This study provides a review of sustainable coffee certifications and results from a quantitative analysis of the effects of Fair Trade, organic and combined Fair Trade/organic certifications on the livelihood strategies of households and cooperatives in Central America and Mexico.
Fair Trade/Organic Coffee, Rural Livelihoods, and the “Agrarian Question”: So...
Effects of Fair Trade and organic certifications on small-scale coffee farmer households in Central America and Mexico
1. Renewable Agriculture and Food Systems: 25(3); 236–251 doi:10.1017/S1742170510000268
Effects of Fair Trade and organic
certifications on small-scale coffee
farmer households in Central America
and Mexico
V. Ernesto Mendez1,*, Christopher M. Bacon2, Meryl Olson3, Seth Petchers4, Doribel Herrador5,
´
Cecilia Carranza6, Laura Trujillo7, Carlos Guadarrama-Zugasti7, Antonio Cordon8,
´
8
and Angel Mendoza
1
Environmental Program and Department of Plant and Soil Science, The University of Vermont,
153 So. Prospect St., Burlington, VT 05401, USA
2
Department of Geography, 513 McCone Hall, University of California-Berkeley, Berkeley, CA 94720-4740, USA
3
Department of Plant and Soil Science, Hills Agricultural Bldg, 105 Carrigan Drive, The University of Vermont,
Burlington, VT 05405, USA
4
1 South Ivy Street, Denver, CO 80224, USA
5
University of Barcelona, Calle Balmes 17-2a-2, Terrassa 08225, Barcelona, Spain
6
´ ´
Ministerio de Medio Ambiente y Recursos Naturales, Unidad de Planificacion—Area de Economıa Ambiental,
Carretera a Santa Tecla Km. 5 1/2, Calle y Colonia Las Mercedes, Edificio MARN, San Salvador, El Salvador
7
Universidad Autonoma de Chapingo, CRUO, Km. 3 Carretera Huatusco-Jalapa, Huatusco, Veracruz, Mexico
8
´
Asociacion CRECER, 20 Calle 14-19 zona 10, Guatemala Ciudad, Guatemala
*Corresponding author: emendez@uvm.edu
Accepted 13 April 2010; First published online 4 June 2010 Research Paper
Abstract
We provide a review of sustainable coffee certifications and results from a quantitative analysis of the effects of Fair Trade,
organic and combined Fair Trade/organic certifications on the livelihood strategies of 469 households and 18 cooperatives
of Central America and Mexico. Certified households were also compared with a non-certified group in each country. To
analyze the differences in coffee price, volume, gross revenue and education between certifications, we used the Kruskal–
Wallis (K–W) non-parametric test and the Mann–Whitney U non-parametric test as a post-hoc procedure. Household
savings, credit, food security and incidence of migration were analyzed through Pearson’s chi-square test. Our study
corroborated the conditions of economic poverty among small-scale coffee farmer households in Central America and
Mexico. All certifications provided a higher price per pound and higher gross coffee revenue than non-certified coffee.
However, the average volumes of coffee sold by individual households were low, and many certified farmers did not sell
their entire production at certified prices. Certifications did not have a discernable effect on other livelihood-related
variables, such as education, and incidence of migration at the household level, although they had a positive influence
on savings and credit. Sales to certified markets offer farmers and cooperatives better prices, but the contribution derived
from these premiums has limited effects on household livelihoods. This demonstrates that certifications will not single-
handedly bring significant poverty alleviation to most coffee-farming families. Although certified coffee markets alone
will not resolve the livelihood challenges faced by smallholder households, they could still contribute to broad-based
sustainable livelihoods, rural development and conservation processes in coffee regions. This can be done by developing
more active partnerships between farmers, cooperatives, certifications and environmental and rural development organ-
izations and researchers in coffee regions. Certifications, especially Fair Trade/organic, have proven effective in supporting
capacity building and in serving as networks that leverage global development funding for small-scale coffee-producing
households.
Key words: farmer cooperatives, political ecology, rural livelihoods, coffee crisis, alternative markets, sustainable coffee
# Cambridge University Press 2010
2. Effects of Fair Trade and organic certifications on small-scale coffee farmer households 237
Introduction “gourmet” or “premium” coffee, specialty coffees are made
from exceptional beans grown only in ideal coffee-
The coffee price crisis starting in the early 1990s, and producing climates. They tend to feature distinctive flavors,
deepening between 1999 and 2003, had enormous social and which are shaped by the unique characteristics of the soil
economic impacts on coffee producers around the world1,2. that produces them’10. Under the specialty coffee umbrella,
Between 1989 and 2004, green coffee prices fell from certified coffees are those which receive a certificate or
US$1.20 lb - 1 to between US$0.45 lb - 1 and US$0.65 lb - 1. label from a third-party, independent agency11, such as
Although prices have rebounded to levels consistent with certified Fair Trade (e.g., FLO label) or certified organic
the market averages during the previous 50 years (generally (e.g., USDA or Naturland labels). ‘Sustainable coffee
between US$1.00 lb - 1 and US$1.25 lb - 1)3, the conse- certification’ is an umbrella term encompassing several
quences of the price crash, and the chronic poverty that types of certifications, and combinations of certifications
already existed, remain. In other words, even though prices that explicitly address social and environmental concerns
have recovered, the coffee crisis continues4,5. and standards. Specialty and sustainable coffee markets
In the midst of this situation, great expectations were have expanded rapidly with increased consumer awareness
placed on the role of several ‘sustainable coffee’ certifi- regarding issues of quality, social equity, taste, health and
cation initiatives, as key alternatives for farmers to confront environment12,13. These trends are evident in the expand-
the crisis6. Certification systems claim to offer a com- ing retail market values for the specialty coffee market
bination of benefits including higher and more stable prices segment, which has increased from less than $4 billion in
to farmers, and increased market access and technical the early 1990s to over $11 billion in 200614. As of 2004,
assistance. In turn, farmers and their organizations must about 10% of global coffee exports could be considered
meet standards and subject themselves to inspections. As specialty coffee, and of these an estimated 20% (or 2%
the number of certification initiatives increases, and con- of the global supply) carried some type of certification
sumers become more aware of different certification (Table 1). The percentage of specialty and sustainable
options, there is a growing concern to accurately document coffees exported from Central America is significantly
the real impact of these alternatives on coffee producers above the global average. During the 2002/2003 harvest,
and their families. Recent research on sustainable coffee an estimated 6% of the region’s total exports were sold
certification points to differentiated impacts on growers for into organic (1.5%), Fair Trade (3.6%) and Rainforest
each type of certification7,8. These studies provide impor- Alliance (0.75%) certifications15. Although small, this
tant comparative examinations of certification standards certified market segment continues to grow rapidly, with
and practices, often focusing on the macro-impacts of each annual rates ranging from 10 to over 30%, depending on
program. However, most lack household- and community- the year and the certification program6,16. If we expand the
level data that document the perceived benefits and limi- 2002/2003 market figures to include Mexico, account for
tations, as expressed by growers and their families9. the rapid growth rates among these certification programs
The general objective of this research was to analyze and document the presence of newer sustainable coffee
the effects of Fair Trade and organic certifications on the certification programs (e.g., Utz certified and Starbucks
livelihoods of coffee producer households of Central CAFE Practices), as of 2003/2004, these certified coffees
America and Mexico. Our specific objectives were: (1) to accounted for 8–13% of the region’s coffee production.
provide a critical review of the different types of sustain-
able coffee certifications; (2) to analyze the effects of Fair
Trade, organic and combined Fair Trade/organic certifi- Certified organic coffee
cations on the livelihood strategies of coffee-producing
Organic certification addresses standards that regulate the
households in Central America (El Salvador, Guatemala coffee production process. Although there are variations
and Nicaragua) and Mexico, including social (e.g., social between certifiers, most organic standards include the fol-
networks) and economic (e.g., gross coffee revenue) lowing: ‘ecological’ management of farms; soil conser-
variables. This information was used to discuss the current vation practices; abstention from the use of synthetic
and potential benefits and limitations of these certifications fertilizers, pesticides and genetically modified crops; and
as a means to support the livelihoods of coffee-producing intensive on-farm record keeping8,17,18. Organic certifi-
households.
cation also establishes a separate chain of custody ensuring
that certified organic food is separated from conventional
products in any processing stages.
Review of Sustainable Certification Organic agriculture has grown quickly in Latin
Initiatives American, expanding from almost 4.9 million ha of cer-
tified organic land in 1998, to 5.8 million ha by 200419,20.
The rise of specialty coffee and sustainability
Expanded production is associated with increasing demand
certifications
in northern markets, including a 20% growth rate in the
The Specialty Coffee Association of America (SCAA) volume of organic coffee imports to the USA, since 200416.
defines specialty coffee as follows: ‘Sometimes called In 2002/2003, more than 2% of the coffee lands in Central
3. 238 ´
V.E. Mendez et al.
Table 1. Differentiated and certified coffee in relation to total volumes sold worldwide (2003–2004).
Volume
Market segment (metric tons) % of total Source
Conventionally traded coffee 4,659,522 90.70 This table
Estimated exported volume of differentiated coffee 480,000 9.30 Lewin et al.2
Total green coffee exported 5,139,522 100 ICO3
Certified coffee exports organic 26,400 0.51 Ponte7
Fair Trade (of which 14,642 is also organic) 28,283 0.55 TransFair USA22
Shade grown 660 0.01 Ponte7
Rainforest alliance 10,000 0.19 Courville11
Utz Kapeh 14,000 0.27 Courville11
Estimated total certified coffee 65,702 1.28 This table
(13.7% of differentiated coffee is also certified)
Most data refer to 2003. Fair Trade lb, converted to metric tons based on 2.2 lb = 1 kg; 1000 kg = 1 metric ton. Data from TransFair
reported originally by FLO. The organic and shade grown data are from Ponte7 and Fair Trade data are from this table. The estimated 10%
growth rate is taken from Ponte7. While Ponte considered the overlap between total fair trade production and the 52% that is also certified
organic, this summary assumes that the organic total estimated by Ponte excludes the Fair Trade organic coffee. This table is a new
estimate addressing the segments in the global coffee trade as of 2003. According to Raynolds et al.8, Utz Kapeh, Rainforest Alliance and
Bird Friendly certified farms have all have substantially increased their production by 2004; however, these three certifications systems
report the total volumes of coffee produced on certified farms although much of this coffee is still exported as conventional coffee and not
necessarily purchased under these terms (see Table 3 in Raynolds et al.8).
America were certified organic, with higher percentages trades directly with smallholder coffee organizations; signs
in Nicaragua and Guatemala21. Organic coffee does not contracts that encourage long-term planning and stability;
establish a minimum price paid to producers or exporters, and provides up to 60% of pre-harvest credit when re-
but sets a premium above that established by the New York quested26.
Board of Trade. In 2002/2003, a survey of importers re- Although low international coffee prices motivated many
ported that organic coffee price premiums in Guatemala farmers and roasters to seek Fair Trade certification, supply
averaged an estimated $0.40 lb - 1 above the market, and continues to outpace demand, and only 20–25% of Fair
in Nicaragua and El Salvador they were $0.25 lb - 1 and Trade certified coffee receives a premium. Recently, the
$0.30 lb - 1, respectively21. Globally, an estimated 52% of USA, a latecomer to Fair Trade certification, has posted
the coffee produced by certified Fair Trade cooperatives is exponential growth rates7. The USA is now the largest Fair
also organic22. Trade certified coffee market in the world (almost 24,000
metric tons per year in 2006), even though this only
FairTrade amounts to 3% of the entirety of the US market27.
According to TransFair USA, since 1998 74.3 million lb
In contrast to organic, Fair Trade certification specifically
of Fair Trade certified coffee have been sold, and these
focuses on smallholder producer organizations, and the
sales have contributed to a cumulative total of $60.4 million
relations, prices and standards associated with the trade
in price premiums above conventional coffee prices22.
process23. Fair Trade advocates believe that fairer trade
can lead to individual and collective empowerment if it
follows standards and enforcement mechanisms that sup-
port human rights, dignity and sustainable development24.
Bird Friendly, Rainforest Alliance and Utz
‘Empowerment’ can be referred to as the ability of indi-
certified coffees
viduals and/or groups to act on their own in order to Shade coffee certifications seek to reward coffee growers
achieve their self-defined goals25. Fair Trade originated that manage their plantations in a way that supports the
as a solidarity movement that supported partnerships be- conservation of tropical species and habitats, with an
tween northern social justice advocates and impoverished emphasis on maintaining a diversified shade tree canopy
southern producers5,6. After these initiatives saw significant (instead of managing ‘full sun’ plantations that have no
growth in the 1980s, leaders in the movement decided to shade trees)28–31. Shade coffee certifications have struggled
establish standards and certification mechanisms that would to compete with other labels, and continue to account for
allow them to expand the volumes of fairly traded goods. market segments smaller than Fair Trade or organic7,32. The
These efforts were consolidated in the early 21st century, dominant labels for shade certification have been The
when the Fair Trade Labeling Organization (FLO) was Rainforest Alliance (RA) label and Smithsonian Migratory
established5. FLO establishes standards and certifies Bird Center’s Bird Friendlyä coffee. Smithsonian also
producers, organizations and traders that pay minimum requires organic certification, which might result in higher
prices; provides a price premium for social development; ecological standards than RA33–35.
4. Effects of Fair Trade and organic certifications on small-scale coffee farmer households 239
RA’s certification standards include using minimum of small-scale coffee farming households and how these
levels of native shade tree biodiversity, limiting the use of are affected by certified coffee markets. The concept of
highly toxic agro-chemicals, maintaining a healthy working livelihoods, which has been extensively applied in devel-
and living environment for workers and enforcing national opment studies49, can be defined as ‘comprising people,
labor laws36. Critics of RA claim that environmental their capabilities and their means of living (e.g., food,
standards are lower than those required by certified organic income and assets)’50, and how they make this living mean-
coffee and that social development criteria are below the ingful51. Our analysis was guided by a political ecology
social objectives of Fair Trade37. approach, with a focus on rural livelihoods. Political eco-
Utz certified (formerly Utz Kapeh) was launched in 1997 logical perspectives have focused on livelihoods to analyze
by Royal Ahold, a Netherlands-based supermarket com- the social reproduction of farmer communities in devel-
pany. It seeks to integrate improved agricultural practices oping countries, and how these are affected by social and
through the ‘Euro-Retailer Produce Working Group- ecological processes associated with natural resources
Good Agricultural Practices’ (EUREP-GAP) with Social access, governance and management52–54. Although poli-
Accountability International’s (SAI) 8000 social condition tical ecology argues for the need to examine social and
guidelines. EUREP-GAP certifies farms for ‘good agricul- ecological processes across multiple geographic scales
tural practices’38, whereas SAI offers a certification to (e.g., local, national and global)55–58, this paper focused
ensure adequate working conditions for laborers39. The mainly on the household and community levels (local/
initial certification criteria were developed in collaboration regional). However, our research does address several
with a number of large coffee estates in Guatemala36. points of interaction at broader scales, primarily through
According to its representative in Guatemala, interviewed our discussion of second- and third-level cooperative
in 2004, Utz certified saw a tenfold increase in global sales, organizations in different countries, and their international
from 2000 certified tons of coffee in 2002 to 21,000 tons social networks. Our findings also provide insights into the
in 2004. local- and regional-level responses of households and
cooperatives to globally driven events, such as the inter-
Assessing the effects of sustainable coffee national coffee price crisis23.
certification Furthermore, our analysis of household livelihood
strategies and their relationship to the coffee commodity
Several macro-level studies have described sustainable chain connects our research to debates on the governance of
coffee certifications7,8,36 and documented premiums where global value chains59,60. An in-depth examination of coffee
certified producers received between US$0.05 lb - 1 and household livelihood conditions and strategies, as well as
US$0.50 lb - 1 above non-certified growers21. Other scholars their relationship to cooperatives, complements the global
have developed empirically rich ethnographic studies on value chain literature on coffee. With a few notable ex-
how international certification systems are translated into ceptions (see Daviron and Ponte61), this body of work has
the daily lives of small-scale farmer communities and focused primarily on regulations, value-added strategies
cultures40–45. Most of this research agrees that higher prices, and rent appropriations62–64. This paper contributes an in-
training and more connections to international networks are depth assessment of locally situated social and economic
some of the advantages for farmers and cooperatives that impacts resulting from participation in certified or conven-
sell to Fair Trade and organic markets37,41,42,46. Some of the tional coffee value chains. Although a deeper discussion
disadvantages include high costs of certification; additional of our results, in terms of global value chain analysis, is
labor investments and lower yields for organic production; beyond the scope of this paper, we believe that our findings
the need for increasingly sophisticated record keeping; will inform future examinations of livelihood outcomes
higher quality standards; and the purchase and payment of associated with the producer end of the value chain.
certified coffee later in the season6,41,47. Conventional
coffee, which refers to all non-certified forms of coffee
production and trade, is generally sold faster and with far Research design and methods
fewer quality requirements23. Several studies have docu- Field surveys were conducted between November 2004 and
mented the economic (e.g., higher prices and financing) and December 2005, providing data for the 2003/2004 coffee
social (e.g., links to support networks and social investment) harvest. In each country we worked with researcher teams
benefits of Fair Trade to farmer cooperatives9, but only a that were based in-country. Two methodological workshops
few have analyzed these issues at the household level in were held previous to field research to develop standardized
an international comparative study (but see Jaffee41 and data collection protocols. Countries, team members and
Arnould et al.48). cooperatives were chosen based on previous collaborative
history among researchers, a team’s established relationship
with coffee farmers and cooperatives, and a cooperative’s
Research Approach and Methods
willingness to participate.
Of particular importance to our research was to develop Our research design was a stratified survey, with
a better understanding of the current livelihood strategies country and certification type as main strata. We chose
5. 240 ´
V.E. Mendez et al.
Table 2. Number of cases and sample sizes for household surveys per certification type and country.
Number of Mean area
Number household under coffee Mean coffee
1
Country Certification of cases surveys cultivation (ha) yield
Guatemala Non-certified 1 cooperative 30 2.1 703
Fair Trade 1 cooperative 30 1.3 960
Fair Trade and organic 2 cooperatives 60 1.4 808
El Salvador Non-certified 1 cooperative 25 1.1 100
Fair Trade 1 cooperative 30 2.3 766
Organic 1 cooperative 30 3.6 418
Fair Trade and organic 1 cooperative 11 3.8 275
Nicaragua Non-certified 1 cooperative 17 1.8 1760
Fair Trade 2 cooperatives 31 2.7 1111
Organic 1 cooperative 39 5.1 696
Fair Trade and organic 1 cooperative 14 3.8 419
Mexico Non-certified 2 cooperatives 55 2.5 1046
Fair Trade and organic 3 cooperatives 97 3.2 501
Totals for the study: 18 cooperatives 469
1
For collectively managed cooperatives we recorded the average area allocated per household.
representative cooperatives from a limited pool of organi- key informants (such as personnel from organizations
zations that researchers had a relationship with. This working in the community)66–68. Within countries, house-
limited the strength of our sampling design, and increased holds were located in different regions, but had similar farm
the stratification effect, and limited the number of cases sizes (Table 2). In the case of collective cooperatives in El
within each certification type, which can result in a loss of Salvador, we calculated area allocations per household as
precision of the statistical analyses65. A similar design was equivalent to tenured land-holdings. Considering that the
also used by the only other quantitative analyses of Fair sample consisted of small-scale producers and coopera-
Trade coffee certification we are aware of48. We chose to tives, we expected coffee yields and production practices
do this because, due to the sensitive nature of some of the to be relatively similar across countries and across cer-
survey questions (e.g., income and food security), it would tifications.
have been difficult to survey cooperatives that the teams did
not know. For organic and Fair Trade certified cooperatives
we sought producer organizations with members that had
Variable selection, processing and
been producing and selling organic and/or Fair Trade coffee
statistical analysis
for at least 3 years prior to the study. After establishing Variables were selected based on their relevance to house-
collaboration agreements with cooperatives, households hold livelihood strategies (e.g., demographics, sources of
were randomly selected from lists of cooperative members. income, capacities, networks and migration)51 and those
As our main criteria, we attempted to sample a target that could demonstrate tangible benefits from certification
population of between 25 and 30 households per certifi- (e.g., gross coffee revenue, education of children, increased
cation type per country (Table 2). However, this proved food security and linkages to support networks)9,41,69.
difficult in some instances, due to a lack of cooperatives Although we included questions related to the costs of
that met the criteria (e.g., at the time, there was only one certification in our survey, we were unable to collect
Fair Trade/organic certified cooperative in El Salvador). In consistent information on this variable at the household
total, we collected data from 469 households that were part level. Most individual members had limited knowledge or
of 18 cooperatives. confused perceptions, and board members were reluctant
We defined a ‘household’ as a single or extended family to share specific figures.
living in the same residence. Household questionnaires In cooperatives where land is collectively owned
were a combination of close-ended and open-ended (all three certified cooperatives in El Salvador, and the
questions. Household-level data were collected through an organic cooperative in Nicaragua), we divided coffee
identical questionnaire for both certified and non-certified production volumes and land areas under coffee cultivation
households, except for the additional questions that only by the number of cooperative members, in order to obtain
pertained to certified households (e.g., certification type, volume and land area figures for each household. These
premium amount, etc.). We triangulated this information figures were used to calculate gross coffee revenue and
with individual semi-structured and informal interviews, yield per hectare figures for households in collective co-
focus groups with cooperative board members and other operatives. While we recognize that these calculations may
6. Effects of Fair Trade and organic certifications on small-scale coffee farmer households 241
not be perfectly accurate (for example, some households ranges instead of means and standard deviations. Catego-
may spend more time working in the collective coffee rical variables (household savings and credit, food security,
plantation and thus earn more income from coffee), we are and migration) were analyzed using Pearson’s chi-square
confident that they are comparable to the volume, gross test with PO0.05 considered significant. For all variables,
revenue and yield figures for the rest of the households in we compared certifications both for the entire sample and
the study. within each of the countries.
In order to compare prices, we calculated a weighted
average price for each household, because households that
sell certified coffee often sell part of their harvest at the Results
certified price and part of the harvest at the conventional
price. We calculated weighted price as
Cooperatives and households in coffee regions
of Central America and Mexico
(farm gate price for certified organic coffee) *
We found three types of cooperatives in our research. First-
(% of coffee harvest sold at organic price) level cooperatives were made up of individual members
+ (farm gate price for certified Fair Trade coffee) * (individuals and/or families), which organized through
(% of coffee harvest sold at Fair Trade price) different processes, such as agrarian reforms or voluntary
initiatives. Within the first level cooperative category, we
+ (farm gate price for conventional coffee) *
worked with two different types. The first type comprises
(% of coffee harvest sold at conventional price) cooperatives, consisting of individual farmers/families,
+ (farm gate price for certified Fair Trade=organic coffee) * who own their land and form a cooperative to facilitate
(% of coffee harvest sold at Fair Trade=organic price) marketing, technical assistance, credit and/or interactions
with external actors. These farmers continue to manage
This more accurately reflected the prices that farmers their farms individually, but follow cooperative guidelines
obtain at the farm gate. Per household gross revenue from (e.g., organic certification rules). The second type com-
coffee sales was calculated as the volume sold per house- prises collectively managed plantations, where members do
hold multiplied by the price received for that volume. In not have individual ownership of land, but have rights or
some coffee cooperatives, such as the Fair Trade and access to land allocations (i.e., residential and agricultural
organic cooperatives in El Salvador, the coffee plantation plot assignations), as part of their membership. These
is collectively owned. In these cases, gross coffee revenues cooperatives are managed as a collective and run by an
were divided equally among member households for the elected board of directors. The second-level cooperatives
purpose of this analysis. were formed by the union of first-level cooperatives. These
Information was processed into a database created in organizations are also called ‘cooperative unions’ or
Foxpro for Windows. Statistical analyses were run using ‘federations’. The third-level cooperatives may include
SPSS software, versions 16.0 and 17.0. To analyze the both first- and second-level cooperatives as members, tend
differences in the coffee price, volume, gross revenue and to have national representation and are sometimes called
education between certifications, we utilized the Kruskal– ‘confederations’.
Wallis (K–W) non-parametric test. In addition, for coffee On average, the 469 households we surveyed were made
price, volume and gross revenue, we used the Mann– up of six members, and had similar educational levels
Whitney U non-parametric test as a post-hoc test for dif- (O5th grade). Men and women were equally represented
ferences between each pair of certification types, with a in numbers, and no significant differences were observed in
significance level of P < 0.01, due to the large number of educational levels by gender. Subsistence agriculture was
pairings. The K–W test was also used to analyze factors an important livelihood strategy for farmer households. In
that might contribute to food security, including a com- El Salvador, 42% of the surveyed group produced staple
parison of coffee revenue, food production and number of foods (primarily corn and beans), buying less than 40% of
income sources between households that did and did not their annual food budget. In the other three countries, only
experience food shortages. We chose these tests because 10% of households reported producing staple foods, and
the data did not show clear normality or equal variances these families bought an average of 65% of their annual
(Levene and Kolmogorov–Smirnov tests), and our house- food budget.
hold samples were uneven for the certification stratum. The mean area of owned or allocated land for coffee
We were unable to sample for Fair Trade cooperatives cultivation was 3.2 ha per household (N = 469). Total mean
in Mexico, or organic in Guatemala and Mexico, and the land areas owned by or allocated to households per country
number of households sampled for each category were also were 5.7, 3.1, 2.9 and 1.5 ha for Nicaragua, El Salvador,
uneven (see Table 2). The K–W test is a recommended Mexico and Guatemala, respectively. Within countries, the
alternative to parametric analysis of variance (ANOVA) farms selected shared comparable geographical conditions.
for populations containing uneven sample sizes70. The All farms managed shaded plantations of Coffea arabica,
K–W test compares medians instead of means; therefore and were located at elevations greater than 500 m above
for these variables we report medians and interquartile sea level. Characterizations of coffee producers in
7. 242 ´
V.E. Mendez et al.
Mesoamerica have found similar yields and outputs as a $0.70 per pound of green coffee produced71. However,
result of farm size and grower type71. Our sample was made most cost studies have been incomplete, since they fail to
up of small-scale farmers and their cooperatives, which include the costs for coffee maintenance, replanting, tools,
according to Flores et al.71 cultivate less than 3.5 ha of land harvest interest payments, transport and time invested
per farmer and produce an average of 1100 lb ha - 1, when in inspections and training73. Subtracting these estimated
coffee is conventionally managed. Average yield for the minimum farm-level production costs from the gross
entire sample (736 lb ha - 1) was lower than those presented average price per pound in our sample of $0.78 resulted
by Flores et al.71 (Table 2). Organic yields were lower than in an estimated average annual net coffee revenue of
conventional yields in Nicaragua and Mexico, but higher in between $0.28 and $0.08 per lb of coffee. By applying this
Guatemala and El Salvador. Yields are affected by the net value to the average reported volume sold (1450 lb
management capacity and available resources of farmers of coffee per year) by the farmers in our sample, we esti-
and cooperatives. Although data comparisons between mated the net coffee revenue per household per year to
organic and conventional coffee yields are scarce, in most be in the range between $116 and $406. This range is for
cases organic plantations tend to produce lower yields and the entire sample and not accounting for differences in
have similar or higher economic costs72. In both El revenues between certifications. However, it represents an
Salvador and Guatemala, the organic cooperatives were important figure because it shows how low the annual
larger and had well-established organic management revenue range from coffee production is for all of these
programs that were more effective than their conventional farmers, including those that sell all their volumes through
counterparts. In Nicaragua and Mexico, where cooperative certified channels.
management and size were very similar across certifica- Farmers generally do not receive payments for all their
tions, organic yields were lower. In these cases, conven- coffee in one lump sum. Instead, some receive advance
tional farmers were able to increase yields through effective payment as credit, partial payment upon coffee delivery
synthetic fertilizer and pesticide management. and a third final payment. Most receive a single payment
We asked the households we surveyed to list the when they sell to intermediaries. Most payments are both
members of their household who generated monetary insufficient and poorly timed, leaving a thin (or hungry)
income and to list the types of income-generating activities season that generally falls between April and August for
undertaken by household members. Half of the members most coffee regions.
within each household reported generating monetary Only 15% of all households interviewed reported
income. The most frequent income-generating activities having monetary savings, these being most frequent
were coffee sales, coffee-related employment, sales of other in Nicaragua, and least frequent in El Salvador. In
agricultural crops and non-agricultural employment. Inter- Guatemala and Mexico, 15 and 16% of households
viewees were asked to estimate the percentage of their reported having some savings. Forty percent of all
income that they earned from coffee sales, coffee-related households reported having access to some form of credit,
employment or from other income sources, although not with a marked difference between countries. Approxi-
all of them provided complete responses to this query mately 34% of households in the four countries reported at
(n = 300). On average, households reported that coffee least one member migrating, either within their countries,
sales and coffee-related employment accounted for 67% or internationally. Mexico (in 30.6% of the sample) and
of their income. Only 37% reported that they earned all of Nicaragua (28.8%) had the highest emigration rates of the
their income from coffee sales and coffee-related employ- sample.
ment. We chose to group coffee sales with coffee-related
employment for this analysis because some households in
the study are members of collectively owned cooperatives,
whereby they do not receive income directly from the sale
Coffee prices, volumes and gross revenue
of coffee but rather are paid by the cooperative for their Gross coffee revenue is determined by three factors: price,
work in the collective coffee plantation. Sales of other total volume (green bean yields) and the proportion of
agricultural crops contributed 15% of household income volume that is sold through certified channels (and thus
on average, while employment in non-agricultural jobs receives the certified price). We first examined how cer-
contributed an average of 11%. Non-coffee agricultural tified cooperatives differ from non-certified cooperatives
employment and income from other sources (primarily with respect to each of these factors and then look at the
support from older children working away from home) resulting differences in gross revenue.
comprised an average of 5 and 3% of household income, Annual international conventional (non-certified) coffee
respectively. prices averaged $0.72 lb - 1 in 2003/20043. As reported in
On average, households sold 1450 lb of coffee for the similar studies, it proved extremely difficult to calculate
2003/2004 harvests. This generated an estimated mean exactly what price households obtain for their coffee23. The
gross coffee revenue of US$1064 for the season. Previous different ways in which coffee is sold and the taxes
studies have estimated the monetary costs of production and discounts that occur in each country were not fully
for smallholders in this region to range from $0.50 to understood by many of the farmers. To compare the prices
8. Effects of Fair Trade and organic certifications on small-scale coffee farmer households 243
1
Table 3. Price, volume and coffee revenue medians (interquartile ranges) for conventional and certified coffee, as reported by all
interviewed households in Mexico and Central America (N = 461).
Fair
Non-certified Fair Trade Organic Trade/organic
4
(n = 124) (n = 98) (n = 68) (n = 175) K–W test
2
Median calculated price 0.51 (0.24)3a 0.68 (0.32)b 0.77 (0.36)c 0.89 (0.32)c c 2 = 186.98, PO0.001**
(US$ lb - 1)
Median volume 700 (1300)a 1094 (1100)b 3550 (3048)b 600 (900)a c 2 = 51.1, PO0.001**
in pounds
Median coffee revenue 332.56 (800.06)a 1006.48 (635.80)b 2733.50 (2164.01)c 534.00 (818.16)a c 2 = 55.78, PO0.001**
in US$
1
The K–W test uses medians, and the figures in parentheses are interquartile ranges (analogous to standard deviation for the median).
2
The calculated price is the median price a household received for their coffee in 2003–2004, weighted by the volumes sold at those
prices.
3
Medians followed by a different letter are significantly different by the Mann–Whitney U test (PO0.01).
4
* Difference is significant at the PO0.05 level; ** difference is significant at the PO0.001 level.
1
Table 4. Median prices (interquartile ranges) by certification and country (N = 461).
Median Median Median Fair
non-certified Fair Trade Median organic Trade/organic
2
calculated price calculated price calculated price calculated price
5
-1
Country (US$ lb ) (US$ lb - 1) (US$ lb - 1) (US$ lb - 1) K–W test
3
Guatemala 0.48 (0.95) a, N = 30 0.64 (0.65)b, N = 30 — 0.81 (0.11)c, N = 57 c 2 = 95.81, PO0.0001**
4 4 4 4
El Salvador 0.51 (0.00) a, N = 25 0.92 (0.00) b, N = 30 1.13 (0.00) c, N = 30 1.06 (0.00) d, N = 11 c 2 = 70.52, PO0.001**
4 4
Nicaragua 0.38 (0.39)a, N = 17 0.54 (0.20)a, N = 31 0.77 (0.00) b, N = 38 1.07 (0.00) c, N = 14 c 2 = 70.52, PO0.001**
Mexico 0.70 (0.27)a, N = 52 — — 1.11 (0.36)b, N = 93 c = 37.30, PO0.001**
1
The K–W test uses medians, and the figures in parentheses are interquartile ranges (analogous to standard deviation for the median).
2
The calculated price is the median price a household received for their coffee in 2003–2004, weighted by the volumes sold at those
prices.
3
Medians followed by a different letter are significantly different by the Mann–Whitney U test (PO0.01).
4
Coffee is produced and sold cooperatively; thus all households in these cooperatives receive the same median price.
5
* Difference is significant at the PO0.05 level; ** difference is significant at the PO0.001 level.
presented in Table 3, we used information gathered through in calculated price between certification types in all of the
household surveys and data from first- and second-level countries, with households that were members of certified
cooperatives. cooperatives receiving higher prices for their coffee
Calculated prices (the sum of the amount of coffee sold (Table 4). Only in the Nicaraguan case were conventional
at each price, multiplied by the % of total coffee sold at this and Fair Trade prices not significantly different. Because
price) were significantly higher for all certified sales in certification types were only represented by members from
comparison to conventional markets (K–W, c 2 = 186.98, one cooperative in several of the countries, we cannot state
df = 3, PO0.0001), with Fair Trade/organic cooperatives conclusively that certifications were solely responsible for
receiving the highest price (Table 3). Price differences these price differences, as other factors unique to these
between each pair of certifications individually were also cooperatives may have contributed to the differences in
significant except between the organic and Fair Trade/ price. However, because price premiums are generally
organic groups (Table 3). based on certifications, it is likely that certifications are
Significant price differences (for both conventional and driving these price differences.
certified) were also observed between countries (K–W, Volumes of coffee produced and sold were highly vari-
c 2 = 60.3, df = 3, PO0.001), so it is possible that the able between households. Fair Trade and organic certified
comparison between certifications was confounded by the households produced significantly more coffee than non-
uneven representation of cooperative types in the four certified households or households holding both certifi-
countries. Thus, we also compared certified and non- cations (Table 3; K–W, c 2 = 51.10, df = 3, PO0.001). It is
certified cooperatives within each of the countries, through important to note that these are total volumes, not volumes
a Mann–Whitney U test. There were significant differences per area, and they are therefore more reflective of farm
9. 244 ´
V.E. Mendez et al.
c 2 = 13.862, PO0.001**
areas per household than per hectare productivity. The
c 2 = 56.66, PO0.001**
organic households and Fair Trade households in this
c 2 = 8.68, PO0.034*
c 2 = 0.474, PO0.491
study tended to cultivate more land per household than
3
K–W test
the other two certification groups, which accounts for the
higher coffee volumes of the organic households despite
their lower per hectare productivity in Nicaragua and
Mexico.
Gross revenue from coffee was significantly different
between households of the four certification types
(K–W, c 2 = 55.78, df = 3, PO0.0001), with all certification
groups earning higher median revenues than the uncertified
(2170.50)a, N = 14
(795.65)b, N = 57
(471.43)c, N = 11
(899.46), N = 93a
group (Table 3). Gross revenues were also significantly
different between certification groups when analyzed by
Trade/organic
Median Fair
country, with the exception of Mexico where revenues in
revenue
(US$)
The K–W test uses medians, and the figures in parentheses are interquartile ranges (analogous to standard deviation for the median). the Fair Trade/organic households were similar to those in
the non-certified households due to lower coffee production
volumes (Table 5).
534.00
1040.92
1076.57
445.65
As discussed by Bacon23, certified farmers are usually
not able to sell all of their production through certified
markets. Reasons for this include quality standards (only
the best cherries), limited market demand (as in the case
2733.50 (228.62)b, N = 38
of Fair Trade) and cooperative quotas (e.g., distribution of
569.49 (0.00)c, N = 30
premiums among the membership). As a result, many
Median organic
certified coffee farmers are forced to sell a portion of
Medians followed by a different letter are significantly different by the Mann–Whitney U test (PO0.01).
revenue
Table 5. Household coffee revenue medians (interquartile ranges) by country and certification (N = 461).
their production in the conventional market. Net revenue
(US$)
—
—
generated through certified sales can decrease considerably
* Difference is significant at the PO0.05 level; ** difference is significant at the PO0.001 level.
when farmers incur the costs of certification, but are not
able to sell their entire production at premium certified
prices.
Various factors influence the percent of production that
households in certified cooperatives are able to sell at
certified prices. The Salvadoran Fair Trade cooperative
2158.00 (2120.00)ab, N = 31
603.13 (1030.85)b, N = 30
was able to sell all of its production at certified prices
1006.48 (0.00)b, N = 30
because of the small number of Fair Trade certified
Trade revenue
Median Fair
cooperatives in this country that produce high-quality
(US$)
coffee. In the Guatemalan and Nicaraguan cases, Fair Trade
—
1
certified cooperatives were limited by a higher number of
Fair Trade cooperatives and higher volumes of production
by certified farms. Thus, most of these households sold
high proportions of their coffee at conventional prices.
Fair Trade/organic certification showed high proportions
of volumes sold at certified prices in all cases, with the
exception of Mexico (only 54%). In this case, where
(2668.26)a, N = 17
(389.25) a, N = 30
(384.38)a, N = 25
(937.89)a, N = 52
farmers were members of large cooperative unions (up to
Median non-certified
20,000 members), the amount of coffee sold at certified
prices was influenced by coffee quality and the standing
revenue
2
(US$)
of first-level cooperatives within their union. A power
struggle was reported between first-level cooperatives and
their unions, in which their lobbying capacity with the
271.84
102.50
752.80
561.09
union is an important factor determining the proportions of
their total production that they are able to sell at certified
prices.
On average, Fair Trade was the certification with the
El Salvador
lowest proportion of coffee sold at certified prices, with
Guatemala
Nicaragua
Country
60% of total volume sold as Fair Trade. The two certified
Mexico
organic cooperatives sold almost 100% of their certified
coffee at certified prices. Fair Trade/organic certification
1
2
3
10. Effects of Fair Trade and organic certifications on small-scale coffee farmer households 245
% Volume Sold as Certified 100.00 Food security
80.00 Household food security has been used extensively as a
Guatemala measure of welfare by NGOs, governments and the UN.
60.00 El Salvador At the 1996 World Food Summit, the Food and Agriculture
Nicaragua Organization of the United Nations defined food security
40.00 Mexico
as ‘all people, at all times, have physical and economic
Average
access to sufficient safe and nutritious food to meet their
20.00
dietary needs and food preferences for a healthy and active
0.00
life’74. Food security results from many livelihood factors,
Fair Trade Organic (N=68) Org & FT including food production, income to purchase food and
(N=91) (N=177) social safety nets75.
Certification To explore the issue of food security, we asked families
Figure 1. Percentage of total volume of coffee produced by if they experienced times when they felt unable to meet
certified households, which was sold through certified channels, their basic food needs. Sixty-three percent of the house-
for the 2003/2004 harvest. holds interviewed reported that they did struggle to meet
their basic food needs. In our sample, households associ-
ated with certifications did not fare any better in terms of
food security. Our statistical analyses of the entire sample
sold an average of 87% of total coffee produced at certified
showed significant associations between food security
prices (Figure 1).
(using household’s reports of difficulties in meeting basic
food needs as a proxy) and certification, but with a higher
Household savings and credit percentage of households in all certified groups reporting
difficulties meeting food needs (Pearson, c 2 = 22.06, df = 3,
Households in our survey population were asked whether
PO0.001) than the non-certified group. When each country
or not they had any monetary savings. While it is generally
uncommon for small-scale coffee-producing households to was examined separately, only in Guatemala was there a
have savings, the percentage of certified households with significant association between certification and food
savings (17%) was significantly higher than the figure for security (Pearson, c 2 = 21.78, df = 2, PO0.001), again
non-certified households (10%) when all certified house- with more certified households reporting difficulties meet-
holds were combined into a single category (Pearson, ing their food needs.
c 2 = 4.113, df = 1, PO0.043). This difference, however, We also asked households to estimate what percentage
of their food they produced on farm and what percentage
was influenced by the households in the Fair Trade co-
they purchased. There was no significant association
operative in Nicaragua, where 55% of respondents reported
having savings. This is a much higher percentage than any between certification and self-sufficiency in food pro-
of the other certified groups and countries. There were no duction (K–W, c 2 = 0.246, df = 1, PO0.620), with all
significant differences between certified and non-certified certification groups purchasing the majority of their food.
households in any of the other countries. On average, households purchased 61% of their food.
Credit access showed a similar pattern as savings; We recognize, however, that food production is only one
however, the association between credit access and cer- livelihood factor that may contribute to food security. To
tease out some of the reasons why households may or may
tification when all certified households were combined into
not be able to meet their food needs, we also examined the
a single group was not quite statistically significant
(Pearson, c 2 = 3.26, df = 1, PO0.073). Overall, 43% of relationships between reported ability to meet food needs
all certified households reported having access to credit, and gross coffee revenue and number of income sources.
compared to 34% of non-certified households. When There was no significant difference in median gross coffee
certification types were considered separately, however, revenue between those households that did and did not
there was a significant association between certification report difficulty meeting their food needs (K–W, c 2 = 0.16,
type and credit access (Pearson, c 2 = 9.81, df = 3, PO df = 1, PO0.690). Having more sources of income was,
however, associated with being able to meet food needs
0.020). Higher percentages of Fair Trade (42%) and Fair
(K–W, c 2 = 7.97, df = 1, PO0.005). Households that did
Trade/organic (48%) households reported having access
to credit than organic (30%) and non-certified households not experience food shortages reported an average of 2.5
(34%). This is not surprising given that Fair Trade contracts income sources, compared with an average of 2.2 income
often include pre-financing for producers, whereas organic sources for households that did report food shortages.
and conventional coffee buyers do not provide this benefit.
This pattern held when each of the countries was examined
Education
separately, with the exception of Guatemala where nearly
all of the non-certified households reported having access We measured differences in educational levels by the
to credit. percent of school-age children in a household that were
11. 246 ´
V.E. Mendez et al.
reported to be currently in school. Children were defined as (ii) second-level cooperatives; (iii) national organizations
of school age if they were over 6 and under 18 years old. or institutions, including NGOs and government insti-
We focused on the current education status of children tutions; (iv) international organizations or institutions,
because coffee certifications are relatively new, and could including NGOs and donors; (v) religious groups, mostly
not be expected to affect the educational outcomes of Evangelical and Catholic church groups; and (vi) local/
adults48. The median for all certification types was approxi- community groups and committees. Households from Fair
mately 50% of school-age children in school, with no Trade/organic cooperatives most frequently reported associ-
significant difference between certification types (K–W, ation with organizations and networks. These households
c 2 = 1.94, df = 3, PO0.58). In none of the four countries cited second-level cooperatives most frequently (50%),
did the percentage of school-age children per family followed by national (46%) and international organizations
attending school differ significantly between certification (44%). Both conventional and Fair Trade households cited
types and non-certified households. international organizations most frequently (26 and 28%,
respectively), while certified organic cooperatives cited
national organizations most frequently (27%).
Migration
Migration can be viewed as a plausible strategy that Knowledge of certification, accountability
rural households undertake to support their livelihoods by and power dynamics
generating income outside of the region. However, some
authors argue that migration can have negative effects on We found that there was still much confusion among
rural communities, as it eventually leads to a depletion of farmers and cooperative members about what certifications
human capital, which is necessary to reproduce existing are, and cooperative member’s understanding of what it
livelihoods76,77. In the case of Fair Trade/organic coffee, means to ‘be certified’. This has also been documented in
some studies and certification advocates have suggested previous studies9, and was consistent with the data provided
that certification has led to decreased migration from coffee by all of the certifiers and some of the other actors inter-
regions77. Hence, our main research objective related to viewed. In general, farmers understood organic certification
migration was to examine if certified cooperatives were much better than Fair Trade. Organic has clear regulations
experiencing a lower incidence of migration than uncerti- regarding agricultural practices, which farmers found easy
fied farmers. To analyze the effects of certification on to comprehend. Understanding Fair Trade is more challen-
emigration, we compared the percentages of households ging, since its standards are based on notions of justice
from whom at least one family member had emigrated and empowerment23,80. On the other hand, certifications
between certified and non-certified groups. ‘Emigration’ were better understood by board members of first-level
was defined as leaving the home community in order to cooperatives, and fully understood by staff and board
work. There was some association between certification members of second-level cooperatives.
and emigration when certification types were compared
individually, with 37% of non-certified households, 31% of
Fair Trade households, 48% of organic households and Discussion
27% of Fair Trade/organic households having had a
member migrate (Pearson, c 2 = 11.01, df = 3, PO0.12).
Effects of certifications on household livelihoods
However, when all certification types were combined into Most studies on coffee certifications support our results that
one group, there was no significant association between Fair Trade, organic or Fair Trade/organic certified co-
certification and migration (Pearson, c 2 = 0.92, df = 3, operatives obtain higher coffee prices. Examples of this
PO0.34). Neither was there any clear association when research include work in Nicaragua23,80,81, Guatemala9,
the data were examined on a country-by-country basis. It is Mexico9,40,41,43 and El Salvador46. Price differences were
unlikely that certifications were contributing to any also observed between countries, which relate to the qual-
differences in emigration in our sample. ity of the coffee (as measured by cupping), but also to
‘symbolic quality’ associated with the marketing of a
‘country’ brand or image61. Households generally have
Participation in social and support networks
very little control over these quality factors.
Through social networks, which can range from the local to Certifications had a positive effect on the incidence
the international, individuals, households and communities of household savings, especially in the Nicaraguan Fair
can expand their access to resources and other opportu- Trade cooperative. Fair Trade and Fair Trade/organic
nities78. For example, Fair Trade certified cooperatives in certifications improved credit access, although credit was
El Salvador received important donations from organiz- usually in the form of pre-financing for coffee and not in
ations associated with Fair Trade during the 2003 earth- the form of loans to households. This is consistent with
quakes79. In our sample, households reported being other studies reporting that Fair Trade/organic certification
linked with at least the six following types of support improves credit access and financial stability for co-
organizations or networks: (i) first-level cooperatives; operative organizations23,37,41, although not necessarily