Journal of Environmental Management 373 (2025) 123545

Contents lists available at ScienceDirect

Journal of Environmental Management

journal homepage: www.elsevier.com/locate/jenvman

Research article

Dynamic analysis of green technology innovation in products and processes

under supply chain competition scenarios—A study based on stochastic
differential game model
Rui Sun, Da-yi He ∗, Jing-jing Yan
School of Economics & Management, China University of Geosciences (Beijing), 29#, Xueyuan Rd., Beijing, 100083, State One, PR China

ARTICLE INFO ABSTRACT

Keywords: This paper investigates the dynamics of market competition among manufacturers pursuing two types of
Green product innovation green technology innovation: product innovation and process innovation, which provides a novel perspective
Green process innovation on how green technological innovations contribute to enhancing product greenness. Using a two-tier supply
Supply chain
chain framework, the study examines the differences in cost-sharing across various types of innovation and
Cost-sharing mode
the impact of game players on product greenness. A stochastic differential game model involving a shared
Stochastic differential game
Chaos
supplier and two competing manufacturers is constructed. Main findings are as follows: (1) In green product
innovation, when market preference for greenness is relatively low, the formation of cost-sharing contracts
is less influenced by the intensity of greenness-based market competition. In such scenarios, government
subsidies emerge as a crucial tool to effectively guiding the market mechanism towards achieving its intended
objectives. (2) As the market competition intensifies, manufacturers’ incentives towards suppliers will shift
from reducing costs to increasing demand. (3) When the intensity of green competitive is low, and supplier
process innovation efficiency is high, manufacturers should bear a greater share of costs; Conversely, suppliers
should bear more costs. When supplier process innovation efficiency is higher and market competition is
stronger, suppliers should bear more costs if consumer green preferences are lower. (4) Over time, considering
consumer preferences, a marginal increase in price exceeds the marginal increment in greenness with respect
to contributing to stabilise price fluctuations. Conversely, it tends to raise the mean value of the price. (5)
The increase in market competition intensity accelerates suppliers’ and manufactures’ strategy adjustments for
green product innovation, but decelerates the strategy adjustment for suppliers’ green process innovation.

1. Introduction directly reduce environmental costs and meet consumer demand for
environment-friendly products (Song et al., 2024; Li et al., 2023).
With the growing awareness of environmental protection and the The Organisation for Economic Cooperation and Development (OECD)
widespread adoption of sustainable development principles, the en- categorises green technology innovation into green product innova-
hancement of a product’s green attributes has become a central focus in
tion and green process innovation. Green process innovation includes
market competition, alongside continuous improvements in functional
cleaner production technology innovation and end-of-pipe treatment
attributes. Greenness reflects the degree of environmental impact of
a product throughout its life cycle, including factors such as resource technology innovation, which aims to reduce harmful substances, de-
consumption, energy utilisation, waste recycling, pollutants emissions, crease pollutant emissions, and increase energy efficiency by improving
and other aspects (Cai et al., 2023). The differentiation in greenness existing production processes or developing new ones (Xie and Zhu,
not only affects consumers’ purchasing choices but also raises higher 2021). Green product innovation emphasises the integration of environ-
requirements for market positioning and competitive strategies of en- mental protection concepts into the selection of raw materials, product
terprises. This will encourage enterprises to continuously carry out design, packaging and other related aspects to minimise the negative
technological innovation by seeking new technologies, materials, and environmental impact throughout the product life cycle (Xie and Zhu,
production methods (Zhou and Duan, 2022). Unlike traditional tech- 2021). Both types of green technological innovation can increase the
nological innovation, the main goal of green technological innovation
greenness of the product (Wang et al., 2024). Green product innovation
is to alleviate the negative impact on the environment, which can

∗ Corresponding author.
E-mail addresses: [email protected] (R. Sun), [email protected] (D.-y. He), [email protected] (J.-j. Yan).

https://doi.org/10.1016/j.jenvman.2024.123545
Received 15 May 2024; Received in revised form 6 November 2024; Accepted 29 November 2024
Available online 14 December 2024
0301-4797/© 2024 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

attract consumers with green preferences by product differentiation green technological innovation is different from traditional technolog-
while green process innovation strategy offers price advantages by ical innovation. Therefore, the path of improving the competitiveness
saving resources (Xie and Zhu, 2021). of enterprises by green technological innovation is also different from
Green innovation depends on collaboration throughout the supply traditional technological innovation because their objects are changed
chain. Greenness flows from upstream to downstream enterprises (Cai from the traditional attributes to the green ones.
et al., 2023; Xu and Lin, 2024). The supplier, often providing raw Therefore, green technological innovations have played an impor-
materials to multiple manufacturers, play a critical role in the supply tant role in the green transformation of supply chains. Unfortunately,
chain. Core manufacturers within the supply chain must collaborate the research on competition between different types of green tech-
with their suppliers to develop effective green strategies. Given the nology innovations is still poor. In addition, due to the uncertainty
limited resources, manufacturers will compete for opportunities to of technological development, the change of market demand and the
cooperate with suppliers. Regardless of type of innovation, a large complexity of supply chain, the green technology innovation in supply
amount of innovation costs need to be invested, and manufacturers and chain has high randomness (Kennedy et al., 2021; Bensoussan et al.,
suppliers will choose different cost-sharing modes in different types of 2019). The stochastic differential game can represent not only the
green technology innovation. dynamics of the innovation process, but also its randomness. Hence,
Product innovation is often related to brand image and product we address the following crucial questions.
uniqueness, so manufacturers often bear part of suppliers’ innovation
costs to incentivise suppliers support to innovation and to ensure (1) How can cost-sharing contracts in green product innovation
product competitiveness in the market (Jeong and Lee, 2024). For and green process innovation increase the greenness of supply
example, in the ‘‘Sustainable Resource Management and Climate Pro- chains?
tection’’ programme, Coca-Cola actively cooperates with its suppliers (2) How does market competition and consumer green preferences
and bears part of the suppliers’ R&D costs while bearing its own green affect the cost-sharing contract of green technological innovation
product R&D costs independently. This mode of cost-sharing ensures between suppliers and manufactures?
the manufacturer’s control over the supplier and investment in the core (3) How does product greenness affect product pricing dynamically?
part of product innovation. To answer these questions, we take supply chain green technology
Process innovation, on the other hand, emphasises the optimisation innovation as the research object and construct a stochastic differen-
of production processes and partnerships, whereby manufacturers and tial game model considering different innovation types and market
suppliers work together to improve the production efficiency and en- competition. The cost-sharing methods and the formation of product
vironmental performance of their products through joint inputs and greenness in different innovation types are distinguished. Then, we
cost sharing (Xie et al., 2024), and improve the production efficiency explore the conditions for the establishment of cost-sharing contracts
and environmental performance of their products. For example, Apple and the mechanism of the role of cost-sharing proposition in improving
reduces pollutants in its manufacturing process by recycling waste product greenness in different innovation modes. Finally, we analyse
and using environment-friendly materials by assessing the costs of the product attributes and market competition that affect product
process innovations with its suppliers and sharing them proportionally. pricing, as well as the factors that influence the cost-sharing rate of
This cost-sharing mode helps to promote cooperation between the two participating subjects.
parties and to leverage their respective professional strengths. The marginal contribution of this paper is to distinguish between
Moreover, it has shown that more than 60% of respondents are the cooperation and cost-sharing approaches of supply chain subjects in
willing to pay a high premium for green products (Tully and Winer, different innovation types, and to explore the mechanism of the role of
2014). Hence, increased greenness of products enhances consumers’ cost-sharing approaches in different innovation types. Moreover, green
willingness to pay (Liu et al., 2012; Laroche et al., 2001; Zhu and technology innovation is a dynamic cumulative process with spillover
He, 2017; Yenipazarli and Vakharia, 2015; Chen, 2001; Yenipazarli effects, where the easiest-to-implement technologies tend to be the
and Vakharia, 2017; Du et al., 2018), necessitating dynamic pricing first to be realised and further improvements become more difficult,
adjustments to reflect these market sentiments. so this paper represents the process of innovation accumulation under
Numerous studies have shown that green technology innovation the change of growth rate with a Logistic function. In addition, this
plays an important role in supply chain green transformation (Yin and paper also expresses the price by stochastic differential equations to
Zhao, 2024a,b) such as improving the production efficiency, product analyse the statistical characteristics, and then builds a link between the
green attributes and market share of enterprises (Li and Gao, 2022; Xu formation mechanism of the price and the innovation efforts, reflecting
et al., 2017; Xu and Lin, 2024), therefore necessitating the coopera- its changes in the stochastic variables. This not only helps enterprises
tion of enterprises in the supply chain (Ranjan et al., 2023). On the to decide the optimal investment in green technology innovation but
one hand, many studies have explored the mechanism of green tech- also provide reliable theoretical support for the adaptation of cost-
nology innovation on greenness by distinguishing between different sharing contract and market strategies. Moreover, it offers reference
supply chain structures and product types (Zhu and He, 2017; Musen value for enterprises to formulate product positioning strategy and
and Jianxiong, 2018), however neglecting the competitive relationship improve market competitiveness.
from the perspective of innovation types. They only regard green The rest parts of this paper are organised as follows. In Section 2, we
technology innovation as a tool to reduce emissions and believe that it discuss related literature. The stochastic differential game models are
can improve the greenness of products (Driessen et al., 2013; Liu et al., established in Section 3. The model is analysed and further extended
2020). On the other hand, the existing studies on cost-sharing across in Section 4 and Section 5, respectively. Section 6 examines numer-
supply chain have considered the simple one-way or two-way cost ical results and Section 7 concludes the research findings and future
sharing between manufacturers and suppliers (Kennedy et al., 2021; directions.
Mohammaditabar et al., 2016; Liu et al., 2020), but not recognising
that various types of innovation have different ways of cooperation and 2. Literature review
cost-sharing. With limited applications find that two-way cost-sharing
is valuable to explore different cost-sharing methods by defining a 2.1. Green technology innovation, product innovation and process innova-
core enterprise. In traditional technological innovation, product inno- tion
vation is often defined as improving product quality; process innovation
is often defined as reducing product costs (Gopalakrishnan, 1999; Unlike traditional innovation, green innovation emphasises the
Lambertini and Mantovani, 2008, 2010). However, the strategic goal of adoption of new technologies and concepts to achieve efficient resource

2
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

use and effective pollution reduction, while simultaneously improving the sustainability of the production process and thus reduce produc-
economic performance. In addition, the aim of green innovation is to tion costs (Ayman, 2022). On the other hand, upstream enterprises
generate good environmental benefits, not just to reduce environmen- may have adopted green technology or materials, while downstream
tal pressures (Driessen et al., 2013). Therefore, green innovation is enterprises may have deeper insights into market demand informa-
regarded as an important strategy for enterprises to gain sustainable tion. Hence, joint cooperation in the supply chain can be improved
competitive advantage (Fernando et al., 2019; Chan et al., 2016). by integrating resources (Li et al., 2020). Moreover, cooperation al-
Green process innovation and green product innovation play dif- lows upstream and downstream enterprises to share costs, risks and
ferent roles in the sustainability performance of enterprises. Green uncertainties in green technology innovation (Liu and Giovanni, 2019).
process innovation can effectively increase energy utilisation and re- Many studies have explored the influencing factors of green input
duce waste generation through the use of alternative energy sources, decision-making and coordination in supply chains. Firstly, the inten-
improved processes and resources recycling, ensuring that manufactur- sity of green competition is positively correlated with the efficiency
ing processes comply with environmental regulations and thus avoiding of green technology innovation among manufacturers, as well as with
environmental pollution penalties (Xie et al., 2016; Yu et al., 2017). the magnitude of green R&D investment (Amin-Naseri and Khojasteh,
Green product innovation reduces the energy consumption of products 2015). Secondly, different power structures of supply chain members
in use by adopting environmentally friendly materials, and builds a have an important impact on their emission reduction strategies. The
better recycling system (Chan et al., 2016), thus reducing the neg- impacts have been explored from the perspectives of supplier-led,
ative impact on the environment throughout the life cycle of the manufacturer-led, retailer-led, and multidimensional-led models (Wang
product (Chen et al., 2006). More importantly, the outstanding en- et al., 2022, 2021; Amin-Naseri and Khojasteh, 2015; Ghosh and Shah,
vironmental characteristics of green products can help enterprises to 2012), respectively. Finally, the game of green supply chain includes
establish differentiated competitive advantages, which in turn help en- not only the game among supply chain members, but also the game
terprises obtain environmental premium (Zong et al., 2014). Moreover, under the involvement of the government (Zhang et al., 2019). The
green process innovation can improve production efficiency and reduce government can regulate product sales by providing retailers with guide
production costs by reducing resource inputs (Yu et al., 2017; Chiou prices and financial subsidies. These governmental intervention can
et al., 2011). harmonise not only the conflict between pricing and green level deci-
Green process innovation and green product innovation pursue dif- sions (Yang and Xiao, 2017), but also the conflict between consumers
ferent environmental goals (Arfi et al., 2018), which leads to differences and suppliers (Li and Gao, 2022). In addition, the cost-sharing contract
in green investments and energy constraints. These differences, in turn is an effective coordination for members in a supply chain to work
have different impacts on the scope for improvement in enterprises’ closely together (Xu et al., 2017), which can not only improve the
product greenness and the supply chain performance, but also realise
financial and social responsibility performance. Green product inno-
the Pareto improvement of the profits of the supply chain as a whole
vation can bring more differentiated products and capital investment
and its members (Kuchesfehani et al., 2023).
portfolios for enterprises (Li et al., 2019). From the perspective of
market demand-driven mechanism, green product innovation develops
2.3. Greenness competition
products in response to market demand, thus strengthening the differ-
entiation advantage of the products. This not only helps enterprises
Rising environmental awareness has increased consumers’ prefer-
to develop new markets and increases market share, but also meets
ence for green products, prompting many enterprises to raise the
the customers’ pursuit of product quality and environmental protection
greenness level of their products. Studies have explored the coordina-
awareness, which in turn brings more environmental premiums. Then,
tion problems of competitive supply chains. By distinguishing between
the capital market’s ‘‘green investment’’ will flow to enterprises that
types of manufacturers, different competitors, and different leadership
are more likely to produce differentiated green products (Xie and
structures in the supply chain, optimal product pricing and greenness
Zhu, 2021). Therefore, compared with green process innovation, green
decisions under the influence of risk preference and inter-chain com-
product innovation can help enterprises attract more investors, which
petition are explored, and coordination contracts are designed (Amin-
will reduce the cost and risk of product development, and thus probably Naseri and Khojasteh, 2015). The effects of risk preference and com-
improving financial performance. Green process innovation can help petition on the profitability of green enterprises are then explored, as
enterprises to break through resource constraints more effectively to well as the strategic choices of competitive green supply chains for
meet government’ requirements of energy-saving and emission reduc- different modes of cost-sharing for green technology innovation (Liu
tion, thus improving their social responsibility performance (Xie et al., et al., 2012; Zhang et al., 2015). It has been found that horizontal
2019). From the perspective of the regulatory mechanism for energy competition between chains can incentivise enterprises to improve
conservation and emission reduction, green process innovation not only the greenness level of their products and thus increase supply chain
focuses on the adoption of clean energy and the introduction of energy- profits (Deng et al., 2020), based on which the introduction of the
saving equipment and other source production technologies to alleviate wholesale price contract and the revenue-sharing contract can effec-
the constraints of non-renewable energy, but also devotes itself to the tively coordinate the supply chain (Wang and Liu, 2019). In addition,
effective management of end-of-pipe pollutants and obtains a good considering the choice between products in the supply chain, studies
enterprise green image. have explored the green technology innovation investment strategies
of manufacturers producing green products and non-green products
2.2. Green technology innovation in supply chain under competitive environments, and found that green products can
not only enable the supply chain to obtain higher revenue, but also
Green technology innovation is supply chain, green supply chain higher greenness (Madani and Rasti-Barzoki, 2017).
and supply chain emission reduction, are important means to realise
enterprises strategy about green development (Xia and Niu, 2021; Yang 2.4. Research gaps and contributions
et al., 2021). Under the requirements of environmental protection poli-
cies, enterprises in the supply chain that have set emission reduction Most studies have concluded that technological innovation can
targets need to increase their revenues through green technological raise production efficiency and quality, and enterprises’ market share,
innovation. Conversely, green technological innovation requires the thereby improving the market competitiveness of enterprises. In ad-
cooperation of enterprises in the supply chain (Ranjan et al., 2023). dition, technological innovation can optimise the allocation of corpo-
On the one hand, enterprises can reduce resource consumption and rate resources and generate spillover effects. Upon examination of the
pollutants emission through green technological innovations to realise existing literature, we identify two major research gaps.

3
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

Table 1
Comparison with related literature.
Papers SC GTI P&P Competition CS Greenness
Xie and Zhu (2021) ✓
Yenipazarli and Vakharia (2017) ✓
Deng et al. (2020) ✓ ✓ ✓
Chenavaz (2012) ✓
Li et al. (2019) ✓ ✓
Ranjan et al. (2023) ✓ ✓
(Li and Gao, 2022) ✓
Kuchesfehani et al. (2023) ✓ ✓
Xu et al. (2017) ✓ ✓
Amin-Naseri and Khojasteh (2015) ✓ ✓
Cai et al. (2023) ✓ ✓
Zhu and He (2017) ✓ ✓ ✓
This paper ✓ ✓ ✓ ✓ ✓ ✓

Note: SC donates supply chain; GTI donates green technology innovation; P&P donates green product innovation and green process innovation;
CS donates cost-sharing.

Firstly, most studies have treated green technology innovation as materials and key components, which are subsequently supplied to
a whole entity when studying the green strategies of supply chain the two manufacturers who are responsible for producing and selling
enterprises. However, it is critical to recognise that different types of the products. The products produced by the two manufacturers do
green technological innovations possess distinct environmental objec- not differ in function and use, but they vary in greenness and price.
tives, contribute uniquely to the sustainable development performance Thus, manufacturers develop green technology innovation programmes
of enterprises, and are associated with varied mechanisms for im- to improve the greenness of their products and gain a competitive
plementing cost-sharing contracts. The differentiation is essential as advantage.
green product innovation and green process innovation pursue differ- Manufacturer 1 mainly implements product innovation strategies
ent environmental goals, which subsequently impacts the scope for to develop green products. Green products requires sustainability at
improvement in financial performance and social responsibility per- the raw material selection stage, so suppliers need to make green
formance of enterprises. Therefore, a nuanced approach is required to product innovation efforts to provide green raw materials. Product
study the behaviour of supply chain entities in the context of varying innovation usually involves core aspects such as product design and
types of green technology innovation. market positioning, which are critical to a manufacturer’s competitive
Secondly, green technological innovation is different from tradi- advantage and brand value. In this case, the manufacturer will usually
tional technological innovation by focusing on environmental protec- bear its own green product R&D costs independently to ensure that
tion and sustainable development instead of economic benefits and it takes the lead in product innovation and maintains a competitive
efficiency improvement. Therefore, it is necessary to distinguish dif- advantage in the marketplace. For the supplier’s R&D costs, the manu-
ferent types of green technology innovation to study the behaviour of facturer will partially bear them as a cooperative incentive to promote
supply chain enterprises. Studies have considered competition among suppliers’ more active participation from the supplier in green prod-
homogeneous manufacturers and retailers, or among different types of uct innovation. Manufacturer 2 mainly implements process innovation
products, but there is a notable gap in considering the competitive strategies focused on resource recycling, consumption reduction and
between types of innovation or the dynamic impact of market com- emission reduction by strengthening supplier management and opti-
petitiveness on firms’ decision-making. Green process innovations tend mising production processes. In this process, the manufacturer and the
to have a cost advantage over green product innovations, which can be supplier work closely together to optimise the production process. The
reflected in a price advantage; conversely, green product innovation innovation costs are jointly funded by both parties, and the proportion
by introducing greenness into products tend to have a competitive dif- of the costs to be borne by each party is negotiated according to the
ferentiation advantage over green process innovations, and the higher specific requirements of the collaborative project.
the market’s green preference the more competitive the differentiation. The spillover effect of green technology innovation can reduce the
In addition, changes in the competition intensity affect the decision incentive of enterprises to innovate. Therefore, the government pro-
making of shared suppliers. vides funds needed by enterprises to operate in a specific way to reduce
Table 1 highlights the contributions of our paper, which are sum- their innovation costs and guide them to increase their innovation
marised to three aspects. First, we categorise green technology inno- efforts. Taking into account the reality of government subsidies for
vation into green product innovation and green process innovation, green factories and green supply chains, we assume that government
and distinguish the operation mode of supply chain based on different subsidies are granted according to the outcome of the innovation.
innovation types. Second, the mechanism of cost-sharing contracts to Specifically, subsidies are given to the manufacturer in the case of
improve supply chain innovation is explored for the cost-sharing con- product innovations and to the entire supply chain in the case of process
tracts applied in different innovation models. Third, we explore the con- innovations.
ditions for the establishment of cost-sharing contracts and the process The supply chain model is shown in Fig. 1 the main notations are
of price adjustment under different market competition intensities. presented in Table 2.
The following are key assumptions.
3. Model
Assumption 1. Assume that the green product innovation effort level
3.1. Assumptions and notations of manufacturer 1 is 𝑓𝑚1 (𝑡), and the cost 𝐶𝑚1 (𝑡) is a convex function of
1 2 (𝑡), and 𝜂
its innovation input, i.e., 𝐶𝑚1 (𝑡) = 𝜂𝑚1 𝑓𝑚1 𝑚1 is the innovation
2
Assume that the supply chain consists of a shared supplier and two cost coefficient of manufacturer 1. A larger 𝜂𝑚1 indicates a higher
manufacturers (𝑖, 𝑖 = 1, 2), and that the core enterprise in the supply innovation cost and a lower innovation efficiency. The government
chain is the manufacturer. The supplier handles the production of raw subsidises product innovations at a rate of 𝜉𝑚1 and 0 ≤ 𝜉𝑚1 ≤ 1, i.e. the

4
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

Fig. 1. Supply chain model.

Table 2
Notations of parameters and variables.
Notations Descriptions
Parameters
1∕𝑣 Efficiency of collaborative green process innovation
𝛾𝑔 Influence coefficient of product greenness on market demand
𝜒 Competition intensity in the market
𝜁 Importance of product greenness in market competition
𝐾𝑚𝑖 Manufacturer 𝑖 maximum parameters for green technology innovation
𝑟𝑚𝑖 Innovation rate of manufacturer 𝑖
𝛿, 𝜎 Natural attenuation rate; noise parameter
𝜉𝑚1 , 𝜉𝑚2 Government subsidies for green technology innovation costs
𝜃1 Proportion of product innovation costs shared by manufacturer 1 for suppliers
𝜃2 Proportion of process innovation costs shared by manufacturer 2
1∕𝜂𝑠1 , 1∕𝜂𝑚1 , 1∕𝜂𝑠2 , 1∕𝜂𝑚2 Green technology innovation efficiency
𝜑𝑠1 , 𝜑𝑚1 , 𝜑𝑠2 , 𝜑𝑚2 Conversion efficiency of green technology innovation into greenness
Variables
𝑊 (𝑡) Wiener process
𝐷𝑚𝑖 (𝑡) Demand function
𝑧𝑚𝑖 (𝑡) The inherent evolutionary law of green technology innovation
𝑔1 (𝑡), 𝑔2 (𝑡) Greenness of products
𝐶𝑚1 (𝑡), 𝐶𝑠1 (𝑡), 𝐶𝑠2 (𝑡) Green technology innovation costs
𝑥𝑚𝑖 , 𝑥𝑠𝑖 Price
Decision variables
𝑓𝑠1 (𝑡), 𝑓𝑚1 (𝑡), 𝑓𝑠2 (𝑡), 𝑓𝑚2 (𝑡) Green process (product) innovation effort level

government covers part of the innovation costs of manufacturer 1. by manufacturer 2 is 𝜃2 and the proportion covered by the supplier is
Supply chain coordination is achieved through a one-way cost-sharing 1 − 𝜃2 . The government subsidises process innovation at a rate of 𝜉𝑚2
contract between suppliers and manufacturer 1, where manufacturer and 0 ≤ 𝜉𝑚2 ≤ 1.
1 shares a certain percentage 𝜃1 of green innovation costs incurred by For simplicity, other costs of daily production are not considered.
suppliers. The supplier’s green product innovation effort level is 𝑓𝑠1 (𝑡)
1 2 (𝑡), where 𝜂 Assumption 2. Product greenness is determined by the technolog-
and the cost is 𝐶𝑠1 (𝑡) = 𝜂𝑠1 𝑓𝑠1 𝑠1 is the supplier’s green
2 ical innovation capability. Innovation is a continuous and dynamic
product innovation cost coefficient.
accumulation process that involves diffusion, and in terms of the inher-
Manufacturer 2 cooperates with suppliers to improve equipment ent dynamics of technological innovation, technological development
and production processes through green process innovation to reduce usually follows some form of S-curve (Frances and William, 2018),
the cost of pollution control and gain a price advantage. The green Therefore Logistic curves are used to describe the process of innovation
process innovation effort levels of supplier and manufacturer 2 are de- accumulation:
noted as 𝑓𝑚2 (𝑡) and 𝑓𝑠2 (𝑡), respectively, and the process innovation cost 𝐾𝑚𝑖
is jointly funded, and their respective cost proportions are negotiated 𝑢𝑚𝑖 (𝑡) = , (1)
1 + 𝑐e−𝑟𝑚𝑖 𝑡
based on the specific requirements of the co-operation project. The and the differential expression is given by:
1
total cost of process innovation is 𝐶𝑠2 (𝑡) = 𝑣(𝜂𝑠2 𝑓𝑠2 2 (𝑡) + 𝜂 𝑓 2 (𝑡)), ( )
𝑚2 𝑚2 𝑢 (𝑡)
2
where 𝜂𝑠2 and 𝜂𝑚2 are the cost coefficients of process innovation of 𝑢̇ 𝑚𝑖 (𝑡) = 𝑟𝑚𝑖 𝑢𝑚𝑖 (𝑡) 1 − 𝑚𝑖 . (2)
𝐾𝑚𝑖
supplier and manufacturer 2 respectively, and 1∕𝑣 is the positive effect
generated by the co-operation, 0 < 𝑣 < 1, and a smaller value of 𝑣 Let 𝑧𝑚𝑖 = 𝑢𝑚𝑖 (𝑡), which represents the internal law followed by
corresponds to higher efficiency in the collaboration between supplier the technological innovation of manufacturer 𝑖, and 𝑖 = 1, 2. 𝐾𝑚𝑖 is
and manufacturer 2. The proportion of process innovation costs covered the maximum value parameter, 𝑟 is the rate parameter and 𝑐 is a

5
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

constant. The influencing factors of innovation can be mainly classi- respectively. Therefore, the supplier’s objective generalised function is
fied into internal and external aspects, the first is the internal factor, given by:
{ 𝑇 [ 𝑃𝑠1 𝑃𝑠2
i.e., its own evolutionary law; the second is the external factor, which
𝐽𝑠 = max E e−𝜌𝑡 𝐷𝑚1 𝑥𝑠1 𝑓 (𝑥𝑠1 )d𝑥𝑠1 + 𝐷𝑚2 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2 −
mainly includes the green technology innovation efforts of suppliers 𝑓𝑠 (𝑡) ∫0 ∫0 ∫0
] }
and manufacturers. External factors operate on the basis of internal
factors. (1 − 𝜃1 )𝐶𝑠1 − (1 − 𝜃2 )(1 − 𝜉𝑚2 )𝐶𝑠2 d𝑡 , (6)
Since manufacturer 1 produces green products by further processing where 𝜌 is the discount factor.
the green raw materials provided by the supplier, the impact of the 𝑃
Manufacturer 1 sells products to obtain revenue 𝐷𝑚1 ∫0 𝑚1 𝑥𝑚1 𝑓 (𝑥𝑚1 )
innovation efforts of both on product greenness is represented by 𝑃
multiplication. Therefore, the greenness 𝑔1 (𝑡) of the product produced d𝑥𝑚1 , pays for ordering from suppliers 𝐷𝑚1 ∫0 𝑠1 𝑥𝑠1 𝑓 (𝑥𝑠1 )d𝑥𝑠1 , and the
by manufacturer 1 through product innovation at time 𝑡 is: green product innovation cost is (1−𝜉𝑚1 )𝐶𝑚1 , and the cost of green prod-
√ uct innovation for the suppliers is 𝜃1 𝐶𝑠1 . Therefore, the manufacturer
d𝑔1 (𝑡) = {𝜑𝑠1 𝑓𝑠1 (𝑡)(1 + 𝜑𝑚1 𝑓𝑚1 (𝑡))𝑧𝑚1 (𝑡) − 𝛿 𝑔1 (𝑡)}d𝑡 + 𝜎 𝑔1 (𝑡)d𝑊 (𝑡), (3) 1’s objective generalised function is given by:
{ 𝑇 [ ( 𝑃𝑚1 𝑃𝑠1 )
where 𝜑𝑠1 and 𝜑𝑚1 are the effect coefficients of transforming green 𝐽𝑚1 = max E e−𝜌𝑡 𝐷𝑚1 𝑥𝑚1 𝑓 (𝑥𝑚1 )d𝑥𝑚1 − 𝑥𝑠1 𝑓 (𝑥𝑠1 )d𝑥𝑠1 −
𝑓𝑚1 ∫0 ∫0 ∫0
product innovation efforts into greenness for supplier and manufacturer ] }
1, respectively, 𝛿 is the natural decay rate, 𝑊 (𝑡) is a Wiener process, (1 − 𝜉𝑚1 )𝐶𝑚1 − 𝜃1 𝐶𝑠1 d𝑡 . (7)
and 𝜎 is a noise parameter.
𝑃
The greenness of the product 𝑔2 (𝑡) achieved through process in- Manufacturer 2 sells products to obtain revenue 𝐷𝑚2 ∫0 𝑚2 𝑥𝑚2 𝑓 (𝑥𝑚2 )
novation is determined by the joint efforts of the supplier and the 𝑃
d𝑥𝑚2 , pays for ordering from suppliers 𝐷𝑚2 ∫0 𝑠2 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2 , and the
manufacturer 2, 𝑔2 (𝑡) is given by: green process innovation cost is 𝜃2 (1 − 𝜉𝑚2 )𝐶𝑚2 . Therefore, the manu-
√ facturer 2’s objective generalised function is given by:
d𝑔2 (𝑡) = [(𝜑𝑠2 𝑓𝑠2 (𝑡) + 𝜑𝑚2 𝑓𝑚2 (𝑡))𝑧𝑚2 (𝑡) − 𝛿 𝑔2 (𝑡)]d𝑡 + 𝜎 𝑔2 (𝑡)d𝑊 (𝑡), (4) { [ ( 𝑃𝑚2 )
𝑇 𝑃𝑠2
𝐽𝑚2 = max E e−𝜌𝑡 𝐷𝑚2 𝑥𝑚2 𝑓 (𝑥𝑚2 )d𝑥𝑚2 − 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2 −
where 𝜑𝑠2 and 𝜑𝑚2 are the effect coefficients of transforming green 𝑓𝑚2 ∫0 ∫0 ∫0
] }
process innovation efforts into greenness for supplier and manufacturer
2, respectively. 𝜃2 (1 − 𝜉𝑚2 )𝐶𝑚2 d𝑡 . (8)

The goal of both supplier and manufacturer is to maximise rev-

Assumption 3. The products produced by two manufacturers compete enue. In process innovation, there is no sequence of decision-making
in the sales market and this competitive relationship is represented by between supplier and manufacturer 2. Unlike process innovation, the
a market demand function. The demand function is based on repre- decision-making process for product innovations between supplier and
sentations from existing study (Chen et al., 2024; Shigui and Yong, manufacturer 1 is sequential, with supplier first makes green product
2023; El and Pasin, 2006; Liu et al., 2023, 2011). Competition can be innovation efforts based on the manufacturer’s proposed product solu-
classified into price competition and non-price competition. We assume tions, and then the manufacturer 1 decides on green product innovation
that non-price competition depends on consumers’ preference for the efforts based on the green raw materials provided by the supplier.
greenness of the product, the market demand function 𝐷𝑚𝑖 (𝑡) for the
product produced by manufacturer 𝑖 is: Proposition 1. The optimal green product innovation effort level of the
𝐷𝑚1 (𝑡) =𝐷01 (𝑥𝑚1 ) + 𝛾𝑔 𝑔1 (𝑡) − 𝜒[𝜁(𝑔2 (𝑡) − 𝑔1 (𝑡)) + (1 − 𝜁)(𝑥𝑚1 − 𝑥𝑚2 )], supplier is:
∗ 𝛷𝜑𝑠1 𝑧𝑚1 𝜂𝑚1 (1 − 𝜉𝑚1 )
𝐷𝑚2 (𝑡) =𝐷02 (𝑥𝑚2 ) + 𝛾𝑔 𝑔2 (𝑡) − 𝜒[𝜁(𝑔1 (𝑡) − 𝑔2 (𝑡)) + (1 − 𝜁)(𝑥𝑚2 − 𝑥𝑚1 )], (5) 𝑓𝑠1 = , (9)
(1 − 𝜃1 )𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 ) − 2𝛷𝛹 (𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2
where 𝐷0𝑖 (𝑥𝑚𝑖 ) is the potential market demand of the product, which
′ (𝑥 ) < 0. 𝛾 > 0 is
where
is related to the price of the product 𝑥𝑚𝑖 , and 𝐷0𝑖 𝑚𝑖 𝑔 ( )
𝑃𝑠1 𝑃 𝑃
the coefficient of influence of product greenness on market demand, 𝛾𝑔 ∫0 𝑥𝑠1 𝑓 (𝑥𝑠1 )d𝑥𝑠1 + 𝜒 𝜁 ∫0 𝑠1 𝑥𝑠1 𝑓 (𝑥𝑠1 )d𝑥𝑠1 − ∫0 𝑠2 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2
i.e., consumers’ green preference. 0 < 𝜒 ≤ 1 is the market’s competitive 𝛷= ,
𝜌+𝛿
intensity, assuming that 0 < 𝜁 < 1 is the proportion of the product’s ( )
𝑃 𝑃
(𝛾𝑔 + 𝜒 𝜁) ∫0 𝑚1 𝑥𝑚1 𝑓 (𝑥𝑚1 )d𝑥𝑚1 − ∫0 𝑠1 𝑥𝑠1 𝑓 (𝑥𝑠1 )d𝑥𝑠1
greenness that occupies the market competition, then 1 − 𝜁 is the 𝛹= .
proportion of the price that occupies the market competition. 𝜌+𝛿

Prices are assumed to be random variables, which will be further The optimal green process innovation effort level of the supplier is:
analysed in Section 5.1. Thus, manufacturer 𝑖’s price 𝑥𝑚𝑖 is a random 𝛩𝜑𝑠2 𝑧𝑚2
∗
variable on interval [0, 𝑃𝑚𝑖 ] with probability density function 𝑓 (𝑥𝑚𝑖 ). 𝑓𝑠2 = , (10)
(1 − 𝜃2 )(1 − 𝜉𝑚2 )𝜂𝑠2 𝑣
The supplier’s supply price to manufacturer 𝑖 is 𝑥𝑠𝑖 , [0, 𝑃𝑠𝑖 ], with a prob-
ability density function of 𝑓 (𝑥𝑠𝑖 ). Product innovations are considered where
( )
revolutionary and process innovations are incremental, so the former 𝑃𝑠2 𝑃 𝑃
𝛾𝑔 ∫0 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2 + 𝜒 𝜁 ∫0 𝑠2 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2 − ∫0 𝑠1 𝑥𝑠1 𝑓 (𝑥𝑠1 )d𝑥𝑠1
tend to require larger cost inputs that need to be compensated for by 𝛩= .
higher prices. Accordingly, it is assumed that 𝑥𝑠1 > 𝑥𝑠2 and 𝑥𝑚1 > 𝑥𝑚2 . 𝜌+𝛿
For brevity, the time unit 𝑡 will be omitted hereafter. The optimal green product innovation effort level of the manufacturer 1
is:
3.2. The model
∗ 𝛷𝛹 𝜑𝑚1 (𝜑𝑠1 𝑧𝑚1 )2
𝑓𝑚1 = . (11)
(1 − 𝜃1 )𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 ) − 2𝛷𝛹 (𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2
𝑃
According to the assumptions, the supplier gains 𝐷𝑚1 ∫0 𝑠1
𝑥𝑠1 𝑓 (𝑥𝑠1 )
𝑃 The optimal green process innovation effort level of the manufacturer 2
d𝑥𝑠1 and 𝐷𝑚2 ∫0 𝑠2 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2 from supplying raw materials to man-
is:
ufacturer 1 and manufacturer 2, respectively, and the costs of green
product innovations and green process innovations incurred by coop- ∗ 𝛤 𝜑𝑚2 𝑧𝑚2
𝑓𝑚2 = , (12)
erating with the manufacturers are (1 − 𝜃1 )𝐶𝑠1 and (1 − 𝜃2 )(1 − 𝜉𝑚2 )𝐶𝑠2 , 𝜃2 (1 − 𝜉𝑚2 )𝜂𝑚2 𝑣

6
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

where Corollary 1. The conditions for the establishment of a cost-sharing contract

( )
𝑃 𝑃
(𝛾𝑔 + 𝜒 𝜁) ∫0 𝑚2 𝑥𝑚2 𝑓 (𝑥𝑚2 )d𝑥𝑚2 − ∫0 𝑠2 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2 for green product innovation in the supply chain are:
𝛤 = . (1)
𝜌+𝛿 √
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2 𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
𝛥𝑝𝑠12 > , 0 < 𝛾𝑔 < , 0 < 𝜒𝑔 < 𝜒𝑔1 ,
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 𝑝𝑠1
Proof. See Appendix A.
where
√
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
4. Analysis −𝛾𝑔 (𝑝𝑠1 + 𝛥𝑝𝑠12 ) + 𝛾𝑔2 (𝑝𝑠1 − 𝛥𝑝𝑠12 )2 + 4𝛥𝑝𝑠12
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1
𝜒𝑔1 = .
2𝛥𝑝𝑠12
4.1. Evolutionary characteristics of product greenness
(2)
To capture the impact of the optimal green innovation efforts on 𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
product greenness, the statistical properties of product greenness are 𝛥𝑝𝑠12 < ,
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1
represented.
and
Proposition 2. The expected value of product greenness is 0 < 𝛾𝑔 < 𝛾𝑔1 , 0 < 𝜒𝑔 < 1,
( )
𝑁 𝑁
E(𝑔 ∗ ) = 𝑖 + e−𝛿 𝑡 𝑔0 − 𝑖 , (13) or
𝛿 𝛿 √
∗ (1 + 𝜑 𝑓 ∗ )𝑧 ∗ ∗ )𝑧 . 𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
where 𝑁1 = 𝜑𝑠1 𝑓𝑠1 𝑚1 𝑚1 𝑚1 and 𝑁2 = (𝜑𝑠2 𝑓𝑠2 + 𝜑𝑚2 𝑓𝑚2 𝑚2 𝛾𝑔1 < 𝛾𝑔 < , 0 < 𝜒𝑔 < 𝜒𝑔1 ,
The variance is 2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 𝑝𝑠1
( )
𝑁 𝑁 2 e−𝛿𝑡 (𝑁𝑖 − 𝛿 𝑔0 )(2𝑁𝑖 + 𝜎 2 ) where
D(𝑔 ∗ ) =e−2𝛿𝑡 𝑔0 − 𝑖 + e𝛿𝑡 𝑖 − √ ( )
𝛿 𝛿 𝛿2 𝜂 𝜂 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
( 2) ) −(𝑝𝑠1 + 𝛥𝑝𝑠12 ) + (𝑝𝑠1 + 𝛥𝑝𝑠12 )2 − 4𝑝𝑠1 𝛥𝑝𝑠12 − 𝑠1 𝑚1
(𝑁𝑖 − 𝛿 𝑔0 )(2𝑁 𝑖 + 𝜎 𝑁 (2𝑁𝑖 + 𝜎 2 ) 𝑁 (2𝑁𝑖 + 𝜎 2 ) 2
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 ) 𝛥𝑝𝑚𝑠1
− e−2𝛿𝑡 𝑔02 + 2
− 𝑖 2
+ 𝑖 . 𝛾𝑔1 = .
𝛿 2𝛿 2𝛿 2 2𝛥𝑝𝑠12
(14)

Proof. See Appendix B. Proof. The cost-sharing contract between manufacturer 1 and supplier
must satisfy certain conditions. Through Eqs. (9) and (11), this con-
Eq. (13) can be rewritten as
dition that enables suppliers and manufacturer 1 to engage in green
𝑁
E(𝑔 ) = (1 − e−𝛿 𝑡 ) 𝑖 + 𝑒−𝛿 𝑡 𝑔0 .
∗
(15) product innovation efforts is:
𝛿
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 (𝛾𝑔 𝑝𝑠1 + 𝜒𝑔 𝛥𝑝𝑠12 )(𝛾𝑔 + 𝜒𝑔 )𝛥𝑝𝑚𝑠1
From Eq. (15), product greenness is increased over time by the level 𝜃1 < 1 − , (17)
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
of innovation effort, which also implies that misbehaviour will generate
more losses. where 𝜃1 ≥ 0, hence:
Eq. (14) can be written as 𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
( ) 𝛥𝑝𝑠12 𝜒𝑔2 + 𝛾𝑔 (𝑝𝑠1 + 𝛥𝑝𝑠12 )𝜒𝑔 + 𝛾𝑔2 𝑝𝑠1 − < 0. (18)
2 − 4e−𝛿 𝑡 2 4𝑔0 e−𝛿 𝑡 𝜎 2 (e−2𝛿 𝑡 + e−𝛿 𝑡 + 1) 2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1
D(𝑔 ∗ ) = 𝑁𝑖 + − 𝑁𝑖
𝛿2 𝛿 2𝛿 2 Let:
𝑔 𝜎 2 (e−2𝛿 𝑡 + e−𝛿 𝑡 ) 𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
+ 0 . (16) 𝑓 (𝜒𝑔 ) = 𝛥𝑝𝑠12 𝜒𝑔2 + 𝛾𝑔 (𝑝𝑠1 + 𝛥𝑝𝑠12 )𝜒𝑔 + 𝛾𝑔2 𝑝𝑠1 − . (19)
𝛿 2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1
8𝑠 𝛿e−𝛿 𝑡 − 𝜎 2 (e−2𝛿 𝑡 + e−𝛿 𝑡 + 1) ln2 Eq. (19) is a quadratic equation about 𝜒𝑔 , and 𝜒𝑔 ∈ (0, 1]. Then, we
In Eq. (16), let 𝑁0 = 0 and 𝑡0 = . (1)
∗ 16e−𝛿 𝑡 − 8 𝛿 can find:
When 𝑡 < 𝑡0 , 𝑁𝑖 and D(𝑔 ) have an inverted U-shaped relationship. If
𝜂 𝜂 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
𝑁𝑖 < 𝑁0 , D(𝑔 ∗ ) increases with 𝑁𝑖 . At the beginning of the innovation 𝛥 = 𝛾𝑔2 (𝑝𝑠1 − 𝛥𝑝𝑠12 )2 + 4𝛥𝑝𝑠12 𝑠1 𝑚1 > 0, (20)
period, enterprise innovation is highly sensitive to the influence of 2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1
external factors and there is a lot of exploration, with the increase in in- so Eq. (19) has two solutions. Since 𝛥𝑝𝑠12 > 0 and 𝛾𝑔 (𝑝𝑠1 + 𝛥𝑝𝑠12 ) >
novation investment, the growth rate of product greenness accelerates, 0, there is one solution less than zero according to the rooting for-
and the level of innovation effort by the enterprise will have a greater mula. Combined with the image characteristics of a quadratic function
impact on it. If 𝑁𝑖 > 𝑁0 , D(𝑔 ∗ ) decreases with 𝑁𝑖 . As the innovation (Fig. 2), the other solution is greater than zero for the Eq. (18) to be
activity proceeds and the innovation effort scales up, the firm gradually realised. That is:
determines the green strategy and the uncertainty decreases. (2) When
𝜒𝑔1
𝑡 > 𝑡0 , 𝑁𝑖 and D(𝑔 ∗ ) have a U-shaped relationship. If 𝑁𝑖 < 𝑁0 , D(𝑔 ∗ ) √ ( )
decreases with 𝑁𝑖 . As time passes and technology matures, enterprises 𝜂 𝜂 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
−𝛾𝑔 (𝑝𝑠1 + 𝛥𝑝𝑠12 ) + 𝛾𝑔2 (𝑝𝑠1 + 𝛥𝑝𝑠12 )2 − 4𝛥𝑝𝑠12 𝑟2𝑔 𝑝𝑠1 − 𝑠1 𝑚1
need only maintain a low level of input effort to achieve a stable output 2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1
= > 0,
of innovation. If 𝑁𝑖 > 𝑁0 , D(𝑔 ∗ ) increases with 𝑁𝑖 . In the later stages of 2𝛥𝑝𝑠12

innovation as the proportion of inputs increases, innovation spillovers (21)

become severe, and the uncertainty of innovation increases. i.e.,
√
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
4.2. Mechanisms of cost-sharing ratios in different innovation models 0 < 𝛾𝑔 < . (22)
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 𝑝𝑠1

(1) Green product innovation Then, we compare the values of 𝜒𝑔1 and 1 to determine the range
𝑃 𝑃
Let 𝑝𝑚1 = ∫0 𝑚1 𝑥𝑚1 𝑓 (𝑥𝑚1 )d𝑥𝑚1 , 𝑝𝑠1 = ∫0 𝑠1 𝑥𝑠1 𝑓 (𝑥𝑠1 )d𝑥𝑠1 , 𝑝𝑠2 = of 𝜒𝑔 . Assuming that 𝜒𝑔1 < 1, and there is:
𝑃𝑠2
∫0 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2 , 𝛥𝑝𝑚𝑠1 = 𝑝𝑚1 − 𝑝𝑠1 , 𝛥𝑝𝑠12 = 𝑝𝑠1 − 𝑝𝑠2 , and 𝜒𝑔 = 𝜒 𝜁, 𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
< (1 + 𝛾𝑔 )(𝛥𝑝𝑠12 + 𝛾𝑔 𝑝𝑠1 ). (23)
which means the market competition intensity in greenness. 2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1

7
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

From Eq. (22), we know that the value of 𝛾𝑔 will not exceed The relationship between 𝜃1 and 𝜒𝑔 when both 𝑓𝑠1 and 𝑓𝑚1 are
√ √
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2 𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2 maximum is:
, so let 𝛾𝑔 = , and we
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 𝑝𝑠1 2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 𝑝𝑠1 ( )−
can get: 2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 (𝛾𝑔 𝑝𝑠1 + 𝜒𝑔 𝛥𝑝𝑠12 )(𝛾𝑔 + 𝜒𝑔 )𝛥𝑝𝑚𝑠1
𝜃1 → 1− . (30)
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2 𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2 𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
< +
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1
√ From Eq. (28)–Eq. (30), it can be found that the effect of 𝜒𝑔 change
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2 on 𝜃1 and 𝑓𝑠1∗ (or 𝑓 ∗ ) is opposite. When 𝜒 increases, manufacturer
𝑔
(𝑝𝑠1 + 𝛥𝑝𝑠12 ) + 𝛥𝑝𝑠12 . (24) 𝑚1
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 𝑝𝑠1 1 will increase its investment in innovation to improve the greenness
√ of its products and gain market share. As a result, manufacturer 1 will
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2 decrease 𝜃1 and the incentives to suppliers shift from cost reduction
Therefore, when 𝛾𝑔 = , 𝜒𝑔1 < 1. But when
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 𝑝𝑠1 to increased demand. When 𝜒𝑔 decreases, increasing 𝜃1 or government
𝛾𝑔 = 0, we are not sure if 𝜒𝑔1 is greater than 1. If 𝜒𝑔1 < 1 even when subsidies can be achieved by the optimal decision, which not only helps
𝜂 𝜂 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2 to incentivise the green product innovation efforts of suppliers and
𝛾𝑔 = 0, there is 𝛥𝑝𝑠12 > 𝑠1 𝑚1 , and then 0 < 𝜒𝑔 ≤ 𝜒𝑔1 .
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 manufacturers, but also helps to maintain the equilibrium in a competi-
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2 tive environment. When 𝜒𝑔 increases, it relies more on the spontaneous
But if not, there is 𝛥𝑝𝑠12 < , let 𝜒𝑔1 > 1, we can
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 regulation of the market to maximise the decision variable, i.e. prices.
get:
Although government subsidies or 𝜃1 can be reduced to maintain the
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2 contract, the resulting impact is not stable, which may reduce the green
𝑝𝑠1 𝛾𝑔2 + (𝑝𝑠1 + 𝛥𝑝𝑠12 )𝛾𝑔 + 𝛥𝑝𝑠12 − < 0, (25)
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 product innovation efforts of suppliers and manufacturers 1.
let
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2 (2) Green process innovation
𝑓 (𝛾𝑔 ) = 𝑝𝑠1 𝛾𝑔2 + (𝑝𝑠1 + 𝛥𝑝𝑠12 )𝛾𝑔 + 𝛥𝑝𝑠12 − , (26)
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 𝑃
Let 𝑝𝑚2 = ∫0 𝑚2 𝑥𝑚2 𝑓 (𝑥𝑚2 )d𝑥𝑚2 , 𝛥𝑝𝑚𝑠2 = 𝑝𝑚2 − 𝑝𝑠2 , 𝛥𝑝𝑠21 = 𝑝𝑠2 − 𝑝𝑠1 ,
according to the root formula, 𝛥 = (𝑝𝑠1 − 𝛥𝑝𝑠12 )2 + 4𝑝𝑠1 and then, 𝛥𝑝𝑠21 < 0.
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
> 0, so Eq. (26) has two solutions and one of
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 Corollary 3. (1) 𝑔2 increased with 𝜃2 when the following conditions are
them is less than zero and the other is:
satisfied:
√ ( )
𝜂 𝜂 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2 1 𝛾𝑔 (𝑝𝑠2 𝜂𝑚2 𝜑2𝑠2 − 𝛥𝑝𝑚𝑠2 𝜑2𝑚2 𝜂𝑠2 ) 𝑝𝑚2 𝜑2 𝜂𝑚2
−(𝑝𝑠1 + 𝛥𝑝𝑠12 ) + (𝑝𝑠1 + 𝛥𝑝𝑠12 )2 − 4𝑝𝑠1 𝛥𝑝𝑠12 − 𝑠1 𝑚1 < 𝜃2 < 1, 𝜒𝑔 < , < 1 + 𝑠2 .
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝛥𝑝𝑚𝑠1 2 2 2
𝛥𝑝𝑚𝑠2 𝜑𝑠2 𝜂𝑚2 − 𝛥𝑝𝑠21 𝜂𝑠2 𝜑𝑚2 𝑝𝑠2 𝜑2𝑚2 𝜂𝑠2
𝛾𝑔1 = ,
2𝛥𝑝𝑠12
(27) (2) 𝑔2 decreased with 𝜃2 when the following conditions are satisfied:
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2 1 𝛾𝑔 (𝑝𝑠2 𝜂𝑚2 𝜑2𝑠2 − 𝛥𝑝𝑚𝑠2 𝜑2𝑚2 𝜂𝑠2 )
and 𝛾𝑔1 > 0 because of 𝛥𝑝𝑠12 < . Therefore, when 0 < 𝜃2 < ,𝜒 > ,
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 𝑚𝑠1 )2 𝛥𝑝 2 𝑔 𝛥𝑝𝑚𝑠2 𝜑2𝑠2 𝜂𝑚2 − 𝛥𝑝𝑠21 𝜂𝑠2 𝜑2𝑚2
0 < 𝛾𝑔 < 𝛾𝑔1 , there is 0 < 𝜒𝑔 < 1, which means that 𝜒𝑔 has no
effect on whether or not a cost-sharing contract is established. In this and
scenario, consumers demonstrate low green preferences, indicating that 𝜑2 𝜂𝑚2
𝑝𝑚2
there is a lack of market incentives for green product innovation in > 1 + 𝑠2 ,
the supply chain, and consequently results in low overall efficiency of
𝑝𝑠2 𝜑2𝑚2 𝜂𝑠2
green product innovation in the supply chain. As a result, the forma- or
tion of cost-sharing contracts is primarily influenced by price factors
𝑝𝑚2 𝜑2 𝜂𝑚2 𝛥𝑝𝑚𝑠2 𝜑2𝑠2 𝜂𝑚2 − 𝛥𝑝𝑠21 𝜂𝑠2 𝜑2𝑚2
rather than the intensity of green market competition. In this case, the < 1 + 𝑠2 , 𝛾𝑔 < .
𝑝𝑠2 2
𝜑𝑚2 𝜂𝑠2 𝑝𝑠2 𝜂𝑚2 𝜑2𝑠2 − 𝛥𝑝𝑚𝑠2 𝜑2𝑚2 𝜂𝑠2
government ought to increase subsidies to enhance the efficiency of
innovation and steer the market towards playing a leading role in the
decisions of innovation agents.
Proof. From Eqs. (10) and (12), the effect of 𝜃2 (or 𝜒𝑔 ) on 𝑓𝑚2 and
𝑓𝑠2 is opposite, so increasing 𝑓𝑠2 by increasing 𝜃2 (or 𝜒𝑔 ) will lead to
Corollary 2. The relationship between 𝜃1 and 𝜒𝑔 when both 𝑓𝑠1 and 𝑓𝑚1 the decrease of 𝑓𝑚2 at the same time. Green product innovations are
are maximum is: priced higher, so as competition for greenness in the market increases,
( )−
2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 (𝛾𝑔 𝑝𝑠1 + 𝜒𝑔 𝛥𝑝𝑠12 )(𝛾𝑔 + 𝜒𝑔 )𝛥𝑝𝑚𝑠1 rational suppliers will tend to invest in green product innovations and
𝜃1 → 1 − . less in process innovations. Manufacturer 2 will increase the level of
𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )(𝜌 + 𝛿)2
process innovation effort in order to capture market share and enhance
co-operation with suppliers. Increasing the level of green process in-
𝜕𝑓 𝜕𝑓 novation efforts is to maximise the greenness of the product, i.e. to
Proof. From Eqs. (9) and (11), we can get 𝜕 𝜃𝑠1 > 0 and 𝜕 𝜃𝑚1 > 0. The
1 1 maximise:
conditions of maximum 𝑓𝑠1 and maximum 𝑓𝑚1 are not contradictory,
𝜑𝑠2 𝑓𝑠2 + 𝜑𝑚2 𝑓𝑚2 =
so the greenness of the product is the highest when both of them are
maximum. [(𝛾𝑔 𝑝𝑠2 + 𝜒𝑔 𝛥𝑝𝑠21 )𝜑2𝑠2 𝜂𝑚2 + 𝜂𝑠2 (𝛾𝑔 + 𝜒𝑔 )𝛥𝑝𝑚𝑠2 𝜑2𝑚2 ]𝑧𝑚2 𝜃2 + 𝜂𝑠2 (𝛾𝑔 + 𝜒𝑔 )𝛥𝑝𝑚𝑠2 𝜑2𝑚2 𝑧𝑚2
.
Eqs. (9) and (11) can be rewritten: 𝜃2 (1 − 𝜃2 )(1 − 𝜉𝑚2 )𝜂𝑠2 𝜂𝑚2 𝑣(𝜌 + 𝛿)

∗ 𝜑𝑠1 𝑧𝑚1 𝜂𝑚1 (1 − 𝜉𝑚1 ) (31)

𝑓𝑠1 = , (28)
(1 − 𝜃1 )𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )
− 2𝛹 (𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 (1) From Eq. (31), when (𝛾𝑔 𝑝𝑠2 +𝜒𝑔 𝛥𝑝𝑠21 )𝜑2𝑠2 𝜂𝑚2 +𝜂𝑠2 (𝛾𝑔 +𝜒𝑔 )𝛥𝑝𝑚𝑠2 𝜑2𝑚2
𝛷
1
and > 0 and < 𝜃2 < 1, 𝑔2 increased with 𝜃2 . 𝛥𝑝𝑚𝑠2 𝜑2𝑠2 𝜂𝑚2 −𝛥𝑝𝑠21 𝜂𝑠2 𝜑2𝑚2 > 0,
2
𝜑𝑚1 (𝜑𝑠1 𝑧𝑚1 )2 so we can get:
∗
𝑓𝑚1 = . (29)
(1 − 𝜃1 )𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 ) 𝑟𝑔 (𝑝𝑠2 𝜂𝑚2 𝜑2𝑠2 − 𝛥𝑝𝑚𝑠2 𝜑2𝑚2 𝜂𝑠2 )
− 2(𝜑𝑠1 𝜑𝑚1 𝑧𝑚1 )2 𝜒𝑔 < , (32)
𝛷𝛹 𝛥𝑝𝑚𝑠2 𝜑2𝑠2 𝜂𝑚2 − 𝛥𝑝𝑠21 𝜂𝑠2 𝜑2𝑚2
8
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

Fig. 2. The image features of Eq. (19).

where 𝜒𝑔 > 0, so 𝑝𝑠2 𝜂𝑚2 𝜑2𝑠2 − 𝛥𝑝𝑚𝑠2 𝜑2𝑚2 𝜂𝑠2 > 0, that is: (i) If 𝑝𝑠2 𝜂𝑚2 𝜑2𝑠2 − 𝛥𝑝𝑚𝑠2 𝜑2𝑚2 𝜂𝑠2 > 0, then:

𝛥𝑝𝑚𝑠2 𝜑2𝑠2 𝜂𝑚2 − 𝛥𝑝𝑠21 𝜂𝑠2 𝜑2𝑚2

𝑝𝑚2 𝜑2 𝜂𝑚2 𝛾𝑔 < , (36)
< 1 + 𝑠2 , (33) 𝑝𝑠2 𝜂𝑚2 𝜑2𝑠2 − 𝛥𝑝𝑚𝑠2 𝜑2𝑚2 𝜂𝑠2
𝑝𝑠2 𝜑2𝑚2 𝜂𝑠2
and
in this scenario, the intensity of competition in the market for
greenness is lower, manufacturer 2 has a price advantage, and the 𝑝𝑚2 𝜑2 𝜂𝑚2
combined efficiency of supplier process innovation is higher. On the < 1 + 𝑠2 , (37)
𝑝𝑠2 𝜑2𝑚2 𝜂𝑠2
one hand, manufacturer 2 can more easily maintain market share and
profitability, and will therefore be more inclined to bear more costs
that is, market green preferences are low, competitive intensity is
to innovate, in order to attract consumers and build brand loyalty. On
focused more on price, manufacturer 2 has less profitability, market
the other hand, in order to incentivise suppliers to better engage in
demand gives incentive signals to suppliers, and suppliers sharing more
innovation and increase their inputs, manufacturer 2 will also choose
of the cost of innovation can enable manufacturers to maintain their
to bear more of the costs of process innovation.
price advantage.
(2) From Eq. (31), when (𝛾𝑔 𝑝𝑠2 +𝜒𝑔 𝛥𝑝𝑠21 )𝜑2𝑠2 𝜂𝑚2 +𝜂𝑠2 (𝛾𝑔 +𝜒𝑔 )𝛥𝑝𝑚𝑠2 𝜑2𝑚2
(ii) If 𝑝𝑠2 𝜂𝑚2 𝜑2𝑠2 − 𝛥𝑝𝑚𝑠2 𝜑2𝑚2 𝜂𝑠2 < 0, then there must be an Eq. (35),
1
< 0 and 0 < 𝜃2 < , 𝑔2 decreased with 𝜃2 . 𝛥𝑝𝑚𝑠2 𝜑2𝑠2 𝜂𝑚2 −𝛥𝑝𝑠21 𝜂𝑠2 𝜑2𝑚2 > 0, and there is:
2
so we can get:
𝑝𝑚2 𝜑2 𝜂𝑚2
> 1 + 𝑠2 , (38)
𝛾𝑔 (𝑝𝑠2 𝜂𝑚2 𝜑2𝑠2 − 𝛥𝑝𝑚𝑠2 𝜑2𝑚2 𝜂𝑠2 ) 𝑝𝑠2 𝜑2𝑚2 𝜂𝑠2
𝜒𝑔 > , (34)
𝛥𝑝𝑚𝑠2 𝜑2𝑠2 𝜂𝑚2 − 𝛥𝑝𝑠21 𝜂𝑠2 𝜑2𝑚2
that is, greenness is more competitive and suppliers are less efficient
where 𝜒𝑔 < 1, so there is: at innovation and therefore deserve a greater share of the costs of
process innovation.
𝛾𝑔 (𝑝𝑠2 𝜂𝑚2 𝜑2𝑠2 − 𝛥𝑝𝑚𝑠2 𝜑2𝑚2 𝜂𝑠2 ) < 𝛥𝑝𝑚𝑠2 𝜑2𝑠2 𝜂𝑚2 − 𝛥𝑝𝑠21 𝜂𝑠2 𝜑2𝑚2 . (35)

9
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

5. Extension in decrease of market share due to higher price. Therefore, before

such a surge, the enterprise should use green products to bring brand
premium to obtain higher pricing space, which will not only allow
5.1. Market competition and product prices
enterprises to better carry out marketing campaigns and enhance the
product’s market awareness, but will also provide a buffer for price
Price as a random variable, mean and variance are two important increases in case of a sudden increase in costs, so that the company
numerical characteristics, but greenness and market competition will can achieve stable price growth in a competitive market. Similarly, in
affect the price in different ways. Therefore, the prices of manufacturer 𝜆𝑔 𝛥𝑝1
1 will be analysed below as an example in terms of influencing changes order to increase the mean of price, there is ≥ , which implies
𝜆𝑝 𝛥𝑔1
in the mean and variance of prices. that consumers prefer to buy cost-effective products. Therefore, the
The stochastic process for price 𝑝(𝑡) is given by: 𝜆𝑔 𝛥𝑝1
= is a relatively stable business strategy for enterprises.
√ 𝜆𝑝 𝛥𝑔1
d𝑝(𝑡) = [𝑝(𝑡) − 𝑔(𝑝(𝑡) − 𝜙)]d𝑡 + 𝜎 𝑝(𝑡)d𝑊 (𝑡), (39)
Therefore, the incentive effect of product greenness and market
where 𝜙 is the market price acceptable to consumers and 𝑔 is the competition on product prices is represented by increasing the mean 𝑜
government’s adjustment factor with 𝑔 ∈ (0, 1). Rewriting Eq. (39): of the price random variable, and its contribution to price stabilisation
√ is represented by decreasing the variance 𝜏 2 . Further, denote the price
d𝑝(𝑡) = (𝐵 − 𝐴𝑝(𝑡))d𝑡 + 𝜎 𝑝(𝑡)d𝑊 (𝑡), (40)
as:
where 𝐴 = 1 − 𝑔 and 𝐵 = 𝑔 𝜙. For brevity, the time unit 𝑡 is similarly 𝑥0 = 𝑚𝑥 + (𝑛 − 𝑚)𝑜, (45)
omitted below.
Similar to the process of proving Proposition 2, we can get the where 0 ≤ 𝑚 ≤ 1, 𝑛 ≥ 1. Define 𝑛 − 1 and 1 − 𝑚 as the incen-
expected value of product prices is: tive coefficients for product marketisation and stabilisation of market
( )
E(𝑝) =
𝐵
+ e−𝐴𝑡 𝑝0 −
𝐵
, (41) prices, respectively, so the mean of 𝑥0 is 𝑛𝑜, the variance is 𝑚2 𝜏 2 , and
𝐴 𝐴 0 ≤ 𝑥0 ≤ 𝑃0 .
and the variance is:
( ) −𝐴𝑡 2
𝐵 𝐵 2 e (𝐵 − 𝐴𝑝0 )(2𝐵 + 𝜎 )
D(𝑝) =e−2𝐴𝑡 𝑝0 − + e𝐴𝑡 −
𝐴 𝐴 𝐴2
( ) 5.2. Stability analysis
(𝐵 − 𝐴𝑝0 )(2𝐵 + 𝜎 2 ) 𝐵(2𝐵 + 𝜎 2 ) 𝐵(2𝐵 + 𝜎 2 )
− e−2𝐴𝑡 𝑝20 + − + .
𝐴 2 2𝐴 2 2𝐴2
(42) When the stability of the system is altered, the system may bifur-
cate or transition directly into a chaotic state. In the actual market,
Eq. (41) can be rewritten:
enterprises under information asymmetry are all bounded rational, and
𝑔𝜙
E(𝑝) = (1 − e−(1−𝑔)𝑡 ) + 𝑝0 e−(1−𝑔)𝑡 , (43) they will adjust the level of effort in the next cycle according to the
1−𝑔
size of their marginal target in the previous cycle. Hence, the key to
where e−(1−𝑔)𝑡 < 1, so E(𝑝) increased with 𝜙. From Eq. (42), lim𝑡→∞ D(𝑝) system disorder is the effect of the different behavioural efforts of the
𝑔 𝜙(2𝑔 𝜙 + 𝜎 2 ) participants on the product attributes, which is an adjustable aspect
= and 𝜙 > 0, which indicates that D(𝑝) decreased with 𝜙
2(1 − 𝑔)2 compared to the stochastic factor. Next, focusing on the decisions and
after a long period of evolution. 𝜙 is given by:
games in technological innovation, the innovation is assumed to be a
𝜙 = 𝜙0 + 𝜆𝑔 𝑔1 − 𝜆𝑝 (𝑝1 − 𝑝2 ), (44) certain quantity disregarding its own laws. Based on the above analysis,
the model is further simplified and the revenue functions of the supplier
where 𝜙0 is the consumer’s initial willingness to pay, and 𝜆𝑔 and 𝜆𝑝 are and manufacturer are, respectively:
the coefficients of the product greenness and the competitor’s price on
the consumer’s willingness to pay, respectively. The price consumers 𝜋𝑠 =𝑝𝑠1 (𝐷01 + 𝛾𝑔 (1 + 𝜑𝑠1 𝑓𝑠1 (1 + 𝜑𝑚1 𝑓𝑚1 ))𝑧𝑚1
are willing to pay for product 1 depends on the uniqueness of the − 𝜒(𝜁((+𝜑𝑠2 𝑓𝑠2 + 𝜑𝑚2 𝑓𝑚2 )𝑧𝑚2 − (1 + 𝜑𝑠1 𝑓𝑠1 (1 + 𝜑𝑚1 𝑓𝑚1 ))𝑧𝑚1 )
product and the competitor’s price, and the product greenness is used − (1 − 𝜁)(𝑝1 − 𝑝2 )))
to represent the uniqueness of the product.
+ 𝑝𝑠2 [𝐷02 + 𝛾𝑔 (𝜑𝑠2 𝑓𝑠2 + 𝜑𝑚2 𝑓𝑚2 )𝑧𝑚2
𝑝1 increased with 𝑔1 , defined as 𝛥𝑔1 and 𝛥𝑝1 , respectively, and
the value of the change in 𝜙 caused by their changes is 𝛥𝜙, in order − 𝜒(𝜁((1 + 𝜑𝑠1 𝑓𝑠1 (1 + 𝜑𝑚1 𝑓𝑚1 ))𝑧𝑚1 − (𝜑𝑠2 𝑓𝑠2 + 𝜑𝑚2 𝑓𝑚2 )𝑧𝑚2 )
𝜆𝑔
to achieve a stable increase in price, there is 𝛥𝜙 < 0, i.e., ≤ − (1 − 𝜁)(𝑝2 − 𝑝1 ))]
𝜆𝑝 1 1
2 2 2
𝛥𝑝1 𝛥𝑝1 − (1 − 𝜃1 ) 𝜂𝑠1 𝑓𝑠1 − (1 − 𝜃2 )(1 − 𝜉𝑚2 ) 𝑣(𝜂𝑚2 𝑓𝑚2 + 𝜂𝑠2 𝑓𝑠2 ) (46)
, where denotes the rate of change of the price after the 2 2
𝛥𝑔1 𝛥𝑔1
𝜆𝑔
increase of greenness by one unit, indicates the sensitivity degree 𝜋𝑚1 =(𝑝𝑚1 − 𝑝𝑠1 )[𝐷01 + 𝛾𝑔 (1 + 𝜑𝑠1 𝑓𝑠1 (1 + 𝜑𝑚1 𝑓𝑚1 ))𝑧𝑚1
𝜆𝑝
𝜆𝑔 − 𝜒(𝜁((𝜑𝑠2 𝑓𝑠2 + 𝜑𝑚2 𝑓𝑚2 )𝑧𝑚2 − (1 + 𝜑𝑠1 𝑓𝑠1 (1 + 𝜑𝑚1 𝑓𝑚1 ))𝑧𝑚1 )
of consumer preference, the larger is, the more consumers tend − (1 − 𝜁)(𝑝1 − 𝑝2 ))]
𝜆𝑝
𝜆𝑔 1 2 1 2
to choose green products, and the smaller is, the more consumers − (1 − 𝜉𝑚1 ) 𝜂𝑚1 𝑓𝑚1 − 𝜃1 𝜂𝑠1 𝑓𝑠1 (47)
𝜆𝑝 2 2
tend to be influenced by price. As consumer sensitivity to greenness
increases, the rate of change in price increases. In other words, the 𝜋𝑚2 =(𝑝𝑚2 − 𝑝𝑠2 )[𝐷02 + 𝛾𝑔 (𝜑𝑠2 𝑓𝑠2 + 𝜑𝑚2 𝑓𝑚2 )𝑧𝑚2
marginal increment in price is greater than the marginal increment in − 𝜒(𝜁((1 + 𝜑𝑠1 𝑓𝑠1 (1 + 𝜑𝑚1 𝑓𝑚1 ))𝑧𝑚1 − (𝜑𝑠2 𝑓𝑠2 + 𝜑𝑚2 𝑓𝑚2 )𝑧𝑚2 )
greenness when considering consumer preferences. On the one hand, − (1 − 𝜁)(𝑝2 − 𝑝1 ))]
the marginal cost of innovation will increase at a later stage after
1 2 2
decreasing in the earlier stage, so that a higher marginal incremental − 𝜃2 (1 − 𝜉𝑚2 ) 𝑣(𝜂𝑚2 𝑓𝑚2 + 𝜂𝑠2 𝑓𝑠2 ) (48)
2
price is needed to compensate for the increase in cost in order to ensure
profitability and maintain price stability. On the other hand, new Taking the first-order partial derivatives of Eq. (46)–Eq. (48), the
technological innovations lead to a surge in costs probably resulting marginal revenue function of the game subject with their innovation

10
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

effort are:
⎧ 𝜕 𝜋𝑠 = 𝜑 𝑧 (1 + 𝜑 𝑓 )[𝑝 (𝛾 + 𝜒 𝜁) − 𝑝 𝜒 𝜁] − (1 − 𝜃 )𝜂 𝑓
𝑠1 𝑚1 𝑚1 𝑚1 𝑠1 𝑔 𝑠2 1 𝑠1 𝑠1
⎪ 𝜕 𝑓𝑠1
⎪ 𝜕𝜋
⎪ 𝑠 = 𝜑𝑠2 𝑧𝑚2 [𝑝𝑠2 (𝛾𝑔 + 𝜒 𝜁) − 𝜒 𝜁 𝑝𝑠1 ] − (1 − 𝜃2 )(1 − 𝜉𝑚2 )𝑣𝜂𝑠2 𝑓𝑠2
⎪ 𝜕 𝑓𝑠2
⎨ 𝜕𝜋 (49)
⎪ 𝑚1 = (𝑝𝑚1 − 𝑝𝑠1 )(𝛾𝑔 + 𝜒 𝜁)𝜑𝑠1 𝑓𝑠1 𝜑𝑚1 𝑧𝑚1 − (1 − 𝜉𝑚1 )𝜂𝑚1 𝑓𝑚1
⎪ 𝜕 𝑓𝑚1
⎪ 𝜕𝜋
⎪ 𝑚2 = (𝑝𝑚2 − 𝑝𝑠2 )(𝛾𝑔 + 𝜁 𝜒)𝜑𝑚2 𝑧𝑚2 − 𝜃2 (1 − 𝜉𝑚2 )𝑣𝜂𝑚2 𝑓𝑚2
⎩ 𝜕 𝑓𝑚2

A gradient adjustment mechanism is introduced, i.e., the subject’s

innovation effort level at moment 𝑡 + 1 is determined by the marginal
𝜕 𝜋𝑖
returns at moment 𝑡. When > 0, the subject raises its innovation
𝜕 𝑓𝑖 Fig. 3. Trajectory of 𝑔1 .
𝜕 𝜋𝑖
effort level at the moment 𝑡 + 1; when = 0, the subject keeps its
𝜕 𝑓𝑖
innovation effort level stable and unchanged at the moment 𝑡 + 1; and
𝜕 𝜋𝑖 so the conditions for this point to be a stable equilibrium solution of
when < 0, the subject reduces its innovation effort level at the
𝜕 𝑓𝑖 the game system should be analysed. The stability of the equilibrium
moment 𝑡 + 1. Under this learning rule, the adjustment strategies of the
point can be determined by the relation between the trace 𝑇 𝑟(𝐽 ) and
game players are:
the determinant 𝐷𝑒𝑡(𝐽 ) of the matrix 𝐽 . According to the Jury stability
⎧𝑓 (𝑡 + 1) = 𝑓 (𝑡) + 𝜇 𝑓 (𝑡) 𝜕 𝜋𝑠 criterion, when satisfied:
1 𝑠1
⎪ 𝑠1 𝑠1
𝜕 𝑓𝑠1
⎪ 𝜕𝜋
⎪𝑓𝑠2 (𝑡 + 1) = 𝑓𝑠2 (𝑡) + 𝜇2 𝑓𝑠2 (𝑡) 𝑠 ⎧1 + 𝑇 𝑟(𝐽 ) + 𝐷𝑒𝑡(𝐽 ) > 0
⎪ 𝜕 𝑓𝑠2 ⎪
(50) ⎨1 − 𝑇 𝑟(𝐽 ) + 𝐷𝑒𝑡(𝐽 ) > 0 (57)
⎨ 𝜕𝜋
⎪𝑓𝑚1 (𝑡 + 1) = 𝑓𝑚1 (𝑡) + 𝜇3 𝑓𝑚1 (𝑡) 𝑚1 ⎪
⎩1 − 𝐷𝑒𝑡(𝐽 ) > 0
⎪ 𝜕 𝑓𝑚1
⎪ 𝐸 is the stable equilibrium solution of the system.
⎪𝑓𝑚2 (𝑡 + 1) = 𝑓𝑚2 (𝑡) + 𝜇4 𝑓𝑚2 (𝑡) 𝜕 𝜋𝑚2
⎩ 𝜕 𝑓𝑚2
6. Numerical results
The game subjects adjust their strategies at a certain rate 𝜇𝑖 . There-
fore, the strategy adjustment equation of the game parties are: We denote the number of green technology innovations by the
number of green patents granted (in millions) from 2000 to 2021, and
⎧𝑓 (𝑡 + 1) = 𝑓 (𝑡) + 𝜇 𝑓 (𝑡){𝜑 𝑧 (1 + 𝜑 𝑓 )[𝑝 (𝛾 + 𝜒 𝜁 ) − 𝑝 𝜒 𝜁 ] − (1 − 𝜃 )𝜂 𝑓 }
⎪ 𝑠1 𝑠1 1 𝑠1 𝑠1 𝑚1 𝑚1 𝑚1 𝑠1 𝑔 𝑠2 1 𝑠1 𝑠1 the data are obtained from the China Research Data Service Platform
⎪
⎪𝑓𝑠2 (𝑡 + 1) = 𝑓𝑠2 (𝑡) + 𝜇2 𝑓𝑠2 (𝑡){𝜑𝑠2 𝑧𝑚2 [𝑝𝑠2 (𝛾𝑔 + 𝜒 𝜁 ) − 𝜒 𝜁 𝑝𝑠1 ] − (1 − 𝜃2 )(1 − 𝜉𝑚2 )𝑣𝜂𝑠2 𝑓𝑠2 } (CNRDS). Logistic regression results yield: 𝐾𝑚𝑖 = 0.58, 𝑟𝑚𝑖 = 0.269,
⎨ 𝑐𝑚𝑖 = 237.46. 𝑅2 = 0.973 and 𝐩 < 0.01, indicating a good fit and
⎪𝑓𝑚1 (𝑡 + 1) = 𝑓𝑚1 (𝑡) + 𝜇3 𝑓𝑚1 (𝑡){(𝑝𝑚1 − 𝑝𝑠1 )(𝛾𝑔 + 𝜒 𝜁 )𝜑𝑠1 𝑓𝑠1 𝜑𝑚1 𝑧𝑚1 − (1 − 𝜉𝑚1 )𝜂𝑚1 𝑓𝑚1 }
⎪ significant results. Herfindahl–Hirschman Index (HHI) can be used to
⎪𝑓𝑚2 (𝑡 + 1) = 𝑓𝑚2 (𝑡) + 𝜇4 𝑓𝑚2 (𝑡){(𝑝𝑚2 − 𝑝𝑠2 )(𝛾𝑔 + 𝜁 𝜒)𝜑𝑚2 𝑧𝑚2 − 𝜃2 (1 − 𝜉𝑚2 )𝑣𝜂𝑚2 𝑓𝑚2 }
⎩ judge the degree of competition, when HHI< 0.01, the market is highly
(51) competitive and market share is dispersed; when 0.01<HHI< 0.15, the
market is highly competitive and market share distribution is relatively
The stability of the equilibrium point of the model and the possible balanced; when 0.15 < HHI < 0.25, the market is generally competitive
chaotic states are analysed and the Jacobian matrix of Eq. (51) is: and market share distribution is relatively concentrated; when HHI
> 0.25, the market is highly concentrated and the distribution of market
⎡𝐽11 𝐽12 𝐽13 𝐽14 ⎤
⎢ ⎥ share is very concentrated. Therefore, let 𝜒 = 0.1. Let 𝑥𝑚1 ∈ [0, 200],
𝐽 𝐽22 𝐽23 𝐽24 ⎥
𝐽 = ⎢ 21 (52) 𝑥𝑚2 ∈ [0, 100], 𝑥𝑠1 ∈ [0, 60], and 𝑥𝑠2 ∈ [0, 40], all of which obey a
⎢𝐽31 𝐽32 𝐽33 𝐽34 ⎥
⎢𝐽 ⎥ uniform distribution.
⎣ 41 𝐽42 𝐽43 𝐽44 ⎦ (1) Trajectory of product greenness
Referring to the relevant studies (Chen et al., 2024; Liu et al.,
Let 𝑓𝑠1 (𝑡 + 1) = 𝑓𝑠1 (𝑡), 𝑓𝑠2 (𝑡 + 1) = 𝑓𝑠2 (𝑡), 𝑓𝑚1 (𝑡 + 1) = 𝑓𝑚1 (𝑡) and 𝑓𝑚2 (𝑡 +
2023; Zhu and He, 2017; Wang et al., 2022), in addition to the above
1) = 𝑓𝑚2 (𝑡), the equilibrium point of the system can be obtained. It can
realistic data, the remaining parameters are relatively flexible, so they
be found that besides the boundary equilibrium point, the system has
∗∗ , 𝑓 ∗∗ , 𝑓 ∗∗ , 𝑓 ∗∗ ), where
are modelled by assignments. Let 𝛿 = 0.0003, 𝜎 = 0.9, 𝜌 = 0.1, 𝛾𝑔 = 0.2,
a unique internal Nash equilibrium point 𝐸(𝑓𝑠1 𝑠2 𝑚1 𝑚2 𝜁 = 0.2, 𝜑𝑠1 = 0.5, 𝜑𝑠2 = 0.6, 𝜑𝑚1 = 0.5, 𝜑𝑚2 = 0.8, 𝜂𝑠1 = 2.5, 𝜂𝑠2 = 2,
∗∗ )[𝑝 (𝛾 + 𝜒 𝜁) − 𝑝 𝜒 𝜁]
𝜑𝑠1 𝑧𝑚1 (1 + 𝜑𝑚1 𝑓𝑚1 𝜂𝑚1 = 2.5, 𝜂𝑚2 = 2, 𝜉𝑚1 = 0.1, 𝜉𝑚2 = 0.05, 𝜃1 = 0.2, 𝜃2 = 0.4, and 𝑣 = 0.8.
∗∗ 𝑠1 𝑔 𝑠2
𝑓𝑠1 = (53) The trajectories of the optimal greenness of the products produced by
(1 − 𝜃1 )𝜂𝑠1
manufacturer 1 and manufacturer 2 are shown as 𝑔10 in Fig. 3 and 𝑔20
𝜑𝑠2 𝑧𝑚2 [𝑝𝑠2 (𝛾𝑔 + 𝜒 𝜁) − 𝜒 𝜁 𝑝𝑠1 ]
∗∗
𝑓𝑠2 = (54) in Fig. 4, respectively.
(1 − 𝜃2 )(1 − 𝜉𝑚2 )𝑣𝜂𝑠2 As can be seen from Figs. 3 and 4, when 𝜒 = 0.2, the intensity
∗∗ of competition increases, leading to a rise in consumer demand for
𝑓𝑚1
and attention to green products, and manufacturers will increase the
(𝑝𝑚1 − 𝑝𝑠1 )(𝛾𝑔 + 𝜒 𝜁 )𝜑𝑚1 (𝜑𝑠1 𝑧𝑚1 )2 [𝑝𝑠1 (𝛾𝑔 + 𝜒 𝜁 ) − 𝑝𝑠2 𝜒 𝜁 ]
= research and development of green technology in order to satisfy the
(1 − 𝜃1 )𝜂𝑠1 (1 − 𝜉𝑚1 )𝜂𝑚1 − (𝑝𝑚1 − 𝑝𝑠1 )(𝛾𝑔 + 𝜒 𝜁 )(𝜑𝑚1 𝜑𝑠1 𝑧𝑚1 )2 [𝑝𝑠1 (𝛾𝑔 + 𝜒 𝜁 ) − 𝑝𝑠2 𝜒 𝜁 ] market demand, so as to gain a competitive advantage and sustainable
(55) development. From Fig. 3, when 𝜃1 increases, the proportion of costs
∗∗
(𝑝𝑚2 − 𝑝𝑠2 )(𝛾𝑔 + 𝜁 𝜒)𝜑𝑚2 𝑧𝑚2 shared by manufacturer 1 for suppliers increases, which encourages
𝑓𝑚2 = (56) suppliers to invest more money and resources in improving the pro-
𝜃2 (1 − 𝜉𝑚2 )𝑣𝜂𝑚2
duction process, reducing environmental pollution and resource waste,
𝐸 indicates that the game subjects have all made innovation inputs, and manufacturers who use green raw materials provided by suppliers
which obviously represents a more significant economic implication, as the basis for innovation will develop products with a higher degree

11
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

Fig. 4. Trajectory of 𝑔2 .

Fig. 7. Initial bifurcation diagram.

Fig. 5. Sensitivity analysis of 𝑓𝑚1 .

Fig. 8. Bifurcation diagram(𝜒 = 0.2).

selling prices incentivise the manufacturers’ innovation effort level.

However, these prices are unstable, making it difficult to maintain
this high level in the long run. The green product innovation effort
level and green technology innovation effort level plotted according to
the initial conditions are 𝑓𝑚10 and 𝑓𝑚20 , respectively. Fig. 5 indicates
that an increase in the market competition intensity or the cost share
of manufacturer 1 will increase the overall green product innovation
effort level. Fig. 6 shows that an increase in the market competition
intensity or a decrease in the cost shared will increase the green process
Fig. 6. Sensitivity analysis of 𝑓𝑚2 . innovation effort level.
(3) Strategy adjustment and model stability
The stability of the system is analysed numerically to visually
of greenness, significantly enhancing the overall greenness of their explore the path of the system to chaos, and the bifurcation dia-
products. In addition, the manufacturer can enhance its reputation and grams of the system and different variables as they vary are shown in
attract more environmentally conscious consumers to purchase their Fig. 7–Fig. 10.
products, which will form a further incentive. From the analysis in
Fig. 7 shows that the system exhibits the typical phenomenon of
Section 4.2, 𝑔2 decreased with 𝜃2 , so let 𝜃2 = 0.3, Fig. 4 shows that with
multiplicative period bifurcation (flip bifurcation), and it eventually
this adjustment, the product greenness increases. This is because the
enters the chaotic state. (1) If the intensity of market competition
supplier’s innovation efficiency, though lower than that of the manufac-
increases, let 𝜒 = 0.2, it can be seen through Fig. 8 that as the
turer, allows them to better improve the overall innovation efficiency
intensity of market competition increases, it shortens the speed of
of the supply chain, and the supplier can increase the greenness of the
strategy adjustment for supplier product innovation, manufacturer 1
product by bearing more costs.
and manufacturer 2, and increases their level of innovation effort, but
(2) Product prices and innovation efforts
delays the speed of strategy adjustment for supplier process innovation.
Considering the relationship between the price of the product and (2) If the share of green innovation costs to suppliers is increased, let
the level of innovative effort of the game subjects, the level of innova- 𝜃1 = 0.6, the bifurcation diagram is shown in Fig. 9. The proportion of
tive effort of the manufacturer is simulated. The results are shown in sharing does not affect the speed of strategy adjustment, but increases
Figs. 5 and 6. the level of supplier effort. (3) If the proportion of costs covered by
As can be seen in Figs. 5 and 6, during the later stages of innovation manufacturer 2 is increased so that 𝜃2 = 0.7, the bifurcation diagram
evolution, 𝑓 increased with 𝑛 − 1 for a given 𝑚; 𝑓 decreased with 1 − 𝑚 is shown in Fig. 10. The change in the proportion of cost sharing does
for a given 𝑛. When both 𝑚 and 𝑛 are large, 𝑓𝑚𝑖 is largest. The high not affect the speed of strategy adjustment, but increases the level of

12
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

is more influenced by price factors than by the competitive intensity

of greenness. The government should increase subsidies in this case,
and guide the market to take a leading role in the decision-making
of innovation subjects by reducing the cost of product innovation
and improving innovation efficiency. When the intensity of market
competition increases, manufacturers need to increase their inputs to
improve the greenness of their products and to gain market share,
so they will minimise their spending in other areas, the proportion
of green product innovation costs shared by suppliers decreases, and
there is a shift in the incentives for suppliers from cost reduction
to increased demand. Subsidies can be appropriately reduced in this
scenario, with the market playing a leading role. (2) When market
competition is low, increasing government subsidies can activate the
market by incentivising the green innovation effort level of suppliers,
which in turn increases market demand and the innovation effort level
Fig. 9. Bifurcation diagram(𝜃1 = 0.6). of manufacturers. It can also help to increase the success rate of cost-
sharing contracts between manufacturers and suppliers. (3) When the
market is more competitive, it relies more on spontaneous market
regulation to maintain the equilibrium that maximises the optimal
decision, and although government subsidies can be reduced or the
proportion of costs that manufacturers share with suppliers can be
lowered, the impacts are unstable, which may reduce the level of green
innovation efforts by both suppliers and manufacturers. This is due to
the fact that there are many enterprises at the start-up stage entering
the market, and the reduction of government subsidies may lead to the
breakage of the capital chain and aggravate the strategic barriers to
entry. At this time, government subsidies should be adjusted differently
according to different types of enterprises, reducing subsidies for more
mature enterprises and maintaining the original subsidy proportion for
start-ups, so as to rationalise the resource allocation.
(2) For green process innovation, there are many studies that base
their research on the same cost-sharing approach as green product
Fig. 10. Bifurcation diagram(𝜃2 = 0.7).
innovation (Mohammaditabar et al., 2016; Chen et al., 2024). However,
we distinguish between the two. Since suppliers and manufacturers
calculate and share the costs of the innovation together, so it is impos-
process innovation effort of the supplier and decreases the level of sible to increase the greenness of the product through the independent
process innovation effort of manufacturer 2. optimum because how to increase the greenness of the product depends
on the relationship between the overall efficiency of the innovation, the
7. Conclusions degree of competition in the marketplace, and the cost-sharing propor-
tion. The cost-sharing proportion affects greenness in different ways in
Green technology innovation plays a positive role in realising the different scenarios: (1) manufacturers bear more of the costs of process
established emission reduction targets of enterprises, and is regarded innovation and their share is positively proportional to greenness. In
as an important strategy for enterprises to gain sustainable competitive this scenario, the competitive intensity of greenness is lower, the man-
advantages in a brand-new arena. Since the positive externality of ufacturer has a price advantage, and the supplier’s overall efficiency in
technological innovation creates an imbalance between inputs and process innovation is higher. On the one hand, manufacturers are more
benefits, integrated green technological innovation based on supply likely to maintain market share and profitability, and therefore will
chain has become a hot research issue. Focusing on the core issue be more inclined to bear more innovation costs to attract consumers
of collaborative green technology innovation in supply chains (cost- and build brand loyalty. On the other hand, in order to incentivise
sharing model), we categorise green technology innovation into green suppliers to better participate in innovation and increase their inputs,
product innovation and green process innovation, and study the mech- manufacturers will also choose to bear more process innovation costs.
anism of cost-sharing model in different scenarios, as well as consumer (2) Suppliers bear more of the process innovation costs, and their share
sensitivity and product premium. is positively related to greenness. When the market green preference is
low, the manufacturer’s innovation efficiency is high and its profit mar-
7.1. Findings and implications gin is low, the intensity of competition is more reflected in the price,
the market demand sends an incentive signal to the supplier, and the
supplier’s sharing of more innovation costs can enable the manufacturer
(1) In green product innovation, the conditions for the formation
to maintain its price advantage. When greenness competition is high
of cost-sharing contracts are closely related to market competition,
and suppliers are less efficient in innovation, suppliers sharing more
consumer green preferences and supply chain innovation efficiency.
of the cost of process innovation can help manufacturers gain more
Studies have explored the mechanism of cost-sharing contracts in green
market share.
technology innovation (Liu et al., 2020), and we further explore the
conditions required for the formation of cost-sharing contracts and (3) In the study related to differential games, due to the complexity
their variations under different market scenarios. (1) when consumers’ of calculation, many studies treat price as a constant parameter and
green preference is low, the lack of market incentives leads to low do not analyse it (Chen et al., 2024; Shigui and Yong, 2023; El and
overall efficiency of green product innovation in the supply chain. Pasin, 2006). We explore the dynamic relationship between price and
Consequently, in this scenario, the formation of a cost-sharing contract market, considering that price is affected by many factors such as

13
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

market environment. In addition, in the actual market, under the infor- Appendix A
mation asymmetry in the actual market, the subjects are all in limited
rationality, and the participating subjects will adjust the next effort
level according to the fulfilment of their goals in the previous cycle, Proof of Proposition 1
so we study them based on this characteristic. As time evolves, taking Construct the HJB equation:
into account consumer preferences, the marginal increment in price ( ) 𝑃𝑠1
𝜌𝑉𝑠 = 𝐷01 + 𝛾𝑔 𝑔1 − 𝜒(𝜁 (𝑔2 − 𝑔1 ) − (1 − 𝜁)(𝑥𝑚2 − 𝑥𝑚1 )) 𝑥𝑠1 𝑓 (𝑥𝑠1 )d𝑥𝑠1
over the marginal increment in greenness helps stabilise the variance ∫0
of prices. On the one hand, the marginal cost of the innovation, after ( ) 𝑃𝑠2
+ 𝐷02 + 𝛾𝑔 𝑔2 − 𝜒(𝜁(𝑔1 − 𝑔2 ) − (1 − 𝜁 )(𝑥𝑚1 − 𝑥𝑚2 )) 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2
decreasing in the first period, will increase in the later period, so ∫0
that a higher price is needed to compensate for the increase in cost, 1 2 1 2 2
− (1 − 𝜃1 ) 𝜂𝑠1 𝑓𝑠1 − (1 − 𝜃2 )(1 − 𝜉𝑚2 ) 𝑣(𝜂𝑚2 𝑓𝑚2 + 𝜂𝑠2 𝑓𝑠2 )
2 2
to ensure profitability and to maintain price stability. On the other
′
(( ) )
hand, if new technological innovations lead to a surge in costs, this + 𝑉𝑠1 𝜑𝑠1 𝑓𝑠1 (1 + 𝜑𝑚1 𝑓𝑚1 ) 𝑧𝑚1 − 𝛿 𝑔1
can reduce market share if prices also surge. Therefore, enterprises 𝜎2 𝜎2
+ 𝑉𝑠2′ ((𝜑𝑠2 𝑓𝑠2 + 𝜑𝑚2 𝑓𝑚2 )𝑧𝑚2 − 𝛿 𝑔2 ) + 𝑔 𝑉 ′′ + 𝑔 𝑉 ′′ , (A.1)
should make use of the brand premium brought by green products to 2 1 𝑠1 2 2 𝑠2
obtain higher pricing space, which not only allows enterprises to carry ( ) 𝑃𝑚1
𝜌𝑉𝑚1 = 𝐷01 + 𝛾𝑔 𝑔1 − 𝜒(𝜁(𝑔2 − 𝑔1 ) − (1 − 𝜁)(𝑥𝑚2 − 𝑥𝑚1 )) 𝑥𝑚1 𝑓 (𝑥𝑚1 )d𝑥𝑚1
out better marketing activities and enhance the market recognition of ∫0
𝑃𝑠1
the product, but also provides a buffer for price increases in case of a 1 2 1 2
− 𝑥𝑠1 𝑓 (𝑥𝑠1 )d𝑥𝑠1 ) − (1 − 𝜉𝑚1 ) 𝜂𝑚1 𝑓𝑚1 − 𝜃1 𝜂𝑠1 𝑓𝑠1
sudden increase in costs, so that enterprises can achieve a stable price ∫0 2 2
(( ) )
increase in a highly competitive market. On the contrary, considering ′
+ 𝑉𝑚11 𝜑𝑠1 𝑓𝑠1 (1 + 𝜑𝑚1 𝑓𝑚1 ) 𝑧𝑚1 − 𝛿 𝑔1
consumer preference, the marginal increment of price is smaller than 𝜎2 𝜎2
′
the marginal increment of greenness which helps to increase the mean + 𝑉𝑚12 ((𝜑𝑠2 𝑓𝑠2 + 𝜑𝑚2 𝑓𝑚2 )𝑧𝑚2 − 𝛿 𝑔2 ) + 𝑔 𝑉 ′′ + 𝑔 𝑉 ′′ , (A.2)
2 1 𝑚11 2 2 𝑚12
value of price, i.e., consumers prefer to buy cost-effective products. ( ) 𝑃𝑚2
In addition, intensified market competition accelerates manufacturers’ 𝜌𝑉𝑚2 = 𝐷02 + 𝛾𝑔 𝑔2 − 𝜒(𝜁(𝑔1 − 𝑔2 ) − (1 − 𝜁)(𝑥𝑚1 − 𝑥𝑚2 )) 𝑥𝑚2 𝑓 (𝑥𝑚2 )d𝑥𝑚2
∫0
and suppliers’ strategy adjustment for green product innovation, but 𝑃𝑠2
1 2 2
decelerates suppliers’ strategy adjustment for green process innovation. − 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2 ) − 𝜃2 (1 − 𝜉𝑚2 ) 𝑣(𝜂𝑚2 𝑓𝑚2 + 𝜂𝑠2 𝑓𝑠2 )
∫0 2
The cost-sharing proportion does not affect the speed of strategic ′
(( ) )
adjustment of the game subjects, but affects their level of innovation + 𝑉𝑚21 𝜑𝑠1 𝑓𝑠1 (1 + 𝜑𝑚1 𝑓𝑚1 ) 𝑧𝑚1 − 𝛿 𝑔1
effort. ′ 𝜎2 𝜎2
+ 𝑉𝑚22 ((𝜑𝑠2 𝑓𝑠2 + 𝜑𝑚2 𝑓𝑚2 )𝑧𝑚2 − 𝛿 𝑔2 ) + 𝑔 𝑉 ′′ + 𝑔 𝑉 ′′ , (A.3)
2 1 𝑚21 2 2 𝑚22
7.2. Limitations and possible directions ′
𝐝𝑉𝑚𝑖1 ′′ =
𝐝2 𝑉𝑚𝑖1 ′
𝐝𝑉𝑚𝑖2 ′′ =
𝐝2 𝑉𝑚𝑖2
where 𝑉𝑚𝑖1 = , 𝑉𝑚𝑖1 , 𝑉𝑚𝑖2 = , 𝑉𝑚𝑖2 ,
𝐝𝑔1 𝐝𝑔1 2 𝐝𝑔2 𝐝𝑔22
Our paper considers the mechanisms by which different types of 𝐝𝑉𝑠1 𝐝2 𝑉𝑠1 𝐝𝑉𝑠2 𝐝2 𝑉𝑠2
𝑉𝑠1′ = , 𝑉𝑠1′′ = , 𝑉𝑠2′ = , 𝑉𝑠2′′ = .
green technology innovations increase the greenness of supply chains. 𝐝𝑔1 𝐝𝑔1 2 𝐝𝑔 2 𝐝𝑔22
However, other aspects need to be considered in the future. First, we From Eq. (A.2), the optimal decision variables is:
only consider cost-sharing contracts in supply chains, and future re- 𝑉 ′ 𝜑𝑠 𝑓𝑠1 𝜑𝑚1 𝑧𝑚1
search could include benefit-sharing mechanisms in the same research
∗
𝑓𝑚1 = 𝑚1 . (A.4)
(1 − 𝜉𝑚1 )𝜂𝑚1
system. Second, we analysed price as a stochastic variable, focusing
on the growth rate of pricing premium with increasing greenness, and Substituting Eq. (A.4) into Eq. (A.1), and then, the optimal decision
future research could systematically analyse the factors affecting their variables is:
decision-making by using price as a decision variable. Third, we con- 𝑉𝑠1′ 𝑧𝑚1 𝜑𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 )
∗
𝑓𝑠1 = ′ 𝑉 ′ (𝑧 𝜑 𝜑 )2
. (A.5)
sidered the competitive advantage between green process innovation (1 − 𝜃1 )𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 ) − 2𝑉𝑚11 𝑠1 𝑚1 𝑠1 𝑚1
and green product innovation, and more competitive scenarios can be
Substituting Eq. (A.5) into Eq. (A.4), we can get:
compared in the future. Last, digital technologies can help enterprises
and governments make good use of resources, so we will consider 𝑉𝑠1′ 𝑉𝑚11
′ 𝜑 (𝜑 𝑧 )2
𝑚1 𝑠1 𝑚1
∗
𝑓𝑚1 = ′ 𝑉 ′ (𝑧 𝜑 𝜑 )2
. (A.6)
constructing models to study the working mechanisms of enterprises (1 − 𝜃1 )𝜂𝑠1 𝜂𝑚1 (1 − 𝜉𝑚1 ) − 2𝑉𝑚11 𝑠1 𝑚1 𝑠1 𝑚1
to promote green transformation through digital means, as well as the
From Eq. (A.1), the optimal decision variables is:
scenarios of introducing digital technologies.
𝑉𝑠2′ 𝜑𝑠2 𝑧𝑚2
∗
𝑓𝑠2 = . (A.7)
(1 − 𝜃2 )(1 − 𝜉𝑚2 )𝜂𝑠2 𝑣
CRediT authorship contribution statement
From Eq. (A.3), the optimal decision variables is:
′ 𝜑 𝑧
𝑉𝑚22
Rui Sun: Writing – original draft, Visualization, Software, Re- ∗ 𝑚2 𝑚2
𝑓𝑚2 = . (A.8)
sources, Methodology, Formal analysis, Data curation, Conceptualiza- 𝜃2 (1 − 𝜉𝑚2 )𝜂𝑚2 𝑣
tion. Da-yi He: Writing – review & editing, Supervision, Conceptual-
Substituting Eq. (A.5)–Eq. (A.8) into Eq. (A.1)–Eq. (A.3). According
ization. Jing-jing Yan: Supervision.
to the structure of equations, there are:

Declaration of competing interest 𝑉𝑠 (𝑔1 , 𝑔2 ) = 𝑥1 𝑔1 + 𝑥2 𝑔2 + 𝑥3 ,

𝑉𝑚1 (𝑔1 , 𝑔2 ) = 𝑦1 𝑔1 + 𝑦2 𝑔2 + 𝑦3 , (A.9)
The authors declare that they have no known competing finan-
𝑉𝑚2 (𝑔1 , 𝑔2 ) = 𝑧1 𝑔1 + 𝑧2 𝑔2 + 𝑧3 , (A.10)
cial interests or personal relationships that could have appeared to
influence the work reported in this paper. where 𝑥1 , 𝑥2 , 𝑥3 , 𝑦1 , 𝑦2 , 𝑦3 , 𝑧1 , 𝑧2 , 𝑧3 are constants to be solved, and
then 𝑉𝑢′ , 𝑉𝑢′′ , 𝑉𝑓′ , 𝑉𝑓′′ can be solved by the method of coefficients to
Acknowledgments be determined, and substituting into Eq. (A.1)–Eq. (A.3), we can get
Eq. (A.11)–Eq. (A.14):
This research was supported by the National Natural Science Foun- 𝑃𝑠1 𝑃𝑠2
(𝛾𝑔 + 𝜒 𝜁) ∫0 𝑥𝑠1 𝑓 (𝑥𝑠1 )d𝑥𝑠1 − 𝜒 𝜁 ∫0 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2
dation of China (NO. 42277484). 𝑉𝑠1′ = 𝑥1 = . (A.11)
𝜌+𝛿

14
R. Sun et al. Journal of Environmental Management 373 (2025) 123545
𝑃 𝑃 ( )
(𝛾𝑔 + 𝜒 𝜁) ∫0 𝑠2 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2 − 𝜒 𝜁 ∫0 𝑠1 𝑥𝑠1 𝑓 (𝑥𝑠1 )d𝑥𝑠1 𝑁𝑖 (2𝑁𝑖 + 𝜎 2 ) + 2(2𝑁𝑖 + 𝜎 2 )(𝛿 𝑠0 − 𝑁𝑖 )
𝑉𝑠2′ = 𝑥2 = . (A.12) e−2𝛿 𝑡 𝑠20 − . (B.13)
( 𝜌+𝛿 ) 2𝛿 2
𝑃 𝑃
(𝛾𝑔 + 𝜒 𝜁) ∫0 𝑚1 𝑥𝑚1 𝑓 (𝑥𝑚1 )d𝑥𝑚1 − ∫0 𝑠1 𝑥𝑠1 𝑓 (𝑥𝑠1 )d𝑥𝑠1
′ since D(𝑠(𝑡)) = E(𝑠2 (𝑡)) − E2 (𝑠(𝑡)), the variance is Eq. (14).
𝑉𝑚11 = 𝑦1 = .
𝜌+𝛿
(A.13) Data availability
( )
𝑃 𝑃
(𝛾𝑜 + 𝜒 𝜁) ∫0 𝑚2 𝑥𝑚2 𝑓 (𝑥𝑚2 )d𝑥𝑚2 − ∫0 𝑠2 𝑥𝑠2 𝑓 (𝑥𝑠2 )d𝑥𝑠2
′ Data will be made available on request.
𝑉𝑚22 = 𝑧2 = .
𝜌+𝛿
(A.14)
References
Therefore, we can get Eq. (9)–Eq. (12)
Amin-Naseri, M.R., Khojasteh, M.A., 2015. Price competition between two leader–
follower supply chains with risk-averse retailers under demand uncertainty. Int.
Appendix B J. Adv. Manuf. Technol. 79 (1–4), 377–393.
Arfi, W.B., Hikkerova, L., Sahut, J.-M., 2018. External knowledge sources, green
innovation and performance. Technol. Forecast. Soc. Change 129 (1), 210–220.
Ayman, W.A.-K., 2022. Big data analytics capabilities and green supply chain per-
Proof of Proposition 2
formance: investigating the moderated mediation model for green innovation and
The evolution of product greenness is: technological intensity. Bus. Process Manage. J. 28 (5–6), 1446–1471.
√ Bensoussan, A., Chen, S., Chutani, A., Sethi, S.P., Siu, C.C., Yam, S.C.P., 2019. Feedback
d𝑠(𝑡) = (𝑁𝑖 − 𝛿 𝑠(𝑡))d𝑡 + 𝜎 𝑠(𝑡)d𝑊 (𝑡). (B.1)
Stackelberg–Nash equilibria in mixed leadership games with an application to
cooperative advertising. SIAM J. Control Optim. 57 (5), 3413–3444.
Rewrite Eq. (B.1) in the form of a stochastic integral equation:
𝑡 𝑡 √
Cai, J., Sun, H., Shang, J., Hegde, G.G., 2023. Information structure selection in a green
supply chain: Impacts of wholesale price and greenness level. European J. Oper.
𝑠(𝑡) = 𝑠0 + (𝑁𝑖 − 𝛿 𝑠(𝑢))d𝑢 + 𝜎 𝑠(𝑢)d𝑊 (𝑢), (B.2)
∫0 ∫0 Res. 306 (1), 34–46.
Chan, H.K., Yee, R.W., Dai, J., Lim, M.K., 2016. The moderating effect of environmental
the expected value of 𝑠(𝑡) is: dynamism on green product innovation and performance. Int. J. Prod. Econ. 330
𝑡 (Part B), 384–391.
E(𝑠(𝑡)) = 𝑠0 + (𝑁𝑖 − 𝛿E(𝑠(𝑢)))d𝑢. (B.3) Chen, C., 2001. Design for the environment: A quality-based model for green product
∫0
development. Manage. Sci. 47 (2), 250–263.
Derive the Eq. (B.3): Chen, Y.-S., Lai, S.-B., Wen, C.-T., 2006. The influence of green innovation performance
on corporate advantage in Taiwan. J. Bus. Ethics 67 (4), 331–339.
E(𝑠′ (𝑡)) = 𝑁𝑖 − 𝛿E(𝑠(𝑡)). (B.4) Chen, Z., Xuan, X., Mark, G., 2024. Differential game analysis of joint emission
reduction decisions under mixed carbon policies and CEA. J. Environ. Manag. 358,
Transforming Eq. (B.4) into an ordinary differential equation: 120913.
( )
𝑁𝑖 𝛿𝑡 Chenavaz, R., 2012. Dynamic pricing, product and process innovation. European J.
𝑦 = e−𝛿 𝑡 e +𝑐 , (B.5) Oper. Res. 222 (3), 553–557.
𝛿
Chiou, T.-Y., Chan, H.K., Lettice, F., Chung, S.H., 2011. The influence of greening
there is E(𝑠(0)) = 𝑠0 , so: the suppliers and green innovation on environmental performance and competitive
𝑁 advantage in Taiwan. Transp. Res. E 47 (6), 822–836.
𝑐 = 𝑠0 − 𝑖 . (B.6) Deng, W., Feng, L., Zhao, X., Lou, Y., 2020. Effects of supply chain competition on
𝛿 firms’ product sustainability strategy. J. Clean. Prod. 275, 124061.
Therefore, Driessen, P.H., Hillebrand, B., Kok, R.A.W., Verhallen, T.M.M., 2013. Green new
( ) product development: the pivotal role of product greenness. IEEE Trans. Eng.
𝑁 𝑁
E(𝑠(𝑡)) = 𝑖 + e−𝛿 𝑡 𝑠0 − 𝑖 . (B.7) Manage. 60 (2), 315–326.
𝛿 𝛿 Du, S., Tang, W., Zhao, J., Nie, T., 2018. Sell to whom? Firm’s green production in
To solve for variance, define a stochastic process 𝑍(𝑡) = 𝑓 (𝑡, 𝑠(𝑡)) = competition facing market segmentation. Ann. Oper. Res. 270 (1–2), 125–154.
El, F., Pasin, F., 2006. Quality improvement and goodwill accumulation in a dynamic
𝑠2 (𝑡), which, by Ito’s Lemma, there is:
[ ] duopoly. IEEE Trans. Eng. Manage. 175 (2), 1021–1032.
𝜕 𝑓 (𝑡, 𝑠(𝑡)) 𝜕 𝑓 (𝑡, 𝑠(𝑡)) 1 ( √ )2 𝜕 2 𝑓 (𝑡, 𝑠(𝑡)) Fernando, Y., Jabbour, C.J.C., Wah, W.-X., 2019. Pursuing green growth in technology
d𝑓 (𝑡, 𝑠(𝑡)) = + (𝑁𝑖 − 𝛿 𝑠(𝑡)) + 𝜎 𝑠(𝑡) d𝑡+
𝜕𝑡 𝜕 𝑠(𝑡) 2 𝜕 𝑠2 (𝑡) firms through the connections between environmental innovation and sustainable
√ 𝜕 𝑓 (𝑡, 𝑠(𝑡)) business performance: Does service capability matter? Resour. Conserv. Recycl. 141,
𝜎 𝑠(𝑡) d𝑊 (𝑡). (B.8)
𝜕 𝑠(𝑡) 8–20.
Frances, B., William, B., 2018. Innovation and diffusion models in policy research.
Thus, 𝑠2 (𝑡) varies is: Theor. Policy Process 173, 253–297.
√ Ghosh, D., Shah, J., 2012. A comparative analysis of greening policies across supply
d𝑠2 (𝑡) = [−2𝛿 𝑠2 (𝑢) + (2𝑁𝑖 + 𝜎 2 )𝑠(𝑢)]d𝑡 + 2𝜎 𝑠(𝑡) 𝑠(𝑡)d𝑊 (𝑡). (B.9) chain structures. Int. J. Prod. Econ. 135 (2), 568–583.
Gopalakrishnan, S., 1999. A reexamination of product and process innovations using a
Rewrite Eq. (B.9) in the form of a stochastic integral equation: knowledge-based view. J. High Technol. Manag. Res. 10 (1), 147–166.
𝑡 𝑡 √ Jeong, D., Lee, J.D., 2024. Where and how does a product evolve? Product innovation
𝑠2 (𝑡) = 𝑠20 + [−2𝛿 𝑠2 (𝑢) + (2𝑁𝑖 + 𝜎 2 )𝑠(𝑢)]d𝑢 + 2 𝜎 𝑠(𝑢) 𝑠(𝑢)d𝑊 (𝑢), pattern in product lineage. Technovation 131, 102958.
∫0 ∫0
Kennedy, A.P., Sethi, S.P., Siu, C.C., Yam, S.C.P., 2021. Cooperative advertising in a
(B.10) dynamic three-echelon supply chain. Prod. Oper. Manage. 30 (11), 3881–3905.
Kuchesfehani, E.K., Parilina, E.M., Zaccour, G., 2023. Revenue and cost sharing contract
there isE(𝑊 (𝑡)) = 0, so: in a dynamic closed-loop supply chain with uncertain parameters. Ann. Oper. Res.
𝑡{ 322 (2), 877.
}
E(𝑠2 (𝑡)) = 𝑠20 + −2𝛿E(𝑠2 (𝑢)) + (2𝑁𝑖 + 𝜎 2 )E(𝑠(𝑢)) d𝑢. (B.11) Lambertini, L., Mantovani, A., 2008. Process and product innovation by a multiproduct
∫0 monopolist: A dynamic approach. Int. J. Ind. Organ. 27 (4), 508–518.
Lambertini, L., Mantovani, A., 2010. Process and product innovation: A differential
Substituting Eq. (B.7) into Eq. (B.11):
game approach to product life cycle. Int. J. Econ. Theory 6 (2), 227–252.
𝑡 { [ ( )]}
𝑁𝑖 𝑁 Laroche, M., Bergeron, J., Guido, B.-F., 2001. Targeting consumers who are willing to
E(𝑠2 (𝑡)) = 𝑠20 + −2𝛿E(𝑠2 (𝑢)) + (2𝑁𝑖 + 𝜎 2 ) + e−𝛿𝑡 𝑠0 − 𝑖 d𝑢, pay more for environmentally friendly products. J. Consum. Mark. 18 (6), 503–520.
∫0 𝛿 𝛿
Li, M., Gao, X., 2022. Implementation of enterprises’ green technology innovation
(B.12) under market-based environmental regulation: An evolutionary game approach. J.
Environ. Manag. 308 (C), 114570.
there is E(𝑠2 (0)) 𝑠20
= and derivation of Eq. (B.12), so: Li, G., Li, L., Choi, T.M., Sethi, S.P., 2020. Green supply chain management in Chinese
𝑁𝑖 (2𝑁𝑖 + 𝜎 2 ) e−𝛿𝑡 (2𝑁𝑖 + 𝜎 2 )(𝑁𝑖 − 𝛿 𝑠0 ) firms: Innovative measures and the moderating role of quick response technology.
2
E(𝑠 (𝑡)) = − + J. Oper. Manage. 66 (7–8), 958.
2𝛿 2 𝛿2

15
R. Sun et al. Journal of Environmental Management 373 (2025) 123545

Li, L., Li, Y., He, Y., Chen, Y., 2023. Hedge fund ownership and the innovation of Xie, J., Abbass, K., Li, D., 2024. Advancing eco-excellence: Integrating stakehold-
high-tech firms: Evidence from the science and technology innovation board in ers’ pressures, environmental awareness, and ethics for green innovation and
China. Econom. Lett. 233, 111420. performance. J. Environ. Manag. 352, 120027.
Li, G., Wang, X., Su, S., Su, Y., 2019. How green technological innovation ability Xie, X., Huo, J., Qi, G., Zhu, K.X., 2016. Green process innovation and financial
influences enterprise competitiveness. Technol. Soc. 59, 101136. performance in emerging economies: Moderating effects of absorptive capacity and
Liu, Z., Anderson, T.D., Cruz, J.M., 2012. Consumer environmental awareness and green subsidies. IEEE Trans. Eng. Manage. 63 (1), 101–112.
competition in two-stage supply chains. European J. Oper. Res. 218 (3), 602–613. Xie, X., Huo, J., Zou, H., 2019. Green process innovation, green product innovation,
Liu, Z., Anderson, D., Cruz, M., 2011. Consumer environmental awareness and and corporate financial performance: A content analysis method. J. Bus. Res. 191
competition in two-stage supply chains. European J. Oper. Res. 218 (3), 602–613. (C), 697–706.
Liu, B., Giovanni, P.D., 2019. Green process innovation through Industry 4.0 Xie, X., Zhu, Q., 2021. How can green innovation solve the dilemmas of ‘‘harmonious
technologies and supply chain coordination. Ann. Oper. Res. 1–36. coexistence’’. Manage. World 37 (1), 128–149+9.
Liu, L., Han, T., Jin, H., 2023. Differential game study of green supply chain based on Xu, X., He, P., Xu, H., Zhang, Q., 2017. Supply chain coordination with green
green technology innovation and manufacturer competition. Chin. J. manage. 20 technology under cap-and-trade regulation. Int. J. Prod. Econ. 183 (Part B),
(1), 116–126. 433–442.
Liu, G., Yang, H., Dai, R., 2020. Which contract is more effective in improving product Xu, B., Lin, B., 2024. Green finance, green technology innovation, and wind power
greenness under different power structures: Revenue sharing or cost sharing? Int. development in China: Evidence from spatial quantile model. Energy Econ. 132,
J. Prod. Econ. 148, 106701. 107463.
Madani, S., Rasti-Barzoki, M., 2017. Sustainable supply chain management with Yang, J., Long, R., Chen, H., Sun, Q., 2021. A comparative analysis of express packaging
pricing, greening and governmental tariffs determining strategies: A game-theoretic waste recycling models based on the differential game theory. Resour. Conserv.
approach. Comput. Ind. Eng. 105 (C), 287–298. Recycl. 168, 105449.
Mohammaditabar, D., Ghodsypour, S.H., Hafezalkotob, A., 2016. A game theoretic Yang, D., Xiao, T., 2017. Pricing and green level decisions of a green supply chain
analysis in capacity-constrained supplier-selection and cooperation by considering with governmental interventions under fuzzy uncertainties. J. Clean. Prod. 149,
the total supply chain inventory costs. Int. J. Prod. Econ. 181 (A), 87–97. 1174–1187.
Musen, X., Jianxiong, Z., 2018. Impacts of heterogeneous environment awareness and Yenipazarli, A., Vakharia, A., 2015. Pricing, market coverage and capacity: Can green
power structure on green supply chain. RAIRO Oper. Res. 52 (1), 143–157. and brown products co-exist? European J. Oper. Res. 242 (1), 304–315.
Ranjan, C., Sheshadri, C., Shivam, G., Sachin, K., 2023. Green supply chain technology Yenipazarli, A., Vakharia, A.J., 2017. Green, greener or brown: choosing the right color
and organization performance: Moderating role of environmental dynamism and of the product. Ann. Oper. Res. 250 (2), 537–567.
product-service innovation capability. Technovation 128 (Suppl C), 102857. Yin, S., Zhao, Y., 2024a. An agent-based evolutionary system model of the transforma-
Shigui, M., Yong, H., 2023. Investment of green technology in a tourism supply chain tion from building material industry (BMI) to green intelligent BMI under supply
under competition. IEEE Trans. Eng. Manage. 70 (3), 1199–1214. chain management. Humanit. Soc. Sci. Commun. 11 (1).
Song, A., Rasool, Z., Nazar, R., Anser, M.K., 2024. Towards a greener future: How Yin, S., Zhao, Y., 2024b. Digital green value co-creation behavior, digital green network
green technology innovation and energy efficiency are transforming sustainability. embedding and digital green innovation performance: moderating effects of digital
Energy 290, 129891. green network fragmentation. Humanit. Soc. Sci. Commun. 11 (1).
Tully, S.M., Winer, R.S., 2014. The role of the beneficiary in willingness to pay for Yu, W., Ramanathan, R., Nath, P., 2017. Environmental pressures and performance: an
socially responsible products: A meta-analysis. J. Retail. 90 (2), 255–274. analysis of the roles of environmental innovation strategy and marketing capability.
Wang, A., Hu, S., Zhu, M., Wu, M., 2024. Customer contagion effects of voluntary Technol. Forecast. Soc. Change 117, 160–169.
environmental regulation: A supplier green innovation perspective. Energy Econ. Zhang, L., Wang, J., You, J., 2015. Consumer environmental awareness and channel
132, 107446. coordination with two substitutable products. European J. Oper. Res. 241 (1),
Wang, J., Liu, J., 2019. Vertical contract selection under chain-to-chain service 63–73.
competition in shipping supply chain. Transp. Policy 81, 184–196. Zhang, S., Wang, C., Yu, C., Ren, Y., 2019. Governmental cap regulation and manu-
Wang, Y.-y., Su, M., Wang, X.-d., 2022. Altruistic decision-making of closed-loop supply facturer’s low carbon strategy in a supply chain with different power structures.
chain under government incentive mechanism. Chin. J. Manage. Sci. 30 (11), Comput. Ind. Eng. 134 (C), 27–36.
105–116. Zhou, H., Duan, Y., 2022. Online channel structures for green products with reference
Wang, Y., Yu, Z., Jin, M., Mao, J., 2021. Decisions and coordination of retailer-led greenness effect and consumer environmental awareness (CEA). Comput. Ind. Eng.
low-carbon supply chain under altruistic preference. European J. Oper. Res. 293 170.
(3), 910–925. Zhu, W., He, Y., 2017. Green product design in supply chains under competition.
Xia, J., Niu, W., 2021. Carbon-reducing contract design for a supply chain with European J. Oper. Res. 258 (1), 165–180.
environmental responsibility under asymmetric information. Omega 102, 102390. Zong, J., Lu, Y., Tang, F., 2014. Environment attitude, WTP and environmental
premium: evidences from laboratory. Nankai Bus. Rev. 17 (2), 153–160.

16