0.00
6.75
10.92
17.67
57.56
Nuts
0.00
1.07
1.68
2.76
8.98
Legumes
0.00
0.28
0.45
0.73
2.38
Grains
0.00
6.78
10.46
17.24
56.17
Vegetables
5.67
10.76
32.56
48.99
159.57
Fruits & Juices
3.24
7.42
15.12
25.78
83.96
Total
8.97
64.62
124.11
197.70
643.95

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
442
Figure 6. Retail values of US national avoidable food waste in 2009 using 2011 prices (billions of dollars)
Animal products have a more moderate influence on economic impact (in contrast to the climate change
impact). They account for about 37% of the economic impact, only about seven percentage points above
their contribution to the total waste. Grains, vegetables and fruits account for 47% of the economic
impact, about nine percentage points below their contribution to the waste.
4.3 Sensitivity analysis
Sensitivity analysis is a necessary part of any modeling endeavor. It is used to test the robustness of
conclusions to uncertainties in assumptions (Sterman, 2000). Of the different types of sensitivities that
models exhibit, numerical sensitivity to parametric assumptions and estimates is important for LCA
models and is routinely tested in LCA studies (Dalgaard et al., 2008; Pelletier et al., 2010).
The results presented in the previous subsections have been tested for sensitivity to parameters in four
major areas: transport distances, storage time in distribution and retail, portion of consumer waste
occurring after cooking (this is referred to in the following discussion as the post-cooking waste), and the
fat/moisture losses in cooking. Baseline values for these parameters have been defined in the Methods
section. The sensitivity analysis varies these four parameters uniformly one at a time (univariate testing)
for all commodities as follows:
• Transport distance from production to retail: +50% and -50% relative to the baseline values of 2400
km domestic transport plus 5000-10,000 km of ocean transport for imported commodities
• Storage time in distribution and retail: +50% and -50% relative to the baseline value of 7 days
• Post-cooking consumer waste fraction: +50% and -50% relative to the baseline value of 0.5 (i.e., half of
the consumer waste occurs after cooking)
• Fat/moisture loss fractions: +25% and -25% relative to the baseline values in Table 1

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
443
Figure 7. Sensitivities of results to parametric assumptions and estimates
Figure 7 summarizes the sensitivities to parametric variations of two key results: (1) total GHG emissions
from all avoidable waste; and (2) total retail value of all avoidable waste. The sensitivity analysis considers
the effect of all four selected parameters on total GHG emissions, and the effect of only the last
parameter (fat/moisture loss fraction) on total retail value. While variations in transport distance and
storage time will ultimately affect the retail price to some extent, those sensitivities could not be tested
because economic values used in this study are not model-generated but instead taken directly from
other sources (USDA AMS, 2011; USDA ERS, 2011; Safeway, 2011). In contrast, GHG emissions from all
life-cycle stages have been computed based on the life-cycle model described in Section 2.1 and therefore
amenable to a more complete sensitivity analysis.
As the transport distances are varied between +50% and -50% relative to baseline values, the total GHG
emissions vary by +/-5.2%, indicating that the results exhibit only mild sensitivity to this parameter. For
storage times and post-cooking consumer waste, as the parameters are varied in the +/-50% range, the
GHG emissions vary by less than 1%, indicating virtually no sensitivity to these parameters.
As fat/moisture losses are varied in the +/-25% range, the total GHG emissions from avoidable waste vary
between -16.1% and +13.7%. The total retail value of the waste varies between -9.4% and +8%. This
suggests that sensitivity to the fat/moisture loss estimates is significant. As fat/moisture losses increase,
correspondingly more of the consumer level waste must be attributed to the unavoidable waste from the
cooking of consumed foods. The reverse is true as fat/moisture losses decrease. Since cooking methods
for foods such as meats, fish and eggs – and the corresponding fat/moisture losses – can vary
considerably, it is reasonable to expect that the GHG emissions attributable to avoidable waste might
have an uncertainty of up to +/-20% and the retail value might have an uncertainty of up to +/-15%.
5 Conclusions
This study has presented, for the first time, a comprehensive analysis of both the climate change and
economic impacts of food waste in the US. Using the loss-adjusted food availability data from the US
Department of Agriculture (USDA ERS, 2009) for 2009, this study has applied a rigorous life cycle
assessment methodology to calculate the annual life-cycle GHG emissions, which quantify the climate
change impact of food waste. The annual economic impact of the waste has been calculated using recent
retail prices for food commodities. The analysis is based on life-cycle modeling and analysis of 134 distinct
food commodities accounting for most of the food consumption in the US, most of which are produced in
North America (except for tropical fruits and tuna).

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
444
The food waste model – developed specifically to fit the USDA ERS (2009) food availability data – uses a
mass balance method to account for all material flows and adjusts the waste at the consumer level so that
only the avoidable waste due to uneaten food is considered in the final analysis.
The total avoidable food waste at the distribution, retail and consumer levels amounts to over 55
MMT/year, representing nearly 29% of annual production by weight. Over 60% of this waste occurs at the
consumer level. The production, processing, packaging, distribution, retail and disposal of this wasted
food results in GHG emissions of at least 113 MMT CO2e/year, which is equivalent to 2% of US national
emissions. Beef is the single largest contributor to this, producing 16% of all wasted emissions, because of
its high emissions intensity. All animal products together contribute 57% of the wasted emissions, even
though they make up only 30% of the waste by weight. Over two-thirds of the emissions occur in the
production and processing of food commodities.
There is a considerable economic cost to this waste. US businesses and consumers lose as much as $198
billion per year because of wasted food. Consumer waste alone amounts to $124 billion, or nearly 63% of
the total value, which works out to about $1600 per year for a family of four. The annual cost to
businesses and organizations at the retail level is nearly $65 billion. There is a promising opportunity here
to show both consumers and businesses that they have much to gain by reducing waste. Waste reduction
can save money as well as reduce emissions.
The total GHG emissions and economic value of food waste reported in this study represent conservative
lower bounds, since the analysis ignores all energy used at the consumer level as well as the cost of waste
disposal. These emissions are also subject to an uncertainty of up to +/-20% due to cooking assumptions.
The economic value of the waste reported here is subject to an uncertainty of up to +/-15%.
The modeling and analysis presented here can be extended in the future in several areas using the
analytical framework established in this study. By modeling cooking processes in more detail, the
uncertainty bands can be tightened significantly. By including the consumer-level energy use attributable
to food waste – due to shopping trips, refrigeration and cooking, using real-world data – the climate
change and economic impacts can be made more realistic. Consideration of the water footprint, land use
and other resource uses attributable to the wasted food would add further value to the analysis and
results. Finally, the methodology developed in this study can be used to monitor the environmental and
economic impacts of food waste on an ongoing basis, not only within the US but also for other regions of
the world.
References
Amani, P., Schiefer, G. (2011). Review on Suitability of Available LCIA Methodologies for Assessing
Environmental Impact of the Food Sector. Journal on Food System Dynamics 2(2): 194-206.
Barclay, E. (2011). How That Food You Throw Out is Linked to Global Warming. National Public Radio. Available
at: http://www.npr.org/blogs/thesalt/2011/10/07/141123243/how-that-food-you-throw-out-is-linked-to-
global-warming (accessed 29 March 2012).
BSI Group. 2008. PAS 2050:2008 - Specification for the Assessment of the Life Cycle Greenhouse Gas Emissions
of Goods and Services. London: BSI Group. Available at: http://shop.bsigroup.com/en/Browse-by-
Sector/Energy--Utilities/PAS-2050 (accessed 15 March 2011).
Chapagain, A., James, K. (2011). The Water and Carbon Footprint of Household Food and Drink Waste in the
UK. Banbury, Oxon: Waste & Resources Action Programme. Available at:
http://www.wrap.org.uk/retail_supply_chain/research_tools/research/report_water_and.html (accessed
22 December 2011).
CleanMetrics. (2011a). CarbonScopeDataTM. Available at http://www.cleanmetrics.com/html/database.htm
(accessed 22 December 2011).
CleanMetrics. (2011b). FoodCarbonScopeTM product technical brief. Available at
http://www.cleanmetrics.com/pages/FoodCarbonScopeProductTechnicalBrief.pdf (accessed 22 December
2011).
Cuellar, A.D., Webber, M.E. (2010). Wasted Food, Wasted Energy: The Embedded Energy in Food Waste in the
United States. Environmental Science & Technology 44(16): 6464-6469.
Dalgaard, R., Schmidt, J., Halberg, N., Christensen, P., Thrane, M., and Pengue, W. A. (2008). LCA of Soybean
Meal. International Journal of Life Cycle Assessment 13: 240-254.

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
445
EL. (2011). Food Industry Details Anti-Waste Initiative. Environmental Leader. Available at:
http://www.environmentalleader.com/2011/08/23/food-industry-details-anti-waste-initiative (accessed
22 December 2011).
EPA. (2006). Solid Waste Management and Greenhouse Gases: A Life-Cycle Assessment of Emissions and Sinks.
Washington, DC: US Environmental Protection Agency. Available at:
http://epa.gov/climatechange/wycd/waste/downloads/fullreport.pdf (accessed 22 December 2011).
EPA. (2010). Municipal Solid Waste in the United States: 2009 Facts and Figures. Washington, DC: US
Environmental Protection Agency. Available at:
http://www.epa.gov/epawaste/nonhaz/municipal/pubs/msw2009rpt.pdf (accessed 27 March 2012).
EPA. (2011). Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2009. Washington, DC: US
Environmental Protection Agency. Available at: http://epa.gov/climatechange/emissions/downloads11/US-
GHG-Inventory-2011-Complete_Report.pdf (accessed 22 December 2011).
Garnett, T. (2008). Cooking Up a Storm: Food, Greenhouse Gas Emissions and Our Changing Climate. Surrey,
UK: Food Climate Research Network, Center for Environmental Strategy. Available at:
http://www.fcrn.org.uk/sites/default/files/CuaS_web.pdf (accessed 29 March 2012).
Gustavsson, J., Cederberg, C., Sonesson, U., van Otterdijk, R., and Meybeck, A. (2011). Global Food Losses and
Food Waste. Rome: Food and Agriculture Organization of the United Nations. Available at:
http://www.fao.org/fileadmin/user_upload/ags/publications/GFL_web.pdf (accessed 22 December 2011).
Hall, K.D., Guo, J., Dore, M., and Chow, C.C. (2009). The Progressive Increase of Food Waste in America and Its
Environmental Impact. PLoS ONE 4(11): e7940.
Hamerschlag, K., Venkat, K. (2011). Meat Eater’s Guide to Climate Change and Health – Life-cycle Assessments:
Methodology and Results. Washington, DC: Environmental Working Group.
IPCC (2006). IPCC Guidelines for Greenhouse Gas Inventories. Geneva, Switzerland: Intergovernmental Panel on
Climate Change. Available at: http://www.ipcc-nggip.iges.or.jp/public/2006gl/index.html (accessed 22
December 2011).
ISO (2006). ISO 14040:2006 - Life cycle assessment - Principles and framework. Geneva, Switzerland:
International Organization for Standardization.
Kantor, L.S., Lipton, K., Manchester, A., and Oliveira, V. (1997). Estimating and Addressing America’s Food
Losses. Food Review 20: 2–12.
MAFF. (2012). Monthly Statistics of Agriculture, Forestry and Fisheries. Tokyo, Japan: Ministry of Agriculture,
Forestry and Fisheries. Available at: http://www.maff.go.jp/e/tokei/kikaku/monthly_e/index.html
(accessed 29 March 2012).
Morisaki, I. (2011). Pattern of Food Losses in Households: A Case Study in Oita-Prefecture, Japan. Oita, Japan:
Ritsumeikan Asia Pacific University. Available at: http://r-
cube.ritsumei.ac.jp/bitstream/10367/2585/1/MORISAKI%20Ikuko.pdf (accessed 29 March 2012).
Mukherji, B., Pattanayak, B. (2011). New Delhi Starts Drive to Root Out Hunger. The Wall Street Journal.
Available at: http://online.wsj.com/article/SB10001424052702304259304576372813010336844.html
(accessed 29 March 2012).
Nassauer, S. (2012). Leftovers: Tasty or Trash? The Wall Street Journal. Available at:
http://online.wsj.com/article/SB10001424052702304724404577293620871384492.html (accessed 29
March 2012).
Pelletier, N., Pirog, R., and Rasmussen, R. (2010). Comparative Life Cycle Environmental Impacts of Three Beef
Production Strategies in the Upper Midwestern United States. Agricultural Systems 103, 380-389.
Rosenberg, L. (1996). International Source Book on Environmentally Sound Technologies for Municipal Solid
Waste Management. Osaka, Japan: UNEP International Environmental Technology Centre. Available at:
http://www.unep.or.jp/Ietc/ESTdir/Pub/msw/index.asp (accessed 22 December 2011).
Safeway (2011). Safeway Online Shopping. Available at: http://shop.safeway.com/superstore (accessed 22
December 2011).
Sealed Air (2011). Cryovac Fresh Food Packaging. Available at: http://www.cryovac.com/en/default.aspx
(accessed 22 December 2011).
Srinivas, H. (2010). Food Waste in Japan. Available at: http://serendipity.gdrc.info/?p=204 (accessed 29 March
2012).

Kumar Venkat / Int. J. Food System Dynamics 2 (4), 2011, 431-446
446
Stenmarck, A., Hanssen, O.J., Silvennoinen, K., Katajajuuri, J., and Werge, M. (2011). Initiatives on Prevention of
Food Waste in the Retail and Wholesale Trades. Copenhagen, Denmark: Nordic Council of Ministers.
Available at: http://www.norden.org/en/publications/publikationer/2011-548 (accessed 29 March 2012).
Sterman, J.D. (2000). Business dynamics: Systems thinking and modeling for a complex world. New York: Irwin
McGraw-Hill.
Stolaroff, J. (2009). Products, Packaging and US Greenhouse Gas Emissions. Athens, GA: Product Policy
I
nstitute. Available at:
http://www.productpolicy.org/ppi/attachments/PPI_Climate_Change_and_Products_White_Paper_Septe
mber_2009.pdf (accessed 22 December 2011).
US Census Bureau. (2011). The 2009 Population Estimate for the United States. Available at:
http://factfinder.census.gov/servlet/SAFFPopulation (accessed 22 December 2011).
USDA ARS (1975). USDA Agriculture Handbook No. 102, Food Yields. Washington, DC: United States
Department of Agriculture, Agricultural Research Service. Available at:
http://www.nal.usda.gov/fnic/foodcomp/Data/Classics/ah102.pdf (accessed 22 December 2011).
USDA AMS (2011). AMS Market News. Washington, DC: United States Department of Agriculture, Agricultural
Marketing Service. Available at http://www.ams.usda.gov/AMSv1.0/marketnews (accessed 22 December
2011).
USDA ERS (1998). A Dietary Assessment of the U.S. Food Supply: Comparing Per Capita Food Consumption with
Food Guide Pyramid Serving Recommendations - Agricultural Economics Report No. (AER772). Washington,
DC: United States Department of Agriculture, Economic Research Service. Available at:
http://www.ers.usda.gov/Publications/AER772 (accessed 22 December 2011).
USDA ERS (2009). Food Availability (Per Capita) Data System. Washington, DC: United States Department of
Agriculture, Economic Research Service. Available at: http://www.ers.usda.gov/Data/FoodConsumption
(accessed 22 December 2011).
USDA ERS (2011). Meat Price Spreads. Washington, DC: United States Department of Agriculture, Economic
Research Service. Available at: http://www.ers.usda.gov/Data/MeatPriceSpreads (accessed 22 December
2011).
Venkat, K. (2012). Comparison of Twelve Organic and Conventional Farming Systems: A Life Cycle Greenhouse
Gas Emissions Perspective. Journal of Sustainable Agriculture (in press).
Walker, M. (2011). Portland Composting Gets Enthusiastic Green Light. Sustainable Business Oregon. Available
at: http://www.sustainablebusinessoregon.com/articles/2011/08/portland-composting-gets-
enthusiastic.html (accessed 22 December 2011).
Xin, Z., Kaihao, W., and Anqi, C. (2012). Waste Not, Want Not. China Daily. Available at:
http://www.chinadaily.com.cn/cndy/2012-01/19/content_14472383.htm (accessed 29 March 2012).