search
Contact Us
HOME

  • Crossref logo
  • Crossref Similarity Check logo
Journal Search Engine

to

Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1225-8504(Print)
ISSN : 2287-8165(Online)
Journal of the Korean Society of International Agriculture Vol.36 No.1 pp.1-9
DOI : https://doi.org/10.12719/KSIA.2024.36.1.1

Diversification And Cocoa Farm Profitability Among Cocoa Farmers in Ghana: A Conditional Mixed Process Approach

Nana Okyir Baidoo*, Osei-Asare Yaw*, Charles Yaw Okyere*, Taeyoon Kim**†, Daniel Bruce Sarpong*
*Department of Agricultural Economics and Agribusiness, University of Ghana
**Graduate School of International Agricultural Technology, Institutes of Green Bio Science and Technology, Seoul National University, Korea
† Corresponding author (Phone) +82-33-339-5706 (E-mail) taeyoonkim@snu.ac.kr
November 16, 2023 December 8, 2023 December 11, 2023

Abstract


The cocoa sub-sector is an important pillar of Ghana’s economic stability and growth prospects; the sub-sector contributes to fiscal stability, infrastructure, and employment. In recent times, diversification has become a common observable feature in the industry. This study uses the conditional mixed process approach to examine the interaction between the diversification status and profitability of cocoa farming simultaneously in a single system. Using data from 401 farmers in the Western and Ashanti regions of Ghana in 2022, the study found that more profitable farmers were more likely to diversify. Conversely, diversified farmers were likely to be more profitable than non-diversified farmers. It also found that the farmer's age, experience, farm area, and location of the farm impacted both the profitability and diversification status of farmers. Thus, it is suggested that policymakers should direct policies and programs to support the creation of alternative farm-based livelihoods for farmers.


P34 , Q12 , Q14

초록

    INTRODUCTION

    Ghana's production of cocoa is crucial to the country's socioeconomic development at the grassroots level. Revenues from cocoa go to support fiscal policy, strengthen the local currency, and finance the country's key infrastructure and social needs, including roads, schools, and scholarships for dependents of cocoa farmers (Ghana Commercial Bank, 2022). In terms of employment, the sector employs about 800,000 smallholders who derive about 70% of their incomes from cocoa production revenue (Nunoo & Owusu, 2017). Therefore, continued and sustained production of cocoa is of great importance to the country. To ensure farmers remain in the business of cocoa and attract even more farmers into the sector, cocoa production must be seen as productive and profitable. In the last decade, it has been observed that farmers are increasingly diversifying into other livelihoods to increase their incomes, a phenomenon which may point to an inadequacy of returns from cocoa (Owusu-Amankwah et al., 2017) or farmers making a concentration risk decision or an expansionary business decision (Irfany et al., 2020;Peprah, 2015).

    Although cocoa is a broadly researched area, only a few studies investigate the direct impacts of diversification status on profitability. There exists some knowledge of the factors that influence diversification status (Agyei-Manu et al., 2020;Aneani et al., 2011;Owusu-Amankwah et al., 2017). However, in most cases, the profitability of the cocoa enterprise itself is not considered. Bellon et al.(2020) touch on the difference in incomes between diversified and non-diversified farmers. However, it is important that the cost associated with earning these incomes is accounted for. Furthermore, it is also important to determine whether the diversification status of farmers has had an impact on the profitability of the farmers’ cocoa farming enterprises, which is of greater national interest.

    When considering cocoa farmers as business units, comparatively larger profits than alternatives are the primary driving force behind the choice to venture into cocoa farming. To better design and target policies that increase the profitability of this group of farmers or value chains, policymakers must first assess the profitability of existing cocoa farming operations. Gross margin/profit, net margin/ profit, return on assets (ROA), and return on equity (ROE) are the most frequently used assessment metrics in this regard (Bierly & Chakrabarti, 1996;de Carolis, 2003;Deephouse, 2000;Fernández et al., 2019;Hull & Rothenberg, 2008;Lin & Wu, 2014;Markides & Williamson, 1996;Wu et al., 2006). Of equal importance is studying the factors that affect the profitability of cocoa farming. Here again, a number of studies (Fawole & Ozkan, 2018;Kongor et al., 2018;Meliko, 2021;Onoja et al., 2012) have empirically identified some of the factors that affect cocoa enterprise’s profitability. Some factors identified include farm area, age of the farmer, the experience level of a farmer, availability and cost of labor, household size, pruning, and fertilizer, among others. Thus, despite the anecdotal evidence to support the strong link between diversification status and profitability, it is seldom featured in studies when modeling the determinants of profitability.

    Whilst it is possible to model the impact of diversification status on profitability and, in the inverse, the impact of profitability on diversification status, modeling these two variables separately and comparing the results may not be as accurate as modeling them together in a system. As these variables appear on the left-hand side of each other’s dependency equations, it is best to model them together as a system rather than as a stand-alone (Roodman, 2011). A number of papers in econometrics attest to the superior efficiency of such system modeling, especially using Maximum Likelihood Estimations to achieve more accurate estimates (Geweke, 1989;Hajivassiliou & McFadden, 1998;Keane, 1992).

    The main objective of this study is to find the relationship between diversification status and profitability simultaneously using a system modeling approach, which in this case is the Conditional Mixed Process Approach. The study provides valuable insight into the effects of these two variables on each other and other socioeconomic variables in cocoa production in the Ashanti and Western regions of Ghana. The study also gives an illustration of the application of a systems approach in modeling closely related variables in agriculture and business. This will be important in exploring new ways of analyzing similar dependent variables in these fields.

    The study is organized as follows. The next section provides the conceptual framework guiding this study. Next, we introduce the methodology adopted for the analysis, including the data collection methodology and model specifications. The next chapter discusses the results of the models used. The final chapter highlights the conclusions reached and provides advice for policymakers.

    CONCEPTUAL FRAMEWORK

    Figure 1 gives a descriptive overview of the concept underpinning this research. The decision to go into cocoa farming is one made based on the availability of farm resources, which may include land, labor, and sufficient capital to acquire or rent land, or in some cases inheritance of the land resource. From this foundation, farmers in the areas under study venture into cocoa farming as their primary livelihood strategy. Based on their set of socioeconomic characteristics and other factors, there is variation in the profitability of farmers/cocoa farm plots. To increase overall profitability, farmers may, again influenced by their unique set of socioeconomic characteristics, venture into other livelihood strategies, including farm-based livelihoods.

    However, this paper theorizes that in addition to socioeconomic factors, diversification status also influences the profitability of cocoa farmers/enterprises. This paper has adopted the Conditional Mixed Process model to model the interaction between cocoa profitability and diversification status to model these two dependent variables as a single system.

    Agricultural diversification has proven to be economically significant in reducing risk and improving the economic vitality of agricultural enterprises. Research conducted in India showed that farms with diversified fields had higher aggregate net earnings compared to those with traditional farming systems (Ryan & Spencer, 2001). Similarly, studies in Ethiopia and South Africa have revealed the positive impact of diversification strategies in reducing rural household income from agriculture (Walo, 2016).

    A Priori Relationship between Profitability and Diversification

    Based on the literature reviewed, a set of variables was identified as having an impact on profitability and diversification. Table 1 shows this study’s hypothesized relationships between these variables and profitability and diversification.

    Model

    For modeling the correlated impacts of diversification and profitability on each other, alongside other explanatory variables, first, we estimate the impact of profitability on farmers’ decision to diversify using a probit model. Here, profitability is proxied as the log of Earnings Before Interest, Tax, Depreciation, and Amortization (EBITDA). Diversification is restricted to diversifying into farm-based livelihoods together with their already established cocoa farms. From the study, there were three main diversification options identified. These were diversification into tree crops, food crops, or livestock. A farmer is considered as diversified if they are engaged in one or a combination of any of the three identified farm-based livelihood options, for which a score of one (1) is assigned or zero (0) in the case of farmers who are engaged in cocoa farming only. Functionally, we model diversification as a dummy, being a function of profitability and other socioeconomic variables based on Wooldridge (Wooldridge & Zhu, 2020). If diversification is affected by a set of explanatory variables, in this case, Profitability (PBT) and a set of other variables denoted as Xi, then the relationship can be determined by a probit model expressed as:

    y i * = β o + β 1 P B T i + ∑ k = 2 K β k X k i + e 1 i , e 1 i , e 1 i ∼ N ( 0 , 1 )
    (1)

    where y i * is the latent variable for dummy diversification (DIVi), βo, β1, ⋯, βK are parameters to be estimated, and ei is assumed to have standard normal distribution.

    Profitability can also be theorized to be affected by the diversification status of farmers and other regressors. Hufnagel et al. (2020) explored the advantages of incorporating crop diversification in agriculture. The authors contend that the implementation of crop diversification has the potential to stabilize the productivity of cropping systems while simultaneously mitigating negative environmental impacts and preserving biodiversity. However, they highlight a notable gap in the collective comprehension of crop diversification, underscoring the absence of a shared understanding and systematic research approaches in this domain.

    In their article, the authors conducted a comprehensive review of 'crop diversification' measures in agricultural research. They identified diversification practices grounded in empirical studies and emphasized the need for a more cohesive approach. To address this, the authors proposed a novel conceptual framework consisting of five key elements. This framework aims to facilitate the comparison and generalization of results across various primary studies. The suggested elements encompass defining the problem associated with existing farming practices and the potential necessity for diversification, characterizing the baseline system targeted for diversification, specifying the scale and target area, outlining the experimental design and target variables, and articulating the expected impacts.

    Furthermore, Sen et al. (2017) conducted a study reinforcing the positive impact of diversification on agriculture, particularly in terms of increased profitability. Building upon previous research, the paper not only replicates this association using more recent and comprehensive data but also delves into investigating the sources of this positive correlation. The study identifies that diversifying farms through the adoption of ancillary activities, horticulture, and other high-value enterprises (HVE), such as mushroom cultivation, contributes to a significant increase in farm income.

    For this study, we transform profitability into the log of EBITDA and use diversification status (represented as DIV) as a dummy with a value of 1 if a farmer is diversified and 0 if otherwise. We recognize here again that a similar set of socioeconomic variables also explains cocoa farmers' profitability. Then, using a Log-linear Ordinary Least Squares (OLS) regression, we can model the relationship between profitability as the dependent or left-hand side (LHS) variable and diversification status and other socioeconomic variables as explanatory or right-hand side (RHS) variables. This can be represented as:

    l n   P B T i = δ 0 + δ 1 D I V i + ∑ k = 2 K δ k X k i + e 2 i
    (2)

    where δ0, δ1, ⋯, δk are parameters to be estimated, and e2i is a randomly distributed error term. In particular, δ1 is the effect of Diversification status (DIV) on profitability.

    To determine the effects of the two dependent variables on each other, the Conditional Mixed Process model is used. The CMP allows running equations (1) and (2) in a system. This helps to overcome possible endogeneity and selectivity concerns that may arise. The CMP framework as proposed by Roodman (2011) enables the joint modeling of two or more equations, cross-equation correlation of the error terms, mixing of these models in multi-equation systems, and individual model variation by observations.

    Hence the two equations, (1) and (2), can be recast into a single system as:

    y 1 * = θ 1 + e 1
    (3)

    y 2 * = θ 2 + e 2
    (4)

    Where θ 1 = β 0 + β 1 P B T i + ∑ k = 2 K β k X k i (eqn. 1) and θ 2 = δ 0 + δ 1 D I V i + ∑ k = 2 K δ k X k i (eqn. 2)

    y = g ( y * ) = ( 1 { y 1 * > 0 } , y 2 * )
    (5)

    Where ε = (e1, e2)' ~ N(0, ∑), and Σ = [ 1 ρ 12 ρ 21 1 ]

    Here ρ12 = ρ21 represents the correlation between the error terms of diversification and profitability equations. y 1 * and y 2 * are the latent factors of diversification status and profitability, respectively. The maximum likelihood function (L) enables to estimates parameters of β and δ together.

    Data

    The data for this analysis was based on primary data collected from 15 communities in 3 districts in the Western and Ashanti regions of Ghana, as part of the Cocoa for Future project (C4F). The C4F project aims to use agroforlnPBT estry to make cocoa farms in Ivory Coast and Ghana less vulnerable by identifying agricultural and socioeconomic sustainability levers while preserving the environment. The project is sponsored by the CIRAD (French Agricultural Research Centre for International Development). The Ashanti and Western regions account for over half of Ghana’s cocoa production and by extension, 10 to 15% of global production. In total, 401 farmers were interviewed for the study. In the Ashanti region, the survey took place in 5 communities in the Afigya Kwabre North District. In the Western region, the survey was conducted in 5 communities each in the Elembelle and Manso Amenfi districts

    A multi-stage, mixed sampling method was applied in selecting the sample of farmers interviewed. In the first stage, a purposive sampling approach was used to select districts and communities with high populations of cocoa farmers for which it was possible to get farmer lists. Based on this, 15 communities in 3 districts were selected. Subsequently, a total of 150 farmers for each district were selected randomly from the list, resulting in a total of 450 selected farmers. Due to attrition, budgetary and time constraints, 401 farmers were eventually interviewed. The data collection was conducted from July 2022 to August 2022.

    RESULTS AND DISCUSSION

    Descriptive Statistics of Socioeconomic Variables

    For the sample, the average age and experience were 48 and 18 years, respectively. The average land area of the farmers surveyed was 5.343 acres (2.163 hectares) with a standard deviation of 5.395 acres (2.183 hectares). This is consistent with the literature that is currently available and shows that majority of farmers are smallholders with land sizes of between 2 and 5 hectares (Attipoe et al., 2020; Ghana Commercial Bank, 2022). Additionally, the survey's average output of 2.512 bags per acre (393.523 kilos per hectare) is comparable with studies conducted in the industry, indicating yields well below the standards that the Ghana Cocoa Board (COCOBOD) has set as attainable (Asante et al., 2022; Wainaina et al., 2021). The pricing structure is an important aspect of Ghana's cocoa industry. Ghana's cocoa industry is a monopsony, meaning that the government, acting through the Ghana COCOCBOD, is the only buyer and determines the output's price. Prices for cocoa were GHS660 for a 64kg bag or GHS10,560 for a ton ($1,320 per ton given the exchange rate of GHS8 to the dollar as of July 2022). As a result, in the farmers' profitability model, cost minimization and output quantity maximization are the main variables within the farmer’s control, which affect how profitable a farm plot is. The distance to the farm is another significant descriptive statistic. Most of the farmers surveyed reside close to their farmland. This is common practice for farmers living in rural areas, where this project’s survey was done. Predominantly, most migrants as well live in farmsteads within or very close to their farms. From the data collected, the variable migrant or non-migrant, had a mean value of 0.5586, indicating that 55.86% of the farmers interviewed were migrants and were not natives of the villages they had set- tled in. Non-agric income has a mean of 0.266833 and a standard deviation of 0.442857. This implies that only 26.68% of the farmers interviewed had non-agricultural income sources. These included formal and informal salaried jobs and trading.

    Results of Conditional Mixed Process Estimation

    As elaborated in earlier chapters of this paper, the study estimated the effects of profitability on diversification, modeled as a probit, and the effects of diversification on profit, modelled as an ordinary least squares regression, in a system using the Conditional Mixed Process (CMP) approach. The rho value rho 12 constant of -1 points to an almost perfect convergence of the error terms of the 2 models used in the system. This implies that running the two models together in a system is best. The high Wald Chi-square value or the low p-value indicates that the model is significant at a 1% level, further lends credence to the suitability of the model in explaining the relationships being investigated.

    The first part of the results table (Table 3) shows that profitability, transformed as log value, is significant in explaining diversification at a 1% level. The positive coefficient of the parameter estimate in the model also indicates that there is a positive relationship between profitability and the diversification status of the farmers interviewed. An increase in profitability of farmers increases farmers’ likelihood of being diversified into other farm-based likelihoods.

    Also, the model shows that age, experience, farm area, and region of farm location were significant in explaining the diversification status of farmers. Age is positively related to diversification, implying that older farmers are more likely to be diversified. As young people age, they are able to generate more resources and have energy. Thus, they are able to diversify into other livelihoods (Dharmadasa & Polkotuwa, 2016). However, as farmers age beyond a certain point, they are less likely to diversify as they become less able to manage multiple enterprises and become more risk averse. This may provide an explanation as to why the experience was found to be negatively related to diversification (Dharmadasa & Polkotuwa, 2016).

    From the results, there is also a negative relation between farm area and diversification, thus, an increase in farm area/size makes it less likely that a farmer is diversified. This means that as farmers increase their land sizes, they are more likely to specialize in cocoa production and less invested in other livelihoods.

    There was also a positive relationship between the region of the farm’s location, measured as a dummy variable for the Western region and 0 for the Ashanti region. This implies that farmers are more likely to be diversified if their farm plots are located in the Western region compared to if the farm plots are located in the Ashanti region. The Elembelle and Wassa Amenfi Central districts in the Western region, which were the districts chosen for this study, are predominantly rural, with a rural population of 71.9% and 84.4%, respectively (Ghana Statistical Service, 2022).

    The Afigya Kwabre North district in the Ashanti region selected for this study is predominantly urban, with an urban population of 60.2% (Ghana Statistical Service, 2022). The livelihood diversification options considered for this study are agriculture. It is likely that farmers in the Western region have more access to land and are more suited to agricultural livelihoods and thus are more likely to diversify, whereas land acquisition may be more difficult to find in an urban setting as exists in Afigya Kwabre Nort district, and also diversification options may be more non-agricultural (Djanibekov & Gaur, 2018).

    Household size, education, access to credit, membership of FBO, and participation in non-agri income sources were expected to have a positive relationship with both profitability and diversification. Being a migrant and distance from dwelling to farm were expected to have a negative relationship with both profitability and diversification. However, all of these variables were found to be insignificant in determining profitability and diversification. This implies that for the sample selected for this study, there were no discernable, statistically significant relationships between these variables and their profitability and diversification outcomes.

    The second part of Table 3 presents the results of the Ordinary Least Squares regression between profitability and farmers’ diversification status. Here, diversification was found to be statistically significant in explaining profitability at a 10% significant level. From the results, diversified farmers are likely to be 54.2% more profitable in terms of EBITDA than their non-diversified colleagues. Also, age, experience, farm area, and region were also found to be statistically significant in determining the profitability of cocoa farmers, all at a 1% level. Age and region were negatively related to profitability, while experience and farm area were positively related. From the coefficients estimated, a year increase in age results in a decrease of 1.3% in profitability. An additional year of experience in cocoa farming is estimated to result in a 5.4% increase in profitability. A one-acre increase in farm area also is expected to lead to a 9.8% increase in profitability. Also, farms in the Western region are expected to be 99.2% less profitable than farms in the Ashanti region.

    Marginal effects are calculated for probit models in Table 4 to assess the impact of a change in an independent variable on the probability of a binary outcome. Ordinary Least Squares (OLS) regression models, which is the second part of Table 3, show that each coefficient represents the estimated change in the mean of the dependent variable associated with a one-unit change in the corresponding independent variable while holding other variables constant, which is as same as marginal effects for profitability. Thus, marginal effects in Table 4 show only for the probit model. From the results in Table 4, a 1% change in profit increases the probability of a farmer being diversified by 11.3%.

    CONCLUSIONS

    The study is based on existing literature, which points to a strong connection between cocoa farmer profitability and the decision to diversify in recent times. However, by modelling this interdependency in a system, given other socioeconomic explanatory variables, there is a clearer relationship between the two. The Conditional Mixed Process (CMP) is able to execute this precisely. The results of the model show a near-perfect relationship between the error terms of the two models, which were a probit model with the diversification status as the dependent variable and an ordinary least squares regression, with profitability as the dependent variable. From the CMP model, it can be concluded that the 2 variables, diversification, and profitability, affect each other. Higher profits are likely to be associated with diversified farmers, and on the other hand, diversified farmers also tend to have higher profitability than their colleagues who are not diversified. In summary, farmers with higher profits are likely to be diversified into other farmbased livelihoods whilst farmers who are diversified into other farm-based livelihoods have more profitable cocoa farm enterprises. Whilst the study’s conclusion that diversification and profitability in cocoa are positively related, this can be argued to feed into other studies that indicate a decline in cocoa productivity leading to and encouraging diversification further from cocoa (Odijie, 2018) that argues that diversification, including activities sponsored by private companies are a response to declining cocoa incomes.

    Age, experience of farmers, the size or area of farms, and the location of farms also significantly affect diversification and profitability. Age and farm area positively affects diversification whilst experience and farm being located in the Western region also negatively affect diversification. On the other hand, experience and farm area positively affect profitability whilst age and farm being in the Western region also negatively affect profitability.

    Based on the results of the study, farmers need to be encouraged to consider diversifying into other farm-based livelihoods such as livestock rearing or other tree crop farming. This will ensure that farmers have enough revenue throughout the year to support the production costs required for a more productive and profitable cocoa farming business. Training and technical support may need to be extended to guide farmers into saving to set up such alternate livelihoods. Also, more attention needs to be paid to farmers in the Western region to identify and correct factors that may lead to low productivity and profitability of cocoa farmers in the region.

    The results of this imply that farmers are more likely to be diversified into other farm-based likelihood if they are profitable. Also, cocoa farmers in more rural districts are the likeliest to diversify into farm-based livelihoods. As a risk mitigation and welfare strategy, cocoa farmers should be encouraged to diversify into both farm-based agricultural livelihoods and non-farm based livelihoods depending on the nature of farmers’ location, whether rural or urban.

    ACKNOWLEDGMENTS

    This study about “Diversification and Cocoa Farm Profitability Among Cocoa Farmers in Ghana: A Conditional Mixed Process Approach” was conducted within the framework of the Cocoa4Future (C4F) project, which is funded by the European DeSIRA Initiative under grant agreement No. FOOD/2019/412-132 and by the French Development Agency. The C4F project pools a broad range of skills and expertise to meet West African cocoa production development challenges. It brings together many partners jointly striving to place people and the environment at the core of tomorrow's cocoa production.

    The lead author is grateful to the Regional Scholarship and Innovation Fund (RSIF) for the scholarship for the PhD program at the University of Ghana under which this study was conducted. Also, the lead author is grateful to the Department of Agricultural Economics and Agribusiness of the University of Ghana and the Green Bio Science and Technology of Seoul National University for the guidance offered during the sandwich program at which time this analysis was conducted.

    Figure

    JKSIA-36-1-1_F1.gif

    Interaction between Cocoa Farming, Profitability and Diversification.

    JKSIA-36-1-1_F2.gif

    Map of Ghana Showing Study Areas.

    Table

    Variables and Expected Relationship with Profitability and Diversification.

    Descriptive Statistics of Main Socioeconomic Variable Variables of Cocoa Farmers.

    Results of Conditional Mixed Process Estimation.

    Source: Produced by the author using Primary Data 2022
    * p<0.1, ** p<0.05, *** p<0.01

    Marginal Effects Estimates of Probit Model Estimation of Diversification and Profitability.

    Reference

    1. Agyei-Manu, K. , Nimoh, F. , Owusu-Peprah, M. , Kyeremateng, G.B. 2020. Cocoa farmers choice of alternative livelihood in mining communities in Upper Denkyira West District, Ghana. Journal of Development and Agricultural Economics. 12:181–197.
    2. Aneani, F. , Anchirinah, V.M. , Owusu-Ansah., Asamoah, F. 2011. An analysis of the extent and determinants of crop diversification by cocoa (Theobroma cacao) farmers in Ghana. African Journal of Agricultural Research. 6:4277–4287.
    3. Angrist, J.D. 2001. Estimation of Limited Dependent Variable Models With Dummy Endogenous Regressors. Journal of Business & Economic Statistics. 19:2–28.
    4. Bellon, M.R. , Kotu, B.H. , Azzarri, C. , Caracciolo, F. 2020. To diversify or not to diversify, that is the question. Pursuing agricultural development for smallholder farmers in marginal areas of Ghana. World Development. 125:104682.
    5. Bierly, P. , Chakrabarti, A. 1996. Generic knowledge strategies in the U.S. pharmaceutical industry. Strategic Management Journal. 17:123–135.
    6. de Carolis, D.M. 2003. Competencies and imitability in the pharmaceutical industry: An analysis of their relationship with firm performance. Journal of Management. 29:27–50.
    7. Deephouse, D.L. 2000. Media reputation as a strategic resource: An integration of mass communication and resource-based theories. Journal of Management. 26:1091–1112.
    8. Dharmadasa, R.A.P.I.S. , Polkotuwa, P.W.T.P. 2016. Income Diversification of Estate Sector in Sri Lanka. Journal of Business Studies. 3:27–39.
    9. Djanibekov, U. , Gaur, V. 2018. Nexus of energy use, agricultural production, employment and incomes among rural households in Uttar Pradesh, India. Energy Policy. 113:439–453.
    10. Fawole, W.O. , Ozkan, B. 2018. Revisiting the profitability and technical efficiency of cocoa production amidst economic diversification program of the Nigerian Government: A case study of Ondo State. Journal of Agribusiness in Developing and Emerging Economies. 8:186–200.
    11. Fernández, E. , Iglesias-Antelo, S. , López-López, V. , Rodríguez- Rey, M. , Fernandez-Jardon, C.M. 2019. Firm and industry effects on small, medium-sized and large firms’ performance. BRQ Business Research Quarterly. 22:25–35.
    12. Geweke, J. 1989. Bayesian Inference in Econometric Models Using Monte Carlo Integration. Econometrica. 57:1317–1339.
    13. Ghana Commercial Bank.2022. SECTOR INDUSTRY ANALYSIS 2022 COCOA SECTOR REPORT.
    14. Ghana Statistical Service.2022. Population and Housing Census 2021 - Ghana. https://www2.statsghana.gov.gh/nada/index.php/ddibrowser/110/export/?format=pdf&generate=yes
    15. Hajivassiliou, V. , McFadden, D. 1998. The Method of Simulated Scores for the Estimation of LDV Models. Econometrica. 66:863–896. https://EconPapers.repec.org/RePEc:ecm:emetrp:v:66:y:1998:i:4:p:863-896
    16. Hufnagel, J. , Reckling, M. , Ewert, F. 2020. Diverse approaches to crop diversification in agricultural research. A review. Agronomy for Sustainable Development. 40:1–17.
    17. Hull, C.E. , Rothenberg, S. 2008. Firm performance: The interactions of corporate social performance with innovation and industry differentiation. Strategic Management Journal. 29:781–789.
    18. Irfany, I. , McMahon, P.J. , Toribio, J.A. , Phan-Thien, K.Y. , Rifai, M.A. , Yusdiyanto, S. , Vinning, G. , Guest, D.I. , Walton, M. , Nuryartono, N. 2020. Determinants of diversification by cocoa smallholders in Sulawesi. International Journal of Social Economics. 47:1243–1263.
    19. Keane, M. 1992. A Note on Identification in the Multinomial Probit Model. Journal of Business & Economic Statistics. 10:193–200. https://EconPapers.repec.org/RePEc:bes:jnlbes:v:10:y:1992:i:2:p:193-200
    20. Kongor, J. , Steur, H. , Walle, D. , Gellynck, X. , Afoakwa, E.O. , Boeckx, P. , Dewettinck, K. 2018. Constraints for future cocoa production in Ghana. Agroforestry Systems. 92:1373–1385.
    21. Lin, Y. , Wu, L.Y. 2014. Exploring the role of dynamic capabilities in firm performance under the resource-based view framework. Journal of Business Research. 67:407–413.
    22. Markides, C.C. , Williamson, P.J. 1996. Corporate Diversification And Organizational Structure: A Resource-Based View. Academy of Management Journal. 39:340–367.
    23. Meliko, M. 2021. Profitability analysis of smallholder cocoa production in South West Region of Cameroon Economics of Production in African Agriculture View project. African Journal of Agricultural Research. 17:991–997.
    24. Nunoo, I. , Owusu, V. 2017. Comparative analysis on financial viability of cocoa agroforestry systems in Ghana. In Environment, Development and Sustainability. 19:83–98.
    25. Odijie, M.E. 2018. Sustainability winners and losers in businessbiased cocoa sustainability programmes in West Africa. International Journal of Agricultural Sustainability. 16:214–227.
    26. Onoja, A.O. , Deedam, N.J. , Achike, A.I. 2012. PROFITABILITY AND YIELD DETERMINANTS IN NIGERIAN COCOA FARMS: EVIDENCE FROM ONDO STATE. Journal of Sustainable Development in Africa. 14:172–182.
    27. Owusu-Amankwah, R. , Ruivenkamp, G. , Essegbey, G. , Frempong, G. 2017. The Nature, Extent and Effect of Diversification on Livelihoods of Farmers: A Case Study of the Cocoa Farmers in Ghana. International Journal of Agriculture Innovations and Research. 5:2319–1473. http://csirspace.csirgh.com:80/handle/123456789/2084
    28. Peprah, K. 2015. Sustainability of cocoa farmers’ livelihoods: A case study of Asunafo District, Ghana. Sustainable Production and Consumption. 4:2–15.
    29. Roodman, D. 2011. Fitting Fully Observed Recursive Mixed-process Models with cmp. The Stata Journal. 11:159–206.
    30. Ryan, J.G. , Spencer, D.C. 2001. Future challenges and opportunities for agricultural R & D in the semi-arid tropics. International Crops Research Institute for the Semi-Arid Tropics (ICRISAT).
    31. Sen, B. , Jha, G. , Raj Singh, D. 2017. Agricultural Diversification and its Impact on Farm Income: A Case Study of Bihar. Agricultural Economics Research Review. 30:77–88.
    32. Walo, M.T. 2016. Bridging the Rural-Urban Divide for Local Economic Development in Nekemte and its Hinterlands, Oromia, Ethiopia. Journal of Agriculture, Food Systems, and Community Development. 6:125–143.
    33. Wooldridge, J.M. , Zhu, Y. 2020. Inference in Approximately Sparse Correlated Random Effects Probit Models With Panel Data. In Journal of Business and Economic Statistics. 38:1–18.
    34. Wu, F. , Yeniyurt, S. , Kim, D. , Cavusgil, S.T. 2006. The impact of information technology on supply chain capabilities and firm performance: A resource-based view. Industrial Marketing Management. 35:493–504.
    35. Yussuf, B.A. , Mohamed, A.A. 2022. Factors Influencing Household Livelihood Diversification: The Case of Kebri Dahar District, Korahey Zone of Somali Region, Ethiopia. Advances in Agriculture. 2022:1-14.
    Copyright © The Korean Society of International Agriculture. All rights reserved.
    Rural Development Administration Bldg., 300 Nongsaengmyeong-ro, Wansan-gu, Jeonju 54875, Republic of Korea