Caught between theft and rapidly rising prices
Smallholder vanilla farmers transitioning into contract farming schemes of certified exporters in Northeast Madagascar during the vanilla market boom of 2016-2018
by Lloyd Johannes Dr. Blum
Date of Examination:2021-10-07
Date of issue:2023-09-08
Advisor:Prof. Dr. Jan Barkmann
Referee:Prof. Dr. Achim Spiller
Referee:Prof. Dr. Theo Rauch
Persistent Address:
https://dx.doi.org/10.53846/goediss-10061
Files in this item
Name:2023_08_29_Ph.D.Dissertation_LloydBlum.pdf
Size:54.0Mb
Format:PDF
The following license files are associated with this item:
Abstract
English
This Ph.D. Dissertation explores smallholder preferences for contract farming options in the trade of vanilla from Madagascar. It seeks to identify possibilities to increase the mutual benefit of contract farming for both smallholders and buyers of vanilla in the context of the remarkable price volatility that characterizes the vanilla market. The SAVA Region in Northeastern Madagascar is the most important origin of vanilla globally. It is inhabited by an estimated 80,000 vanilla farmers who produce roughly 10,000 tons of unprocessed (green) vanilla each year. Ignoring fluctuations in supply, this production results in around 2,000 tons of annual exports of processed (black) vanilla. The SAVA Region thus contributes at least 40% of annual global vanilla production, next to supplying the world with the most popular of all vanilla varieties known as "Bourbon Vanilla" (V. planifolia). Vanilla production in the SAVA Region is achieved almost exclusively through family farming. Smallholders typically cultivate areas of less than one hectare per household. Plantation production of vanilla is rather uncommon, by contrast. Contract farming (CF) of vanilla, however, has gained traction among international traders since the mid-1990s when the vanilla market was liberalized; CF has picked up over the last 15 years in particular. Contractual buyers are mostly export companies or professional vanilla processors (so-called "preparators” of vanilla) who transform fresh (green) vanilla into dried (black) vanilla before trading it on the international market. This piece of research found that today, nearly 20% of smallholder households produce vanilla under contract in the SAVA Region. Quantitative studies which explore the perceived costs and benefits of CF from the smallholder perspective are still lacking for the vanilla business, however. The knowledge gap is to be closed by this book. Vanilla is subject to remarkable price volatility. Price fluctuations result from the interplay of global demand and supply. Supply shortages in crucial producer countries, natural disasters, and speculative trade all influence fluctuations in the price of vanilla. Speculative trade is a phenomenon that characterizes spot market transactions in Madagascar whenever vanilla prices are on the rise. Increasing prices tend to result in an influx of informal traders in Northeastern Madagascar as well as in a concurrent uptick in the incidents of vanilla theft. As a result of vanilla theft, smallholders, in turn, harvest their crops prematurely, causing problems of declining vanilla quality whenever prices boom. In this context of market boom, it proves difficult for international buyers to access good quality (ripe) vanilla from Madagascar. Traders extend CF offers to vanilla farmers to incentivize quality production. Besides product quality considerations, vanilla exporters and preparators also pursue certified CF schemes to implement private voluntary sustainability standards (such as Organic, Fair Trade, and/or Rainforest Alliance) that are in high demand by international clients. Respecting such process standards allows exporters and preparators to differentiate into higher-value niche markets. However, the extent to which smallholders benefit from certified CF is a topic of intense debate. The production of quality (ripe) crops and the transition toward more sustainable processes of production (i.e., pesticide- and contaminant-free, child-labor-free, and rainforest-conserving, etc.) not only brings benefits but can also imply major economic costs to smallholders. The production of mature vanilla, for example, is not self-evident against the backdrop of intensifying vanilla theft and associated violence, as these phenomena pose risks in terms of harvest failure and personal safety. Likewise, production restrictions associated with certified crops can be perceived as limiting the entrepreneurial freedom of farmers. Exploring smallholder preferences for CF is regarded as highly relevant since research on the topic can potentially improve the design of CF offers and resolve smallholder acceptance issues. Economic experiments provide a means to quantify the perceived costs and benefits of CF from the smallholder perspective. This study is based on a Choice Experiment (CE) that was embedded in a mixed methods approach. The mixed methods that were deployed included exploratory participatory research, a representative cross-sectional survey characterizing vanilla farming households and their marketing strategies at the level of the SAVA Region (comprising N=1,291 HHs from 60 villages). It also included the CE which captured smallholder preferences for CF options (comprising N=604 HHs from 14 villages), as well as pre- and post-experimental qualitative interviews with contract farmers, vanilla traders, and development workers to validate the findings elicited by the CE. The two quantitative household surveys (i.e., the baseline study with N=1,291 randomly selected HHs, and the CE with N=604 randomly selected stratified HHs) allowed us to validate stated preferences for CF options (i.e., the stated preferences elicited by smallholders during the choice experiment) with reported marketing behaviors of the farmers (i.e., revealed preferences reported from market interactions in the real world), thereby reducing the possibility of hypothetical bias of presented results. Reported marketing behaviors underlined the logic of action of smallholder vanilla farmers in their day-to-day realities. The smallholder logic expressed by the CE results was remarkably consistent with behavioral choices that farmers reported based on their real-life market actions. Moreover, the inclusion of qualitative interviews as part of the mixed methods research approach enables us to understand the quantitative data in light of typical challenges faced by farmers and buyers who implement CF schemes. The academic literature on CF highlights a high degree of context specificity regarding smallholder preferences for CF options (i.e., for different CF models, for certified vs. non-certified contracts, for specific sustainability standards pursued by the buyers, etc.). Contractual preference patterns also tend to vary between different groups of farmers (e.g., between male vs. female farmers, young vs. old farmers, etc.). What is more, preference heterogeneity may also exist within specific groups of farmers, too (e.g., among male or female farmers, etc.). In this book, we highlight smallholder preferences for certified CF options and specifically differentiate between Organic vs. Fair Trade vs. Organic & Fair Trade vs. Rainforest Alliance certified contracts vs. simple market specification contracts. The study also focuses on the heterogeneous preference patterns exhibited by contracted vs. non-contracted and by male vs. female farmers. Studying preference heterogeneity enables us to identify the groups of farmers who represent the biggest obstacles to a successful implementation of each type of studied contract as well as to understand related smallholder acceptance issues. At the time of our research (2016-2018), CF of vanilla in Northeastern Madagascar was largely characterized by loose, non-written, and unstable contractual arrangements between vanilla buyers and their signatory farmers. The price boom had evoked opportunistic and speculative side-selling behavior among vanilla farmers and among (informal) traders. Moreover, there was the incapacity of many farmers to supply acceptable vanilla qualities to their contractual buyers. Both circumstances typically led to the rapid termination of contractual relationships. (Market specification contracts lasted on average for 2.18 years only). Stated economic preferences, however, suggested that most vanilla farmers perceived CF offers in a positive light. Prices paid for unprocessed (green) vanilla through CF schemes were significantly better (on average 27% higher) than prices paid on local spot markets. Nonetheless, considerable preference heterogeneity existed between different groups of farmers regarding their willingness to pay for offered CF benefits. Male vanilla farmers, for example, showed almost twice the willingness to pay (WTP male = 13,000 MGA per kg of green vanilla sold) than female farmers (WTP female = 7,000 MGA per kg of green vanilla sold) for meeting the requirements of a Rainforest Alliance certified contract. Rainforest Alliance certification affected approximately 37% (n=44/119) of all contracted vanilla farming HHs (n=119/604) in our sample (N=604) in 2018. Next, we also found that contracted farmers possessed, on average, significantly more productive resources than non-contracted farmers. In terms of preference heterogeneity, contracted vanilla farmers showed more than ten times the willingness to pay (WTP contracted = 34,000 MGA per kg of green vanilla sold) for the Rainforest Alliance CF option than the resource-poorer non-contracted farmers (WTP non-contracted = 3,000 MGA per kg green vanilla sold). Cross-checking for reported marketing behaviors of the latter two groups indicated that the production of quality vanilla was not equally feasible for these two groups of farmers. The resource-poorer (non-contracted) farmers generally sold lower quantities of vanilla and also reported more considerable challenges in producing vanilla of acceptable quality. Yet, the resource-poorer (non-contracted) farmers typically pursued both the sale of unprocessed (green) and processed (black) vanilla for a substantial proportion of their total harvest during the price boom. Most CF schemes, however, limited signatory farmers to selling unprocessed (green) vanilla during the price boom in the SAVA Region. Indeed, less than 1% of contracted farmers (n=1/119) were able to sell processed (black) vanilla to their contractual buyers in 2018. Consequently, defaulting on contractual supply commitments was frequently observed among the resource-poorer (non-contracted) farmers who newly entered into CF schemes, according to our data. This finding was qualitatively validated by the observations of interviewed exporters. Moreover, a high contractual default was observed during the studied price boom years of 2016-2018. Contractual default causes large fluctuations of farmers in contracted producer groups. Organizational costs associated with re-mobilizing and re-training new signatory farmers to meet contractual and certified production requirements were considered very high by interviewed vanilla traders. Thus, many traders hesitated to invest in CF to formalize their commitments toward smallholder vanilla farmers in the years of booming prices. The intermediary model of CF (in which buyer-assembled producer groups are serviced by collectors who maintain buyer-to-farmer relations) was the most common form of vertical coordination (representing n = 6/9 CF schemes) captured by our CE among the (n=119) contracted vanilla farmers. Improving CF arrangements to include more benefits for farmers can theoretically increase the acceptance of CF in the eyes of signatory smallholders. Likewise, improving the contractual design can theoretically give traders a competitive advantage by (a) attracting more suppliers or (b) incentivizing more exclusive supply commitments. This study demonstrates that both multiple certifications and the use of strategic corporate social responsibility (strategic CSR) are two options available to vanilla buyers to increase the perceived benefits of CF offers in the perception of smallholder producers. Our results suggest that buyers can avoid paying higher vanilla prices by pursuing either of these two contractual innovations. Moreover, public-private partnerships between international development organizations and private vanilla companies can help contractual vanilla buyers to manage some of the logistical challenges and costs involved in implementing CF offers. In conclusion, this book suggests that improving contractual design is valid as a concept and advisable in practice. Yet, contractual design alone constitutes just one of several important success factors that need to be considered for a successful implementation of CF. Moreover, preference heterogeneity among smallholder farmers for CF options makes targeting committed households difficult for contractual buyers. In practice, an ex-ante screening of smallholder households for specific socio-economic characteristics is costly and challenging for any agribusiness trader before setting up a CF scheme. Thus, the deliberate targeting of smallholders with group-specific CF offers – which, in theory, could facilitate farmer self-selection and reduce the need for costly production controls – is rarely practiced by contractual buyers. Instead, standardized (one-size-fits-all-type) CF offers are typically made by contractual vanilla buyers in their day-to-day operations. Smallholder selection and retention thus happen by following a reactive, performance-based identification process that establishes whether the contracted farmers deliver upon agreed product qualities in adequate quantities year by year. This selection process implies the need for regular production controls and requires buyers to tolerate many dropouts among the signatory farmers. Traders typically require several years to identify smallholders who reliably supply vanilla under contract. The adaptive management protocol that traders practice to select and retain contracted farmers is, however, also suited to improve contractual design. An essential recommendation of this study thus suggests that the systematic testing of our data-driven preference patterns revealed for existing CF options in the SAVA Region could be worthwhile in the attempt to raise CF popularity and smallholder supply commitments. Improvements in smallholder acceptance levels would need to be monitored and evaluated by following an adaptive management logic. Due to significant preference heterogeneities for CF features, which exist between as well as among the analyzed groups of smallholders, it is to be expected that contractual buyers would need several attempts (i.e., several years), sufficient financial resources, and patience to improve their CF offers. As the price boom context in Madagascar is typically characterized by intense buyer competition and high contractual default rates, this study advises smaller traders (e.g., exporters and preparators trading less than 50 tons of cured vanilla per year - who typically lack the financial resources to make risky investments) to wait until the vanilla price boom ends before venturing into CF.
Keywords: Madagascar; choice experiment; smallholder preferences; contract farming; vanilla; private voluntary sustainability standards; corporate social responsibility; gender; mixed logit
