Business Models in a Regulated Commodity Sector: A Structural Analysis of Cocoa Agribusiness in Ghana

Defining business model identity

Business Model Identity (BMI) requires conceptual grounding to distinguish it from organizational identity while clarifying its role within business model scholarship. In the classical organizational identity literature, identity addresses the question “Who are we as an organization?” and refers to the collective self-understandings and distinctive attributes that organizational members believe characterize their organization (Albert & Whetten, 1985). Organizational identity therefore concerns social meanings, shared interpretations, and perceptions of the organization.

In contrast, BMI concerns the identity of the business model rather than the organization itself. It refers to the relatively enduring, distinctive, and coherent configuration of primary value-creation logic, strategic asset configuration, and structural composition through which an enterprise creates, coordinates, and captures value within an institutional context. BMI therefore captures the structural character of the business model and the mechanisms through which economic activity is organized. Organizations with different cultures, ownership structures, or organizational identities may consequently exhibit the same BMI, while organizations with similar organizational identities may operate fundamentally different business model architectures.

This conceptualization is consistent with business model research that views the business model as an integrated architecture of value creation rather than a collection of discrete components. Teece (2010) conceptualizes the business model as the architecture through which firms create, deliver, and capture value, while Amit and Zott (2010) describe it as a system of interdependent activities whose content, structure, and governance jointly determine value creation. Similarly, Richardson (2008) and Bocken et al. (2014) emphasize the importance of coherence among value propositions, activity systems and value-capture mechanisms. Although business models evolve, their core logic often remains recognizable over time, allowing certain structural characteristics to persist despite adaptation and innovation (Demil & Lecocq, 2010). This suggests that business models possess an identifiable and relatively stable architecture that can be analyzed independently of organizational self-perceptions.

The use of the term identity is therefore deliberate. Identity conveys the notions of centrality, distinctiveness, coherence and continuity (Albert & Whetten, 1985; Whetten, 2006), all of which are relevant to understanding business model architecture. Centrality refers to the dominant value-creation logic that defines the model's economic role. Distinctiveness reflects the unique combination of strategic assets and structural arrangements that differentiate one model from another. Coherence captures the alignment among value-creation activities, asset configurations and governance structures, while continuity reflects the persistence of these relationships despite incremental adaptation. In regulated commodity sectors, where organizations operate within established institutional frameworks, these characteristics provide a useful basis for understanding why certain business models emerge, persist and evolve.

The term identity is preferred to related concepts such as business model type, archetype or configuration. While these concepts primarily classify or describe observable arrangements of business model elements, BMI seeks to identify the underlying structural architecture that makes a business model recognizable and enduring despite changes in particular activities, assets or relationships. BMI therefore extends beyond classification by linking value-creation logic, strategic asset configuration, structural composition and institutional embeddedness into a single analytical construct. As such, it provides a diagnostic lens for examining how different business model architectures shape behavior, innovation potential and value capture within regulated commodity systems.

Theoretical foundations and operational dimensions of the BMI framework

The BMI framework rests on three complementary theoretical pillars, operationalized through three interrelated dimensions. The first pillar is archetypal value-creation logic, based on Malone et al.'s (2005) typology of Creator, Distributor, Landlord and Broker business models, which isolates the fundamental economic role through which a firm participates in value creation and exchange. Its transferability enables comparison between heterogeneous actors independent of sector-specific labels, making it valuable for analyzing complex agrifood chains. The second pillar is the configurational perspective, emphasizing alignment among the value proposition, activity system, and value capture mechanism (Amit & Zott, 2010; Bocken et al., 2014; Richardson, 2008). Rather than viewing business models as discrete components, this approach interprets them as integrated architectures where strength depends on coherence among internal elements; misalignment often explains underperformance or limited adaptability. The third pillar is institutional embeddedness, a critical but often underdeveloped dimension in business model research. In highly regulated commodity sectors, formal institutions define permissible activities, oversee pricing and quality, allocate rights, and structure market access (Kolavalli & Vigneri, 2017). As Barrett et al. (2022) observe, agrifood systems are shaped as much by governance structures and coordination mechanisms as by market incentives.

These three perspectives are operationalized through three interrelated dimensions. Value-creation logic identifies the primary economic role of the model. In Ghana's cocoa sector, few actors fit neatly into a single archetype; Licensed Buying Companies, for example, function primarily as Distributors but often incorporate financial activities associated with hedging or liquidity management (Staritz et al., 2023; Kolavalli et al., 2012). Strategic asset configuration, building on the resource-based view (Barney & Clark, 2007), emphasizes how physical, intangible, financial, and human assets shape the feasible activity system (Barrett et al., 2022). In Ghana's cocoa chain, firms display highly asymmetrical asset bases: farmers are constrained by limited fixed capital and liquidity (Boansi et al., 2024; Abdulai et al., 2018), LBCs rely on warehouses, haulage assets, procurement financing and licensing agreements (Ansah et al., 2018), while processors depend on heavy fixed capital investments and strict quality certification (Naydenov et al., 2022; Codjoe et al., 2025). Temporal-structural embeddedness captures both stability and evolution of business models (Sosna et al., 2010), with adaptability conditioned by institutional frameworks and regulatory routines, pricing mechanisms, and quality standards (Teye et al., 2025). Together, these perspectives justify BMI as a structural identity embedded in economic logic, resource configuration and institutional constraints, moving analysis beyond classification to diagnostic insight.

Conceptual framework: Diagnosing identity within a regulated system

The BMI lens offers a three-stage diagnostic framework for regulated commodity sectors (Figure 1). Stage one maps business model identities across the chain in Ghana's cocoa sector against the three BMI dimensions, providing a structured view of identity diversity. Stage two, which involves structural profiling, deepens this mapping through analysis of asset structures, financing arrangements and institutional dependencies, recognizing that actors face distinct financial constraints, regulatory obligations and capital requirements (Kolavalli et al., 2012), with institutional pressures shaping opportunity recognition and strategic choices (Teye et al., 2025). Stage three, interaction mapping, examines how identities relate through vertical coordination and horizontal dynamics, revealing the sector's operative architecture and mechanisms through which constraints propagate (van Paassen et al., 2022). This integrated approach yields a system map of business model identities and interdependencies, enabling identification of leverage points for innovation and policy intervention, aligning with insights on value chain governance, price setting and financialization that shape distributional outcomes (Staritz et al., 2023; Baines, 2017).

Figure 1.

Phase 1: Identity identification – data collection and analytical grid

The objective was to map key sector actors against the three BMI dimensions: Primary Value Creation Logic, Strategic Asset Configuration, and Structural Composition.

Sampling framework

On-Farm Actors (Producer Households)

A multi-stage sampling procedure was used to collect data on cocoa-farming households. Selection of the sample size was done at the following stages:

• Stage I: Purposive selection of the Eastern Region in Ghana (See Appendix A1 for the justification for the selection of the region).

• Stage II: Purposive selection of three districts in the Eastern Region from clusters of districts in the region. The selected districts included the West Akim, Upper West Akim, and the Asene Manso-Akroso districts.

• Stage III: Random selection of farming communities in the districts.

• Stage IV: Random sampling of respondent cocoa agribusiness units (producer households) in the farming communities.

Each selected community in the study area (Figure 2) was zoned, and cocoa agribusiness units were selected from each zone based on who was available in a particular zone at the time the researcher and/or his agents got to the zone in the community, at times when operators were expected to be at home.

Figure 2.

Post-Farmgate Actors

A purposive sampling strategy was used to select the entities that define the core regulated value chain architecture:

1. Ghana Cocoa Board (COCOBOD): The statutory regulator and market coordinator.

2. PBC Ltd: A Licensed Buying Company (LBC), and the only one purposively sampled for this category.

3. Cocoa Processing Company (CPC) Ltd: A downstream processor.

See Appendix A2 for the justification for post-farmgate actor selection for the study.

Phase 2: Structural profiling – quantitative asset analysis

This phase aimed to quantify the Strategic Asset Configuration dimension of each BMI, providing a material fingerprint.

Phase 3: Interaction mapping – integrative systemic analysis

The objective was to map the vertical and horizontal interactions between the profiled BMIs. Data from Phases 1 and 2 were synthesized. Using flow diagrams derived from documentary rules (e.g., the COCOBOD-PBC Ltd agreement) and survey data on market outlets, the formal vertical interactions and horizontal power relationships were charted to produce a diagnostic system map of the sector's architecture.

Identification and characterization of business model identities

Applying the BMI construct reveals four core identities, each representing a stable configuration of value creation logic, strategic assets, and structural form. Their characterization provides a diagnostic map of the sector's foundational elements.

Statutory composite cocoa agribusiness (SCCA) model: The architect-player

COCOBOD's activities extended across several business-model archetypes, including Intellectual Property Landlord, Inventor, Wholesaler/Retailer and Financial Trader. The designation Statutory Composite Cocoa Agribusiness (SCCA) Model reflects its statutory basis, combination of value-creation logics and sectoral focus. Figure 3 records between 48 and 55 marketing licenses issued annually during 2016/2017–2019/2020, while the number of active Licensed Buying Companies declined from 44 to 38. Licensing therefore constituted a central component of the SCCA's operating configuration.

Figure 3.

The SCCA displayed a mixed tangible and intangible asset structure. Trade and other receivables constituted the largest share of total assets at 35.81%, followed by property, plant and equipment at 26.49%, contract assets at 12.81%, cash and cash equivalents at 8.09%, and inventories at 7.49%. The remaining assets comprised intangible assets, financial investments, investment properties, and rights-of-use assets. Overall, intangible and contractual claims exceeded tangible holdings, distinguishing the SCCA from the other models.

Comparative structural profiling

Table 7 compares the tangible-to-intangible, fixed-to-current, and operating-to-non-operating asset ratios of the four models. The BIP contained no valued intangible or non-operating assets, producing infinite tangible-to-intangible and operating-to-non-operating ratios. Its fixed-to-current ratio of 0.12 was the lowest among the models, reflecting the dominance of cocoa beans within its valued asset base. The SCCA recorded the lowest tangible-to-intangible ratio, at 0.52, and a fixed-to-current ratio of 0.93. It was therefore the only model in which intangible and contractual assets exceeded tangible assets and fixed and current holdings were approximately balanced. The CACD ratios of 1.62 and 1.55 indicated greater reliance on tangible and fixed assets, respectively. The CM recorded the highest tangible-to-intangible ratio of 13.96, fixed-to-current ratio of 6.13, and operating-to-non-operating ratio of 118.79.

The profiles consequently differentiated the models according to their material foundations. Current tangible production assets dominated the BIP; intangible and contractual claims were most prominent in the SCCA; tangible fixed distribution assets characterized the CACD; and fixed operating assets overwhelmingly defined the CM.

Interaction mapping

Figure 4 presents the vertical and horizontal relationships among the four BMIs. The SCCA forms the regulatory envelope within which the other models operate. Its vertical relationships with the BIP, CACD, and CM involve the downward transmission of rules, licenses, and prices and the corresponding upward movement of cocoa beans and regulatory compliance.

Figure 4.

Horizontal market transactions follow the sequence BIP → CACD → CM. The BIP supplies cocoa beans to the CACD, which aggregates and transfers them through the regulated system to the CM for processing. No direct horizontal relationship was identified between the BIP and CM. The CACD therefore occupies the intermediary position between primary production and processing, while the SCCA coordinates the system through its regulatory and commercial relationships.

Distinctive business model configurations

The BIP and CM models demonstrate that actors sharing the same archetype may possess different identities. Both are Creators of Physical Assets, but the BIP is organized around biological production and current operating assets, whereas the CM depends on industrial transformation and fixed processing assets. BMI therefore provides greater analytical precision than archetypal classification by identifying how a value-creation logic is enacted through a distinctive asset and activity configuration.

The SCCA and CACD illustrate composite identities. The SCCA combines Landlord, Distributor and Financial Trader logics under a statutory mandate, while the CACD integrates a dominant distribution logic with supplementary financial activities. These findings support Wirtz's (2020) view that value-creation logic forms the core of a business model, while showing that this core may combine several logics in regulated sectors.

Assets as the material expression of identity

The contrasting asset profiles provide observable evidence of the four BMIs. This supports the resource-based view that resource configurations shape feasible organizational activities and strategies (Barney & Clark, 2007), as well as configurational perspectives emphasizing alignment among assets, activities, and value-capture mechanisms (Amit & Zott, 2010; Richardson, 2008).

The BIP's dependence on tangible production assets is consistent with documented capital and liquidity constraints among Ghanaian cocoa farmers (Abdulai et al., 2018; Boansi et al., 2024). The CACD's fixed-asset structure corresponds with the storage, transport, and procurement infrastructure required by Licensed Buying Companies (Ansah et al., 2018), while the CM's concentration in processing assets reflects the capital intensity of cocoa manufacturing (Naydenov et al., 2022; Codjoe et al., 2025).

The SCCA extends conventional resource-based interpretations because much of its structural power derives from licenses, contractual rights, receivables, and financial claims. This is consistent with evidence that governance, pricing authority and financialization shape value capture in Ghana's cocoa sector (Staritz et al., 2023; van Huellen & Abubakar, 2021). Institutionally conferred rights may therefore constitute strategic assets alongside physical resources.

Institutional and relational identity

The interaction map shows that BMI is shaped not only by internal assets and activities but also by permitted relationships within the regulated system. The SCCA defines the regulatory envelope, while transactions follow the BIP–CACD–CM sequence. The absence of a direct BIP–CM relationship makes the CACD an institutionally required intermediary.

This finding is consistent with accounts of Ghana's board-managed cocoa system (Kolavalli & Vigneri, 2017) and with studies emphasizing governance and coordination in agrifood value chains (Barrett et al., 2022; Baines, 2017; van Paassen et al., 2022). It extends firm-centered business-model perspectives by showing that institutions do not merely surround business models; they partly constitute their activities, assets, and relationships.

Overall, the results support BMI through four properties: centrality, reflected in dominant value-creation logic; distinctiveness, expressed through asset and governance configurations; coherence, shown by alignment among logic, resources and institutional role; and relative continuity, reflected in the persistence of recognizable structural patterns. Consistent with Demil and Lecocq's (2010) concept of dynamic consistency, such continuity does not imply immutability. BMI therefore offers an initial empirical operationalization of how enterprises characteristically create, coordinate, and capture value within regulated systems.

Conclusions

This study applied the Business Model Identity framework to Ghana's regulated cocoa sector and identified four structurally distinct models: the Bean Input Producer (BIP), Statutory Composite Cocoa Agribusiness (SCCA), Composite Agricultural Commodities Distributor (CACD) and Chocolate Manufacturer (CM). Their differences arise not only from their positions in the value chain but from the distinctive combinations of value-creation logic, strategic assets, structural composition and institutional relationships through which they operate.

The sector follows a regulated BIP-CACD-CM sequence within an institutional architecture governed by the SCCA. The BIP is concentrated in tangible, seasonal production assets; the CACD combines commodity distribution with supplementary financial activities; and the CM depends heavily on fixed processing assets. The SCCA occupies a composite regulator–commercial operator position supported by licensing authority, contractual claims, and financial assets. The absence of a direct BIP–CM relationship confirms the institutionally mandated separation between primary production and processing.

These findings show that business-model identity is expressed through the alignment of value-creation logic, asset configuration, and institutional position. The BMI framework therefore provides a structured basis for diagnosing differences in capabilities, constraints and opportunities across regulated commodity systems.

Policy and practice implications

Policy interventions should recognize that the four BMIs face different structural constraints. For the BIP, the limited financial recognition of biological assets and the absence of direct processing relationships restrict opportunities for investment and value addition. Strengthening producer cooperatives, developing appropriate mechanisms for valuing productive biological assets, and enabling farmer participation in logistics or processing ventures could broaden producers' asset and income bases.

The SCCA's simultaneous regulatory and commercial functions warrant clearer institutional safeguards to minimize potential conflicts of interest. Greater separation, transparency or independent oversight of these roles could improve accountability and create a more predictable operating environment for other sector participants.

The CACD's mandatory intermediary position could be used to expand services beyond bean aggregation. Licensing arrangements could encourage traceability, climate-smart input provision, certified-sale financing and production-data services. For the CM, policy support should address the constraints associated with capital-intensive processing through more reliable bean-sourcing arrangements, research and development incentives, and investment in cocoa by-product and waste utilization.

Limitations and future research

The findings should be interpreted in light of three limitations. First, the post-farmgate asset analysis covers 2015–2020 and therefore provides a structural profile rather than a longitudinal assessment of adaptation to recent shocks. Second, the CACD identity is based on one Licensed Buying Company and may not represent the full diversity of firms within this category. Third, the interaction map identifies formal relationships but does not quantify power, trust or value distribution across them.

Future research should apply the BMI framework longitudinally, compare multiple firms within each identity, and test it across other countries and regulated commodities. Combining BMI analysis with network analysis, value-appropriation measures and qualitative power mapping would also clarify how structural identities evolve and how their relationships shape the distribution of value.