Planting subsidies or brand building? Policy thresholds and goal conflict in agribusiness-led contract farming

1 Introduction

Contract farming is an important organizational arrangement linking agricultural production with downstream processing, marketing, and food distribution (Yang et al., 2025). Through pre-harvest agreements, agribusiness firms can coordinate procurement with agricultural production, while farmers obtain more predictable access to downstream markets. Recent evidence from the vegetable and livestock sectors shows that contract farming can strengthen market linkages, mitigate market and transaction risks, and improve farmer income and production performance (Liang et al., 2023). These functions make contract farming important not only for farmer livelihoods but also for the continuity and economic sustainability of agricultural supply (Yang et al., 2026). However, these benefits are not automatically sustained. Farmers bear planting costs and yield uncertainty, while the value created by government intervention may accrue mainly to the agribusiness firm and consumers. A policy that raises total welfare but provides farmers with limited gains may weaken their production incentives and raise concerns about the economic attractiveness of continued participation in the contractual arrangement. This raises a central policy question: when the government selects an intervention to maximize social welfare, does the socially preferred policy also generate higher profit for farmers, or can the two policy rankings diverge? Such a divergence may arise because planting subsidies support farmers directly, whereas the benefits of brand building reach farmers indirectly through the firm's endogenous procurement-price decision, creating the possibility of a goal conflict between aggregate social welfare and farmer profit.

To address this question, we compare two representative forms of government intervention on different sides of the agricultural market. Planting subsidies support the production side by lowering farmers' planting costs and strengthening their incentives to plant (Alizamir et al., 2019; Ye et al., 2021; Tang et al., 2024). Government-supported brand building works on the demand side by expanding consumer demand and increasing agricultural product value through regional branding, quality certification, and market promotion (Xu et al., 2024). Together, these policies represent two basic approaches: supporting agricultural production directly or creating value in the downstream market. The key question is whether the government should support farmers directly or rely on the procurement price to pass some of the additional market value back to them.

The corporate social responsibility (CSR) of agribusiness firms adds another layer to this policy choice. Corporate participation in rural poverty alleviation and industrial development has become an important form of social responsibility. Recent evidence shows that firms' targeted assistance can increase household income and reduce poverty through financial empowerment and industrial development (Zhou et al., 2026). Related supply-chain research also shows that a downstream firm's concern for farmer welfare can affect purchasing decisions and the benefits received by farmers (Yu et al., 2024). This issue is especially relevant in contract farming because the procurement price determines how the value generated by market expansion is divided between the firm and the farmer. A firm with stronger social responsibility concerns may offer a higher procurement price and transfer a larger share of market gains to the farmer. At the same time, such voluntary support may substitute for part of the production incentive otherwise provided by the government. CSR may therefore affect not only pricing and surplus allocation within the supply chain, but also the relative performance of planting subsidies and brand building.

Existing studies have examined agricultural subsidies, contract farming, and socially responsible supply chains from largely separate perspectives. Research on agricultural policy compares price protection, income protection, input subsidies, and output subsidies, showing that alternative policy designs may have different effects on planting decisions, farmer profit, consumer surplus, and social welfare (Alizamir et al., 2019; Tang et al., 2024). Studies of contract farming further investigate how yield uncertainty, farmer risk preferences, and government subsidies shape contract design and supply-chain performance (Peng and Pang, 2019). Research on socially responsible supply chains finds that a firm's concern for upstream or downstream partners can alter pricing, output, and coordination outcomes (Liu et al., 2019).

Despite these advances, three issues remain. First, few studies compare production-side planting subsidies with demand-side brand building within a single analytical framework. Second, contract-farming models often take market demand as given. As a result, they say little about how the market value created by government-supported brand building is passed to farmers through the procurement price set by the agribusiness firm. Third, CSR research has mainly examined firms' pricing and supply-chain coordination. Less is known about whether an agribusiness firm's concern for farmer welfare changes the relative performance of alternative government policies.

To address these questions, we develop a sequential game involving a government, a downstream agribusiness firm (hereafter, the firm), and a farmer. The firm enters into a contract with the farmer and sets the procurement price, after which the farmer chooses the planting area under stochastic yield. The firm's objective function includes both its own expected profit and a positive weight assigned to farmer profit, thereby capturing its social responsibility concern. The government maximizes social welfare, defined as the sum of farmer profit, firm profit, and consumer surplus net of policy costs, and chooses either a per-unit-area planting subsidy or a brand-building level. By comparing the equilibrium outcomes under the two policy regimes, we examine their effects on the procurement price, planting area, supply-chain profits, and social welfare, and identify the conditions under which each policy is preferable.

The analysis yields three main findings. First, planting subsidies and firm social responsibility interact as partially substitutable production incentives. A planting subsidy induces the firm to lower its procurement price, but this crowding-out response weakens as the firm becomes more socially responsible. Once the government optimally adjusts the subsidy, stronger firm social responsibility raises the procurement price and reduces public support without changing the farmer's equilibrium planting area. Second, the two policies generate markedly different procurement-price responses. Planting subsidies allow the firm to reduce the price paid to the farmer, whereas brand building increases the value of securing agricultural output and leads to a higher procurement price. Thus, the same government objective can produce different distributions of incentives between public support and firm procurement. Third, the policy preferred from a social-welfare perspective need not coincide with that preferred by the farmer. The social-welfare threshold for brand building is lower than the farmer-profit threshold, creating a parameter region in which brand building improves aggregate welfare while planting subsidies remain more favorable to the farmer. The size of this region varies systematically with production risk, cost conditions, and firm social responsibility.

This study makes three theoretical contributions. First, we broaden the policy scope of agribusiness-led contract farming by placing production-side support and demand-side market development within the same analytical framework. Existing agricultural supply-chain research has largely compared alternative forms of subsidies or financial support. By introducing brand building as an alternative to acreage-based subsidies, we shift the policy question from how production should be subsidized to whether public resources should support production directly or create market value downstream. Second, we introduce an endogenous procurement-price channel into the comparison of agricultural policies. This allows the value created by demand-side intervention to be separated from the value ultimately captured by the farmer. Consequently, the framework distinguishes aggregate policy effectiveness from the distribution of policy benefits within the supply chain and identifies conditions under which these two criteria lead to different policy choices. Third, this study links firm social responsibility to public policy design. Prior research mainly examines how socially responsible preferences affect firms' pricing, coordination, or surplus allocation. We show that the same behavioral preference can also alter the effective boundary between alternative government interventions. This extends firm social responsibility from an internal supply-chain decision factor to a determinant of public-policy effectiveness.

The remainder of the paper is organized as follows. Section 2 reviews the literature. Section 3 presents the model framework. Section 4 derives the optimal decisions under both policies. Section 5 compares their performance. Section 6 provides numerical analysis to illustrate the main results. Section 7 discusses the theoretical and sustainable food-system implications. Section 8 concludes the paper. All proofs are provided in the Appendix.

2 Literature review

This study draws on three related streams of research: contract farming, government intervention, and corporate social responsibility. We review representative studies in these three streams and position our analysis relative to them.

2.1 Contract farming

Contract farming provides an ex-ante coordination mechanism between farmers and downstream buyers by specifying procurement conditions before agricultural production occurs. Early studies mainly focus on contract structure, bargaining power, and coordination efficiency. Alternative farmer–buyer arrangements can generate different production incentives and coordination outcomes (Niu et al., 2016). Under farmer heterogeneity and supply uncertainty, contract design also affects procurement efficiency and the allocation of production responsibilities (Federgruen et al., 2019). Chen and Chen (2021) further show that buyers' procurement commitments can significantly influence agricultural supply decisions. These studies establish contract farming as an important mechanism for aligning production with downstream demand. More recent research incorporates production uncertainty and incentive alignment into this framework. Yield uncertainty and farmer risk preferences can substantially alter planting decisions and the effectiveness of government-supported contract farming (Peng and Pang, 2019). Alternative cooperation modes likewise produce different outcomes in planting acreage and supply-chain performance when agricultural output is uncertain (Cai et al., 2021). Moving beyond static contracting, Zhang et al. (2023) consider dynamic productivity improvement and private information, highlighting the importance of intertemporal incentives. The scope of uncertainty has also expanded. By jointly considering yield, quality, and price uncertainty, Nematollahi et al. (2025) find that incentive-based contracts can improve both supply-chain performance and farmer benefits. Contract farming is also affected by the marketing environment in which farmers and buyers operate. Du and Xu (2026) compare direct farmer–e-tailer transactions with cooperative-based channels and find that the choice of channel leads to different outcomes for farmers and buyers. Weng et al. (2026) show that farmers' use of digital marketing channels depends on policy support, perceived risk, and facilitating conditions. These studies suggest that farmer–buyer relationships are shaped not only by contract terms but also by market access and channel conditions.

Overall, the literature has progressed from contract coordination to uncertainty, incentive design, and digital market access. However, downstream demand is generally treated as given. Less attention has been paid to how market value created through firm-related activities is transferred back to farmers through procurement prices. This issue motivates our analysis of brand building and firm pricing decisions.

2.2 Government intervention

Government intervention is widely used in agriculture because farmers face production uncertainty, market risk, and relatively weak bargaining positions. Early research primarily evaluates alternative forms of income and price protection. The welfare consequences of such policies differ because benefits to farmers and consumers must be weighed against public expenditure and implementation costs (Alizamir et al., 2019). Similar trade-offs arise when comparing minimum support prices with other support instruments. Chintapalli and Tang (2021, 2022) show that the preferred policy depends on fiscal constraints, implementation costs, and the welfare objective pursued by the government.

The literature has subsequently shifted from asking whether support is beneficial to asking how intervention should be designed. One important issue concerns the target of subsidies. Ye et al. (2021) compare support directed to farmers with support directed to downstream producers and find that subsidy allocation changes supply-chain outcomes. Policy instruments can also rank differently depending on the evaluation criterion. Input and output subsidies, for instance, have distinct effects on farmer profit, income disparity, and consumer surplus (Tang et al., 2024). Huang and Cao (2025) compare farmer and retailer subsidy schemes in agricultural supply chains and show that their effectiveness varies with farm size, fiscal constraints, and farmer risk preferences. A similar lack of universal dominance appears in agricultural supply-chain finance: the relative performance of interest and direct subsidies depends on production characteristics and market conditions (Yi et al., 2025). Government intervention is not limited to direct monetary transfers. Recent studies increasingly examine policies that change farmers' operating environments or support conditions. Ray et al. (2025) compare cost subsidies with indirect support under yield uncertainty and farmer heterogeneity and show that the preferred intervention depends on implementation conditions and farmer responses. Government support can also affect operational decisions such as freshness investment, pricing, and supply-chain coordination (Zhang et al., 2026). Taken together, these studies suggest that policy effectiveness is mechanism-specific rather than universal.

Existing work nevertheless remains concentrated on production-oriented or financial support. Our study instead compares two interventions that operate through different sides of the market. A planting subsidy directly changes farmers' production incentives, whereas brand building increases market value through demand expansion. The benefits of the latter can reach farmers only if the firm transfers part of the additional return through the procurement price. This distinction raises a broader policy question that has received limited attention: should governments stimulate agricultural production directly or create market value downstream, especially when the policy preferred from a social-welfare perspective may differ from that preferred by farmers?

2.3 Corporate social responsibility

Corporate social responsibility research relaxes the conventional assumption that firms maximize only their own profits. Responsible supply-chain decisions may incorporate supplier welfare, disadvantaged producers, environmental outcomes, or broader social value. Supply chains involving low-income suppliers provide an important setting in which commercial and social objectives interact (Sodhi and Tang, 2014). CSR effort can also respond strategically to market conditions; competition among purchasing firms, for example, affects their incentives to undertake socially responsible activities (Shi et al., 2020).

A related stream captures such behavior through altruistic and fairness preferences. When a dominant retailer attaches value to another supply-chain member's payoff, distributional outcomes and coordination efficiency can change (Wang et al., 2021). These behavioral considerations also interact with public intervention. In poverty-alleviation supply chains, government subsidies and fairness concerns jointly affect decentralized decisions and coordination (Kang et al., 2021). More directly related to the agricultural setting of this study, Yu et al. (2024) model a retailer with altruistic preferences toward a farmer cooperative and show that such preferences influence purchasing decisions, farmer benefits, and supply-chain coordination. Recent studies have also examined CSR in platform-based supply chains. Yi et al. (2023) conceptualize ecosystem social responsibility, under which a platform considers the welfare of participants in its wider ecosystem. In agricultural platform financing, social responsibility concerns affect farmer decisions and stakeholder outcomes (Wang et al., 2023). Ma et al. (2024) and Xiao et al. (2023) further show that platform CSR can change pricing and sales-mode decisions. Although these studies focus on platforms, they support a broader point: the social responsibility of a powerful downstream actor can influence pricing, stakeholder outcomes, and the distribution of supply-chain gains.

Existing research therefore shows that social responsibility can reshape pricing, coordination, and surplus allocation. Its role in government policy choice, however, remains less clear. In agribusiness-led contract farming, the firm sets the procurement price, which determines how market gains are shared with farmers. A stronger concern for farmer welfare may raise the procurement price and partly substitute for government production support. It may also affect how much of the value created by brand building reaches farmers. The firm's concern for farmer welfare may therefore change the relative performance of planting subsidies and brand building.

2.4 Research gap and positioning

The three streams point to a common gap. Contract-farming research explains coordination and risk allocation but generally treats downstream demand as exogenous. Agricultural-policy studies show that policy performance depends on the instrument and welfare criterion, yet most comparisons remain among production-side or financial support mechanisms. CSR research shows that responsible firms can alter pricing and value distribution, but rarely examines whether CSR changes the relative performance of alternative government policies.

To address these issues, we develop a unified government–firm–farmer framework with stochastic yield. The firm is an agribusiness firm that purchases agricultural products from the farmer and sells them in the downstream market. The government can use either a planting subsidy, which directly affects production incentives, or brand building, which expands market demand. The firm sets the procurement price and places a positive weight on farmer profit, representing its CSR concern. This framework allows us to compare the two policies in terms of social welfare and farmer profit and to identify when the socially preferred policy is not the one that benefits the farmer more. Table 1 compares our framework with several closely related studies.

Table 1. Comparison of this study with closely related literature.

Table 1 shows that existing studies have examined agricultural subsidies, yield uncertainty, agricultural branding, and CSR, but these issues are usually studied separately. Most policy comparisons remain within production-side or financial support mechanisms. Xu et al. (2024) examine government support for regional agricultural brands but do not compare brand building with a production-side policy. This study brings the two policy approaches together in an agribusiness-led contract-farming supply chain and incorporates CSR into the comparison. By distinguishing between social welfare and farmer profit, we identify two policy thresholds and a goal-conflict region in which the policy generating higher social welfare does not provide the farmer with higher profit.

3 Model setting

We consider an agricultural supply chain consisting of a government, an agricultural firm, and a single farmer. The single farmer represents an aggregate production unit, such as a farmer cooperative or a group of homogeneous contract farmers, rather than a literal individual farmer.

The government intervenes in the agricultural supply chain with the objective of maximizing social welfare. The firm and the farmer engage in the production and sale of a single agricultural product. All model parameters and relevant information are assumed to be common knowledge, and information asymmetry is therefore excluded. This study focuses on how planting subsidies and brand building affect contract farming. We first describe the basic structure of the contract-farming arrangement and then introduce the two government intervention policies.

Under contract farming, the firm and the farmer sign a procurement contract before the planting season. The contract specifies a procurement price w, at which the firm agrees to purchase the farmer's entire harvest during the harvesting season. After the contract is signed, the farmer determines the planting area q. The farmer incurs a planting cost 12cFq2, where c_*F* denotes the planting cost coefficient. Agricultural production is inherently exposed to natural uncertainties, including weather conditions, pests, and diseases. We therefore model the yield per unit of planting area, denoted by u, as a random variable. Let G(u) and g(u) denote its cumulative distribution function and probability density function, respectively, with Eu = μ, Du = σ², where μ is the expected yield and σ is the standard deviation of yield, which serves as the yield-risk parameter. Accordingly, the realized quantity procured by the firm is qu. After cleaning and processing the agricultural product, the firm sells the entire quantity in the market at price p(qu), which decreases with the realized market supply qu. The inverse demand function can be expressed as p = a−βqu, where a represents the market size of the agricultural product.

Under the planting subsidy policy, the government provides a subsidy s for each unit of planting area. The farmer therefore receives a total planting subsidy sq. Under the brand-building policy, the government undertakes brand-building activities to expand the market for the agricultural product. The government determines the brand-building level φ and incurs a corresponding cost 12cGφ2, where c_*G* is the brand-building cost coefficient. Brand building enhances the market potential of the agricultural product. Accordingly, under the brand-building policy, the market price is given by p = a − βqu + bφ, where a denotes the market size, β measures the sensitivity of the market price to agricultural output, and b captures the effect of brand building on the market price. We define the brand-building cost-effectiveness index as κ=b2cG. Thus, a higher κ may result either from greater brand-building effectiveness or from a lower brand-building cost. Following Alizamir et al. (2019), consumer surplus is expressed as CS=12βq2(μ2+ σ2).

It is important to clarify how the market value generated by brand building reaches the farmer. Brand building does not directly transfer government expenditure to the farmer. Instead, its effect operates through an endogenous value-transmission mechanism. A higher brand-building level increases downstream market potential and raises the marginal value of agricultural output to the firm. The firm consequently adjusts its procurement price w, through which part of the additional downstream value is transmitted upstream to the farmer. The farmer then adjusts the planting area in response to the procurement price. This pass-through is neither assumed to be automatic nor complete. Its extent depends on the firm's optimal pricing decision and, in particular, on its social responsibility concern r. A firm that places greater weight on farmer profit has a stronger incentive to transfer part of the additional market value upstream through a higher procurement price.

Agricultural yield and market prices are economically required to be non-negative. Accordingly, the analysis is restricted to the feasible region in which u≥0 and the market price remains non-negative under both policy regimes. Specifically, a − βqu ≥ 0 under the planting subsidy policy and a − βqu + bφ ≥ 0 under the brand-building policy. Although the normal distribution is adopted as a tractable approximation for stochastic yield, it is interpreted over the economically relevant range of non-negative agricultural yield. All subsequent equilibrium results, comparative-static results, and policy comparisons are derived and interpreted subject to these feasibility conditions.

Consistent with Liu et al. (2019), we consider a socially responsible firm that places a positive value on the farmer's profit. Rather than maximizing its own expected profit alone, the firm maximizes the following expected utility: U = π^*C*+rπ^*F*, where π^*C* and π^*F* denote the expected profits of the firm and the farmer, respectively. The parameter r(0 < r < 1) measures the firm's social responsibility concern toward the farmer. A larger r indicates that the firm assigns greater importance to the farmer's profit when making its decisions.

Figure 1 captures the timing of strategic decisions and operational activities in the contract farming mechanism. The government moves first by selecting policy interventions, followed by the firm's contract offering and procurement price decision. The farmer then chooses the planting area under the announced contract terms. After yield realization, the firm procures the harvested output and sells it in the market.

Throughout the paper, subscripts PS and BB denote the planting subsidy policy and the brand-building policy, respectively, while superscripts F, C, and G refer to the farmer, the firm, and the government. Accordingly, πPSF and πBBF denote the farmer's expected profits under the planting subsidy and brand-building policies, respectively, whereas πPSCand πBBC denote the corresponding expected profits of the firm. Similarly, SW_*PS* and SW_*BB* represent social welfare under the planting subsidy and brand-building policies, respectively. Table 2 summarizes the main notation used throughout the analysis.

Table 2. Notation and definitions.

4 Model analysis

4.1 Planting subsidy policy

Under the planting subsidy policy, the firm purchases the farmer's entire harvest at the procurement price specified in the contract, while the government provides an acreage-based subsidy to encourage agricultural production. Accordingly, the farmer's expected profit is given by Equation 1:

πPSF(q)=E[wqu+qs-12cFq2](1)

**Theorem 1.**Under the planting subsidy policy, the farmer's optimal planting area is given by Equation 2:

qPS*=wμ+scF(2)

The firm purchases the agricultural product from the farmer and resells the realized output in the market. Its expected profit therefore consists of sales revenue net of procurement expenditure and is given by Equation 3:

πPSC(w)=E[(a-βqu)qu-wqu](3)

As specified in Section 3, the firm is socially responsible and takes the farmer's profit into account when making its procurement decision. Specifically, the firm maximizes the expected utility UPSC=r πPSF+πPSC, where πPSC and πPSF denote the expected profits of the firm and the farmer, respectively, and r measures the firm's social responsibility concern toward the farmer. The firm's expected utility is given by Equation 4:

UPSC=r πPSF+πPSC=aqμ-βγq2-wqμ+r(wqμ+qs-12cFq2)(4)

where γ = Eu² = μ²+ σ².

**Theorem 2.**Under the planting subsidy policy, the firm's optimal procurement price that maximizes its expected utility is given by Equation 5:

wPS*=-2sβγ+((-1+r)s+aμ)cF2βγμ+(2-r)μcF(5)

It follows that ∂wPS(s)∂s<0 and ∂wPS(s)∂r>0. Therefore, Theorem 2 yields an intuitive implication. A higher planting subsidy induces the firm to reduce the procurement price, indicating that part of the government support is offset by the firm's pricing response. However, this crowding-out effect becomes weaker as the firm places greater weight on the farmer's profit. Thus, firm social responsibility moderates the negative effect of the planting subsidy on the procurement price.

The government chooses the planting subsidy to maximize social welfare. Social welfare consists of four components: the farmer's expected profit πPSF, the firm's expected profit πPSC, consumer surplus 12βγq2, and government subsidy expenditure sq. Accordingly, the government's objective function is given by Equation 6:

SWPS=πPSF+πPSC+12βγq2-sq=aqμ-12βγq2-12cFq2(6)

In practice, the subsidy and procurement price must satisfy the feasibility conditions imposed in the model (s≥0, w>0). These constraints are incorporated when solving the government's optimization problem so that the resulting equilibrium remains economically meaningful.

**Theorem 3.**Under the planting subsidy policy, the government's optimal planting subsidy that maximizes expected social welfare is given by Equation 7:

sPS*=aμ(βγ+(1-r)cF)βγ+cF(7)

subject to cF>cF0= βγr.

When the parameters satisfy c_*F*>c_*F*0, the optimal government decision simultaneously satisfies the subsidy and procurement-price constraints. These conditions rule out economically implausible outcomes, such as a subsidy exceeding the effective planting cost or a negative procurement price. Unless otherwise stated, the subsequent analysis of the planting subsidy policy is conducted under these feasibility conditions.

**Theorem 4.**Under the planting subsidy policy, the equilibrium planting area and procurement price are given by Equations 8, 9, respectively:

qPS*=aμcF+βγ(8)

wPS*=a(rcF-βγ)βγ+cF(9)

A notable result is that the farmer's equilibrium planting area is independent of the firm's social responsibility concern. This result reflects the coordinating role of the planting subsidy. Because the government chooses the subsidy from the perspective of social-welfare maximization, it can adjust the subsidy in response to changes in the firm's procurement decision. Consequently, stronger firm social responsibility changes the division of production incentives between the government and the firm, but does not alter the farmer's equilibrium planting area. In this sense, government production support and firm social responsibility are partially substitutable.

**Corollary 1.**

∂sPS∂a>0, ∂sPS∂r<0, ∂sPS∂cF<0, ∂sPS∂μ>0, ∂sPS*∂σ> 0.

Corollary 1 shows that the optimal planting subsidy increases with market size. As the market expands, a larger quantity of agricultural products can be absorbed by demand. A welfare-maximizing government therefore has a stronger incentive to stimulate production by increasing the subsidy.

By contrast, the optimal subsidy decreases as the firm's social responsibility concern increases. A more socially responsible firm assigns greater importance to the farmer's profit and consequently offers a higher procurement price. This strengthens the farmer's production incentive even in the absence of additional government support. The government therefore reduces the planting subsidy, revealing a substitution effect between firm social responsibility and public production support.

The planting subsidy also decreases with the planting cost coefficient. A higher planting cost implies a greater production burden and reduces the efficiency with which public expenditure can be translated into agricultural supply. From a social-welfare perspective, the government consequently becomes less willing to expand the subsidy when production is particularly costly.

The effects of expected yield and yield risk operate differently. The government provides greater support when expected yield is higher and also increases the subsidy as yield risk rises. The latter result reflects the role of planting subsidies in stabilizing agricultural production under uncertainty: when farmers face greater yield risk, stronger government support is required to sustain production incentives and maintain agricultural supply.

**Corollary 2.**

∂wPS∂a>0, ∂wPS∂r>0, ∂wPS∂cF>0, ∂wPS∂μ<0, ∂wPS*∂σ< 0.

The equilibrium procurement price increases with market size. When market demand is larger, the firm has a stronger incentive to secure agricultural supply and therefore offers a higher procurement price to encourage the farmer to expand production. The procurement price also increases with the firm's social responsibility concern. As the firm places greater weight on the farmer's profit, it transfers a larger share of the supply-chain surplus upstream through a higher procurement price.

A higher planting cost coefficient likewise induces the firm to increase the procurement price. When agricultural production becomes more costly, the farmer's willingness to plant declines. The firm responds by offering a higher procurement price to maintain an adequate supply. In contrast, the procurement price decreases with expected yield and yield risk. A higher expected yield increases the quantity obtained from a given planting area, reducing the need for the firm to stimulate additional planting through a high procurement price. Greater yield risk also leads the firm to adopt a more conservative procurement strategy, thereby lowering the equilibrium procurement price.

**Corollary 3.**

∂qPS∂a>0, ∂qPS∂cF<0, ∂qPS∂σ<0, ∂qPS∂μ> 0.

The equilibrium planting area increases with market size because a larger market raises both the procurement price and the government's optimal subsidy, thereby strengthening the farmer's incentive to plant. In contrast, a higher planting cost coefficient suppresses production and leads to a smaller planting area. Yield risk also reduces the equilibrium planting area. Greater uncertainty lowers the attractiveness of agricultural production and, through the firm's pricing response, further weakens the farmer's incentive to expand planting.

The effect of expected yield is more subtle and depends on the parameter conditions imposed by the model. Ordinarily, a higher yield per unit of area increases output from a given planting area and may therefore reduce the need to expand acreage. Under the feasible parameter region considered here, however, planting costs are sufficiently high that agricultural supply remains constrained. In such circumstances, an increase in expected yield improves the production potential without immediately creating excessive supply, and an expansion in planting area can still be sustained by market demand. This result suggests that planting subsidies are particularly effective in supporting production when farmers face relatively high planting costs.

4.2 Brand-building policy

Under the brand-building policy, the government invests in agricultural branding to expand market demand and enhance the market value of the agricultural product. The government, the firm, and the farmer make sequential decisions, and the equilibrium is derived by backward induction.

Given the procurement price offered by the firm, the farmer chooses the planting area. The farmer's expected profit under the brand-building policy is given by Equation 10:

πBBF(q)=E[wqu-12cFq2](10)

Following the same optimization procedure as under the planting subsidy policy, we obtain the farmer's optimal planting decision.

**Theorem 5.**Under the brand-building policy, the farmer's optimal planting area is given by Equation 11:

qBB*=1cFwμ(11)

Government-supported brand building expands market demand and allows the agricultural product to be sold at a higher market price. The firm earns sales revenue from the realized output while incurring the corresponding procurement expenditure. Its expected profit is given by Equation 12:

πBBC(w)=E[(a-βqu+bφ)qu-wqu](12)

As under the planting subsidy policy, the firm incorporates the farmer's profit into its objective function through its social responsibility concern.

**Theorem 6.**Under the brand-building policy, the firm's optimal procurement price that maximizes expected utility is given by Equation 13:

wBB*=cFμ(a+bϕ)2cF(1-r)μ+(cFr+2βγ)μ(13)

Under the brand-building policy, social welfare consists of the farmer's expected profit πBBF, the firm's expected profit πBBC, consumer surplus 12βγq2, and the government's brand-building expenditure 12cGφ2. Hence,

SWBB=πBBF+πBBC+12βγq2-12cGφ2(14)

Substituting the farmer's and firm's optimal decisions into Equation 14 yields the government's social-welfare function. Let cG>cG0=μ2b2(3βγ+(3-2r)cF)(2βγ+(2-r)cF)2. The optimal brand-building level can then be obtained.

**Theorem 7.**Under the brand-building policy, the government's optimal brand-building level that maximizes social welfare is given by Equation 15:

φBB*=abψ(2βγ+(2-r)cF)2cG-b2ψ(15)

where ψ=μ2(3βγ+(3- 2r)cF).

**Theorem 8.**Under the brand-building policy, the equilibrium planting area and procurement price are given by Equations 16, 17, respectively:

qBB*=aμ(2βγ+(2-r)cF)cG(2βγ+(2-r)cF)2cG-b2ψ(16)

wBB*=acF(2βγ+(2-r)cF)cG(2βγ+(2-r)cF)2cG-b2ψ(17)

**Corollary 4.**

∂φBB∂a>0, ∂φBB∂b>0, ∂φBB∂cF<0, ∂φBB∂cG<0, ∂φBB*∂σ< 0.

Corollary 4 shows that the government's optimal brand-building level increases with market size. When the underlying market is larger, brand-building investment can generate greater market returns, making the government more willing to strengthen its brand-building effort.

In contrast, the optimal brand-building level decreases with both the brand-building cost coefficient and the planting cost coefficient. A higher brand-building cost directly reduces the return from public investment. Meanwhile, a higher planting cost weakens the farmer's ability to respond to the additional demand created by brand building. In both cases, the effectiveness of downstream market expansion is reduced, leading the government to choose a lower brand-building level.

The optimal brand-building level increases with the effectiveness of branding. When brand-building activities have a stronger effect on market demand, a given level of government expenditure creates greater market value, encouraging the government to invest more heavily in brand development.

Yield risk has the opposite effect. As production uncertainty increases, the agricultural supply chain becomes less able to convert market expansion into stable realized output. Since the government must still bear the cost of brand building, greater yield risk reduces the attractiveness of further demand-side investment and leads to a lower optimal brand-building level.

To facilitate the interpretation of the following comparative-static results, we introduce three critical brand-building cost thresholds, denoted by c_*G*1, c_*G*2, and c_*G*3. Specifically, c_*G*1 determines whether an increase in the planting cost raises or lowers the procurement price, while c_*G*2 distinguishes the conditions under which expected yield has a positive or negative effect on the procurement price. Together with the relevant planting-cost condition, c_*G*3 identifies the parameter regions in which expected yield has different effects on the equilibrium planting area. Economically, these thresholds reflect the balance between the demand-expansion effect of brand building and the production-incentive effect of the procurement price. Their exact expressions and derivations are provided in Appendix A.

**Corollary 5.**

(1) ∂wBB∂a>0, ∂wBB∂b>0, ∂wBB∂cG<0, ∂wBB∂σ< 0.

(2) If c_*G*0 < c_*G* < c_*G*1, then ∂wBB*∂cF<0; if c_*G* ≥ c_*G*1, then ∂wBB*∂cF> 0.

(3) If c_*G*0 < c_*G* < c_*G*2, ∂wBB*∂μ>0; if c_*G* ≥ c_*G*2, ∂wBB*∂μ< 0.

The procurement price increases with market size. A larger market raises the potential return from agricultural sales and strengthens the firm's incentive to secure more supply. The firm therefore offers a higher procurement price to encourage greater production.

The procurement price also rises as brand-building effectiveness improves and falls as the cost of brand building increases. More effective brand building induces the government to undertake stronger brand-building efforts, expands market demand, and increases the value of additional agricultural supply. The firm consequently raises its procurement price. Conversely, a higher brand-building cost discourages government investment, weakens demand expansion, and results in a lower procurement price.

Greater yield risk lowers the procurement price. As production uncertainty increases, the government reduces its brand-building effort, while the firm becomes more cautious about committing to a high procurement price under uncertain supply conditions.

The effect of the planting cost coefficient is conditional. A higher planting cost directly weakens the farmer's production incentive, which gives the firm an incentive to raise the procurement price. At the same time, however, higher planting costs reduce the government's optimal brand-building effort and thereby weaken market demand. When the demand-side effect dominates, the procurement price decreases with the planting cost coefficient. When the production-incentive effect dominates, the firm instead raises the procurement price to secure sufficient supply.

The effect of expected yield is similarly non-monotonic. In the absence of brand building, a higher expected yield would normally reduce the need for the firm to stimulate production through a high procurement price. Under the brand-building policy, however, expected yield also affects the government's optimal branding decision. When brand building is relatively inexpensive, the government responds to higher expected yield by increasing brand-building effort, and the resulting market expansion can dominate the direct supply effect. In this case, the procurement price increases with expected yield. When brand-building costs are sufficiently high, the demand-expansion channel is weaker and the direct supply effect dominates, causing the procurement price to decrease with expected yield.

**Corollary 6.**

(1) ∂qBB∂a>0, ∂qBB∂b>0, ∂qBB∂cG<0, ∂qBB∂σ<0, ∂qBB*∂cF< 0.

(2) If cF0<cF<2βγr and c_*G*3 < c_*G*, then ∂qBB∂μ<0; if cF0<cF<2βγr and _{c_G*0 < cG} < c_*G*3, then ∂qBB∂μ>0; if cF≥2βγr and _{c_G*0 < cG}, then ∂qBB*∂μ> 0.

The equilibrium planting area decreases with the planting cost coefficient. Higher production costs weaken the farmer's incentive and ability to expand planting. Yield risk also reduces the planting area because greater production uncertainty lowers the attractiveness of agricultural production and induces the firm to offer a lower procurement price.

By contrast, a larger market and more effective brand building increase the planting area. Both factors enhance the market return from agricultural production and strengthen the incentives transmitted upstream to the farmer. A higher brand-building cost reduces government branding effort and therefore leads to a smaller planting area.

The relationship between expected yield and planting area depends on planting costs and the cost of brand building. When the government's brand-building cost is relatively low, a higher expected yield induces stronger brand-building effort and a higher procurement price, which can stimulate the farmer to expand planting. When brand building is costly, this market-expansion effect becomes weaker, and the direct effect of higher productivity dominates, reducing the acreage required to supply the market.

5 Comparative analysis

**Corollary 7.**

The procurement price under the brand-building policy is always higher than that under the planting subsidy policy, i.e., wPS<wBB.

The intuition follows from the different mechanisms through which the two policies operate. Under the planting subsidy policy, government support directly strengthens the farmer's production incentive. Because the farmer receives an acreage-based subsidy, the firm can reduce the procurement price while still maintaining agricultural production. Moreover, the expansion of planting increases potential supply and reinforces the firm's incentive to lower the procurement price to limit the risk of excessive output and declining market prices.

Brand building operates differently. By expanding demand and increasing the market value of the agricultural product, it raises the firm's marginal benefit from securing additional supply. The firm therefore has a stronger incentive to increase the procurement price and encourage the farmer to expand production. Consequently, demand-side brand building generates a systematically higher procurement price than production-side planting subsidies.

**Corollary 8.**

There exists a farmer-profit threshold, denoted by κ_*F*, such that when κ ≥ κ_*F*, then πBBF≥πSF, qBB≥qPS; when κ < κ_*F*, the opposite ranking holds. Moreover κ_*F* satisfies ∂κF∂σ>0, ∂κF∂r<0, ∂κF∂β>0, ∂κF∂a= 0.

Corollary 8 shows that brand building does not necessarily make the farmer better off than a planting subsidy. Brand building produces a higher planting area and greater farmer profit only when its cost-effectiveness exceeds the threshold κ_*F*. Below this threshold, the demand expansion generated by brand building is insufficient to transmit enough additional value upstream, whereas a planting subsidy directly strengthens the farmer's production incentive and therefore provides greater benefits.

Yield risk raises κ_*F*, indicating that production uncertainty makes it more difficult for demand-side market expansion to translate into higher farmer profit. Consequently, brand building must become more cost-effective before it can outperform a planting subsidy from the farmer's perspective.

In contrast, stronger firm social responsibility reduces κ_*F*. A firm that places greater weight on farmer profit is more willing to transfer the value created by brand building upstream through a higher procurement price. Firm social responsibility therefore strengthens the transmission of market gains to the farmer. Market size does not affect κ_*F*. Although a larger market increases farmer profit under both policies, it does not change their relative ranking.

**Corollary 9.**

There exists a social-welfare threshold, denoted by κ_*SW*, such that when κ ≥ κ_*SW*, then SW_*BB* ≥ SW_*PS*; when κ < κ_*SW*, then SW_*BB*<SW_*PS*. The threshold κ_*SW* satisfies ∂κSW∂σ>0, ∂κSW∂r<0, ∂κSW∂a=0, ∂κSW∂β> 0.

Corollary 9 shows that the welfare advantage of brand building depends on whether its cost-effectiveness is sufficiently high. When the market value created through brand building is large enough to compensate for government expenditure, the brand-building policy generates higher social welfare. Otherwise, the planting subsidy is more effective because it directly stimulates agricultural production.

Yield risk raises the social-welfare threshold, implying that production uncertainty weakens the welfare gains from demand-side intervention. Stronger firm social responsibility, in contrast, lowers the threshold because the firm's concern for farmer profit improves the transmission of demand-side gains into production incentives and supply-chain outcomes.

Market size does not affect the social-welfare threshold. It increases the absolute level of social welfare under both policies but leaves their relative ranking unchanged. Thus, the welfare advantage of brand building is governed by its cost-effectiveness and the production environment rather than by market size per se.

Corollaries 8 and 9 identify the conditions under which brand building outperforms planting subsidies from two different perspectives: farmer profit and social welfare. Examining these thresholds separately, however, does not reveal whether an improvement in aggregate welfare is necessarily accompanied by an improvement in farmer profit. The two policy rankings may not always coincide. We therefore compare κ_*SW* and κ_*F* and incorporate the feasible upper bound of brand-building cost-effectiveness to characterize how policy preference changes across different parameter regions.

**Corollary 10.**

(1) When 0 < κ ≤ κ_*SW*, SW_*BB* ≤ SW_*PS* and πBBF<πPSF. The planting subsidy policy therefore dominates brand building in terms of both social welfare and farmer profit.

(2) When _{κ_*SW* < κ ≤ κF}, SW_*BB*>SW_*PS* whereas πBBF≤πPSF. In this region, the social-welfare and farmer-profit rankings favor different policies.

(3) When _{κ_*F* < κ < κmax}, SW_*BB*>SW_*PS* and πBBF>πPSF. Brand building therefore dominates the planting subsidy policy in terms of both social welfare and farmer profit.

Corollary 10 reveals a three-stage transition in policy preference as the cost-effectiveness of brand building improves: a planting-subsidy-dominant region, a goal-conflict region, and a brand-building-dominant region. Because the social-welfare threshold is strictly lower than the farmer-profit threshold, brand building first becomes desirable from an aggregate-welfare perspective and only later becomes preferable from the farmer's perspective. When brand-building cost-effectiveness lies between the two thresholds, its demand-expansion benefits are already sufficient to improve overall social welfare, but these gains have not yet been transmitted upstream strongly enough to make the farmer better off than under a planting subsidy.

This intermediate region therefore represents a goal conflict: aggregate social welfare is higher, whereas the farmer would still prefer the production-side subsidy. Only after brand-building cost-effectiveness exceeds the farmer-profit threshold can the policy simultaneously increase social welfare and farmer profit. Hence, the welfare-maximizing government chooses brand building once κ exceeds κ_*SW*. However, when κ lies between κ_*SW* and κ_*F*, this choice leaves the farmer worse off than under the planting subsidy, revealing a distributional tension that aggregate welfare alone does not capture.

**Corollary 11.**

Let Δκ = κ_*F*−κ_*SW* denote the width of the goal-conflict region. Then:

(1) ∂Δκ∂σ>0, implying that greater yield risk widens the goal-conflict region.

(2) ∂Δκ∂r<0, indicating that stronger firm social responsibility narrows the goal-conflict region.

(3) ∂Δκ∂β>0, which means that greater price sensitivity widens the goal-conflict region.

(4) ∂Δκ∂cF>0. Thus, higher planting costs widen the goal-conflict region. When 0<r≤34, both κ_*SW* and κ_*F* increase with c_*F*, causing the goal-conflict region to shift to the right and expand. When r>34, the effect of c_*F* on κ_*SW* depends on the other model parameters, but Δκ still increases with c_*F*.

(5) ∂Δκ∂μ<0, indicating that a higher expected yield narrows the goal-conflict region. When 0<r≤34, both thresholds decrease with μ, so the goal-conflict region shifts to the left and contracts. Under r>34, the effect of μ on κ_*SW* is conditional, but the width of the goal-conflict region still decreases.

Corollary 11 further shows that the extent to which the market gains generated by the brand-building policy are transmitted to the farmer depends critically on production conditions and supply-chain characteristics. Greater yield risk, stronger price sensitivity, and higher planting costs widen the gap between the two thresholds. Under these conditions, the brand-building policy may already generate higher social welfare while still failing to provide the farmer with greater profit than the planting subsidy policy.

A higher expected yield has the opposite effect. Greater productivity makes it easier to convert the additional demand created by the brand-building policy into realized supply and farmer profit, thereby narrowing the gap between the social-welfare and farmer-profit thresholds. Firm social responsibility plays a similar role. As the firm places greater weight on farmer profit, a larger share of the market value created by brand building is transferred upstream through the procurement price, reducing the range over which social welfare improves without a corresponding increase in farmer profit.

Hence, the goal-conflict region is not an inherent consequence of the brand-building policy. Rather, it arises from the interaction among the effectiveness of brand building, agricultural production conditions, and the mechanism through which market gains are distributed within the supply chain.

6 Numerical analysis

Sichuan is one of China's major tea-producing regions, where local governments have adopted both production-side and demand-side measures to support the development of the tea industry. These practices provide an appropriate setting for examining the planting subsidy policy and the brand-building policy considered in this study.

The Sichuan tea industry is used as a field-informed illustrative setting rather than as a fully calibrated empirical case study. The benchmark yield parameters are informed by field observations collected in 2025 from several dozen local tea farmers and tea farms in Ya'an, Sichuan Province. Yield per unit planting area was measured in metric tons per mu (mu is a customary Chinese unit of land area; 1 mu ≈ 0.0667 hectares), and the observations were statistically summarized using SPSS to characterize the central yield level and its dispersion. Accordingly, we use μ = 1.10 tons/mu and σ = 0.15 tons/mu as the benchmark expected yield and yield-risk parameters, respectively. For the remaining parameters, directly comparable field measurements were not available. These parameters are therefore normalized benchmark values selected within the theoretical feasibility region. This distinction reflects the different nature of the model parameters: μ and σ characterize directly observable agricultural yield, whereas the remaining parameters are structural coefficients that cannot be directly identified from the present field survey. This mixed parameterization is also consistent with related analytical agricultural supply-chain studies, in which empirically or literature-informed inputs are combined with selected benchmark values for numerical illustration (Peng and Pang, 2019; Tang et al., 2024). Accordingly, the numerical analysis is intended primarily to illustrate the analytical mechanisms and comparative-static results rather than to reproduce the full economic scale of the Sichuan tea industry. Specifically, we set a = 10000, c_*F* = 100, c_*G* = 0.5, b = 5.6, β = 0.01, r = 0.1. Detailed definitions, units, sources, and determination procedures for all benchmark parameters are reported in Appendix B, Table B1.

To examine how the cost-effectiveness of brand building affects the relative performance of the two policies, Figure 2 presents the ratios of social welfare, farmer profit, and planting area under the brand-building policy (BBP) to those under the planting subsidy policy (PSP). Specifically, SWR, FRR, and PAR denote the social-welfare ratio SWBBSWPS, farmer-profit ratio πBBFπPSF, and planting-area ratio qBBqPS, respectively. The horizontal line (Ratio of BBP to PSP) y = 1 indicates equal performance under the two policies.

6.1 Effect of the brand-building cost-effectiveness index κ

Figure 2 shows that SWR, FRR, and PAR all increase with κ, although they cross the benchmark line y = 1 at different points. When 0 < κ ≤ κ_*SW*, none of the three ratios exceeds 1. At this stage, the demand expansion generated by the brand-building policy is insufficient to offset its implementation cost, and the planting subsidy policy performs better in terms of both social welfare and farmer profit. When _{κ_*SW* < κ ≤ κF}, SWR rises above 1, while FRR and PAR remain below 1. Thus, the brand-building policy already generates higher social welfare, but its additional market value has not yet been sufficiently transmitted upstream to the farmer. This interval therefore corresponds to the goal-conflict region identified in Corollary 10, where social-welfare improvement and farmer income growth favor different policies. When _{κ_*F* < κ < κmax}, all three ratios exceed 1. The brand-building policy then outperforms the planting subsidy policy in social welfare, farmer profit, and planting area. Notably, FRR and PAR cross the benchmark line at the same threshold because, in equilibrium, farmer profit is proportional to the square of the planting area. As κ increases further, FRR grows faster than PAR, suggesting that improved brand-building cost-effectiveness not only stimulates production but also strengthens its effect on farmer profit.

Overall, the numerical results are consistent with Corollaries 8–10 and illustrate the transition in policy advantage as κ increases. The planting subsidy policy is preferable when brand building is relatively inefficient. At intermediate levels of κ, brand building improves social welfare without necessarily making the farmer better off. Only after κ exceeds the farmer-profit threshold does the brand-building policy achieve both higher aggregate welfare and greater farmer benefit. This result highlights the importance of whether the value created through market expansion is effectively transmitted to the production side.

6.2 Effect of the planting cost coefficient c_**F**

Figure 3 shows that, as c_*F* increases, planting area, farmer profit, firm profit, and social welfare decline under both policies. Higher planting costs therefore weaken the overall performance of the contract-farming supply chain. The decline is more pronounced under the brand-building policy. This is because brand building creates value primarily through demand expansion, and its effectiveness depends on the farmer's ability to respond by increasing production. As planting becomes more costly, this supply response is constrained, making it more difficult for the additional market value created by brand building to reach the production side. By contrast, the planting subsidy policy directly improves the farmer's production incentive and is therefore less sensitive to rising planting costs.

The policy rankings also change at different values of c_*F*. As planting costs increase, the planting area and farmer profit under the brand-building policy fall below those under the planting subsidy policy before the social-welfare ranking reverses. Hence, over an intermediate range of planting costs, the brand-building policy can still generate higher social welfare even though the farmer is already better off under the planting subsidy policy. This again reflects a divergence between aggregate welfare and farmer profit. Brand building is therefore more advantageous when planting costs are relatively low, whereas the relative advantage of the planting subsidy policy becomes stronger as production costs rise.

6.3 Effect of expected yield per unit area μ

Under the baseline parameter setting, Figure 4 shows that a higher expected yield per unit area μ increases planting area, farmer profit, firm profit, and social welfare under both policies. Greater productivity allows a given planting area to generate more marketable output, expanding the profit opportunities of both the farmer and the firm while enabling a larger share of market demand to be satisfied. The increase is more pronounced under the brand-building policy, indicating a stronger complementarity between agricultural productivity and demand expansion. When expected yield is higher, the additional demand created by brand building can be converted more effectively into realized sales, thereby amplifying its contribution to supply-chain performance.

The crossover points further reveal that the policy ranking in social welfare changes before those in farmer profit and planting area. At relatively low levels of μ, the planting subsidy policy provides stronger production incentives and therefore yields a larger planting area and higher farmer profit. As μ increases, the brand-building policy first overtakes the planting subsidy policy in social welfare, while farmer profit remains lower, giving rise to the goal-conflict region. Once expected yield increases further, the planting area and farmer profit under the brand-building policy also exceed those under the planting subsidy policy.

These results show that expected yield affects not only the absolute performance of the supply chain but also the relative attractiveness of the two policies. Higher productivity makes it easier to translate demand expansion into realized supply and farmer profit, thereby narrowing the gap between social-welfare improvement and farmer benefit. This numerical pattern is consistent with Corollary 11, which shows that a higher μ reduces the width of the goal-conflict region.

6.4 Sensitivity of policy thresholds to yield risk and firm social responsibility

Table 3 reports the sensitivity of the two policy thresholds and the width of the goal-conflict region to yield risk and firm social responsibility. When σ increases from 0.05 to 0.25, both κ_*SW* and κ_*F* increase, and Δκ increases monotonically. The quantitative change in the width of the goal-conflict region is relatively modest around the benchmark parameterization, but its direction is fully consistent with Corollary 11. This suggests that, within the range considered around the field-informed benchmark, yield risk affects the location and width of the policy thresholds in the theoretically predicted direction, although the magnitude of this effect is limited.

Table 3. Sensitivity of the policy thresholds and the goal-conflict region to yield risk and firm social responsibility.

Δκ = κ_*F* − κ_*SW* denotes the width of the goal-conflict region. The relative change is calculated as (Δκ − Δκ₀) /Δκ₀ × 100%, where Δκ₀ denotes the value under the benchmark setting σ = 0.15 and r = 0.1. When one parameter is varied, all other parameters are held at their benchmark values.

By contrast, firm social responsibility has a more pronounced quantitative effect. Asr increases, both thresholds decline and the goal-conflict region becomes progressively narrower. For example, relative to the benchmark r = 0.1, increasing r to 0.30 reduces the width of the goal-conflict region by approximately 9.26%. This result supports the theoretical prediction that a stronger concern for farmer profit facilitates the upstream transmission of the market value created by brand building and thereby reduces the divergence between social-welfare and farmer-profit policy preferences.

Overall, the sensitivity analysis confirms the comparative-static results in Corollary 11: greater yield risk widens the goal-conflict region, whereas stronger firm social responsibility narrows it.

7 Discussion

7.1 Theoretical implications

Most studies of agricultural support compare different types of subsidies or financial assistance (Alizamir et al., 2019; Ye et al., 2021; Chintapalli and Tang, 2022; Tang et al., 2024; Zhang et al., 2026). The present analysis makes a different comparison. It places a policy that acts directly on production alongside one that works through market demand. This distinction matters because the two policies reach farmers in different ways.

A planting subsidy immediately changes the return from planting. Its effect on the farmer's production incentive is partly offset by the firm's lower procurement price. Once the subsidy is introduced, the firm has an incentive to lower the procurement price. Part of the public support is consequently offset by the firm's pricing response. Brand building has no such direct effect on planting costs. It first expands demand and raises the market value of agricultural output. The firm then responds by increasing the procurement price to secure more supply. Farmer benefit under this policy thus depends on a two-step process: value must first be created in the downstream market and then passed back to the farmer.

This value-transmission process adds to the contract-farming literature. Earlier studies show that procurement arrangements affect planting and agricultural supply (Federgruen et al., 2019; Chen and Chen, 2021). Our results further suggest that the procurement price determines how government-created market value is divided within the supply chain. A successful branding program may increase total market value without producing a similar increase in farmer profit. The effect of demand-side intervention cannot therefore be judged from market expansion alone.

The role of firm social responsibility also differs across the two policies. Under a planting subsidy, a stronger concern for farmer profit leads the firm to offer a higher procurement price. The welfare-maximizing government can then reduce the subsidy without changing the equilibrium planting area. In this sense, firm social responsibility partially substitutes for the production support that would otherwise come from the government. Under brand building, the same concern helps pass a larger share of the additional market value to the farmer. It lowers both policy thresholds and reduces the gap between them. This differs from earlier CSR studies, which mainly examine pricing, coordination, and surplus allocation within the supply chain (Liu et al., 2019; Wang et al., 2023; Yi et al., 2023; Ma et al., 2024). Here, CSR also changes the conditions under which one government policy performs better than another.

The difference between the two thresholds is central to the analysis. Brand building can raise social welfare before it raises farmer profit above the level provided by a planting subsidy. This does not mean that brand building fails to create value. Rather, the additional value is not yet large enough, or is not passed upstream strongly enough, to make the farmer better off. Yield risk, price sensitivity, and planting costs make this gap wider. Higher expected yield and stronger firm social responsibility make it narrower. Policy rankings are therefore shaped by both production conditions and the way market gains are shared.

7.2 Implications for sustainable food systems

The implications for sustainable food systems considered here are primarily economic and institutional. In particular, the model examines farmer profit and planting incentives, which reflect the economic incentives provided to the production side of the supply chain. Because farmer participation, contract renewal, and exit decisions are not explicitly modeled, these outcomes should not be interpreted as direct predictions of long-term participation behavior. Farmers commit land and other inputs before output is realized and bear the consequences of uncertain yield. If their return remains limited, the contractual arrangement may provide only a weak basis for continued production. Farmer profit and planting incentives therefore indicate the economic support provided to the production side rather than predicting long-term participation behavior.

The goal-conflict region is important in this respect. Within this region, brand building is preferred under the government's social-welfare objective, yet the farmer earns more under the planting subsidy. Some of the value created by brand building is retained downstream or reflected in consumer surplus rather than transferred to the farmer. A rise in total welfare may therefore coexist with a relatively weak improvement in the farmer's economic position. From a policy perspective, such a distributional pattern may raise concerns about the economic attractiveness of continued participation, although actual contract-renewal and exit decisions are beyond the scope of the present model.

These policy implications should not be interpreted as suggesting that planting subsidies are inherently more sustainable than brand building. When brand building is sufficiently cost-effective, it produces higher social welfare and higher farmer profit. The relevant issue is whether the extra market value reaches the farmer. Government evaluation of branding programs should therefore consider the procurement price and farmer profit, rather than relying only on changes in demand, sales value, or aggregate welfare.

Production conditions also matter. Yield risk raises the level of brand-building cost-effectiveness needed to benefit the farmer and widens the conflict between the two policy rankings. In a high-risk production environment, a higher level of brand-building cost-effectiveness is required for demand-side intervention to outperform direct production support in terms of farmer profit. Stronger firm social responsibility can reduce this problem because more of the market gain is passed upstream. Even so, CSR does not remove the need to evaluate policy outcomes. Its effect still depends on the production environment and the effectiveness of the branding program.

The present analysis does not directly address environmental performance. Issues such as resource use, food loss, environmental externalities, and repeated contract participation could be incorporated in future research.

8 Conclusions

This paper compares a planting subsidy with government-supported brand building in an agribusiness-led contract-farming supply chain. Agricultural yield is uncertain, the government maximizes social welfare, and the agribusiness firm considers farmer profit when setting the procurement price. The farmer then chooses the planting area.

The two policies work in different ways. A planting subsidy supports production directly, but the firm responds by lowering the procurement price. Brand building expands demand and makes agricultural output more valuable to the firm, leading to a higher procurement price. Its advantage nevertheless depends on the cost of creating this additional market value.

The comparison produces two thresholds. The social-welfare threshold is lower than the farmer-profit threshold. When brand-building cost-effectiveness lies between them, brand building generates higher social welfare, while the farmer earns more under the planting subsidy. Yield risk, price sensitivity, and planting costs widen this region. Higher expected yield and stronger firm social responsibility narrow it. Market size changes the outcomes under both policies but not their relative ranking.

The main policy lesson is straightforward. Creating market value does not guarantee that farmers receive more value. A demand-side policy becomes preferable for both the government and the farmer only when brand building is sufficiently cost-effective and the resulting gains are passed upstream through the procurement price. Policy assessment should therefore consider farmer outcomes alongside social welfare, because the distribution of policy benefits is an important consideration when evaluating the economic support provided to the production side of the agricultural supply chain.

Several limitations provide opportunities for future research. The analysis considers an agribusiness firm and a representative farmer under symmetric information and a linear demand structure. In practice, farmers may differ in production costs, yield risk, risk preferences, and bargaining power, while agribusiness firms may compete for agricultural supply or possess private information. Future research could also introduce dynamic participation, contract-renewal, and exit decisions to examine more directly how alternative policies affect the long-term stability of contract-farming relationships.

Generative AI statement

The author(s) declared that Generative AI was used in the creation of this manuscript. During the preparation of this manuscript, the author(s) used ChatGPT (OpenAI, GPT-5.6) for the purposes of text polishing and language refinement. The author(s) have reviewed and edited the output and take full responsibility for the content of this publication.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fsufs.2026.1964914/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Justice Gameli Djokoto, Dominion University College, Ghana

Reviewed by: Jian Hu, Lanzhou University of Finance and Economics, China

Octavio Barrera, Universidad Autónoma de Baja California, Mexico