BREWING A SUSTAINABLE FUTURE: CERTIFYING KENYA'S SMALLHOLDER TEA FARMERS, 2007-2017 SYNOPSIS In 2007, multinational consumer goods company Unilever launched a partnership with the Kenya Tea Development Agency (KTDA) to help bring Kenya's more than 500,000 small-scale tea farmers up to the certification standard set by the Sustainable Agriculture Network, a global coalition of environmental organizations. To participate, farmers had to fulfill dozens of criteria related to worker safety, environmental management, and agricultural practices. The KTDA, a private company that had been government run until 2000, was able to roll out certification quickly and on an unprecedented scale, thanks to its large market share, its rapport with farmers, the willingness of multinational companies to support high-quality sustainably grown tea, and funding by donor organizations. By mid 2016, all of Kenya's smallholders had met certification standards, and Unilever's flagship Lipton brand was selling 100%-certified tea. Soon after, other major global brands met the same target. Farmers pointed to increased yields, stronger health and safety procedures, and improved livelihoods as benefits of the certification initiative. The British Academy-Department for International Development Anti-Corruption Evidence (ACE) Program funded the development of this case study. Blair Cameron drafted this case study based on interviews conducted in Kenya in January and February 2017 Case published May 2017. INTRODUCTION On any given day in 2007, more than a billion people around the world sipped a morning cup of tea, but few of them appreciated the impact the custom had on workers and the environment in Kenya, the world's biggest tea exporter. They were not alone. "Farmers did not know how some of the things they did actually affected them and their future," recalled Patrick Kiguru, who cultivated a small farm in the heart of Kenya's tea-growing region. At the time, many farmers grew crops and exotic trees on riversides, which reduced water flow, and they used dangerous agrochemicals haphazardly, which was harmful to the health of workers and detrimental to the environment. Kiguru was one of more than 500,000 Kenyans who grew tea as a cash crop on farms smaller than 40 hectares (and often less than 1 hectare).1 Collectively, those smallholders produced about 60% of all the tea grown in Kenya. Most also grew other crops like maize, potatoes, beans, and fruit trees, and they kept farm animals. Best known for its vast plains where tourists flocked to catch glimpses of elephants, leopards, and lions, Kenya also had highlands, where ample rainfall, cool temperatures, and abundant sunlight helped the tea industry thrive. Lush green tea bushes covered many mountain slopes-especially in areas with fertile volcanic soil. Throughout the year, Kiguru, his family, and the casual workers he employed plucked leaves from his 6,700 tea bushes and delivered them for sale through a farmer-managed buying center run by an elected committee. At the center, workers collected and weighed the leaves before transporting them to the nearby Makomboki tea factory, where the leaves were dried, cut into fine pieces, and placed in containers to ferment. After quality grading, the packaged tea was ready for sale. Makomboki was one of 54 factories that made up the Kenya Tea Development Agency (KTDA), which had a long history of working closely with small-scale farmers. Founded as a government marketing board shortly after Kenya gained independence from Britain in 1963, the KTDA became a private company in the early 2000s (see Textbox 1). The organization had a unique structure: Smallholder farmers owned shares in local factories that processed their tea and prepared it for auction in Mombasa, Kenya's major port city. There buyers bid on tea from all 54 KTDA factories as well as dozens of other private factories. Textbox 1: Kenya Tea Development Agency Shortly after independence from Britain in 1963, Kenya’s new government established the Kenya Tea Development Authority to support smallholder tea farmers and market their tea internationally. At the time, there were about 20,000 smallholder farmers in Kenya cultivating about 400 hectares of tea. In June 2000, the government privatized the organization and renamed it the Kenya Tea Development Agency (KTDA). By then, there were more than 500,000 smallholder tea farmers across Kenya. Those farmers all became shareholders in their local KTDA factories, and the 45 KTDA factories that existed at the time became shareholders in the new company. The privatized KTDA also established several subsidiary companies, which provided insurance, electricity, microfinance, and other services for smallholders. The KTDA’s structure helped facilitate the dissemination of information throughout the organization. The company’s 54 factories were divided into 12 zones, and farmers from each zone elected a representative to the KTDA’s national board. The area surrounding the factories also was divided into zones. Each factory had six or seven zones, and farmers in each zone elected one farmer to represent them on the factory board. Each zone was further divided into about 10 collection areas around buying centers, where farmers delivered their tea leaves. Each farmer was a member of a specific buying center, and a committee elected by the members ran each center. As of 2017, the KTDA encompassed about 600,000 smallholders with farms covering about 120,000 hectares. KTDA factories operated 67 processing facilities (although most factories had a single tea-processing facility, several operated satellite facilities). Factories paid farmers for the quantity of tea leaves they delivered to their buying centers as well as an annual bonus that reflected the price their factory’s tea had fetched at auction. Tea from each factory differed in quality and price, and the product was bagged and labeled at each individual factory. Tea packers knew exactly where tea had been sourced. Because the certification took a factory-by-factory approach, packages of KTDA tea that arrived at the port of Mombasa to be auctioned were either 100% certified or 0% certified. As factories sourced only leaves from their own buying centers in their own catchment areas, there was little possibility of certified bags containing tea from noncertified farms. Kenya was the largest tea exporter in the world. (Although India and China produced more tea overall, most of it was for their domestic markets.) At the time, the most popular tea brand in the world was Lipton, which was owned by Unilever, the world's largest producer of consumer goods. The Dutch-British company bought about 12% of all tea produced globally and about 30% of all tea produced by KTDA farmers. Unilever also owned several tea plantations, called estates, in Kenya. In 2006, senior Unilever managers decided on a new marketing strategy for the Lipton brand, which had been around for more than a century. The company had invested in making its own tea estates more environmentally friendly and had partnered with the KTDA to run programs with small-scale farmers. The Lipton team wanted to communicate those initiatives to consumers in a way that would improve the tea brand's image. Unilever staff began looking at certification systems that permitted companies to promote their products as responsibly produced if their suppliers complied with specified standards for farm management and fair treatment of workers. Several nongovernmental organizations (NGOs) and industry associations had published such standards, and the question was which of them to choose.2 After considering the available options, Unilever chose the Rainforest Alliance, a New York-based NGO that aimed to harness market forces to reduce environmental degradation. Although no tea farm had ever become Rainforest Alliance certified, the organization's green frog seal was common on other products like coffee and cocoa. The NGO had gained prominence in the 1990s for its work with United States-based Chiquita to raise environmental standards and working conditions across the company's vast banana plantations in Latin America.3 To earn Rainforest Alliance certification, farms had to adhere to a standard set by the Sustainable Agriculture Network (SAN), a global network of environmental organizations to which the alliance belonged. In May 2007, Unilever announced that it wanted its suppliers to become Rainforest Alliance certified and that by 2015, it would stop buying from tea factories that did not comply with the SAN standard (see Textbox 2). Unilever could instruct managers on its own tea estates to comply with the new standard, but the KTDA faced a far tougher task to ensure that smallholder farmers, too, met SAN requirements. Textbox 2: The SAN standard The SAN’s Sustainable Agriculture Standard had 100 criteria divided into 10 principles. • Social and environmental management system • Ecosystem conservation • Wildlife protection • Water conservation • Fair treatment and good working conditions for workers • Occupational health and safety • Community relations • Integrated crop management • Soil management and conservation • Integrated waste management To earn certification, farms had to comply with 22 critical criteria as well as at least 50% of the criteria under each principle and 80% of all criteria. For the KTDA factories, all smallholder farmers had to meet 50% of criteria under each principle and all critical criteria; and the average compliance rate for noncritical criteria among all farmers at the factory had to be above 80%. In addition, each KTDA factory had to comply with the 16 criteria under SAN’s group certification standard, which was designed for smallholders that applied for certification in groups. To comply with the group standard, factories had to have smallholder training programs in place, a risk assessment system to identify and address criteria that smallholders found difficult to comply with, and an internal management system to organize and monitor all farmers in the group. 1 Sustainable Agriculture Standard. Sustainable Agriculture Network, July 2010; http://www.san.ag/biblioteca/docs/SAN-S-1-4_Sustainable_Agriculture_Standard.pdf. 2 Group Certification Standard. Sustainable Agriculture Network, March 2011; http://www.san.ag/biblioteca/documento.php?id=176 THE CHALLENGE The KTDA was in no position to rebuff Unilever's push for certification. "Unilever was one of our major buyers; it bought over 60% of our tea," said Peter Mbadi, senior manager of agriculture services at the KTDA. "We were also one of Unilever's major suppliers. So they needed us, and we needed them. In terms of sustainability, they needed to have a stable supplier for the tea they wanted." Despite having an eight-year time frame, Mbadi and his team faced a formidable task. They had to persuade the operators of 54 factories and more than 500,000 smallholder farmers to embrace the certification concept, and then they had to lead the difficult implementation of the new standard. Mbadi said he believed that meeting the sustainability standards could help farmers improve the quality, quantity, and consistency of their tea production. But the undertaking would require innovation, resources, and leadership. First, KTDA management had to assess how its factories and farmers had to change in order to meet the goal. Although all 54 KTDA factories complied with the ISO (International Organization for Standardization) standard for quality management systems, SAN standards were more stringent-particularly with regard to environmental impact and worker safety. To pass a certification audit, factories had to have sound waste management systems, and all workers had to wear appropriate protective equipment. Although the KTDA held training events where agency staff taught farmers on techniques related to sustainability, the organization had no idea how many of the farmers followed through on what they'd learned. Further, that past training had focused mainly on how to boost production of tea. SAN standards, however, required that farmers change the ways they managed their entire farms, not just their tea crops. Because most farmers kept animals and grew other crops in addition to tea, the KTDA had to broaden the scope of training programs. Certification also implied investment costs that could potentially discourage participation. Factories had to keep better records, expand health and safety procedures for workers, and pay fees for compliance audits. Farmers had to stop cultivating land that was next to rivers, build new storage facilities for pesticides and fertilizers, and take other measures to fully comply with the standard. Certification could move ahead only if the individual board of each factory endorsed the idea. Each factory's board comprised six or seven members elected by the farmers who brought their tea to the factory's buying centers and shared ownership of the operation. Without clear evidence of the benefits of certification, factory boards might be reluctant to support the program. If factory boards agreed to support the effort, the KTDA still faced the challenge of working directly with the organization's more than 500,000 farmers, most of whom had not completed high school. Effective communication was critical in order to ensure understanding and compliance. Frayed relations between some farmers and the KTDA complicated the task. "At the time, there was a [communication] gap between KTDA management and farmers," said Washington Ndwiga, a staffer at one of the KTDA's factories. "The farmers kept complaining that they had limited access to managers and that KTDA staff were overly demanding on tea quality." Successful implementation also required an audit system to make sure that participating factories and farmers followed through on their commitments. The SAN required two levels of auditors to check and monitor farmers for compliance. First, each factory had to create an internal monitoring system and train its own auditors to check all farms for compliance with the standard. Second, each factory had to hire an independent audit company to conduct an annual inspection of its own facilities as well as a sample of its members' farms. Still, there were some unusual aspects of the way smallholders operated in Kenya that eased the rollout of certification. In many countries, small-scale farmers were poorly organized and scattered across remote areas. In Kenya, smallholders clustered around factories in favorable tea-growing areas to the east and west of the Rift Valley, which cut through Kenya's highlands. Farmers were shareholders in their local factories and were contractually obligated to sell all of their tea to their local buying center. Further, the factories were tightly regulated by the Kenya Tea Board, a government agency that handled licenses for tea factories. The highly organized nature of the Kenyan tea industry also reduced the risk of mixing certified and uncertified products-a major challenge for most supply-chain certification systems. Even if they were willing to sidestep their contracts with factories, farmers had little opportunity to sell their tea on the local market because more than 95% of tea produced in Kenya was exported. Further, packs of tea were separated and clearly labeled before being transported to Mombasa for sale at auction. "The packers know exactly where they source the tea from because the name of the tea factory is on the sack when it reaches the warehouse," said Marc Monsarrat, who led the Rainforest Alliance's training programs in Africa. Mixing became an issue when the tea got packaged later for retail use, however. Most teas sold to consumers were blends of several varieties. It would take years to get all of Unilever's suppliers in Kenya up to certification standards, so in the interim, the company had to carefully track which of the teas it bought were certified and which were not. To ensure the integrity of the supply of certified tea, Unilever had to guarantee that packs sold with the Rainforest Alliance label contained at least 30% certified leaves and the percentage was correctly labeled on the packs. FRAMING A RESPONSE The KTDA had to work out a plan that would get all of its factories on board with certification, and it had to develop a training program that would teach farmers and factory workers how to comply with SAN standards. A pilot project already in the works was the logical starting point. In 2006, Unilever and the KTDA had formed a joint venture-funded by the United Kingdom's Department for International Development-to introduce on KTDA farms Unilever's own sustainability guidelines, which had already been adopted by the company's own estates. Through the project, the KTDA planned to introduce "farmer field schools," starting with a pilot project at four factories. The curriculum would emphasize practical lessons and enable farmers to try out new techniques and strategies, and participants would have a voice in shaping the courses. The project offered significant benefits for both the pilot factories and their smallholder members: free training for farmers that would likely help them improve the quality and quantity of tea they produced. Although Rainforest Alliance certification was not part of the original planning for the field schools initiative, Unilever and the KTDA saw the value of incorporating training on the SAN standard into the field school program. The four KTDA factories chosen for the original pilot-Momul, Mungania, Ngere, and Nyansiongo-were geographically spread across Kenya's tea-growing regions. Beginning in late 2006, plans called for each factory to introduce six field schools, with 30 farmers attending each school. Farmers would attend classes twice a month for two hours and graduate after one year. Mbadi of the KTDA said the idea was that the pilot factories would serve as models. Neighboring factories could learn from the pilots if and when the KTDA wanted to expand the program across all of its factories. The KTDA used its existing staff to begin the pilot project. Each factory had a field services coordinator and up to six assistants who dealt directly with the farmers whose tea the factory processed. Ndwiga, who at the time was coordinator at the Mungania factory, described his job this way: "My work was to train the farmers. . . . It included field preparation, planting, tea husbandry, harvesting, and maintenance of the tea bushes through fertilizing and pruning." Ndwiga stressed that instruction at the field schools entailed much more than book learning. "It was not us telling the farmers what to do," he said. Instead, "we would be letting the farmers learn firsthand through practical experience the benefits of the [new practices] we were discussing. . . . For example, the farmers would have two trial plots alongside each other: The [traditionally farmed] plot and the recommended-method plot. The farmers would harvest and weigh the [tea leaves] from each of those plots and record their findings. That way they would be able to experience and learn [the practices] they were being taught." The pilot project provided a financial incentive to help farmers deal with the added costs of meeting the strict standards of Rainforest Alliance certification. To encourage the boards of the four pilot factories to embrace the program, Unilever agreed to pay a premium of US$0.10 for every kilogram of certified tea bought from the four pilot factories. Because tea sold for about US$2.00 per kilogram at the time, the premium implied a 5% boost for farmers' incomes. "That's why the boards said yes," Ndwiga said. After the schools were launched, farmers were trained, and the pilot factories earned certification, the managers at KTDA and Unilever would face the challenge of expanding the initiative to the other 50 factories. Moreover, Mbadi and other KTDA managers quickly realized they had to develop a complementary approach to speed up implementation. "Farmer field schools were going to be the vehicle to Rainforest Alliance certification," Ndwiga said. "But we soon realized we could not wait for the farmers to undergo the entire training session for one year before aligning with the Rainforest Alliance. We realized we had to do it immediately." KTDA managers decided to train a corps of so-called lead farmers, who would have responsibility for teaching other farmers how to meet the standard's criteria. The lead farmers also would serve as internal auditors who checked farms for compliance with the SAN standard. GETTING DOWN TO WORK By late 2007, Unilever had trained its own staff on the SAN standard and brought its own tea estates in line with Rainforest Alliance certification requirements. Representatives from SAN headquarters in Costa Rica traveled to Kenya to train the employees of a local company, Africert, how to audit factories and farms according to the SAN standard. Lead farmers would be responsible for internal audits to check compliance, and Africert would handle the independent third-party audits required to earn certification. Launching farmer field schools Momul factory, located near Unilever's flagship estate in Kericho, had launched the first farmer field school in late 2006-several months before Unilever's public commitment to buy Rainforest Alliance-certified tea. Momul and the three other pilot factories launched more field schools in 2007, and training on the SAN standard was incorporated into all of them. Ndwiga, the field services coordinator at Mungania, one of the four pilot factories, said he and the factory manager explained the project to the committees of farmers who operated the buying centers that served the factory. "We told them what was involved and explained that we needed to have some buying centers and farmers volunteer for the farmer field schools and [that] soon we would need everyone to start with Rainforest Alliance certification," he said. Admission to the farmer field schools was selective. "First, you had to be willing to train other farmers what you learned," Ndwiga said. "The second criterion was that you had to be a farmer who was actually on the farm. Some farmers had other businesses and traveled all over, but we said that to become a member of the farmer field school, you must be on the farm all the time. . . . We wanted those who could communicate well with other farmers and who were accepted by the community. The other consideration was that we had to have a gender balance: at least 30% of each gender." At Mungania, there were more volunteers than places available. "In consultation with our field staff, we shortlisted the required number of volunteers based on the criteria," Ndwiga said. "It may have been a little subjective, but it was the best thing we could do at that particular time." The 30 selected farmers at each buying center chose a chairman, who was in charge of making sure the farmers knew the time and place of meetings. At the first meeting, field services coordinators and their assistants helped the farmers draw up priorities. Farmers were asked to "define their challenges and rank them," according to Ndwiga. He said each of the 24 sessions during the year included instruction in SAN criteria and followed a basic outline: a recap of what had been taught at the previous class, discussion of a new topic, and a practical application of the learning. Ndwiga said most farmers wanted to learn more about tea husbandry and management techniques. "For example, they wanted to be trained on how best to prune," he said. "You have to prune tea bushes once they grow to a certain level. . . . They wanted to be sure exactly how to do that." After focusing on topics related to tea growing, the field schools branched out into other topics that would help farmers improve their livelihoods. "The farmers wanted to learn about other crops, not just tea. They wanted to learn about milk production from their cows. They also wanted to learn about HIV and AIDS," which were prevalent in Kenya. "Those topics came maybe three, four, five months down the line, after we had exhausted the tea-related topics that the farmers were interested in." Although Ndwiga and his staff had the expertise to teach many of the topics, they invited experts when they could. "I'd been training these farmers for many years, so [it was better] to get somebody new-an expert from a research organization like the Tea Research Institute [a government agency, at the time called the Tea Research Foundation]-to come and talk on the tea-related topics," he said. It did not take long before farmers attending the schools began to see the benefits. "After about six months, they weighed the tea from their trial plots where they'd used the traditional methods and the tea from the plots where they'd used the recommended method," Ndwiga recalled. "They realized that they were getting better tea quality and higher quantities from the plots using the recommended method." Training lead farmers Initially, KTDA managers wanted the field school graduates to become the lead farmers, but because they were in the middle of a 12-month curriculum, the pilot factories decided to recruit new farmers to be trained for the lead farmer role. Factory staff used the same selection criteria as they had for the farmer field school participants, and one lead farmer was recruited for every 300 farmers at the factory. The number of farmers at each factory varied from about 2,000 to more than 25,000. Financial incentives were low. Participants received the equivalent of about US$2.50 a day for the four-day training period, and factories paid the lead farmers about US$0.50 for each farm they inspected. "We tried to pay lead farmers a reasonable rate so that they would find it attractive as a source of income, but we also needed to make sure it was not too expensive, so that the factories could sustain the payments after the project funding finished," said Winnie Mwaniki, who worked on Unilever's pilot project and who became Rainforest Alliance regional manager for eastern and southern Africa in 2009. KTDA staff tried to recruit individuals who were motivated to help their fellow farmers increase yields and become more environmentally friendly. Each of the pilot factories hosted a training course for its lead farmers. "The first two days were theory about the 10 principles" of the SAN standard, which ranged from conservation of soil, water, and wildlife to crop management, worker safety, and community relations, Ndwiga said. "We had pictures showing what was needed, and we simplified the presentation as much as possible. Then we did a practical with them on how to collect data from the farmers." KTDA managers designed a checklist of SAN criteria, and during the four-day session, the lead farmers learned how to check for compliance. The checklist narrowed down and simplified SAN standards, and lead farmers could check all of the criteria by inspecting the property and interviewing the farmer. The final day of training was devoted to pest management. No pesticides were necessary to grow tea in Kenya because no pests or diseases threatened the crop, but many tea farmers used chemicals on other crops, and they rarely had adequate storage facilities or protective equipment to handle them safely. Safe use of agrochemicals was a critical criterion for certification, so lead farmers learned which agrochemicals the SAN standard allowed and which were banned, what the correct procedures for storing fertilizers and pesticides were, and what safety steps farmers had to take when applying agrochemicals on their farms. After the training, factory managers and field service coordinators visited the buying centers and explained to farmers the next steps on the path toward certification. "We introduced the lead farmer, and told them that that lead farmer would assist them with the certification program," Ndwiga said. "Then we showed them the [checklist] and explained that the lead farmer would collect data from each farm as required . . . so all of the farmers were aware of what was required and the need to cooperate with the lead farmers and factory field staff." The lead farmer had to identify which criteria farmers did not comply with and inform them what was required in order to comply. Meeting the standard Each lead farmer had the responsibility to check 300 properties for compliance. During 2008, lead farmers at the pilot factories visited their assigned farms and went through checklists. On larger properties with absentee owners, the lead farmers worked with farm managers. If a farm failed on certain criteria, the lead farmer provided advice and suggestions on what could be done in order to comply. The lead farmers then returned the checklists to the factory, where the field services coordinator and assistants went through them to identify problem areas that might require further training. Although compliance with some of the SAN criteria was mandatory, other aspects of the program were optional. Farmers had to meet at least half of the optional criteria, referred to as noncritical criteria, under each of the standard's 10 principles. For a factory to become certified, its member farmers had to achieve an average compliance rate of at least 80%. Mbadi said the initial checklists showed that KTDA's farmers were in compliance with about 75% of the SAN standard's criteria even before the formal training began. "With the extra trainings, we just had to increase that figure from 75% to over 80%," said Mbadi. "It was not difficult for the farmers to change a few small things." Field services coordinators addressed any major issues-especially criteria that had low levels of farmer compliance-during field days, when they called together farmers from a buying center or from a group of neighboring buying centers. With the increased focus on farmer activities, factory staff began spending more time working directly with farmers to help them meet the standard's requirements. The SAN standard's strong focus on personal safety was novel to many farmers, some of whom stored agrochemicals in their homes, posing a threat to children. Ndwiga said he scheduled field days to teach farmers how to build chemical storage facilities that met SAN requirements. "We would have a demonstration on how to construct them, and show them an example," he said. "We would go to a farmer who had one and congregate there to show how the farmer had constructed the store cheaply." Ndwiga and other field service coordinators used the same approach to assist farmers in understanding and complying with other challenging criteria, such as safe use of pesticides, energy conservation, the use of protective equipment, and waste management. SAN's waste management criteria, although not mandatory, represented another issue that was foreign to many farmers. "Farmers were burning their waste, and the standard said they should not burn," Ndwiga said. "The farmers had to separate biodegradables, recyclables, and solid waste." Factory staff had to explain the reasoning, and progress was usually difficult: "It was like changing a culture." Persuading farmers of the need to protect riparian zones-areas along rivers and streams-was another challenge. Ndwiga said farmers in many areas had planted crops or exotic trees along streams, where the plants absorbed water and reduced flow. "The farmers would plant right to the edge of the river, because, they said, when it gets dry, that is the only place they could grow food, which was true," he said. "But the standard asked them to leave six meters from the stream uncultivated." The protection of riparian zones was a noncritical SAN criterion, and many farmers were reluctant to stop cultivating along waterways. Ndwiga said factory staff and lead farmers encouraged farmers to take small steps toward the six-meter goal. "It was gradual," Ndwiga said. "I told them to start with one foot, and I would be happy. Leave one foot, and then see how the water flows; see how the water becomes cleaner and cleaner every day as more farmers leave the river banks uncultivated and instead protect them [from erosion]." In many cases, protecting stream banks involved cutting down eucalyptus trees that farmers had planted to produce firewood, because the nonnative trees consumed significant amounts of water. Lead farmers encouraged farmers to stop planting exotic trees in riparian areas and instead plant Napier grass and indigenous trees, which used less water and helped prevent erosion of riverbanks. They also encouraged farmers to plant trees elsewhere on their property. Factories generally had to make few changes to fulfill the certification criteria. Some had to pay workers back wages for overtime hours (required for certification), and others had to provide more-protective equipment. Alfrick Sang, the field services coordinator at Momul factory, said that having ISO certification helped prepare his factory to pass the SAN audit. The facility already had strong management systems in place, he said, and kept good records about what went on in the factory, as required by both ISO and SAN standards. Factory managers hired Africert to conduct independent certification audits after the field services coordinators were confident that the farmers had met compliance requirements. Third-party audits usually took four to six days depending on factory size. Auditors checked payrolls and other factory records, inspected the factory floor, and interviewed factory staff. Then auditors visited a sample of farmers and interviewed them. Momul, the first factory to be audited, failed on one critical criterion: None of the farmers used protective equipment when they applied pesticides and other chemicals. Auditors found the same problem at the three other pilot factories. "We had to go back to the drawing board as far as personal protective equipment was concerned," said Ndwiga. "It was a bottleneck for the project. All of the pilot factories were audited, and all of them were told they could not pass." Many farmers did not understand why protective equipment was necessary, and most were unwilling to pay for the masks, gloves, and boots required. KTDA senior staff in Nairobi had to develop a solution. Eventually, KTDA managers decided to buy the equipment in bulk and provide a set to each farmer on credit. A few months after the initial audit, Momul distributed protective equipment to the farmers and explained that the cost would be deducted from their pay over time. After the auditors returned and verified that the farmers had the protective equipment, Momul became the first KTDA factory to earn Rainforest Alliance certification-in mid 2009. Ngere, Nyansiongo, and Mungania followed soon after. Expanding farmer field schools and certification The field schools were popular with both farmers and factory managers. Sang said that word spread to farmers at neighboring buying centers and even neighboring factories about how newly trained farmers enjoyed increased yields of tea leaves. To meet growing demand, the KTDA began to roll out field schools across more buying centers at the four already-certified factories, and it also introduced the schools to the 50 remaining KTDA factories. Further, all farmers at the pilot factories-not just those involved in the field schools-enjoyed the financial benefits of Rainforest Alliance certification because of the premium that Unilever paid on each kilogram of tea purchased from certified factories "It was good money, and when the premium was distributed to all the farmers, they were happy," Ndwiga said. Because they competed for buyers at the tea auction in Mombasa, KTDA factories worked to improve the quality of the leaves their farmers grew. Those that developed a reputation for high quality and flavor could command higher prices. Ndwiga said news spread "like bushfire" about the financial advantages of certification, including the roughly 5% Unilever premium that certified factories received on top of the tea's sale price. Attention sharpened further as other major companies started to prioritize purchasing from factories that had been certified. "Mungania had never beaten the neighboring factory in profits," Ndwiga said. But after getting certified, "Mungania beat the neighboring factory for the first time ever, which created a lot of excitement." The remaining factories soon got in line to join the project and earn certification. "No factory wanted to miss out on selling to Unilever," said Mbadi. Although funding from the UK Department for International Development ran out at the end of 2008, a new donor-IDH, the Sustainable Trade Initiative, a Dutch organization-agreed to finance the next phase of the project. Unilever contributed further funding, and other multinational tea companies quickly made commitments, too. "There was a flurry of meetings in late 2008 and early 2009," said Monsarrat of the Rainforest Alliance. "Very quickly, we received interest and commitments from Tetley, Taylors of Harrogate, and Twinings"-three of the biggest tea buyers in the world. Soon after, Tesco and Typhoon, two more major tea buyers, also committed to purchasing Rainforest Alliance-certified tea. Funding was channeled through the Rainforest Alliance and other environmental NGOs operating in Kenya. Those NGOs hired specialist staff to train factory workers and lead farmers at KTDA factories and at other private factories across Kenya. IDH matched any funding contributed by the multinational companies involved and later began directly funding the expansion of farmer field schools through the KTDA. Certifying new factories By 2010, with the pilot project successful and more factories eager to earn certification, the KTDA was ready to scale up quickly. Mbadi put a specialized team together to lead the push to get the KTDA's remaining 50 factories certified. Sang, the field services coordinator at Momul, the first factory to be certified, was in charge of getting all factories west of the Rift Valley certified, and Jared Kebaya, another KTDA manager, was in charge of factories to the east. For the next stage in the project, Ndwiga left his position at the KTDA and joined Africa Now (which later changed its name to Partner Africa), an NGO that the Rainforest Alliance contracted to help new factories meet SAN criteria. Ndwiga worked with Kebaya on factories east of the Rift Valley; his counterpart working with Sang in the west was Mark Omondi. The first step for the team in charge of the scale-up was to explain the standard and the certification process to KTDA staff at each factory. Mbadi began visiting factories with Sang, Kebaya, Ndwiga, Omondi, and Mwaniki from the Rainforest Alliance. After the initial visit at each factory, Ndwiga and Omondi conducted research on the factory floor. "There are different sections for cutting, fermenting, sorting, grading, and packing," said Ndwiga. "We would go to all of them and check everything against the principles of the standard." Ndwiga and Omondi checked that staff used protective equipment, that workers were treated fairly and paid overtime, and that all factory staff followed safety procedures at each stage in the tea production process. After observing factory workers go about their day-to-day jobs and interviewing some of them, Omondi and Ndwiga issued a report outlining the SAN criteria the factories already complied with and what they still had to do in order to pass a certification audit. Sometimes factory managers pushed back. "At first, they were on the defense," said Ndwiga. "But I alerted them that the auditors would come and point out the same nonconformities. Also, KTDA headquarters would find out: Peter Mbadi would see my reports and follow up with the factory management." The project staff then visited a sample of farms to determine what farmers had to do in order to pass a certification audit. In general, the issues were similar at every farm and at each factory. After Ndwiga and Omondi completed their reports, Mbadi and Mwaniki met with the board of each factory to explain each aspect of certification and seek approval to move forward. Using the report, they could explain to board members exactly what had to be done at both the factory and the farms to meet SAN requirements. At those meetings, the project staff would explain how certification had benefited the four pilot factories. In some cases, the Rainforest Alliance link had opened up opportunities. "In my presentation, I would use the Nyansiongo Tea Factory as an example," Ndwiga said. "A buyer from the US came to buy tea from the Nyansiongo factory after seeing it on the Rainforest Alliance website. The buyer wanted the tea to come in banana leaves, packed by the farmers themselves. So the farmers provided banana leaves as the packaging for the tea, and the buyer paid them a premium." When a factory board agreed to pursue certification, the rollout followed much the same procedure as it had at the pilot factories. In many cases, field school graduates went on to become lead farmers responsible for training other farmers and verifying compliance with the standard. To overcome the protective-equipment issue, which came up at every factory, Mbadi told field service coordinators to try a new approach: Instead of buying protective equipment for each farmer, the factories would purchase a few sets of protective equipment and train a few farmers at each buying center on how to safely use agrochemicals. Every time farmers had to apply pesticides or fertilizer, they could hire one of the designated sprayers for a small fee. The new system worked well, and more factories earned Rainforest Alliance certification. Ensuring continued compliance Over time, as more factories set up farmer field schools and more graduates got trained as lead farmers, the number of lead farmers increased and their workloads eased. Kiguru, who became a lead farmer at Makomboki factory after graduating from a field school and attending the four-day training session, was responsible for 100 farmers. (KTDA management had reduced the number from the original 300). Each year, Kiguru dedicated about 10 days to visiting his 100 farmers and running through the SAN checklist with them. If any of them failed on any criteria, he helped them work out a plan for compliance and would check on them whenever he passed by their property. Kiguru said that factories chose lead farmers who were dedicated to sustainability, and it was their own commitment that drove them to ensure compliance with the SAN criteria. "They are very careful about selecting lead farmers and take only those with a passion for the job," he said. "I sometimes go back and visit a farmer three times to see if he is complying and to assist him in doing whatever he finds difficult." If Kiguru identified issues he couldn't resolve, he could notify the field services coordinator and the coordinator's assistants. "The lead farmers are not working alone; they are working with the field staff at the factory level," said Ndwiga. "The manager is continually monitoring what is happening by looking at the reports. If he feels there are issues not being well addressed, he calls the lead farmers again and addresses them." Each year, auditors from Africert or RA-Cert (the Rainforest Alliance's offshoot certification company, which opened an office in Kenya in 2012) returned to conduct an independent audit of each factory. Each annual audit cost about US$3,000 depending on the size of the factory and the number of farmers in the factory's catchment area. The factory and the audit firm organized a mutually convenient week for the audit to take place, and a team of auditors traveled to the factory to check it for compliance and to visit a sample group of farmers. Factory staff provided a list of all farmers, and auditors could visit any of them. Following a common audit procedure recommended by the SAN, the number of farms visited was the square root of the total: if a factory had 10,000 farmers, the auditor would visit 100. Overseeing certified tea sales The Rainforest Alliance required Unilever to provide quarterly reports on the tea it sold using the organization's green frog seal. If less than 100% of the leaves inside the package were certified, the company had to label the package with the percentage that was certified; the minimum percentage required to use the seal was 30%. For every tea product produced, Unilever kept a blend sheet detailing the tea leaves used and when those leaves had been purchased. Initially, Unilever sold Lipton tea as a 50%-certified blend and had to carefully track which teas went into Lipton products. Rainforest Alliance staffers kept a list of certified factories and the dates the factories had earned their certifications, and they checked Unilever's reports against that list to make sure at least half of the tea that went into Lipton packets had come from certified sources. Later, when more factories became certified, Unilever bought exclusively Rainforest Alliance certified teas. At that point, tea blenders no longer had to track the percentage of certified teas in each blend, because all of the leaves they used were certified. Unilever paid a royalty fee of about US$12 to the Rainforest Alliance for each metric ton of tea sold with the green frog seal. Although the company sourced all of its tea from certified factories, only a small percentage of tea was sold under the label. Unilever staff in individual countries or regions decided to market tea products using the Rainforest Alliance seal only when they perceived they would gain an advantage by doing so. Unilever did not sell tea with the Rainforest Alliance seal in the Middle East, Africa, and Russia, where there was little consumer demand for certified products. "In Western markets [like the United Kingdom and the United States], we see the benefit of putting the seal on the pack," said Erwin Vroom, Unilever's tea procurement manager. "But it is a marketing choice. . . . In Indonesia, for example, we found that consumers didn't want to buy products that had a frog on them." In a separately negotiated deal with the KTDA, the company also paid a certification fee of US$0.10 per kilogram to factories from which it bought tea. Unilever originally paid a premium for all certified tea it bought from KTDA factories, but after 2010, it paid the premium only for tea it later labeled with the Rainforest Alliance seal. Mbadi said that about 3% of all KTDA tea went on to be sold with the seal and that the small amount of money the factories earned from the certification fees helped pay annual audit costs. OVERCOMING OBSTACLES Once farmers were complying with all of the SAN standards' critical criteria and had reached 80% compliance with noncritical criteria, there was less incentive and less urgency to continue fulfilling the standards' remaining levels of compliance. The system provided neither rewards for following through nor penalties for failing to do so. "We find through the audits that certain criteria are never complied with," Mbadi observed. "And because they are noncritical, it keeps repeating from one year to the next." Not surprisingly, the criteria that farmers most often struggled to meet were those that required the greatest investment, such as the construction of storage facilities and the planting of indigenous trees. Many farmers also resisted standards that required changing entrenched practices, such as switching from burning waste to separating and recycling it. In 2016, SAN released a new standard designed to address the problem. Scheduled to take effect in July 2017, the new standard eased the initial requirements for certification but stipulated compliance with more criteria over time. It divided all noncritical criteria into three categories. Crucially, it also set deadlines to make the improvements, and farms and factories risked losing their certification if they failed to meet the requirements. To become certified initially, farmers had to meet all critical criteria and 50% of the criteria in the first noncritical category. By the third year, farmers had to be in compliance with 90% of the first category of noncritical criteria and 50% of the second. After six years, farmers had to be in compliance with 90% of the first and second categories and 50% of the third, which included more-difficult and more-complex requirements such as monitoring waste management service providers to ensure that waste disposal did not harm the environment or neighboring communities. A second problem with the certification initiative involved the quality of audits. Because audit visits were scheduled in advance and took place only once a year, factories could demonstrate compliance during the brief audit period and then revert to old practices when the auditors left. Some criteria were particularly hard to police, such as the fair treatment of workers. A 2011 investigation by Dutch NGO SOMO (the Dutch acronym for Centre for Research on Multinational Corporations) into the treatment of workers on Unilever's Kericho estate found evidence of sexual harassment and coercion on the estate.4 Such issues, although covered by the SAN standard, were difficult to detect during periodic audits. Workers might be unwilling to disclose such information during interviews, or they might be under pressure from managers or farmers to withhold comments that could threaten a factory's certification. In response to the Kericho investigation, Rainforest Alliance auditors conducted a surprise visit to the estate but were unable to find specific cases of harassment. The alliance's Monsarrat said there were always risks of discrimination and harassment in factories and on farms that employed thousands of people. He said that although there is a possibility that audits might fail to pick up on those issues, Rainforest Alliance-certified factories received closer attention from watchdog NGOs and journalists and that such investigations sometimes resulted in improved practices. In 2013 and 2014, Unilever introduced new procedures for reporting complaints about managers and employed new staff members tasked with ensuring the welfare of workers on its Kericho tea estate. The company said the new staff and policies resulted in "a substantial improvement in the way sexual harassment issues are addressed."5 ASSESSING RESULTS By mid-2016, all KTDA factories in Kenya had earned Rainforest Alliance certification, and all remained certified in early 2017 despite changing financial incentives. Unilever quit paying the price premium it had paid to the four factories during the initial pilot project, but Mbadi said there was no backlash because all of the factories were aware it had been a short-term means to get the long-term program up and running. Although Kenya's two largest tea export markets, Egypt and Pakistan, did not demand certified tea, Kenyan tea factories still had good reasons to earn certification. At the Mombasa auction, many of the major buyers would bid only on certified tea. As a result, noncertified tea often sold for lower prices than certified tea-irrespective of quality. Almost all of Kenya's more than 100 tea factories chose to follow the KTDA's lead and became certified. "Now, you find that almost all the factories in Kenya are certified-apart from one or two," Mbadi said. "So [almost] all the tea sold in Mombasa is Rainforest Alliance certified. So now you cannot say a certain tea is fetching more because it is Rainforest Alliance certified. I think it is the quality that determines the price differential. But what Rainforest Alliance certification does mean is that your tea will not be disadvantaged. Not having certification would be a disadvantage because the bidding competition would not be there, so the prices would be lower. Buyers would say, 'No, I'm not bidding on that tea because it is not Rainforest Alliance certified.'" Although as of 2017 no studies had fully assessed the environmental impact of certification in Kenya's tea-growing regions, anecdotal evidence suggested the effect was positive and growing. Monsarrat of the Rainforest Alliance said that many farmers now worked harder to conserve forest areas and protect waterways, and as a result, "streams that were previously dry have begun flowing again." Some farmers said they have seen long-term benefits to the program. "Now, farmers do the right thing and think about their future," said Kiguru, lead farmer at the Makomboki factory. "For example, they used to think that farming along the river would be more profitable, but today they know that that would threaten their future, so they don't do it." Although many of the farmers Kiguru worked with had taken steps to protect riparian areas, other farmers lagged behind. Still, compliance rates for most criteria were improving slowly over time, according to Mbadi. Some unsustainable practices were so ingrained, however, that it was near impossible to change norms-especially with no enforcement mechanism. Although lead farmers often identified farmers who failed to meet certain criteria and later followed up, there was no way to compel farmers to comply with specific noncritical criteria. That was set to change after July 2017, when farmers would risk losing their certification under the new SAN standards if they failed to comply with a greater percentage of noncritical criteria every three years. The KTDA continued to expand the farmer field schools, and by early 2017, more than 3,400 had been set up. More than 90,000 farmers, representing about one-sixth of all KTDA farmers, had either graduated from the schools or were attending them. From 2013 to early 2017, the IDH directly funded the KTDA's creation of more field schools. Although the KTDA had to pay for any additional schools, Mbadi said, the organization planned to embed the concept into its operations. "Our assumption was that each of the graduates could pass on their knowledge to their neighbors," Mbadi said. "But we can't be sure that it will happen, so we are planning to continue training more farmers through the field schools. . . . Those who have never attended school will be allowed to participate. It will be like an education system where every year, new people are brought in and trained." An impact evaluation on the first factories to launch farmer field schools found that more than 90% of participants said they were satisfied or very satisfied with the schools, and 96% said they had benefited from attending. Advantages included new farming skills, an understanding of sustainable energy, and knowledge of soil conservation. The evaluation-conducted by Wageningen University in the Netherlands and commissioned by the KTDA, IDH, and Unilever-also found that farmers who had graduated from field schools achieved higher tea yields than farmers who had not attended, and they experienced greater increases in income during the study period.6,7 The increase in yields was "a combination of certification and farmer field schools," Mbadi said. "The field schools train you how to implement, but it is the certification that keeps you on your toes because you're audited every year. So it is like you are in school: you know every time there is a test, you have to work hard." REFLECTIONS In many countries, smallholder certification initiatives failed to gain traction at the speed and scale that Rainforest Alliance tea certification did in Kenya. Certain important preexisting conditions helped facilitate the rapid expansion of tea certification in Kenya. Because more than 95% of tea produced in Kenya was exported, producers were tied closely to global rather than domestic demand. In countries and industries in which domestic demand was strong, producers could choose to serve the local market rather than bear the cost of responding to shifting international standards. Kenya's tea industry also was highly concentrated both geographically and structurally. The Kenya Tea Development Agency (KTDA) accounted for 60% of the market, and as a result, it was relatively easy to win agreement on standards and to coordinate large numbers of farmers. In addition, the KTDA's structure was highly integrated, with strong relationships between farmers and factories. This organization facilitated the effective rollout of certification. "It was easy because of our structure," said Peter Mbadi, senior manager of agriculture services at the KTDA. "Each factory has its catchment zones, and in each catchment there are buying centers. . . . At the buying center there are committee members who are elected. So all you need to do is call the committee members and tell them we are having a meeting. Then we can discuss with them and tell them what is required, and they will go ahead and implement it. The buying centers are very well organized." Kenya also regulated the tea industry more strongly than did the governments of many other developing countries. Each KTDA factory had to be registered and licensed by the Tea Directorate, which replaced the Kenya Tea Board in a 2013 government reshuffle. In addition, to meet licensing requirements, tea factories had to be ISO certified for quality management systems. Kenyan law barred unlicensed factories from selling domestically or internationally, and unlicensed factories were excluded from the Mombasa auction, where about 80% of Kenyan tea was sold. Other countries, such as Tanzania, Rwanda, and Malawi, launched similar certification initiatives with NGOs, tea companies, and the Rainforest Alliance. But smallholder certification was far more challenging outside Kenya. "Smallholders in the rest of the world are not organized like they are in the KTDA," said Erwin Vroom, Unilever's tea procurement manager. "In the rest of the world, most smallholders sell to collectors, who then onsell the tea to factories. One day they might sell to one collector, and one day they might sell to a different collector." In such a system, enlisting factories to help smallholders comply with changing standards was far more difficult, as was monitoring the supply chain. Finally, the driving force behind the initiative was Unilever's follow-through on its 2007 commitment to buy only Rainforest Alliance-certified tea. The company had such a large market share in Kenya that it was able to shift the whole industry. "We were one of the largest buyers in Kenya, which made it easier to convince suppliers to move toward certification," said Vroom. "We had enough buying power to make that happen." References 1 "Inclusive Business Case Study: Kenya Tea Development Agency Ltd." International Finance Corporation. June 2014; http://www.ifc.org/wps/wcm/connect/5fa58180445b4102a1dbadc66d9c728b/KTDA.pdf?MOD=AJPERES. 2 Aileen Ionescu-Somers and Ralf W. Seifert, "Leapfrogging to Mainstream; Unilever Makes Lipton Tea Sustainable." Supply Chain Forum: An International Journal, 15:1(2014), 68-75; http://www.tandfonline.com/doi/pdf/10.1080/16258312.2014.11517334?needAccess=true. 3 "Chiquita Top Banana in Rainforest Alliance Project." GreenBiz, November 16, 2000; https://www.greenbiz.com/news/2000/11/16/chiquita-top-banana-rainforest-alliance-project. 4 Sanne Van der Wal, Certified Unilever Tea: Small Cup, Big Difference? Centre for Research on Multinational Corporations, October 2011; https://www.somo.nl/wp-content/uploads/2011/10/Certified-Unilever-Tea.pdf. 5 "Kericho Tea Estates." Unilever: Sustainable Living, April 2015; https://www.unilever.com/sustainable-living/what-matters-to-you/kericho-tea-estates.html. 6 Yuca Waarts, Lan Ge, and Linda Puister, For All the Tea in Kenya: Impact Assessment and Baseline Situation of Farmer Field Schools. Wageningen University and Research, 2014; http://edepot.wur.nl/310209. 7 Yuca Waarts, Lan Ge, Giel Ton, and Don Jansen, Sustainable Tea Production in Kenya: Impact Assessment of Rainforest Alliance and Farmer Field School Training. Wageningen University and Research, June 2012; http://edepot.wur.nl/214044 Terms of Use In downloading or otherwise employing this information, users indicate that: a. They understand that the materials downloaded from the website are protected under United States Copyright Law (Title 17, United States Code). This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/. b. They will use the material only for educational, scholarly, and other noncommercial purposes. c. They will not sell, transfer, assign, license, lease, or otherwise convey any portion of this information to any third party. Republication or display on a third party's website requires the express written permission of the Princeton University Innovations for Successful Societies program or the Princeton University Library. d. They understand that the quotes used in the case study reflect the interviewees' personal points of view. Although all efforts have been made to ensure the accuracy of the information collected, Princeton University does not warrant the accuracy, completeness, timeliness, or other characteristics of any material available online. e. They acknowledge that the content and/or format of the archive and the site may be revised, updated or otherwise modified from time to time. f. They accept that access to and use of the archive are at their own risk. They shall not hold Princeton University liable for any loss or damages resulting from the use of information in the archive. Princeton University assumes no liability for any errors or omissions with respect to the functioning of the archive. g. In all publications, presentations or other communications that incorporate or otherwise rely on information from this archive, they will acknowledge that such information was obtained through the Innovations for Successful Societies website. Our status (and that of any identified contributors) as the authors of material must always be acknowledged and a full credit given as follows: Author(s) or Editor(s) if listed, Full title, Year of publication, Innovations for Successful Societies, Princeton University, http://successfulsocieties.princeton.edu/ ISS is program of the Woodrow Wilson School of Public and International Affairs: successfulsocieties.princeton.edu. ISS invites readers to share feedback and information on how these cases are being used: iss@princeton.edu. (c) 2017, Trustees of Princeton University. This case study is made available under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. (c) 2017, Trustees of Princeton University Innovations for Successful Societies makes its case studies and other publications available to all at no cost, under the guidelines of the Terms of Use listed below. The ISS Web repository is intended to serve as an idea bank, enabling practitioners and scholars to evaluate the pros and cons of different reform strategies and weigh the effects of context. ISS welcomes readers' feedback, including suggestions of additional topics and questions to be considered, corrections, and how case studies are being used: iss@princeton.edu. GLOBAL CHALLENGES: Certification Blair Cameron (c) 2017, Trustees of Princeton University Terms of use and citation format appear at the end of this document and at successfulsocieties.princeton.edu/about/terms-conditions.