Blockchain Adoption Strategies in Dual-Channel Supply Chains Under Different Leadership Structures and Asymmetric Market Shares

Abstract

Many manufacturers operate dual-channel supply chains, selling through both direct online and reselling channels, with market shares that are generally asymmetric. At the same time, consumers increasingly demand trustworthy traceability information. Although traceability systems based on blockchain technology (TSBT) can meet this demand, they require substantial investment, making large manufacturers or powerful retailers more likely to lead TSBT adoption. This paper investigates how different leadership structures (manufacturer leadership vs. retailer leadership) and asymmetric market shares influence blockchain adoption strategies. This study reveals that under retailer leadership, blockchain adoption occurs once consumer preference for blockchain-based traceability information (hereafter, consumer preference) in the reselling channel exceeds a threshold, increasing profits for both supply chain members. Under manufacturer leadership, adoption depends on consumer preference in the reselling channel, the initial market share of the reselling channel, and the intensity of channel competition, generally benefiting the manufacturer but potentially harming the retailer. Comparing equilibrium results across leadership structures reveals that blockchain is more likely to be adopted under manufacturer leadership, which may yield higher profits for both members.

1. Introduction

Traditional centralized traceability systems in supply chains are vulnerable to data manipulation and often fail to provide trustworthy product information. Such limitations undermine consumer trust and increase operational and reputational risks, thereby threatening supply chain sustainability [1,2]. For example, Anutric has been reported to have discrepancies between the labeled and actual levels of docosahexaenoic acid in its products (http://food.china.com.cn/2021-09/26/content_77775255.htm (accessed on 25 December 2025)). By enabling transparent and tamper-resistant record keeping, blockchain technology has been widely regarded as a promising tool for improving traceability reliability and supporting supply chain sustainability [1,3]. In practice, many well-established retailers and manufacturers, such as Nestle, Yili, Louis Vuitton, Walmart, and Alibaba, have implemented TSBT. However, developing such systems requires substantial investments in sensor equipment (e.g., Internet of Things technologies for detection and tracking), data computation, and storage facilities [4,5,6]. As a result, only large manufacturers or powerful retailers are likely to take the lead in building these systems and bearing the associated investment costs. We refer to such firms as supply chain leaders. This raises an important question: how do different leadership structures (manufacturer leadership and retailer leadership) shape blockchain adoption strategies?

Addressing this question requires careful consideration of the interactions among supply chain members [7]. In practice, many manufacturers, such as Nike (https://www.nike.com.cn/ (accessed on 25 December 2025)) and BYHEALTH (https://www.by-health.com/en/product/ (accessed on 25 December 2025)), operate dual-channel supply chains, selling products through a reselling channel operated by an independent retailer as well as a direct channel in which the manufacturer sells directly to consumers. Market shares across these channels are often asymmetric. For instance, in 2024, Nike’s direct channel sales reached 21,519 million dollars, while its reselling channel sales were 27,758 million dollars (https://investors.nike.com/investors/news-events-and-reports/investor-news/investor-news-details/2024/NIKE-Inc.-Reports-Fiscal-2024-Fourth-Quarter-and-Full-Year-Results/default.aspx (accessed on 25 December 2025)). Such asymmetries directly influence firms’ pricing decisions and profit margins, which in turn affect blockchain adoption incentives [8,9]. However, existing studies on blockchain adoption in dual-channel supply chains under different leadership structures largely overlook how asymmetric channel market shares affect adoption strategies [6,10,11].

Furthermore, blockchain effectiveness may differ across channels in dual-channel supply chains. As Isaja et al. [12] emphasizes, the key feature of blockchain lies in decentralized, consensus-based validation among multiple peer nodes. In the reselling channel, blockchain adoption typically corresponds to a consortium (permissioned) architecture. In this architecture, multiple supply chain members jointly operate data submission and validation based on consensus mechanism [13]. Records are accepted only after cross-party verification. This consensus mechanism makes it difficult for any single participant to unilaterally alter historical information and thereby enhances the credibility of traceability data [12,14]. By contrast, in the direct channel, traceability information is provided solely by the manufacturer. In the absence of decentralized consensus, data submission and validation are centralized within a single firm, which allows records to be modified ex post [13]. As a result, a single-participant blockchain functions similarly to a traditional centralized database, undermining credibility and consumer trust [12] (https://hbr.org/2017/01/the-truth-about-blockchain (accessed on 25 December 2025)). These differences in blockchain effectiveness across channels may influence the leader’s blockchain adoption strategy [9]. However, most existing studies assume that blockchain effectiveness is uniform across channels [6,15,16].

These observations highlight a research gap in the literature. Existing studies have overlooked how asymmetric channel market shares and heterogeneous blockchain effectiveness across channels influence blockchain adoption strategies under different leadership structures in dual-channel supply chains. This paper aims to fill this gap by addressing the following research questions in this setting:

(1)

When should blockchain technology be adopted under different leadership structures?

(2)

How does blockchain adoption affect prices, demands, and profits of supply chain participants?

(3)

How do leadership structures shape blockchain adoption strategies?

To address these research questions, we develop an analytical model of a dual-channel supply chain consisting of a manufacturer and a retailer. Specifically, consistent with Niu et al. [14] and as discussed above, we adopt a benchmark assumption that traceability systems based on blockchain technology (TSBT) with a single participant (e.g., the direct channel) function like traditional databases and cannot provide trustworthy product information to consumers. We then examine blockchain adoption decisions under retailer and manufacturer leadership. Finally, we relax this benchmark assumption in an extension by distinguishing between traditional databases and single-participant TSBT. This allows us to further evaluate how heterogeneous blockchain effectiveness across channels affects adoption strategies.

The main findings are as follows. First, under retailer leadership, blockchain adoption occurs once consumer preference in the reselling channel exceeds a threshold. This adoption leads to higher demand, prices, and profits for both supply chain members. Second, under manufacturer leadership, adoption depends on consumer preference in the reselling channel, the initial market share of the reselling channel, and the intensity of channel competition. Adoption generally benefits the manufacturer but may harm the retailer when competition is intense or the reselling channel’s initial market share is small. Third, a comparison of equilibrium outcomes across leadership structures shows that blockchain adoption is more likely under manufacturer leadership. Moreover, when both leaders are willing to adopt blockchain under their respective leadership structures, manufacturer leadership yields higher demand, prices, and profits. Finally, an extension that distinguishes between traditional databases and single-participant TSBT indicates that the main results remain valid when consumer preference in the direct channel is low. When such preference is moderate, the equilibrium strategy under manufacturer leadership depends on consumer preference in the reselling channel.

This paper makes two main contributions. First, this study examines how asymmetric channel market shares and heterogeneous blockchain effectiveness across channels jointly influence blockchain adoption strategies under different leadership structures in dual-channel supply chains. Second, this study identifies the conditions under which blockchain is adopted and examines how adoption shapes equilibrium price, demand, and profit under manufacturer versus retailer leadership.

The remainder of this paper is organized as follows. Section 2 reviews the relevant literature, and Section 3 presents the model setting. Section 4 introduces a benchmark model without blockchain. Section 5 derives the equilibrium results and analyzes the effects of blockchain adoption under retailer leadership and manufacturer leadership structures. Section 6 compares the outcomes across different leadership structures. Section 7 discusses extensions of the main model, and Section 8 concludes the study.

2. Literature Review

This study relates to three primary streams of literature: (1) pricing and operations management in dual-channel supply chains, (2) product traceability, and (3) applications of blockchain technology in supply chain management. This section provides a focused review of the related literature.

The first stream of literature focuses on supplier encroachment [17,18], channel cooperation [19], channel coordination [20], and information sharing [21,22], as well as pricing, sales efforts, and service management [23,24,25,26]. Within this stream, our study is most closely related to research on pricing, sales effort, and service management. For instance, Pi et al. [23] explore pricing and service strategies in competitive and cooperative settings among retailers in a dual-channel supply chain. Taleizadeh et al. [26] further investigate the coordination of pricing, emission reduction, and sales effort decisions under carbon emission constraints. They show that cost-sharing contracts can achieve supply chain coordination. These studies highlight the important role of service and sales efforts in shaping consumer behavior when such investments are made by a specific supply chain member that directly interfaces with consumers. Traditional sales efforts are therefore typically undertaken by the downstream firm and can often be unilaterally enhanced with limited additional operational costs. By contrast, blockchain adoption requires the participation of multiple supply chain members and involves non-negligible operational costs, which directly affect adoption incentives. Moreover, blockchain adoption can be initiated and led by either upstream or downstream firms, rather than being restricted to the member directly facing consumers. Addressing this distinction, our study examines joint blockchain adoption decisions under different leadership structures, with particular attention to the role of unit operational costs.

Supply chain traceability, defined as the ability to track the history, application, or location of products [27], is crucial across various industries [28]. Empirical studies show that traceability systems can increase consumers’ willingness to pay [29,30]. Analytical models highlight their role in reducing inventory inaccuracies [31,32], improving operational efficiency and security [33], and lowering variable costs [34]. However, traditional traceability systems often suffer from limited data integrity and transparency. To address these limitations, blockchain technology has been widely regarded as a more reliable solution because it provides immutable and transparent records. Rao et al. [35] report that consumers are willing to pay a 28.2% premium for blockchain-based traceability. This premium is substantially higher than that associated with conventional traceability methods. Consistent with this finding, Doanh et al. [36] and Duan and Zhu [37] show that blockchain-based traceability enhances consumer trust and purchase intention. These studies highlight blockchain’s distinctive advantage in strengthening consumer trust and increasing willingness to pay through credible and transparent traceability. Building on this literature, our study explicitly models how blockchain-induced consumer trust affects firms’ pricing and operational decisions in the supply chain.

Blockchain technology in supply chain management has been extensively examined in the literature. In particular, a growing stream of studies investigates blockchain adoption strategies under specific leadership structures, such as manufacturer leadership or retailer/platform leadership. Within this stream, several studies focus on linear supply chains or competitive environments with symmetric market shares. For example, Zou et al. [2] examine how an e-commerce platform’s blockchain adoption strategy interacts with a manufacturer’s sales channel choice and carbon emission reduction effort. Zhang et al. [15] investigate a platform’s blockchain adoption strategy in a dual-channel supply chain while accounting for demand uncertainty and seller risk aversion. Yu et al. [38] analyze how a high-quality supplier’s blockchain adoption decision interacts with a low-quality supplier’s selling contract choice in a platform supply chain. Another group of studies examines blockchain adoption strategies in competitive environments with asymmetric market shares. Zhong et al. [9] analyze a manufacturer’s blockchain adoption strategy under different subsidy policies, accounting for channel competition, market share asymmetry, and heterogeneous blockchain effectiveness across channels. Shao et al. [16] examine blockchain adoption and green investment in a dual-channel supply chain with asymmetric market shares. They show that a joint strategy of green investment and blockchain adoption becomes optimal only when consumers exhibit strong low-carbon preferences or low green trust, and when the e-commerce channel’s market share is moderate. Similarly, Zhu et al. [8] show that, in a dual-channel supply chain, blockchain adoption always benefits the retailer through market expansion and channel reallocation.

Other studies focus on different leadership scenarios to examine how leadership structures influence blockchain adoption decisions and to identify the optimal blockchain investor. In linear supply chains, several studies investigate blockchain adoption strategies under different leadership structures. For example, Wu et al. [39] study blockchain adoption in fresh product supply chains and show that adoption decisions depend on consumer acceptance, product deterioration rates, and traceability cost allocation. Xia et al. [40] further examine manufacturer and retailer blockchain investment decisions under information asymmetry regarding consumers’ traceability preferences. They find that when the proportion of high-type consumers is moderate, low blockchain implementation costs lead to manufacturer-led adoption, whereas higher costs induce the retailer to take the initiative. Li et al. [41] further consider supplier-led, manufacturer-led, and balanced power structures and propose a coordination mechanism for blockchain adoption. Sun et al. [42] study blockchain adoption under different supply chain leadership structures, accounting for three types of consumer attitudes: interested, neutral, and averse. They show that consumer aversion significantly dampens retailers’ willingness to adopt the technology. Moreover, Fang et al. [7] explore the interaction between retailers’ pricing model choices and supply chain members’ blockchain investment decisions. Extending beyond linear supply chains, another stream of literature examines blockchain adoption under different leadership structures in competitive environments. Chang et al. [43], for instance, analyze equilibrium blockchain adoption and online channel strategies under both manufacturer and retailer leadership. Similarly, Zhou et al. [6] investigate blockchain adoption in a dual-channel supply chain under manufacturer and online platform leadership. They show that blockchain adoption can improve supply chain performance. Zhang et al. [10] study digital technology adoption and supply chain coordination in a dual-channel supply chain and demonstrate that cost-sharing mechanisms can achieve a win–win outcome. Yuan et al. [11] incorporate price-matching policies into a dual-channel supply chain under supplier-led and retailer-led scenarios. They find that suppliers exhibit a stronger inclination toward blockchain adoption. Gong et al. [44] study blockchain adoption to fight counterfeiting under manufacturer-led, platform-led, and cooperative governance structures. They find that adoption hinges primarily on implementation cost and counterfeit quality, irrespective of the investor.

Our contribution is to examine how asymmetric channel market shares and heterogeneous blockchain effectiveness across channels influence blockchain adoption strategies in dual-channel supply chains under different leadership structures. Unlike existing studies that typically assume uniform blockchain effectiveness across channels, we consider channel-specific blockchain effectiveness. In particular, in the direct channel, a single-participant TSBT provides limited credibility due to the absence of decentralized consensus. By incorporating asymmetric market shares and heterogeneous blockchain effectiveness into an analytical model, our study offers a more precise and realistic understanding of blockchain adoption decisions.

Table 1. Major differences between this paper and the most relevant papers.

Paper	Channel Features		Features Related to Blockchain			Leadership Structure	Key Contribution
	Competition	Market Share	Information Level	Operational Cost	Varied Impacts
Yuan et al. [11]	✓		✓		✓	✓	Incorporate the price-matching policy into the blockchain adoption model for dual-channel supply chains under different leadership structures.
Zhou et al. [6]	✓		✓			✓	Examine blockchain adoption strategies in dual-channel supply chains under cost-sharing schemes and different leadership structures considering customer information traceability preferences.
Zhang et al. [10]	✓		✓			✓	Study digital technology adoption and supply chain coordination in a dual-channel supply chain.
Zhong et al. [9]	✓	✓	✓		✓		Investigate blockchain adoption under government subsidies, incorporating heterogeneous effects across online and offline channels and asymmetric market shares.
Gong et al. [44]	✓			✓			Explore blockchain adoption strategies under manufacturer-led, platform-led, and cooperative structures to combat counterfeiting.
Sun et al. [42]				✓		✓	Consider heterogeneous consumer preferences for blockchain-enabled products (i.e., interested, averse, and neutral) under different leadership structures (i.e., supplier–led, retailer-led, and vertical Nash).
Shao et al. [16]	✓	✓					Investigate manufacturers’ green investment decisions and retailers’ blockchain adoption in dual-channel supply chains with asymmetric market shares.
Xia et al. [40]				✓		✓	Examine blockchain investment and pricing strategies under manufacturer-led and retailer-led structures, considering asymmetric consumer traceability preferences between the retailer and the manufacturer.
Li et al. [41]				✓		✓	Consider supplier-led, manufacturer-led, and balanced power structures and propose a coordination mechanism for blockchain adoption
This work	✓	✓	✓	✓	✓	✓	Consider heterogeneous blockchain effectiveness across channels and asymmetric market shares in a dual-channel supply chain under manufacturer-led and retailer-led structures.

summarizes our contributions relative to the existing literature.

3. The Model

We consider a supply chain comprising a manufacturer (m) and retailer (r), as shown in Figure 1. The manufacturer sells its products through two channels: the reselling channel and the direct channel. In the reselling channel, the manufacturer supplies products to the retailer at a wholesale price w, after which the retailer sets the retail price $p_r$. In the direct channel, the manufacturer directly decides the retail price $p_m$. We denote the demand for the reselling and direct channels as $q_r$ and $q_m$, respectively. The total market potential is normalized to 1. The market share of the reselling channel is denoted by $s\in (0,1)$. Hence, the market share of the direct channel is $1-s$. Furthermore, we use b to denote the cross-price sensitivity to reveal the price competition intensity between the two channels. Table 2 provides a summary of all parameters and variables employed in the main model.

Adopting blockchain can fulfill consumer needs for trustworthy and accurate product information. This occurs when blockchain records data from multiple independent members of the supply chain (e.g., the reselling channel) [12,14]. When only one participant is involved (e.g., the direct channel), transactions can be accepted without consensus on their correctness. In this case, a single participant can unilaterally modify recorded information without cross-verification. This weakens the credibility and accuracy of traceability information. Therefore, similar to Niu et al. [14], we initially assume that a TSBT with a single participant (e.g., the direct channel) functions like a traditional database, and the resulting traceability information cannot satisfy consumer needs for trustworthy and accurate product information. This assumption will be relaxed in subsequent extensions. Furthermore, we incorporate the follower’s participation decision when the leader constructs the TSBT. This helps illustrate the interplay between the manufacturer and the retailer in blockchain adoption and highlights the importance of full supply chain participation.

Table 2. Notations.

Notation	Description
Decision variables
B	Blockchain-based information level
w	Wholesale price
$p_r$	Retailer’s retail price
$p_m$	Manufacturer’s retail price
Parameters
$\gamma$	Consumer preference in the reselling channel
b	The intensity of channel competition, $0<b<1$
$\mu$	The cost coefficient of blockchain investment
c	Unit operational cost of using blockchain
s	The market share of the reselling channel, $0<s$ < 1
$1-s$	The market share of the direct channel, $0<s<1$
Functions
$q_r$	The reselling channel’s demand
$q_m$	The direct channel’s demand
${\pi }_r$	The retailer’s profit
${\pi }_m$	The manufacturer’s profit
${\pi }_{sc}$	The profit of the whole supply chain

Table 2. Notations.

Notation	Description
Decision variables
B	Blockchain-based information level
w	Wholesale price
$p_r$	Retailer’s retail price
$p_m$	Manufacturer’s retail price
Parameters
$\gamma$	Consumer preference in the reselling channel
b	The intensity of channel competition, $0<b<1$
$\mu$	The cost coefficient of blockchain investment
c	Unit operational cost of using blockchain
s	The market share of the reselling channel, $0<s$ < 1
$1-s$	The market share of the direct channel, $0<s<1$
Functions
$q_r$	The reselling channel’s demand
$q_m$	The direct channel’s demand
${\pi }_r$	The retailer’s profit
${\pi }_m$	The manufacturer’s profit
${\pi }_{sc}$	The profit of the whole supply chain

In this model, one supply chain member acts as the leader, who establishes the TSBT and bears the associated investment costs, while the other member acts as the follower and decides whether to participate. Consistent with JD.com’s practice of offering its “JD Blockchain Anti-Counterfeiting and Traceability Platform” free of charge to partners, we assume that the follower can join the TSBT without incurring additional costs. When both members adopt blockchain, consumers gain access to more reliable, cross-verified product information. A higher level of disclosed information enhances consumer trust and purchase probability. We denote the blockchain-based information level by B (hereafter, the information level) and consumer preference in the reselling channel by $\gamma$. However, providing a higher information level requires more investments in sensor equipment, data computing, and storage facilities [9,39]. Following prior studies [9,39], we assume that the information level B incurs a quadratic cost $\frac{\mu B^2}{2}$, which is borne by the leader, where $\mu$ denotes the cost coefficient. In addition, blockchain adoption entails a unit operational cost c, reflecting expenses related to verification, commitments, and transaction monitoring [4]. Consistent with Zhang et al. [5] and Choi [45], we assume that production and sales activities incur operational cost. Taken together, the leader faces a trade-off: increasing the information level can stimulate demand but also raises construction and operational costs. As a result, choosing the optimal information level becomes a central decision.

Figure 2 and Figure 3 illustrate the decision sequence. The leader determines whether to adopt blockchain in Stage 1. If the leader adopts it, the information level is confirmed in Stage 2. Moreover, in Stage 3, upon observing the information level, the follower makes up its participation mind. The wholesale price is determined in Stage 4. Finally, in Stage 5, the retail prices in both channels are simultaneously determined by the retailer and manufacturer.

4. No Blockchain Adoption (Scenario N)

We denote the benchmark model without blockchain by N, and use an asterisk (∗) to denote optimal results. Following Chen et al. [46] and Zhong et al. [9], market demand in the reselling and direct channels is given by: $q_r^N=s-p_r^N+b{p}_m^N$, $q_m^N=(1-s)-p_m^N+b{p}_r^N$. Based on these demand functions, the profit functions of the retailer and the manufacturer are:

$${\pi }_r^N=(p_r^N-w^N)q_r^N,$$

(1)

$${\pi }_m^N=w^N{q}_r^N+p_m^N{q}_m^N.$$

(2)

The retailer’s profit comprises the revenue and procurement cost of the reselling channel. The manufacturer’s profit function includes revenues derived from both the reselling and the direct channels.

Lemma 1.

Under the benchmark model, the optimal results are given as follows:

(1) 

the optimal price strategies are $w^{N\ast }=\frac{b(b^2+8)+(b^4-b^3-8b+8)s}{2(8-b^4-7b^2)}$, $p_m^{N\ast }=\frac{b^2-(2-b)(1-b)(b+4)s+8}{2(8-b^4-7b^2)}$ and $p_r^{N\ast }=\frac{b(b^2+8)+(12-b^4-b^3-2b^2-8b)s}{2(8-b^4-7b^2)}$;

(2) 

the demand and profit are $q_r^N=\frac{(b^2+2)s}{b^2+8}$, $q_m^{N\ast }=\frac{1}{2}(1-\frac{(2-b)(b^2+b+4)s}{b^2+8})$, ${\pi }_r^{N\ast }=\frac{{(b^2+2)}^2{s}^2}{{(b^2+8)}^2}$ and ${\pi }_m^{N\ast }=\frac{(1-b)(2-b)(b^2+b+6)s^2-2(1-b)(b^2+8)s+b^2+8}{4(8-b^4-7b^2)}$.

As the initial market share in the reselling channel s increases, demand in the reselling channel rises. This leads the retailer to raise its retail price and, in turn, induces the manufacturer to increase the wholesale price. At the same time, a higher market share in the reselling channel reduces demand in the direct channel, prompting the manufacturer to lower its retail price accordingly. Overall, an increase in the initial market share of the reselling channel increases demand and raises prices in the reselling channel, while reducing demand and lowering price in the direct channel.

5. Equilibrium Results of Blockchain Adoption Decision

This section examines the equilibrium of blockchain adoption strategies under retailer and manufacturer leadership.

5.1. Retailer Leadership (Scenario R)

We first derive the equilibrium results under retailer leadership, which we refer to as Scenario R. When the retailer adopts blockchain, the TSBT is implemented exclusively in the reselling channel. If the retailer adopts blockchain, the manufacturer then decides whether to participate in the blockchain system. Accordingly, two cases arise: (1) the manufacturer joins the system in the reselling channel ($Ry$), and (2) the manufacturer does not join the system ($Rn$). Therefore, under retailer leadership, three strategies are possible: N, $Ry$, and $Rn$.

Under scenario $Ry$, consumers can acquire cross-verified and trustworthy product information, which increases product demand and consumers’ purchase probability [9]. Following Zhong et al. [9] and Song et al. [47], the demand functions are given by: $q_r^{Ry}=s-p_r^{Ry}+b{p}_m^{Ry}+\gamma B^{Ry}$, $q_m^{Ry}=1-s-p_m^{Ry}+b{p}_r^{Ry}$. We refer to the term $\gamma B^{Ry}$ as the expansion effect. The profit functions under scenario $Ry$ are:

$${\pi }_r^{Ry}=(p_r^{Ry}-w^{Ry}-c)q_r^{Ry}-\frac{\mu {\left(B^{Ry}\right)}^2}{2},$$

(3)

$${\pi }_m^{Ry}=(w^{Ry}-c)q_r^{Ry}+p_m^{Ry}{q}_m^{Ry}.$$

(4)

Under scenario $Rn$, only the retailer participates in the TSBT. As a result, consumers cannot obtain cross-verified and fully trustworthy information. Hence, the demand functions remain the same as those without blockchain adoption. Because the manufacturer does not participate in the TSBT, it does not incur unit operational cost of using blockchain associated with production activities. By contrast, the retailer bears investment costs and unit operational cost of using blockchain associated with sales activities. The profit functions under scenario $Rn$ are:

$${\pi }_r^{Rn}=(p_r^{Rn}-w^{Rn}-c)q_r^{Rn}-\frac{\mu {\left(B^{Rn}\right)}^2}{2},$$

(5)

$${\pi }_m^{Rn}=w^{Rn}{q}_r^{Rn}+p_m^{Rn}{q}_m^{Rn}.$$

(6)

The analysis shows that the retailer does not adopt the TSBT without the manufacturer’s participation. Accordingly, strategy $Rn$ does not arise in equilibrium. The optimal outcomes under scenario $Ry$ are detailed in

Table A1. The optimal results under scenario $Ry$.

Results
$B^{Ry\ast }=\frac{-E_2}{2E_1}$
$w^{Ry\ast }=\frac{(b^4+8)\gamma B^{Ry\ast }+2(1-b^2)b^{2}c+(b^2+8)b+(b^4-b^3-8b+8)s}{2(8-b^4-7b^2)}$
$p_m^{Ry\ast }=\frac{(10-b^2)b\gamma B^{Ry\ast }-4(1-b^2)bc+b^2+(-b^3-b^2+10b-8)s+8}{2(8-b^4-7b^2)}$
$p_r^{Ry\ast }=\frac{b^3+8(1-b^2)c+(-b^4-2b^2+12)B^{Ry\ast }\gamma +(-b^4-b^3-2b^2-8b+12)s+8b}{2(8-b^4-7b^2)}$
$q_m^{Ry\ast }=\frac{(b^2+2)b\gamma B^{Ry\ast }+((b-1)b^2+2b-8)s+b^2+12bc+8}{2(b^2+8)}$
$q_r^{Ry\ast }=\frac{(b^2+2)(\gamma B^{Ry\ast }-2c+s)}{b^2+8}$
${\pi }_m^{Ry\ast }=\frac{{(b^2+2)}^2{\gamma }^2{\left(B^{Ry\ast }\right)}^2}{4(-b^4-7b^2+8)}+\frac{\gamma B^{Ry\ast }((b-1)(b^3+4b-4)s+b^3+8b^{2}c+8b-8c)}{2(-b^4-7b^2+8)}+\frac{4c(c-s)}{b^2+8}+\frac{(b-2)(b-1)(b^2+b+6)s^2+2(b-1)(b^2+8)s+b^2+8}{4(-b^4-7b^2+8)}$
${\pi }_r^{Ry\ast }=E_1{\left(B^{Ry\ast }\right)}^2+E_2{B}^{Ry\ast }+\frac{{(b^2+2)}^2{(s-2c)}^2}{{(b^2+8)}^2}$

and satisfy the constraints $\gamma <{\gamma }_1=\sqrt{\frac{{(b^2+8)}^2\mu }{2{(b^2+2)}^2}}$ and $0<c<\frac{s}{2}$. These constraints restrict consumer preference in the reselling channel and the unit operational cost to economically meaningful ranges [4,9], ensuring positive demand and profit in both channels as well as a concave retailer profit function.

Lemma 2.

When the retailer adopts blockchain, if the information level exceeds a certain threshold (i.e., $B>B_m$), the manufacturer joins the TSBT; otherwise, it does not. The threshold $B_m$ is defined in Appendix A.

The threshold information level required for the manufacturer’s participation ($B_m$) decreases with consumer preference in the reselling channel ($\gamma$) but increases with the initial market share of the reselling channel (s). When s is low, profit is mainly generated from the direct channel, so participation in the TSBT can mitigate price competition and improve overall profitability even at a relatively low information level. As s grows, the manufacturer’s profit becomes more dependent on the reselling channel. Hence, participation becomes attractive only if it benefits the reselling channel, which requires a higher information level.

Proposition 1.

When the retailer leads the implementation of TSBT, its adoption decision depends on consumer preference in the reselling channel γ:

(1) 

If consumer preference in the reselling channel exceeds a certain threshold (i.e., $\gamma >e_1$), the retailer adopts TSBT with the manufacturer’s participation (strategy $Ry$) and sets the information level as $B^{Ry\ast }=\frac{2{(b^2+2)}^2(s-2c)}{{(b^2+8)}^2\mu -2{(b^2+2)}^2}$.

(2) 

Otherwise, the retailer opts out of TSBT (strategy N).

The threshold $e_1$ is defined in Appendix A.

Proposition 1 shows that the retailer establishes the TSBT only when consumer preference in the reselling channel $\gamma$ exceeds a certain threshold. In this case, the information level is sufficient to attract the manufacturer’s participation. Under retailer leadership, the retailer bears both the investment cost and the unit operational cost, while the manufacturer incurs only the unit operational cost. The resulting expansion effect generates additional revenue that covers these costs for both parties. Consequently, the retailer initiates the TSBT and thereby induce the manufacturer’s participation. From a managerial perspective, consumer preference in the reselling channel plays a key role in incentivizing blockchain adoption and supporting supply chain sustainability from the demand side. Accordingly, before deciding whether to adopt blockchain, firms should assess consumer preference. Firms may communicate the benefits of TSBT to consumers to strengthen this preference. They can then monitor its evolution over time and adjust information disclosure and pricing decisions accordingly.

From Lemma 2 and Propositions 1 and 2 can be derived.

Proposition 2.

When the retailer adopts blockchain (i.e., $\gamma >e_1$), an increase in the unit operational cost of using blockchain (c) leads to the following outcomes:

(1) 

The optimal information level decreases.

(2) 

The manufacturer’s retail price decreases. The retailer’s retail price and the wholesale price also decrease when consumer preference in the reselling channel is high (i.e., $\frac{\mathsf{\partial }{p}_r^{Ry\ast }}{\mathsf{\partial }c}<0$ for $\gamma >max\{e_1,{\gamma }_{pp}\}$ and $\frac{\mathsf{\partial }{w}^{Ry\ast }}{\mathsf{\partial }c}<0$ for $\gamma >max\{e_1,{\gamma }_w\}$).

(3) 

Demand in the reselling channel decreases. Demand in the direct channel decreases when consumer preference in the reselling channel is high (i.e., $\frac{\mathsf{\partial }{q}_m^{Ry\ast }}{\mathsf{\partial }c}<0$ for $\gamma >max\{e_1,{\gamma }_{qm}\}$).

(4) 

The profits of both the retailer and the manufacturer decrease.

The thresholds ${\gamma }_w$, ${\gamma }_{pp}$, and ${\gamma }_{qm}$ are defined in Appendix A.

Proposition 2 shows the effect of the unit operational cost on the optimal outcomes. The optimal information level decreases with the unit operational cost. This allows the retailer to reduce overall expenses and mitigate demand pressure. However, the impact of the unit operational cost on prices in the reselling channel and demand in the direct channel depends on consumer preference in the reselling channel. When consumer preference in the reselling channel is low, an increase in the unit operational cost induces the retailer to pass part of the cost to consumers through a higher retail price. The higher retail price reduces demand in the reselling channel. In response, the manufacturer lowers the price in the direct channel, thereby stimulating demand in that channel. When consumer preference in the reselling channel is high, prices in that channel are already high, and further increases would substantially dampen demand. Consequently, the retailer lowers the retail price. However, the reduction in the information level dominates this price adjustment, leading to a decline in the reselling channel’s demand. At the same time, although the manufacturer also lowers its retail price, the negative cross-price effect is not fully offset, causing demand in the direct channel to decline as well. Consequently, both supply chain members experience lower profit. From a managerial perspective, rising unit operational cost optimally leads retailers to reduce the information level. However, pricing decisions should not be adjusted mechanically. Firms should not uniformly raise prices to pass costs on to consumers, nor should they lower prices merely because the information level declines. Instead, pricing strategies should be tailored to consumer preference and prevailing market conditions.

We then examine the impact of blockchain adoption.

Proposition 3.

When the retailer adopts blockchain (i.e., $\gamma >e_1$), both the retailer and the manufacturer set higher prices than in the no-blockchain benchmark and simultaneously experience higher demand and profit.

Blockchain adoption meets consumers’ demand for authentic and verifiable product information, thereby increasing product attractiveness and overall demand. Anticipating higher demand, the retailer raises its retail price. This price adjustment in turn induces the manufacturer to increase the wholesale price. At the same time, the unit operational cost is partially passed on to consumers through higher prices in the reselling channel. The higher price mitigates channel price competition and, through cross-price effects, stimulates demand in the direct channel, allowing the manufacturer to raise its retail price without reducing demand. Consequently, both the reselling and direct channels benefit from blockchain adoption. Hence, the manufacturer’s profit increases after blockchain adoption. The improved profitability for both supply chain members increases overall supply chain profit. From a managerial perspective, when consumer preference is high, retailers should strategically accompany blockchain adoption with price increases to capture the value created by trustworthy traceability information.

5.2. Manufacturer Leadership (Scenario M)

In this subsection, we derive the equilibrium results under manufacturer leadership and discuss the impacts of blockchain adoption. We denote this leadership structure as Scenario M. When the manufacturer decides to construct the TSBT, blockchain effectiveness differs across channels. Accordingly, three primary scenarios arise: (1) adoption in the reselling channel only ($M1$); (2) adoption in the direct channel only ($M2$); and (3) adoption in both channels ($M3$). Moreover, in scenarios $M1$ and $M3$, two sub-cases arise depending on the retailer’s participation decision in the reselling channel: (1) with retailer participation ($M1y$ and $M3y$); and (2) without retailer participation ($M1n$ and $M3n$).

If the manufacturer adopts the TSBT in the direct channel, it incurs an operational cost of $2c$ related to production and sales activities. Only the manufacturer participates in the direct channel, consumers cannot obtain cross-verified and trustworthy information. Consequently, there is no expansion effect in the direct channel. Following the same approach as in Scenario R, we formulate the demand and profit functions for each scenario, which are summarized in Table 3.

Similar to scenario R, our analysis shows that $M1n$, $M2$, and $M3$ do not arise in equilibrium. This implies that the manufacturer does not adopt blockchain in the direct channel. The optimal results under scenario $M1y$ are summarized in

Table A2. The optimal results under scenario $M1y$.

Results
$B^{\ast }=\{\frac{\gamma ((1-b)(4-b^3-4b)s+b^3+8b^{2}c+8b-8c)}{2(8-b^4-7b^2)\mu -{(2+b^2)}^2{\gamma }^2},\frac{c}{\gamma }\}$
$w^{M1y\ast }=\frac{(b^4+8)\gamma B^{\ast }+2(1-b^2)b^{2}c+(b^2+8)b+(b^4-b^3-8b+8)s}{2(8-b^4-7b^2)}$
$p_m^{M1y\ast }=\frac{(10-b^2)b\gamma B^{\ast }-4(1-b^2)bc+b^2+(-b^3-b^2+10b-8)s+8}{2(8-b^4-7b^2)}$
$p_r^{M1y\ast }=\frac{b^3+8(1-b^2)c+(-b^4-2b^2+12)B^{\ast }\gamma +(-b^4-b^3-2b^2-8b+12)s+8b}{2(8-b^4-7b^2)}$
$q_m^{M1y\ast }=\frac{(b^2+2)b\gamma B^{\ast }+((b-1)b^2+2b-8)s+b^2+12bc+8}{2(b^2+8)}$
$q_r^{M1y\ast }=\frac{(b^2+2)(\gamma B^{\ast }-2c+s)}{b^2+8}$
${\pi }_m^{M1y\ast }=\frac{1}{4}{B}^{\ast 2}(\frac{{(b^2+2)}^2{\gamma }^2}{-b^4-7b^2+8}-2\mu )+\frac{\gamma B^{\ast }((b-1)(b^3+4b-4)s+b^3+8b^{2}c+8b-8c)}{2(-b^4-7b^2+8)}+\frac{4c(c-s)}{b^2+8}+\frac{(b-2)(b-1)(b^2+b+6)s^2+2(b-1)(b^2+8)s+b^2+8}{4(-b^4-7b^2+8)}$
${\pi }_r^{M1y\ast }=\frac{{(b^2+2)}^2{(\gamma B^{\ast }-2c+s)}^2}{{(b^2+8)}^2}$

. These results hold when $\gamma <{\gamma }_M=\sqrt{\frac{2(8-b^4-7b^2)\mu }{{(b^2+2)}^2}}$ and $c<\frac{s}{2}$.

Lemma 3.

When the manufacturer adopts blockchain in the reselling channel ($M1$), if the information level exceeds a certain threshold (i.e., $B>\frac{c}{\gamma }$), the retailer participates in the TSBT; otherwise, it does not.

Table 3. Demand and profit functions under different scenarios.

Scenario	Member	Demand	Profit
$M1y$	r	$q_r^{M1y}=s-p_r^{M1y}+b{p}_m^{M1y}+\gamma B^{M1y}$	${\pi }_r^{M1y}=(p_r^{M1y}-w^{M1y}-c)q_r^{M1y}$
	m	$q_m^{M1y}=1-s-p_m^{M1y}+b{p}_r^{M1y}$	${\pi }_m^{M1y}=(w^{M1y}-c)q_r^{M1y}+p_m^{M1y}{q}_m^{M1y}-\frac{\mu {\left(B^{M1y}\right)}^2}{2}$
$M1n$	r	$q_r^{M1n}=s-p_r^{M1n}+b{p}_m^{M1n}$	${\pi }_r^{M1n}=(p_r^{M1n}-w^{M1n})q_r^{M1n}$
	m	$q_m^{M1n}=1-s-p_m^{M1n}+b{p}_r^{M1n}$	${\pi }_m^{M1n}=(w^{M1n}-c)q_r^{M1n}+p_m^{M1n}{q}_m^{M1n}-\frac{\mu {\left(B^{M1n}\right)}^2}{2}$
$M2$	r	$q_r^{M2}=s-p_r^{M2}+b{p}_m^{M2}$	${\pi }_r^{M2}=(p_r^{M2}-w^{M2})q_r^{M2}$
	m	$q_m^{M2}=1-s-p_m^{M2}+b{p}_r^{M2}$	${\pi }_m^{M2}=w^{M2}{q}_r^{M2}+(p_m^{M2}-2c)q_m^{M2}-\frac{\mu {\left(B^{M2}\right)}^2}{2}$
$M3y$	r	$q_r^{M3y}=s-p_r^{M3y}+b{p}_m^{M3y}+\gamma B^{M3y}$	${\pi }_r^{M3y}=(p_r^{M3y}-w^{M3y}-c)q_r^{M3y}$
	m	$q_m^{M3y}=1-s-p_m^{M3y}+b{p}_r^{M3y}$	${\pi }_m^{M3y}=(w^{M3y}-c)q_r^{M3y}+(p_m^{M3y}-2c)q_m^{M3y}-\frac{\mu {\left(B^{M3y}\right)}^2}{2}$
$M3n$	r	$q_r^{M3n}=s-p_r^{M3n}+b{p}_m^{M3n}$	${\pi }_r^{M3n}=(p_r^{M3n}-w^{M3n})q_r^{M3n}$
	m	$q_m^{M3n}=1-s-p_m^{M3n}+b{p}_r^{M3n}$	${\pi }_m^{M3n}=(w^{M3n}-c)q_r^{M3n}+(p_m^{M3n}-2c)q_m^{M3n}-\frac{\mu {\left(B^{M3n}\right)}^2}{2}$

Table 3. Demand and profit functions under different scenarios.

Scenario	Member	Demand	Profit
$M1y$	r	$q_r^{M1y}=s-p_r^{M1y}+b{p}_m^{M1y}+\gamma B^{M1y}$	${\pi }_r^{M1y}=(p_r^{M1y}-w^{M1y}-c)q_r^{M1y}$
	m	$q_m^{M1y}=1-s-p_m^{M1y}+b{p}_r^{M1y}$	${\pi }_m^{M1y}=(w^{M1y}-c)q_r^{M1y}+p_m^{M1y}{q}_m^{M1y}-\frac{\mu {\left(B^{M1y}\right)}^2}{2}$
$M1n$	r	$q_r^{M1n}=s-p_r^{M1n}+b{p}_m^{M1n}$	${\pi }_r^{M1n}=(p_r^{M1n}-w^{M1n})q_r^{M1n}$
	m	$q_m^{M1n}=1-s-p_m^{M1n}+b{p}_r^{M1n}$	${\pi }_m^{M1n}=(w^{M1n}-c)q_r^{M1n}+p_m^{M1n}{q}_m^{M1n}-\frac{\mu {\left(B^{M1n}\right)}^2}{2}$
$M2$	r	$q_r^{M2}=s-p_r^{M2}+b{p}_m^{M2}$	${\pi }_r^{M2}=(p_r^{M2}-w^{M2})q_r^{M2}$
	m	$q_m^{M2}=1-s-p_m^{M2}+b{p}_r^{M2}$	${\pi }_m^{M2}=w^{M2}{q}_r^{M2}+(p_m^{M2}-2c)q_m^{M2}-\frac{\mu {\left(B^{M2}\right)}^2}{2}$
$M3y$	r	$q_r^{M3y}=s-p_r^{M3y}+b{p}_m^{M3y}+\gamma B^{M3y}$	${\pi }_r^{M3y}=(p_r^{M3y}-w^{M3y}-c)q_r^{M3y}$
	m	$q_m^{M3y}=1-s-p_m^{M3y}+b{p}_r^{M3y}$	${\pi }_m^{M3y}=(w^{M3y}-c)q_r^{M3y}+(p_m^{M3y}-2c)q_m^{M3y}-\frac{\mu {\left(B^{M3y}\right)}^2}{2}$
$M3n$	r	$q_r^{M3n}=s-p_r^{M3n}+b{p}_m^{M3n}$	${\pi }_r^{M3n}=(p_r^{M3n}-w^{M3n})q_r^{M3n}$
	m	$q_m^{M3n}=1-s-p_m^{M3n}+b{p}_r^{M3n}$	${\pi }_m^{M3n}=(w^{M3n}-c)q_r^{M3n}+(p_m^{M3n}-2c)q_m^{M3n}-\frac{\mu {\left(B^{M3n}\right)}^2}{2}$

The information level required for retailer participation decreases with consumer preference in the reselling channel but increases with the unit operational cost. Accordingly, when consumer preference in the reselling channel is high, the retailer is willing to participate in the TSBT even at a relatively low information level. In contrast, a high unit operational cost discourages the retailer from participating.

Proposition 4.

Under manufacturer leadership, the equilibrium strategy depends on the intensity of channel competition b, the initial market share of the reselling channel s, and consumer preference γ:

(1) 

If $b<max(b_1,b_2)$ and $s<min(s_1,s_2)$, or $max(b_1,b_2)<b<1$, then

(i) 

for $e_3<\gamma \le e_2$, the manufacturer adopts the TSBT in the reselling channel and sets $B^{M1y\ast }=\frac{c}{\gamma }$ to ensure retailer participation (strategy $M1y$);

(ii) 

for $\gamma >e_2$, the manufacturer adopts the TSBT in the reselling channel and sets $B^{M1y\ast }=\frac{\gamma ((1-b)(4-b^3-4b)s+b^3+8b^{2}c+8b-8c)}{2(8-b^4-7b^2)\mu -{(2+b^2)}^2{\gamma }^2}$ with retailer participation (strategy $M1y$);

(iii) 

otherwise, the manufacturer does not adopt the TSBT (strategy N).

(2) 

If $b<max(b_1,b_2)$ and $s\ge min(s_1,s_2)$, then

(i) 

for $\gamma >e_4$, the manufacturer adopts the TSBT in the reselling channel and sets $B^{M1y\ast }=\frac{\gamma ((1-b)(4-b^3-4b)s+b^3+8b^{2}c+8b-8c)}{2(8-b^4-7b^2)\mu -{(2+b^2)}^2{\gamma }^2}$ with retailer participation (strategy $M1y$);

(ii) 

otherwise, the manufacturer does not adopt the TSBT (strategy N).

The thresholds $b_1$, $b_2$, $s_1$, $s_2$, $e_2$, $e_3$, and $e_4$ are defined in Appendix A.

We illustrate Proposition 4 in Figure 4. When channel competition is weak (Figure 4a) and the initial market share of the reselling channel is small ($s<min(s_1,s_2)$), the manufacturer adopts the TSBT when consumer preference in the reselling channel is medium or high. The intuition is that a small s implies the wholesale price and demand in the reselling channel are low, whereas the retail price and demand in the direct channel are high. Under these conditions, adopting the TSBT allows the manufacturer to strengthen the direct channel by mitigating price competition and enhancing brand image. However, this comes at the cost of the reselling channel, which suffers as a result. Consequently, although the manufacturer’s profit margin from the reselling channel declines ($(w^{M1y}-c)q_r^{M1y}<w^N{q}_r^N$), the gains in the direct channel more than offset the losses. As consumer preference in the reselling channel increases further, the manufacturer raises the information level. This adjustment mitigates the negative impact of the unit operational cost on the reselling channel and may even benefit both channels. The blue and green regions in Figure 4a reflect differences in the manufacturer’s information disclosure strategy at different consumer preference levels. In the blue region, the manufacturer strategically sets a low information level to favor the direct channel. However, this low level deters retailer participation, necessitating an adjustment of the information level to $B=\frac{c}{\gamma }$ to secure retailer engagement.

When the initial market share of the reselling channel is large ($s>min(s_1,s_2)$ in Figure 4a), the wholesale price and demand in the reselling channel are high, while the retail price and demand in the direct channel are low. In this case, the incentive to expand the direct channel diminishes. The manufacturer adopts TSBT only when consumer preference in the reselling channel is high. Under such conditions, blockchain adoption either reduces the harm to the reselling channel or generates gains in both channels.

However, when channel competition is intense (Figure 4b), the manufacturer may adopt the TSBT even when the reselling channel has a high initial market share and consumer preference in the reselling channel is low. The rationale is that, in a highly competitive environment, demand in the direct channel is more sensitive to the retail price in the reselling channel. By adopting the TSBT, the manufacturer can raise the retailer’s retail price, thereby mitigating price competition and amplifying the cross-price effect. As a result, both demand and the retail price in the direct channel increase more sharply than under weak competition. This generates additional revenue, which offsets potential losses in the reselling channel and renders blockchain adoption profitable. From a managerial perspective, manufacturers must consider the reselling channel’s initial market share, channel competition intensity, and consumer preference. They also need to ensure retailer cooperation. In particular, when the reselling channel’s initial market share is small or competition is intense, setting a sufficiently high information level can secure retailer participation, prevent channel conflict, and allow manufacturers to capture the value of blockchain adoption.

Based on the equilibrium results, we examine the impact of blockchain adoption.

Proposition 5.

In equilibrium, when the manufacturer adopts blockchain, the retailer and the manufacturer set higher prices, and the manufacturer simultaneously achieves higher demand in the direct channel and higher profit. However, the reselling channel’s demand and the retailer’s profit increase with blockchain adoption if either (1) the intensity of channel competition b is low and the initial market share of the reselling channel is high (i.e., $b<b_m$, $s>max(min(s_1,s_2),s_3)$), or (2) consumer preference in the reselling channel is high (i.e., $\gamma >e_5$). The thresholds $b_m$, $s_3$, and $e_5$ are defined in Appendix A.

Proposition 5 shows that the direct channel’s demand consistently increases with blockchain adoption. However, the reselling channel’s demand depends on consumer preference in the reselling channel, the initial market share of the reselling channel, and the intensity of channel competition. Specifically, when competition is weak and the reselling channel’s initial market share is high, harming the reselling channel is unprofitable. As a result, the manufacturer adopts blockchain only when consumer preference in the reselling channel is high. This raises demand in both channels. By contrast, when the initial market share of the reselling channel is low or channel competition is intense, the manufacturer may adopt TSBT when consumer preference in the reselling channel is medium with a low information level. In this case, blockchain adoption boosts demand in the direct channel but may reduce demand in the reselling channel due to higher retail prices and a limited expansion effect. As consumer preference in the reselling channel increases further, the manufacturer raises the information level. This leads to higher demand in both channels. Hence, blockchain adoption has different impacts on demand and profit under different leadership structures. This difference arises because the retailer earns profit exclusively from the reselling channel and therefore adopts blockchain only when that channel benefits. By contrast, the manufacturer may adopt blockchain even if one channel is harmed, provided that total profit increases. This finding implies that, by responding to consumer preference through blockchain adoption, manufacturers can better manage multi-channel demand and enhance the supply chain’s overall operational sustainability. In doing so, they need to account for heterogeneous blockchain effectiveness across channels.

Corollary 1.

In equilibrium, when the manufacturer adopts blockchain, the supply chain achieves a higher profit when consumer preference γ is high.

The overall profitability of the supply chain depends critically on the retailer’s profit. Specifically, when consumer preference in the reselling channel exceeds a certain threshold, both the retailer and the manufacturer benefit from blockchain adoption. As a result, total supply chain profit increases. By contrast, when consumer preference in the reselling channel is below a certain threshold, the retailer may incur substantial losses. These losses more than offset the manufacturer’s gains. As a result, overall supply chain profit decreases.

6. Comparative Analysis Across Different Leadership Structures

In this section, we examine the similarities and differences between two leadership structures to evaluate the impact of leadership structures on blockchain adoption decisions and supply chain members. Our analysis focuses on condition $\gamma <{\gamma }_M$ to avoid trivial outcomes.

We first examine the leaders’ adoption decisions across various leadership structures.

Proposition 6.

The adoption decisions under different leadership structures are as follows:

(1) 

Leaders under both leadership structures do not adopt blockchain if consumer preference γ is low.

(2) 

Under scenario M, the manufacturer adopts blockchain, while under scenario R, the retailer withdraws it if consumer preference is medium.

(3) 

Leaders under both leadership structures adopt blockchain if the unit operational cost of using blockchain is low and consumer preference is high (i.e., $0<c<c_0$ and $\gamma >e_1$).

The threshold $c_0$ is defined in Appendix A.

Proposition 6 demonstrates that leadership structure plays a critical role in blockchain adoption, especially when consumer preference in the reselling channel is medium. When consumer preference in the reselling channel is low, both leadership structures refrain from adopting blockchain. As consumer preference increases, however, adoption incentives diverge. A manufacturer leader tends to adopt blockchain because it increases the manufacturer’s profit, even at the expense of the reselling channel. The profit increase comes from the enhancement of the direct channel through blockchain. By contrast, a retailer leader shows weaker incentives to adopt under the same conditions. When consumer preference in the reselling channel is high and the unit operational cost is low, blockchain adoption occurs under both leadership structures. If the operational cost is high, the retailer’s profit margins are eroded, and the incremental gains from improving the information level are limited. This discourages the retailer from adopting blockchain. By contrast, the manufacturer captures a larger share of the gains from blockchain adoption. These gains are large enough to offset the higher operational cost. Thus, when consumer preference in the reselling channel is high, a manufacturer leader consistently adopts blockchain. Hence, when consumer preference in the reselling channel is below a certain threshold, manufacturers are more likely to benefit from taking the initiative to lead blockchain adoption, since retailers are unlikely to adopt on their own.

Proposition 7.

When blockchain is adopted under both leadership structures ($0<c<c_0$ and $\gamma >e_1$), manufacturer leadership yields a higher information level, higher prices, higher demand, and higher profits for both supply chain members than retailer leadership does.

Proposition 7 indicates that, when blockchain is adopted under both leadership structures, manufacturer leadership yields higher outcomes. Under retailer leadership, blockchain adoption increases consumer trust and supports higher retail prices. Anticipating this effect, the manufacturer adjusts the wholesale price to appropriate a large share of the additional value created by blockchain adoption. As a result, the retailer captures only limited marginal benefits from further increasing the information level. Consequently, the retailer has little incentive to set a high information level. By contrast, under manufacturer leadership, a higher information level directly stimulates demand in the reselling channel. This increase in demand provides greater pricing flexibility, enabling both the manufacturer and the retailer to raise prices. In addition, the cross-channel effect further increases demand in the direct channel, because a higher retail price in the reselling channel shifts part of demand toward the direct channel. This mechanism further strengthens the manufacturer’s incentive to set a high information level. Hence, the retailer benefits from the elevated information level while avoiding the construction cost. The manufacturer gains from higher information level that more than offset the construction cost. Consequently, profits for both the retailer and the manufacturer are higher under manufacturer leadership. This finding implies that when consumer preference exceeds a certain threshold, firms should favor manufacturer leadership in blockchain adoption. By enabling a higher information level and a more credible TSBT, manufacturer leadership contributes to the long-term sustainable operation of the dual-channel supply chain.

7. Extension

This section examines the differences between traditional databases and TSBT with only one participant. In TSBT, data immutability ensures that modifications leave traces, thereby increasing the cost of falsification. Moreover, consumers can access comprehensive traceability information through the direct channel, which enhances consumers’ trust and purchase likelihood. Therefore, when the manufacturer leads the implementation of TSBT, blockchain may satisfy consumers’ demand for authentic product information in the direct channel. Let $\theta$ denote consumer preference in the direct channel. Since this channel involves only one participant, blockchain information lacks cross-verification. We assume $\gamma ⩾\theta$. To indicate the extended setting, we denote the corresponding decision variables and functions with a hat (e.g., $\widehat{B}$, $\widehat{{\pi }_r}$, etc.). By contrast, in the reselling channel, a single-participant TSBT cannot provide end-to-end information, so it is not considered here.

Table 4. Demand and profit functions under scenario M in the extension.

Scenario	Member	Demand	Profit
$M1y$	r	${\widehat{q}}_r^{M1y}=s-{\widehat{p}}_r^{M1y}+b{\widehat{p}}_m^{M1y}+\gamma {\widehat{B}}^{M1y}$	${\widehat{\pi }}_r^{M1y}=({\widehat{p}}_r^{M1y}-{\widehat{w}}^{M1y}-c){\widehat{q}}_r^{M1y}$
	m	${\widehat{q}}_m^{M1y}=1-s-{\widehat{p}}_m^{M1y}+b{\widehat{p}}_r^{M1y}$	${\widehat{\pi }}_m^{M1y}=({\widehat{w}}^{M1y}-c){\widehat{q}}_r^{M1y}+{\widehat{p}}_m^{M1y}{\widehat{q}}_m^{M1y}-\frac{\mu {\left({\widehat{B}}^{M1y}\right)}^2}{2}$
$M1n$	r	${\widehat{q}}_r^{M1n}=s-{\widehat{p}}_r^{M1n}+b{\widehat{p}}_m^{M1n}$	${\widehat{\pi }}_r^{M1n}=({\widehat{p}}_r^{M1n}-{\widehat{w}}^{M1n}){\widehat{q}}_r^{M1n}$
	m	${\widehat{q}}_m^{M1n}=1-s-{\widehat{p}}_m^{M1n}+b{\widehat{p}}_r^{M1n}$	${\widehat{\pi }}_m^{M1n}=({\widehat{w}}^{M1n}-c){\widehat{q}}_r^{M1n}+{\widehat{p}}_m^{M1n}{\widehat{q}}_m^{M1n}-\frac{\mu {\left({\widehat{B}}^{M1n}\right)}^2}{2}$
$M2$	r	${\widehat{q}}_r^{M2}=s-{\widehat{p}}_r^{M2}+b{\widehat{p}}_m^{M2}$	${\widehat{\pi }}_r^{M2}={\widehat{q}}_r^{M2}({\widehat{p}}_r^{M2}-{\widehat{w}}^{M2})$
	m	${\widehat{q}}_m^{M2}=1-s-{\widehat{p}}_m^{M2}+b{\widehat{p}}_r^{M2}+\theta {\widehat{B}}^{M2}$	${\widehat{\pi }}_m^{M2}={\widehat{q}}_m^{M2}({\widehat{p}}_m^{M2}-2c)+{\widehat{w}}^{M2}{\widehat{q}}_r^{M2}-\frac{\mu {\left({\widehat{B}}^{M2}\right)}^2}{2}$
$M3y$	r	${\widehat{q}}_r^{M3y}=s-{\widehat{p}}_r^{M3y}+b{\widehat{p}}_m^{M3y}+\gamma {\widehat{B}}^{M3y}$	${\widehat{\pi }}_r^{M3y}={\widehat{q}}_r^{M3y}({\widehat{p}}_r^{M3y}-{\widehat{w}}^{M3y}-c)$
	m	${\widehat{q}}_m^{M3y}=1-s-{\widehat{p}}_m^{M3y}+b{\widehat{p}}_r^{M3y}+\theta {\widehat{B}}^{M3y}$	${\widehat{\pi }}_m^{M3y}={\widehat{q}}_m^{M3y}({\widehat{p}}_m^{M3y}-2c)+({\widehat{w}}^{M3y}-c){\widehat{q}}_r^{M3y}-\frac{\mu {\left({\widehat{B}}^{M3y}\right)}^2}{2}$
$M3n$	r	${\widehat{q}}_r^{M3n}=s-{\widehat{p}}_r^{M3n}+b{\widehat{p}}_m^{M3n}$	${\widehat{\pi }}_r^{M3n}={\widehat{q}}_r^{M3n}({\widehat{p}}_r^{M3n}-{\widehat{w}}^{M3n})$
	m	${\widehat{q}}_m^{M3n}=1-s-{\widehat{p}}_m^{M3n}+b{\widehat{p}}_r^{M3n}+\theta {\widehat{B}}^{M3n}$	${\pi }_m^{M3n}={\widehat{q}}_m^{M3n}({\widehat{p}}_m^{M3n}-2c)+({\widehat{w}}^{M3n}-c){\widehat{q}}_r^{M3n}-\frac{\mu {\left({\widehat{B}}^{M3n}\right)}^2}{2}$

summarizes the corresponding demand and profit functions.

By solving the game, we find that strategy $M3n$ is dominated by strategy $M2$, while $M1n$ is dominated by strategy N. These dominance relations underscore the importance of partner participation in determining optimal blockchain adoption strategies in dual-channel supply chains. Given the model’s complexity, we therefore conduct numerical studies to identify the manufacturer’s equilibrium strategy. In practice, the unit operational cost of using blockchain typically ranges from $0.01 to $0.10 per transaction or per product unit (see https://thetraceabilityhub.com/blockchain-tracking-system-when-and-why-its-needed/ (accessed on 26 November 2025) and https://antdigital.com/solutions/mytc (accessed on 26 November 2025)). Consistent with Ju et al. [48], Zhong et al. [9], Zhang et al. [49], and Liu et al. [50], we set $b=\frac{1}{2}$, $\mu =2$, and $c=\frac{1}{20}$. Regarding market shares, industry reports indicate that in the luxury goods market, the direct channel accounted for approximately 50% of total sales in 2022. Leading luxury groups such as Kering derived about 80% of their revenue from the direct channel (https://www.hangyan.co/reports/3250077243392656688?utm_source=dj (accessed on 25 December 2025)). Motivated by these observations, we consider representative cases with $s=\frac{1}{2}$ and $s=\frac{1}{5}$ in the numerical analysis to capture different degrees of market share asymmetry. The corresponding equilibrium outcomes are presented in Figure 5 and illustrate how consumer preferences across the two channels influence manufacturer’s equilibrium strategy.

When consumer preferences in both channels are low, the manufacturer refrains from blockchain adoption. If consumer preference in the reselling channel is moderate, the equilibrium strategy depends on consumer preference in the direct channel. Specifically, when consumer preference in the direct channel is low, adopting blockchain in both channels brings only limited demand expansion in the direct channel. Meanwhile, the additional operational costs raise retail prices and suppress demand. As a result, overall profitability decreases. Under this condition, adopting blockchain only in the reselling channel avoids these drawbacks. As consumer preference in the direct channel increases, the joint preferences across channels support a higher information level. Consequently, blockchain adoption in both channels enhances demand and prices in each channel. The resulting gains offset the higher costs, so full adoption is optimal.

Figure 5. Equilibrium considering differences under scenario M: (a) $s=\frac{1}{2}$, (b) $s=\frac{1}{5}$.

Figure 5. Equilibrium considering differences under scenario M: (a) $s=\frac{1}{2}$, (b) $s=\frac{1}{5}$.

When consumer preference in the reselling channel is high, the manufacturer is motivated to adopt blockchain in both channels even when consumer preference in the direct channel is low. The rationale is that blockchain adoption in both channels enables a higher information level, which better satisfies consumer demand in the reselling channel. Meanwhile, although consumer preference in the direct channel is low, the expansion effect remains significant because of the elevated information level. As a results, the benefits generated across both channels are sufficient to compensate for the increased costs.

Notably, blockchain is never adopted only in the direct channel. This is because if consumer preference in the direct channel is sufficient to justify adoption, consumer preference in the reselling channel is already non-negligible. In this case, joint adoption across both channels generates higher profits by supporting a higher information level. Moreover, Figure 5a,b indicate that a smaller initial market share of the reselling channel strengthens the manufacturer’s incentive to adopt blockchain at lower consumer preference levels. By strategically timing blockchain adoption according to channel characteristics, manufacturers can improve resource allocation efficiency and avoid unnecessary technological investment. Such behavior aligns with the eco-efficiency principle underlying sustainable operations.

Consistent with the main model, the comparison of equilibrium outcomes across leadership structures yields qualitatively similar results, as illustrated in Figure 6 ($b=\frac{1}{2}$, $s=\frac{1}{2}$, $c=\frac{1}{20}$, $\mu =2$, $\theta =\frac{1}{100}$). Specifically, blockchain adoption is more likely to occur under manufacturer leadership. Moreover, when blockchain adoption arises under both leadership structures, both supply chain members achieve higher profits under manufacturer leadership. The above findings align with the practices of Nestlé and Louis Vuitton, both of which operate dual-channel supply chains and lead blockchain adoption. Due to differences in the market shares of their reselling channels and in consumer preferences for traceability between food-beverage and luxury products, the two firms adopt different blockchain strategies. Louis Vuitton adopts blockchain in both channels, while Nestlé mainly adopts it in the reselling channel (https://us.louisvuitton.com/eng-us/faq/services/aura-consortium-blockchain-the-lv-diamond-certificate (accessed on 25 December 2025) and https://www.blocktempo.com/nestle-carrefour-adopt-ibm-blockichain/ (accessed on 25 December 2025)).

8. Conclusions

8.1. Summary

This study investigates blockchain adoption strategies in a dual-channel supply chain with asymmetric market shares under different leadership structures. Our findings highlight several key insights. The analysis shows that under retailer leadership, blockchain adoption occurs once consumer preference in the reselling channel exceeds a threshold. Adoption increases demand and prices in the supply chain. As a result, both the retailer and the manufacturer earn higher profits. Under manufacturer leadership, blockchain adoption depends jointly on consumer preference in the reselling channel, the initial market share of the reselling channel, and the intensity of channel competition. Under manufacturer leadership, adoption generally benefits the manufacturer but may harm the retailer when channel competition is intense or when the reselling channel’s initial market share is small. A comparison of equilibrium outcomes across leadership structures indicates that blockchain adoption is more likely under manufacturer leadership. Moreover, when both leaders are willing to adopt blockchain under their respective leadership structures, manufacturer leadership leads to higher demand, prices, and profits. Finally, considering the differences between traditional databases and TSBT with only a single participant, the extension analysis shows that the main results remain valid when consumer preference in the direct channel is low. However, when such preference is at a moderate level, the equilibrium strategy under manufacturer leadership depends on consumer preference in the reselling channel.

8.2. Theoretical and Managerial Implications

This study offers several theoretical implications for the literature on blockchain adoption in supply chains. Most existing studies assume that blockchain adoption exerts uniform effects across sales channels. In contrast, we consider that blockchain effectiveness differs across channels and show that incorporating channel-specific blockchain effectiveness alters pricing decisions, profit distribution, and the leader’s blockchain adoption strategy under different leadership structures. Second, our study underscores the theoretical significance of asymmetric market shares in shaping blockchain adoption strategies. Asymmetric market shares interact with leadership structures and with heterogeneous blockchain effectiveness across channels. This interaction affects leader’s strategic incentives for blockchain adoption. Hence, market share asymmetry is a crucial theoretical dimension that should be considered in the analysis of blockchain adoption in dual-channel supply chains. Finally, by jointly considering heterogeneous blockchain effectiveness across channels and asymmetric market shares within a unified analytical framework, our study complements existing theories that typically examine these factors separately.

Our findings also provide actionable insights for practitioners operating in dual-channel supply chains. First, supply chain leaders should recognize that blockchain adoption decisions depend critically on the initial market shares, channel competition, consumer preference, the costs of TSBT, and partners’ participation incentives. Accordingly, firms should carefully assess these factors before they adopt blockchain. In practice, some firms actively collect relevant information to support these assessments. For example, Tracr, a natural diamond traceability platform, has conducted consumer surveys in the Chinese market to assess consumer preference for blockchain-based traceability information. This action enables firms to adjust the information level accordingly. Prior case studies also emphasize that clearly identifying the costs of TSBT is a prerequisite for effective implementation [28]. Moreover, firms can leverage historical sales data to evaluate the initial market shares of reselling and direct channels. They can then develop channel-specific blockchain adoption strategies instead of implementing blockchain uniformly across all channels. For instance, Nestle has primarily implemented TSBT in collaboration with Carrefour in its reselling channel, while its online direct channel has not been observed to adopt TSBT.

Second, our results suggest that manufacturers are more likely to benefit from leading blockchain adoption. Under manufacturer leadership, adoption occurs at lower levels of consumer preference and supports a higher information level than under retailer leadership. This pattern is consistent with observed industry practices. For example, Beingmate, a dairy manufacturer operating a dual-channel supply chain, has taken the lead in blockchain adoption. Similarly, in the luxury industry, well-known brands such as Louis Vuitton and Cartier lead blockchain initiatives to meet consumers’ demand for authentic and transparent product information.

Third, our results suggest that the manufacturer’s incentive design should depend on consumer preference and the initial market share of the reselling channel. When consumer preference is not high and the initial market share of the reselling channel is small, manufacturer-led blockchain adoption may harm the retailer’s payoff. In this case, manufacturers may need to offer subsidies to retailers to sustain cooperation and achieve a win-win outcome. By contrast, when consumer preference is high or the reselling channel’s initial market share is large, manufacturer leadership may induce retailer free riding. Under such conditions, manufacturers can instead consider cost-sharing arrangements with retailers to jointly develop the TSBT. For example, Louis Dreyfus Company has led the development of TSBT in collaboration with downstream partners, including the juice bottler Refresco and the supermarket Albert Heijn. Similarly, the food and beverage manufacturer Nestle has collaborated with Carrefour to provide transparent food traceability to consumers through TSBT.

8.3. Limitations and Further Research

While our work provides insights into firms’ responses to blockchain under different leadership structures, several limitations remain. First, we focus on a simple supply chain with a single manufacturer and a single retailer. In reality, however, supply chains are often more complex, involving multiple competing manufacturers, retailers, and platforms. As a result, our findings may not fully capture the strategic interactions and adoption incentives that arise in more complex supply chain networks. Future research could extend the current study to include competitive platforms or retailers to better understand how leadership structures influence blockchain adoption. Second, this study primarily examines the direct effects of blockchain adoption. Future work could explore cross-channel spillovers and cost-sharing mechanisms among supply chain members, particularly in the context of green and sustainable supply chain management.