An Analysis of Driving Factors of Collaboration in Handling Used Cell Phones as a Waste Management Practice

The handling of used cell phones in Indonesia is mostly carried out by informal actors, starting from second-hand market actors. However, the activities of the informal actors often endanger the environment and human health. To reduce the impact, some of the activities should be transferred to formal parties. This requires collaboration of both parties as a form of waste management, which previously has never been established. The objective of this study is to explore the driving factors of collaboration intention of informal actors in handling used cell phones with the formal ones. Data were collected using questionnaires distributed to second-hand market actors in five districts in the Special Region of Yogyakarta Province. In this study, three internal driving factors are considered, which are environmental attitude, management commitment, and financial benefits, as well as two external driving factors: government support and competitor pressure. The regression analysis in each region revealed that the most significant driving factors vary across different regions, such as government support in Gunungkidul, management commitment in Bantul, competitive pressure and government support in Kulonprogo, government support, management commitment, and financial benefit in Sleman, as well as environmental attitude and financial benefit in Yogyakarta City. From the structural equation modeling at the provincial level, it was found that financial benefit and government support were the most significant factors influencing the collaboration intentions of all informal actors. The results of this study can be used as a reference.


Introduction
The number of mobile phone users in Indonesia, especially in the Special Region of Yogyakarta Province (DIY Province), is increasing. Data from the Central Bureau of Statistics (BPS) generally show that the percentage of the population using mobile phones has increased from 2012 to 2019. However, there has been a slight decline in 2020, possibly due to the COVID-19 pandemic. The increase in mobile phone users is shown in Figure 1.
The increase in the number of cell phone usage indicates that there is also an increase in cell phone waste, which is e-waste. The development of technology is followed by an increase in the amount and complexity of waste. Since the waste contains toxic materials that might have a negative impact on both human and environmental health, waste management requires accurate identification and an awareness of the risks involved [1]. Hazardous metals in e-waste are lead, cadmium, mercury, hexavalent chromium, and refractory materials [2][3][4][5][6]. In the long term, these hazardous metals can have an impact on human health and the environment. According to Robinson (2009) [7], in every 1 kg of ewaste, there is 180 mg of cadmium and 0.8 mg of mercury, especially in battery components. Furthermore, the BBC Waste is generally defined as something that is no longer used and thrown away by its owner. Furthermore, Dadzie et al. (2020) [10] explain various definitions of waste. It is understood that all forms of waste must be managed and handled correctly not to burden the environment and disturb public health. E-waste is waste from various electronic and electrical products. The definition and classification of e-waste depend on the regulation or standards that apply in each country. Gollakota et al. (2020) [11] present a detailed classification of e-waste based on several standards. Likewise, Shittu et al. (2021) [12] explained the definition of e-waste and its classification. In contrast to developed countries that explicitly define and classify e-waste and have standardized rules for its management, Indonesia has no specific rules regarding e-waste management. Generally, e-waste is classified as toxic and hazardous material regulated in Minister of Environment and Forestry Regulation No. 6 of 2021 concerning procedures and requirements for managing hazardous and toxic waste.
One of the activities that can be used to manage End of Life (EoL) or End of Use (EoU) products such as used cell phones is waste management. Waste management is a step or various strategies for managing and disposing waste. It can be done by disposing, destroying, processing, recycling, reusing, or controlling waste. Waste management aims to reduce unusable materials while preventing potential environmental damage and threats to human health. The management of used cell phones as electronic waste is known as e-waste management.
Several authors described effective e-waste management, including Isernia et al. (2019) [13] stated that the collection point is the key to e-waste management. The effectiveness of the collection process is influenced by the distribution of collection points in an area. Rautela et al. (2021) [14] stated that for effective implementation of sustainable e-waste management, government supervision is needed in the reciprocal relationship between manufacturers, producers, wholesalers, traders, consumers, and recyclers. Furthermore, Shittu et al. (2021) [12] suggested that one part that needs attention in effective e-waste management is to apply and enforce e-waste management rules for informal actors. As for e-waste management in developing countries, Ilankoon et al. (2018) [15] mentioned that it is necessary to apply a strict legislative framework to realize an e-waste management strategy. These regulations can be developed through modification of the EPR scheme, which provides an e-waste management system that is easy to adopt, provides benefits to all stakeholders, and is adapted to the local economy.
Several obstacles caused the implementation of e-waste management not to run optimally. Gollakota et al. (2020) [11] explained that effective e-waste management is not yet available in most developing countries, and the focal point of successful e-waste management is debatable. One of the shortcomings of e-waste management in developing countries is that the integration of the formal and informal sectors has not been integrated. In addition, there are several other factors, such as the unavailability of special rules for e-waste, the influence of socio-cultural aspects, and the lack of responsibility of producers and consumers. Meanwhile, Rautela et al. (2021) [14] stated that e-waste management in developing countries does not effectively run because e-waste is treated and managed informally in an illegitimate way. Moreover, the unavailability of policies, rules, and regulations, lack of law enforcement, and the implementation of the legal framework for EPR have not yet been maximized.
In Indonesia, mobile phone waste is mostly managed by informal actors, through the starting point of the secondhand market. Informal actors will obtain economic benefits from the activities of managing used cell phones. However, if used cell phone management activities are carried out without safe technology, the impact can affect the health of informal actors and damage the environment. Several researchers such as Chatterjee and Kumar (2009) [2], Chi et al. (2011) [3], Joseph (2007) [4], Li et al. (2011) [16], Kyere et al. (2018) [17], and Robinson (2009) [7] have elaborated the contamination of the environment due to e-waste management activities by informal parties. As Wilson (2007) [18] stated, one of the drivers of waste management is the remaining value of waste, which encourages people to use it as a source of income, especially in developing countries.
On the other hand, Original Equipment Manufacturers (OEMs) as formal actors have technological capabilities to handle used cell phones. However, the offer of a used cell phone return program in Indonesia is not carried out routinely and according to Budijati et al. (2015) [19], the program is not well known by the public, so that after cell phone usage, consumers sell the used mobile phones in the second-hand market, dispose, store, or give them to other people.
To reduce or eliminate the negative impact of the management of used cell phones by informal parties and still provide economic benefits for them, it is necessary establish collaboration in the management of used cell phones involving informal and formal parties. This collaboration aims to regulate the distribution of stages in the management of used cell phones to the disposal process that is safe for the environment and health.
Several authors have provided the definition of collaboration within the supply chain framework. Simatupang and Sridharan (2002) [20] define collaboration as two or more independent companies that work together to plan and implement supply chain operations in order to get a better success rate. Dung (2015) [21], Hudnurkar et al. (2014) [22], Soita (2015) [23], and Wu and Chiu (2018) [24] elaborated the definition of collaboration based on the definitions of other researchers. On the other hand, Maheswari (2019) [25] proposed an engagement model involving the government and intermediary businesses in handling e-waste problems in Indonesia that included empowerment, collaboration, and participation.
The potential or possibility of collaboration between informal and formal actors was conveyed by [26]. They reviewed the situation in several countries and proposed the integration of ISR (Informal Sector Recycling) into the formal sector by taking into account contexts and local conditions. Furthermore, Sasaki et al. (2014) [27] investigated the possibility of integrating the informal sector into formal waste management in Indonesia. Meanwhile, based on an analysis of the situation in four countries regarding informal sector business processes, Wilson et al. (2009) [28] stated that there was a clear potential for mutually beneficial cooperation between the formal and informal sectors. Furthermore, Li and Tee (2012) [29] suggested that to minimize the negative impact of informal channel activities, RL activities and the integration of IWS (Informal Waste Sector) into the formal sector are required.
To the best of the researchers' knowledge, no formal and informal forms of cooperation have been found in Indonesia in the handling of used cell phones. However, there are communities of second-hand market actors in some regions. This community has regular meetings to strengthen their relationship.
This study aims to explore the intention for collaboration of informal parties in managing used cell phones with the formal ones and the driving factors of the collaboration intentions. The intention of collaboration in this study is the intention to carry out management activities for used cell phones so that used cell phones can be returned to their origin point for the handling process or if the disposal is required, it does not damage the environment or endanger human health.
Factors driving collaboration intentions are based on factors that can encourage informal actors to carry out activities that lead to the prevention of environmental damage or environmental behaviours in general. Several researchers who explain the environmental behaviour of managers include Leszczynska (2010) [30] who examines the environmental awareness of managers and further investigates whether this awareness is related to socio-economic development. This study involved 200 managers in Australia and Ukraine and 250 managers in Poland. Lopez-Gamero et al. (2011) [31] examined the environmental attitudes of hotel managers in Spain in the form of perceptions of the natural environment which are influenced by internal and external factors of the company. Next, Nambiar and Chitty (2014) [32] examined the views of business managers in India on the relationship between sustainability and the environment. Ye et al. (2013) [33] examined the attitudes of top managers in 209 companies in China regarding the implementation of reverse logistics in the form of product returns and product recovery. Therefore, the research objectives in this study are:  To explore the collaboration intention of informal actors to manage used cell phones with formal actors so that cell phone waste remains safe for the environment and human health;  To identify the driving factors of the collaboration intention of informal actors in handling used cell phones;  To examine the influence of the driving factors on the collaboration intention of informal actors in handling used cell phones.

Literature Review and Hypotheses Development
This sub-section discusses the driving factors that can influence collaboration intentions for informal actors. This collaboration intention refers to the intention to behave in the environment. The driving factors for environmental behaviour comes from the internal or external. Internal driving factors are factors that come from individuals as part of the company or factors that describe the company's internal conditions. On the other hand, the external driving factors are factors that come from outside the company. The factors are outside the company's control but are able to affect the company's performance.
Fraj-Andrés et al. (2008) [34] stated that the environmental behaviour of firms depends on some internal and external forces. He et al. (2018) [35] also proposed the existence of internal and external pressures on corporate environmental behaviour in their study of 702 paper-making companies in China. Testa et al. (2016) [36]conducted a study to determine the effect of external pressure, internal factors, and environmental attitudes of entrepreneurs on a small and micro-scale company's proactive environmental strategy. Yen & Yen (2012) [37] explored internal and external motivations for green purchasing activities in electronic companies in Taiwan.
The factors considered in this research include internal and external factors, which consist of three internal factors (environmental attitude, management commitment, and financial benefits) and two external factors (government support and competitor pressure). These factors were determined based on a literature review of factors that can motivate environmental intentions and behaviour in a company and adjustments of the case studies in this research were done through field validation by asking several respondents in each research area whether the informal actors felt these factors involved.
The driving factors considered in this study are explained as follows.

Environmental Attitude
Environmental attitude in this study refers to the positive attitude of informal actors (second hand cellphone actors) towards the environment, commonly referred to as an environmentally friendly attitude. Janmaimool & Khajohnmanee (2019) [38] define environmental attitude as a person's belief in the relationship between humans and the environment. This belief also includes understanding the consequences when environmental damage occurs. Meanwhile, Chan et al., (2017) [39] stated that environmental attitude is often equated or interchanged with environmental concern. Arshad et al. (2022) [40] and Okumus et al. (2019) [41] explored environmental attitudes in environmental knowledge, awareness, and concern.
He et al. (2018) [35] showed that internal pressure could affect corporate environmental behaviour, where employees' and top management's environmental awareness are part of internal pressures. Then, Okumus et al. (2019) [41] proposed that the environmental attitude (in terms of environment concern) of hotel employees in Turkey is the best predictor of ecological behaviour. Furthermore, Arshad et al. (2022) [40] stated that employees with an excellent environmental attitude would encourage organizations to implement environmental management system policies. Moreover, Long et al. (2017) [42] examined the impact of attitude toward environmental behaviour on environmental innovation intention in 182 companies of various types in China. Then, Tariq et al. (2020) [43] examined the relationship between employees' environmental attitudes and employees' ecological behavior of employees in 65 small and mediumsized hotels operating in Pakistan's tourist areas. Next, Testa et al. (2016) [36] conducted a study to determine the effect of environmental awareness of managers to adopt a proactive corporate environmental strategy in 355 small and microscale businesses in Liguria, a region in central Italy. Then, Zientara & Zamojska (2018) [44] examined the relationship between environmental values or beliefs held by hotel employees in Poland with organizational citizenship behaviour for the environment (OCBE), and the results showed a positive relationship. From the previous studies reviewed, it is expected that someone with a high environmental attitude will show positive environmental behaviour. Therefore, the hypothesis of this research is as follows.

H1
: Environmental attitude has a positive and significant effect on the intention to collaborate with informal actors in handling used cell phones.

Management Commitment
Management commitment is a form of responsibility and commitment from the owners and employees of the secondhand cell phone market to carry out activities that support the environment. In this case, it can be realized through collaboration in the management of used cell phones among formal parties so that the activities of handling used cell phones do not damage the environment or endanger workers' health.
In general, as stated by El-Kassar and Singh (2019) [45], management commitment is the encouragement given by a company to carry out environmentally friendly activities and to incorporate the ideas into the corporate culture. Yusliza et al. (2019) [46] argued that to achieve the successful implementation of green activities, top management must provide a high commitment so that the implementation of green activities can offer a competitive advantage for the company. In addition, Ates et al. (2012) [47] stated that organizational capability plays an important role in facilitating the implementation of the company's environmental strategy and impacts environmental performance, where one source of organizational capability is organizational commitment.  [47] demonstrated that organizational commitment positively impacts the extent to which firms adopt a proactive environmental strategy in manufacturing firms in Turkey. Ghazilla et al. (2015) [48] stated that management commitment is one of the drivers for implementing green manufacturing practices of SMEs in Malaysia. Likewise, Nordin et al. (2014) [49] showed that top management commitment is one of the main driving factors of sustainable manufacturing in manufacturing companies in Malaysia, involving respondents from operation managers, manufacturing managers, and the environmental, safety and health managers. In addition, He et al. (2018) [35] stated that internal and external pressure affects corporate environmental behaviour, where one form of internal pressure is commitment management. Tariq et al. (2020) [43] found that managers' environmental commitment strengthens the relationship between employees' environmental attitudes and their ecological behavior. Yen and Yen (2012) [37] show a positive and significant effect of top management commitment on environmental collaboration with suppliers and green purchasing activities in the electronics industry in Taiwan. Yusliza et al. (2019) [46] found that top management commitment influences various green human resource management (GHRM) activities, which include green analysis and job description of job position, green performance, green recruitment, green rewards, green selection, and green training in 400 Malaysian manufacturing and service organizations.
Furthermore, Bhatia and Jakhar (2021) [50] studied the effect of top management commitment (TMCO) on green product innovation (GPI) in Indian automotive manufacturing companies empirically. Burki et al. (2019) [51] examined the relationship between top management commitment and process innovation in the green supply chain (GSC), in the form of green process innovation and green managerial innovation in selected ISO 14000 certified Turkish exporting firms located in the Izmir region (Turkey). Meanwhile, Lee and Joo (2020) [52] investigated whether support from top management can significantly increase the level of environmental collaboration with participating companies in upstream and downstream green supply chains and their impact on environmental work in companies in manufacturing industries in South Korea.
It is necessary to have awareness from the internal parties in implementing environmental conservation activities. The management commitment factor explains organizational commitment from top management to employees in behaviour that supports environmental preservation. Therefore, related to this research, the proposed hypothesis is as follows: H2: Management commitment has a positive and significant effect on the intention to collaborate with informal actors in handling used cell phones.

Financial Benefit
The financial benefit as a driving factor in this study is the potential benefits obtained when informal actors carry out environmental conservation activities in the form of collaboration in handling used cell phones. Maheswari, et al. (2020) [53] investigated that finance was one of the informal e-waste business performance measurements using a sustainable reverse logistics scorecard. Financial was one of the dimensions/driving factors that influenced the green supply chain collaboration [54] and green manufacturing practice in small medium enterprises [48].
Likewise, Nordin et al. (2014) [49] identified that one of the driving factors of sustainable manufacturing activity is economic benefit; the research was conducted in Malaysian manufacturing industries. Aside from the improper disposal, the challenges of the current EoL electrical and electronic equipment recycling program were the home storage and the informal actors. The reason is the lack of economic incentives for the proper return of used electronic equipment, especially for expensive and quickly obsolete products [55]. Henriques and Catarino (2016) [56] conducted a preliminary study on small and medium-sized companies in Portugal adopting energy efficiency improvements. One of the identified motivators is financial factor, which provide benefits in strengthening capacity, providing financing, external parties for financial and technical guidance, access to capital, financial resources to develop bankable projects, and financial support for investment. Then, Kudlak (2017) [57] examined the drivers for implementing environmental management systems in companies in Poland. These drivers include efforts to reduce costs, increase sales, and increase market share. Next, Meath et al. (2016) [58] identified the key factors that motivate and hinder the design process of a voluntary energy efficiency program for SMEs in Queensland. Factors related to economic profit are one of the motivators, namely financial interests in the form of reducing energy costs, opportunities to obtain funding, and obtaining marketing opportunities due to the promotion of environmental performance achievements. Testa et al. (2016) [36] examined managers' opinions about motivators in the adoption of proactive corporate environmental strategies. One of the motivators is the internal factor which is cost-saving. Meanwhile, Wang et al. (2018) [59] stated that one of the reasons companies support green supply chain management activities is the cost driver, where companies can reduce costs while helping environmental sustainability. This research was applied to plants in three industries: machinery, electronics, and transportation in various countries. It was found that the cost drivers significantly influence internal and external green practices. Referring to the results of previous studies, the hypothesis in this study is as follows: H3: Financial benefit has a positive and significant effect on the intention to collaborate with informal actors in handling used cell phones.

Government Support
As one of the stakeholders in environmental control, the government plays a role in motivating environmental conservation. Government support in this study is the role of the government in supporting informal actors to reduce used cell phone management activities that are not safe for the environment and health. This support can be in the form of regulation, incentives provided, and facilities and infrastructure that informal actors can utilize in carrying out their activities.
The previous studies showed the government's role in encouraging companies to carry out environmental conservation activities. As stated by Lee (2008) [60], the involvement of local and central governments in green supply chain (GSC) initiatives is in the forms of coordinating the GSC initiatives, increasing funds for the activities, providing information and technical assistance to small and medium-sized firms, popularizing knowledge of environmental management, and build infrastructure for facilitating GSC initiatives. Meanwhile, Tatoglu et al. (2015) [61] examined the relative importance of each dimension of the Corporate Environmental Policies (CEP). One of the dimensions is stakeholder pressure, including government policies and regulations.
Several studies are related to an initial study on the importance of government support in environmental activities, including [48]. They conducted a preliminary study to determine the driving and inhibiting factors for applying green manufacturing practice in Malaysian SMEs. They found that the legislation factor in which there is a financial incentive from the government is one form of government support. Then, Henriques and Catarino (2016) [56] identified the situation in small and medium-sized companies in Portugal adopting energy efficiency improvements. Government policy is considered as a motivator, where government policies include the obligation of the state and government to develop effective energy programs and the need to design energy efficiency programs. There are government representatives in suppressing energy efficiency, providing fiscal subsidies, and providing grants for technology investment. Next, Moktadir et al. (2018) [62] identified the primary motivators in adopting sustainable manufacturing practices for the Bangladesh leather industry. One of the identified drivers is governmental support and legislation, which the government requires to provide funds to implement these activities smoothly.
Meanwhile, studies examining how government support influences corporate environmental activities include [35]. They examined government pressure which influenced corporate environmental behaviour, which consists of environmental defensive behaviour, environmental accommodative behaviour, and proactive environmental behaviour. Ye et al. (2013) [33] examined government pressure on the attitude of top managers to implement RL in the form of product returns and product recovery and found that government pressure had a significant effect on managers' attitudes.
From these previous studies, it is clear that government support is one of the driving factors of the company's environmental activities. For this reason, the hypothesis developed regarding government support is as follows: H4: Government support has a positive and significant effect on the intention to collaborate with informal actors in handling used cell phones.

Competitor Pressure
According to Dai et al. (2018) [63], competitive pressure influences a company's response to aggressive environmental strategies. It is the key role in adopting small medium enterprises because it is sensitive to the competition they have [64]. The competitor pressure factor means competitor activities such as obeying existing regulations, being committed to environmental activities, and establishing cooperation in environmental conservation which will affect informal parties. The influence of competitor pressure related to environmental preservation will encourage informal actors to do the same.  [65] examined the effect of competitor pressure in applying green innovation in manufacturing and service firms in Taiwan. This study found that competitor pressure had a positive and significant impact on the company's green innovation activities. Furthermore, Ye et al. (2013) [33] examined the effect of competitor pressure on the attitude of top managers to carry out RL activities in the form of product returns and product recovery. It was found that competitor pressure had a significant effect on the attitude of managers toward the implementation of the RL. From the previous research reviewed, the proposed hypothesis in this study is as follows: H5: Competitor pressure has a positive and significant effect on the intention to collaborate with informal actors in handling used cell phones.
Based on the explanation of the factors that influence the intention of second-hand market actors to collaborate in handling used cell phones, the research model proposed in this study is shown in Figure 2.

Research Object
The research object is the second-hand mobile phone market actors who are the initial players in the informal channel of handling used cell phones. This research started from the phenomenon of the rise of second-hand cell phone market actors in Indonesia who carry out selling, buying, and repairing used cell phones. It is because Indonesia has not obliged cell phones manufacturers to take back used cell phones that consumers no longer use. This condition is an opportunity for the informal sector to buy and sell used mobile phones that can provide economic benefits.

Management commitment
Collaboration intention Financial benefit Government support Competitor pressure On the other hand, the activities of informal actors in handling used cell phones, such as burning, throwing the remaining components into the trash, and taking precious metals by chemical processes, can pollute the environment and endanger their health. To reduce the danger level due to informal actors' activities, it is necessary to propose management of mobile phone waste involving informal and formal actors. Based on these reasons, this research was conducted to identify the intentions of informal actors to collaborate in handling cell phones with formal actors.
The research respondents were market actors for second-hand mobile phones in five regencies/cities in the Special Region of Yogyakarta Province (DIY Province), consisting of the Regencies of Gunungkidul, Bantul, Kulonprogo, Sleman, and the City of Yogyakarta.

Measurement
The measurement of collaboration intention was conducted using questionnaire with a Likert scale of 1 to 5. The initial questionnaire consisted of 32 items. Those factors are adapted from the following previous studies:  Environmental attitude was adapted from [36,42].

 Collaboration intention 3 items adapted from Ajzen
The complete questionnaire items are shown in Appendix I.

Research Stages
The stages of the research included:  Distribution of the initial questionnaire to the respondents;  The validity and reliability test of the initial questionnaire using the SPSS software version 16;  Distribution of the formal questionnaire to the respondents;  Conducting a classic assumption test for data in each district/city using the SPSS software version 16;  Performing regression analysis for data in each district/city using the SPSS software version 16;  Conducting confirmatory factor analysis (CFA) using AMOS 25 for data in all areas of DIY Province;  Developing structural equation modelling (SEM) development using AMOS 25 software for data in all areas of DIY Province; confirmatory factor analysis (CFA) used AMOS 25.

Demographic Characteristic
The questionnaire was distributed to second hand market actors (that includes buying and selling, cell phone service or cannibalization) in Yogyakarta province (424 respondents from September to October 2018). It consists of several regencies, such as Bantul with 75 respondents, Sleman with 160 respondents, Yogyakarta municipality with 90 respondents, Gunungkidul with 50 respondents, and Kulonprogo with 49 respondents. The respondents' characteristics are age, gender, household member, income, education level, marital status, position at work, and treatment of used components as shown in Figure 3  The majority of respondents are male and 71% of the respondents is between 20 and 30 years old. Interestingly, 92% of respondents have senior high school background and above for education level; almost 55% of the respondents have an income of approximately 1-2 million per month.
In addition, as shown in Figure 4, the majority of the respondents are unmarried and the position at work is an employee. Almost 60% of the respondents save the used components. The reason is that the respondents often reuse those parts for other broken cell phones and repair those parts for resale.

The Location of Second-Hand Market Actors
Second-hand market actors as research objects are spread across five regencies/ cities within the DIY Province. DIY Province is one of the provinces in Indonesia, located on the South-Central side of Java Island. It is known as a Special Region because it is the territory of the Yogyakarta Palace. The location of the DIY Province is between 7.33-8.12 South Latitude and 110.00-110.50 East Longitude, with an area of 3185.80 km 2 or 0.17% of the total area of Indonesia.  Gunungkidul region has the position 007°46'00"-008°09'00" south latitude & 110° 21' 00"-110° 50' 00" east longitude;the border on the north is Klaten region, the south is Indian Ocean, the west is Bantul region and Sleman region, and the east is Wonogiri region. It is the biggest area compared to other areas; it is about 1,431 km 2 [68]. Then, a total population about 747,161 per 2020 and population growth rate of 0.88% [69]. This region has 18 sub-districts, 144 villages, and 1,431 hamlets. Geographically, Gunungkidul region is located in the southeastern part of the Special Region of Yogyakarta. Gunungkidul Regency has neither inland nor remote areas. According to the geographical conditions, there are 18 coastal villages, 56 villages located on the slopes/ridges of the hills and 70 villages located on the plains. The capital of this region is Wonosari.
The position of Bantul region is 14º 04' 50" -27º 50' 50" South Latitude and 110º 10' 41" -110º 34' 40" East Longitude. This region is surrounded by Gunungkidul region on the east, Yogyakarta municipality and Sleman region on the north, Kulonprogo region and Indian Ocean are on the west and the south, respectively. The area is about 508.13 km 2 [68].Then, in 2020, the population was about 985,770. This region had the highest growth rate compared to other regions, which is 1.14% [69]. Bantul region has a plain area located in the middle and hilly areas located in the east and west, as well as a coastal area in the south. This region consists of 17 sub-districts divided into 75 villages and 933 hamlets [70]. The capital of this region is Bantul.
The position of Kulonprogo region is 007 º 38' 42" -007 º 59' 3" South Latitude and 110 º 01' 37" -110 º 16' 26" East Longitude. It is surrounded by Bantul region and Sleman region on the west; Magelang, Indian Ocean, and Purworejo are on the north, the south, and the west, respectively. Not much different from Bantul region, the area of Kulonprogo region is 586 km 2 [68]. In 2020, the population was about 436,395 with the growth rate at about 0.99% [69]. This region consists of 12 sub-districts and 88 villages [71]. The capital city of this region is Wates.
The position of Sleman region is 110º 33' 00" and 110º 13' 00" East Longitude, 7º 34' 51" and 7º 47' 30" South Latitude. It is surrounded by Boyolali region and Central Java on the north; Yogyakarta municipality, Bantul region, and Gunungkidul region on the south; Kulonprogo region and Magelang region on the west; and Klaten region and Central Java on the east. The area of the Sleman region is about 574 km 2 [68]. Compared to other regions, Sleman region has the highest population which is about 1,125,804 with the growth rate of 1.06% [69]. It consists of 17 sub-districts with 86 villages and 1212 hamlets [72]. The capital of this region is Sleman.
The position of Yogyakarta municipality is 110º 24' 19" to 110º 28' 53" East Longitude and 7º 15' 24" to 7º 49' 26" South Latitude with an average elevation of 114 m above sea level. The borders are Sleman region and Bantul region on the north and the south, respectively. Then, Bantul region and Sleman region are both on the west and the east. Compared to other regions, Yogyakarta municipality has the smallest area which is about 32.50 km 2 [68]. Even though the total population was about 373,589 people and the growth rate was 1.06%, this region is the most densely populated area [69]. There are fourteen districts and 45 villages [73]. The capital of this region is Yogyakarta city.

Regression Analysis Result
This section discusses the results of collaboration intention and regression analysis of driving factors on handling the collaboration of used cell phones for the second-hand market actors. Before the regression analysis was applied, the reliability and validity tests were carried out on the questionnaires distributed in each region. The tests for the city of Yogyakarta, Bantul, Gunungkidul, and Kulonprogo districts used 30 initial respondents' answers, while for Sleman Regency used 40 respondents' answers. Reliability tests were conducted to measure the consistency of the questionnaire which is an indicator of the variables. A questionnaire is said to be reliable if a respondent's answers to the questions are consistent from time to time. According to Barr and Gilg (2007) [74], the questionnaire is said to be reliable if the value of Cronbach α is more than 0.6. The test results for each region showed that the Cronbach α coefficient of six variables was more than 0.6, which means that all the factors were reliable.
In addition, the validity test was required to show the extent to which the questionnaire items used in a study were able to measure what it aimed to measure. Validity test was used to measure the validity of questionnaire items. The validity test in this study was carried out by comparing the total Pearson correlation value with the R

Gunungkidul Region
Respondents in Gunungkidul were 50 informal actors. From the respondents' answers, the average value of each factor studied and the value of collaboration intentions were obtained, which is presented in Table 1. The calculation shows that the collaboration intention of informal actors in the Gunungkidul region is 3.46, which means the informal actors argue that they are neutral and tend to collaborate. Before performing regression analysis, it is necessary to test the classical assumptions. The purpose of classical assumption testing is to provide certainty that the regression equation obtained is accurate in estimation, unbiased, and consistent. This classic assumption test is a prerequisite test that is carried out before carrying out further analysis of data collected. Classical assumption test in this research consists of normality test, multicollinearity test, and heteroscedasticity test. The summary of classical assumption tests is displayed in Table 2. The classical assumption test results for each region are presented in Appendix II. It can be seen that for Gunungkidul area, all classical assumption tests are met. The regression result of driving factors on handling collaboration of used cell phones for second-hand market informal actors in Gunungkidul region is presented in Table 3. It can be seen that the government support is the most significant driving factor compared to others. Factors that affect collaboration intentions are seen if the significance value is smaller than 0.05. Therefore, these factors affect collaboration intentions in handling used cell phones. In Gunungkidul region, one factor influences collaboration intentions in handling used cell phones that is government support with a significance value of 0.004. It was also found that management commitment had a moderate effect on collaboration intentions with a significance value of 0.158. Based on the results of multiple determination (R 2 ) of 0.462 (see Table 4), all predictors (independent variables) of collaboration can explain the variation of collaboration intention by 46.2% while 53.8% is influenced by other factors outside the model.

Bantul Region
In Bantul area, 75 informal actors were involved as respondents. The average value of each factor and the value of collaboration intentions, based on the respondents' answers, are presented in Table 5. The results show that the value of collaboration intentions of informal actors in Bantul Regency is 3.87, meaning that informal actors have the intention to collaborate. As this research uses regression analysis, it is preceded by the classical assumption test. The results of classical assumption tests is displayed in Table 6. The classical assumption test results in detail are shown in Appendix II. The results of the classical assumption test for the Bantul region show that it is fulfilled for all types of tests. The regression result of driving factors on handling collaboration of used cell phones for second-hand market informal actors in the Bantul region is presented in Table 7. It can be seen that the management commitment is the most significant driving factor compared to others. The calculation shows that the collaboration intention of informal actors in Bantul region is 3.87. Factors that affect collaboration intentions are seen if the significance value of α is smaller than (0.05). Then, these factors affect collaboration intentions in handling used cell phones. In Bantul region, one factor influences collaboration intentions in handling used cell phones, which is management commitment with a significance value of 0.010. Also, a factor that has a moderate influence on collaboration intentions, namely environmental attitude with a significance value of 0.154, was obtained. Based on the results of multiple determination (R 2 ) of 0.270 (see Table 8), all predictors (independent variables) of collaboration can explain the variation of collaboration intention by 27% while 73% is influenced by other factors outside the model.

Kulonprogo Region
In the Kulonprogo area, there are 49 informal actors involved as respondents. Table 9 shows the average value of respondents' answers for each factor and collaboration intention. The value of collaboration intention is 3.783, meaning that informal actors in Kulonprogo Regency intend to collaborate in handling used cell phones with formal actors. Classical assumption test needs to be done before performing regression analysis. The summary of classical assumption tests for Kulonprogo region is displayed in Table 10. The complete results of the classical assumption test are presented in Appendix II. For Kulonprogo region, all classical assumption tests are fulfilled. The regression result of driving factors on handling collaboration of used cell phones for second-hand market informal actors in the Kulonprogo region is presented in Table 11. It can be seen that the competitors' pressure is the most significant driving factor compared to other factors. The calculation shows that the collaboration intention of informal actors in the Kulonprogo region is 3.783. Factors that affect collaboration intentions are seen if the significance value of α is smaller than (0.05). Then, these factors affect collaboration intentions in handling used cell phones. In Kulonprogo region, one factor influences collaboration intentions in handling used cell phones that is competitor pressure with a significance value of 0.013 and government support with significance of 0.015. Based on the results of multiple determination (R2) of 0.364 (see Table  12), all predictors (independent variables) of collaboration can explain the variation of collaboration intention by 36.4% while 63.6% is influenced by other factors outside the model.

Sleman Region
Respondents in Sleman Regency were 160. The average value of the respondents' answers about the five factors studied and the average value of collaboration intentions are presented in Table 13. It can be seen from the table that the value of collaboration intentions of informal actors in Sleman is 4,008. This value is the highest intention value compared to other regions. That means they really intend to collaborate. Before performing regression analysis, it is necessary to test the classical assumptions. The results of classical assumption tests for Sleman region is displayed in Table 14. The complete results of the classical assumption test are shown in Appendix II. The results of the classical assumption test for the Sleman region show that it is fulfilled for all types of tests. The regression result of driving factors on handling collaboration of used cell phones for second-hand market informal actors in the Sleman region is presented in Table 15. It can be seen that the government support is the most significant driving factor compared to others. The calculation shows that the collaboration intention of informal actors in the Sleman region is 4.008. Factors that affect collaboration intentions are seen if the significance value of α is smaller than (0.05). Then, these factors affect collaboration intentions in handling used cell phones. In the Sleman region, one factor influences collaboration intentions in handling used cell phones that is government support with a significance value of 0.002, followed by management commitment and financial benefit significance level of 0.040 and 0.056. Based on the results of multiple determination (R 2 ) of 0.141 (see Table 16), all predictors (independent variables) of collaboration can explain the variation of collaboration intention by 14.1% while 85.9% is influenced by other factors outside the model.

Yogyakarta Municipality
For the Yogyakarta city area, 90 informal actors were involved as respondents. Table 17 presents the average value of each factor and the value of collaboration intention, which was obtained from the answers of the respondents. It can be seen that the value of collaboration intention is 3.60. This means that the level of collaboration intention of informal actors in the city of Yogyakarta is neutral and close to agreeing to collaborate. The summary of classical assumption test that was carried out before the regression analysis is shown in Table 18. The detailed results of the classical assumption test are shown in Appendix II. The results for all types of tests are fulfilled for the Yogyakarta city area, so it can be continued with regression analysis. The regression result of driving factors on handling collaboration of used cell phones for second-hand market informal actors in the Yogyakarta municipality is presented in Table 19. It can be seen that the environmental attitude is the most significant driving factor compared to others. The calculation shows that the collaboration intention of informal actors in the Yogyakarta municipality is 3.60. Factors that affect collaboration intentions are seen if the significance value of α is smaller than (0.05). Then, these factors affect collaboration intentions in handling used cell phones. In the Yogyakarta municipality, one factor influences collaboration intentions in handling used cell phones, which is environmental attitude with a significance value of 0.002 and financial benefit has a moderate impact with a significance level of 0.081. Based on the results of multiple determination (R 2 ) of 0.274 (see Table 20), all predictors (independent variables) of collaboration can explain the variation of collaboration intention by 27.4% while 72.6% is influenced by other factors outside the model. The most influential factors on the intention to collaborate in each region are presented in Figure 10. From the figure, the position of one region compared to other region in the Special Region of Yogyakarta Province can be seen.

The Special Region of Yogyakarta Province
The results for the Special Region of Yogyakarta Province were obtained by processing the data from five regions, which included 424 respondents. The value of this collaboration intention for all actors was 3.744. It means that generally the respondents had the collaboration intention in handling used cell phones with formal actors. The values for each factor at the provincial level, which were obtained from the average values of the five regions, are shown in Table 21. Next, to obtain the factors that influence the collaboration intentions of all actors at the province level, the Structural Equation Modelling (SEM) approach was used. Before the structural model was formed, the CFA (Confirmatory Factor Analysis) was conducted on exogenous variables. For endogenous variables, CFA is not necessary, because there is only one variable. CFA is intended to check whether all questionnaire items can be used. There is no general rule in determining the cut-off value for loading factor (Doll et al., 1995). This paper specified the loading factor as more than 0.5. In the exogenous CFA, the fit model was achieved with the p-value=0.665, χ2=210.536, GFI=0.960, AGFI=0.931, and RMSEA=0.000. Based on the value of the loading factor, there were two items deleted; the first item was the government support factor and the sixth item was competitor pressure.
The structural model result is shown in Table 22. The structural fit model was obtained with p-value=0.051, χ2=343.389, GFI=0.945, AGFI=0.915, and RMSEA=0.019. It shows that the most significant driving factors in the Special Region of Yogyakarta Province is financial benefit and government support with the influence values of 0.231 dan 0.150, respectively and the significance levels are 0.014 and 0.041, respectively. The coefficient of determination is 0.284, shown in Table 23, meaning that the five factors considered in this study describe collaboration intentions of 28.4%, and other factors form 71.6% of collaboration intentions.

Discussion
The results of each factor influencing collaboration intentions and their managerial implications are discussed as follows.

Environmental Attitude
The environmental attitude factor is the dominant factor influencing the intention to collaborate with informal actors in the Yogyakarta municipality and a moderate driving for informal actors in the Bantul Region. This factor is related to the attitude of informal actors towards the environment. It means that informal actors with high environmental attitudes will also show high collaboration intentions.
The result shows that the average environmental attitude of informal actors in both Yogyakarta municipality and Bantul Region is 4, which means that the actors have a high environmental attitude. Furthermore, in terms of the level of education, the education background of informal actors in Yogyakarta is high school with 93.3%, while in Bantul, it is 92%. According to Latif et al. (2012) [75], education level has a significant impact on pro-environmental intention and behaviour. As  [77] stated that differences in the education level will result in differences in the desire to carry out environmentally friendly behaviour. Thus, the higher a person's education level, the higher his concern for the environment.
The results of this study are in line with Arshad et al. (2022) [40], where environmental concern significantly affects the ecological behaviour of employees in small and medium hotels in Pakistan. According to Chan et al. (2017) [39], environmental concern was positively related to ecological behaviour of international tourist hotel employees in Hong Kong. Likewise, He et al. (2018) [35] showed that employee and top management environmental awareness could affect corporate environmental behaviour. In addition, Long et al. (2017) [42] demonstrated the positive and significant impact of the attitude toward environment factors on the environment. Still related to the influence of environmental attitude, the results of Okumus et al. (2019) [41] showed that the environmental concern of hotel employees in Turkey is the best predictor of ecological behaviour. Then, Testa et al. (2016) [36] showed that environmental awareness has a positive and significant effect on proactive environmental strategy. This also aligns with Zientara and Zamojska's (2018) [44] research which demonstrated that environmental values were positively related to organizational citizenship behaviour for the environment (OCBE).
The managerial implication of this study is that to maintain environmental attitudes, informal groups of actors should often hold discussions on environmental issues so that the understanding of the environment becomes even and equal among the actors. Through the Department of Trade and Cooperatives, the government can also provide information about the environment and its relation to used cell phones. In addition, the formal actors should conduct their social responsibilities, such as coaching informal actors to understand how to handle used cell phones so that they are safe for the environment and human beings.

Management Commitment
Management commitment in this study refers to the commitment of owners and employees in carrying out proenvironmental activities, in this case, collaborating with formal parties in managing used cell phones. The results showed that the management commitment factor strongly influences the intention to collaborate of informal actors in Bantul and Sleman regions. As for the Gunungkidul Region actors, this factor has a moderate influence. The value of management commitment in the three regions is 3.8; 3.979; and 3.4 for Bantul, Sleman, and Gunungkidul, respectively. It can be seen that the management commitment of informal actors in Bantul and Sleman Regions is higher than that of the informal actors in Gunungkidul, so it can be said that it is in line with the level of influence.
The strong influence of management commitment is in line with the research of Ates et al. (2012) [47], which showed that organizational commitment has a positive impact on the adoption of a proactive environmental strategy. It is also in line with the research of He et al. (2018) [35] which stated that one of the internal pressures in the form of commitment management affects corporate environmental behaviour. Research by Tariq et al. (2020) [43] found that a manager's environmental commitment strengthens the relationship between employees' environmental attitude and employees' ecological behavior. Yen and Yen (2012) [37] showed a positive and significant effect of top management commitment on environmental collaboration with suppliers and green purchasing activities. In addition, Yusliza et al. (2019) [46] found that top management commitment positively and significantly affects various green human resource management (GHRM) activities. Lee and Joo (2020) [52] show that top management is an essential factor which influences the level of collaboration between suppliers and customers in a green supply chain.
Meanwhile, Burki et al (2019) [51] found that top management commitment has a positive and significant effect on green process innovation, while its influence on green managerial innovation is moderate. In contrast, the results of Bhatia and Jakhar (2021) [50] are not in line with this study, where top management commitment (TMCO) has no significant effect on green product innovation (GPI).
With the results found in Bantul, Sleman, and Gunungkidul Region, the managerial implication that can be emphasized is that informal actors with a high level of management commitment need to be maintained, so that owners and employees of informal actors are always committed to carrying out activities that support environmental conservation. This commitment can be transmitted to other informal actors through meetings held in informal actors' associations in several areas so that owners and employees will understand the importance of being committed to environmental conservation. Local and central governments and formal actors can also contribute to fostering and enhancing this management commitment factor by providing additional information and education related to environmental problems and their handling.

Financial Benefit
The strong influence of financial benefits on the collaboration intention occurred in informal actors in the Sleman Region and studies at the provincial level. It means that informal actors will intend to collaborate in managing used cell phones if they feel there are economic benefits for them. The effect of moderate financial benefits was found in actors in the Yogyakarta municipality area.
The average value of the respondents' answers regarding the financial benefit factor was 3.983 for actors in Sleman Regency, and 3.781 for all provinces, which is the average value of financial benefits in all regions. The value of financial benefits to actors in the Yogyakarta municipality area is 3.7. The value of financial benefits that has a strong impact is more significant than those with a moderate influence.
The results of this study are in line with the results of Wang et al. (2018) [59] found that cost factor significantly influences internal and external green practices. In addition, the used cell phones are usually sold through the informal sector for cashback [55]. For managerial insight, the financial benefits for informal actors are one of the most important reasons for running their business. However, the role of these informal actors has not received adequate attention. Therefore, the formal actors should support the informal ones to collaborate in handling used cell phones.

Government Support
There are three regions and a study at the provincial level. It was found that the government support factor had a strong influence on the intention to collaborate in Kulon Progo, Sleman, and the Gunungkidul Region. Meanwhile, there was no moderate influence of the government support factor for actors in any region. The value of the government support factor in each region is 4.438, 4.019; and 3.7 for Kulonprogo, Sleman, and Gunungkidul, respectively. Meanwhile, the value of government support for actors in all provinces is 4.113.
The government support questionnaire items show that the actors expect the government to support the implementation of pro-environment activities. It will encourage the actors to intend to collaborate in handling used cell phones with formal parties. Therefore, by looking at the value of government support from the Kulon Progo and Sleman regions, as well as at the provincial level, it seems that this value is very high. So, it can be interpreted that these actors expect the government to condition, provide information and technical assistance, popularize environmental management, and provide infrastructure for facilitating environmental activities.
The result of the study is that government support significantly encourages collaboration intentions, which is in line with studies by Lee (2008) in which government involvement plays an important role in the willingness of suppliers to participate in the green supply chain. Also, in the research of He et al. (2018) [35], government pressure influenced corporate environmental behaviour. Ye et al. (2013) [33] showed that government pressure has a significant effect on managers' attitudes to RL implementation. Next, Nguyen et al. (2018) [78] investigated that laws and regulations play the most significant impacts on recycling behavioural intention, compared to environmental awareness and attitude toward recycling, social pressure, the cost of recycling, and the inconvenience of recycling.
Furthermore, the management implication of this result is that the government is authorized to provide support for collaboration between formal and informal actors. This support can be in the form of rules, policies, facilities, resources, and information. So far, regulations related to e-waste in Indonesia can be found in [25]. However, these regulations, namely Government Regulation no. 101 of 2014, do not explicitly mention e-waste, but the waste in question is hazardous and toxic material waste in general. Thus, there are no specific regulations regarding e-waste management in Indonesia.

Competitor Pressure
The results show that the competitor pressure factor strongly influences informal actors in the Kulon Progo Region only and is also not seen as a moderate driving factor. It means the informal actors in Kulon Progo will be encouraged to do collaborative activities in handling used cell phones when their competitors carry out activities related to the environment.
The value of competitor pressure for informal actors in the Kulon Progo Region is 3.982, which means this value is relatively high and higher than the value of competitor pressure in all provinces. From the number of informal actors in Kulon Progo, there are fewer informal actors compared to the other regions, and they are not spread throughout the region, so there is a possibility that the level of competition between informal actors will be high.
The results of research related to competitor pressure are in line with the study of Weng et al. (2015) [65], which found that competitor pressure had a positive and significant effect on the company's green innovation activities. This is also in line with Ye et al. (2013) [33], where competitor pressure significantly affects managers' attitudes toward RL implementation. This finding is supported by previous research conducted by Riva and Gani (2020) [79], demonstrating that competitor pressure also positively affects the environmental performance of upscale hotels. Competitor initiatives and strategies guide the hotels to adopt green marketing practices. The managerial implication of this research is the need for the government and formal actors to provide counseling and training to upgrade the knowledge and skills of informal actors.
Generally, the results of this study can be considered by stakeholders who may be involved in handling e-waste, mainly used cell phones in Indonesia. For the government as part of the policymakers, these results can be used as input for setting rules, providing information, providing assistance, funding assistance, etc., for the safe management of used cell phones. For formal actors such as mobile phone manufacturers, the results of this study can be considered in redesigning and promoting a program to take back used cell phones from consumers as a form of corporate social responsibility. In the end, informal actors' handling of used cell phones will not harm health and the environment but still provides economic benefits for informal actors.
Taking into account the results of the coefficient determination in each region and the study at the provincial level, which is below 50%, it is still necessary to explore other factors that can motivate players in the second-hand cell phone market to collaborate in cell phone management. Furthermore, it is also essential to study the factors that can hinder informal actors from collaborating with formal actors in handling used cell phones. Understanding the factors driving and inhibiting collaboration intentions among informal actors will make it easier to design policy-making and appropriate forms of collaboration.

Conclusions
Based on the analysis of collaboration intentions among informal actors, it was found that:  Informal actors intend to collaborate with formal actors in managing used cell phones, with an average intention value of 3.744.  There is no form of collaboration between informal and formal actors in Indonesia in handling used cell phones.
The results of this study can be used as a consideration for policymakers to regulate e-waste management, which mainly used cell phones.
 Formal actors can also use the results of this study to promote the take-back program of used cell phones as a form of waste management through corporate social responsibility and collaboration with informal actors.

Data Availability Statement
The data presented in this study are available on request from the corresponding author.

Funding
This article is part of a postdoc research financially supported by a Post-Doctoral Grant from the Directorate General of Higher Education, Ministry of Education and Culture of Indonesia with contract number: PDD-001/SKPP/III/2018.

Acknowledgements
This paper is part of a postdoc research financially supported by a Post-Doctoral Grant from the Directorate General of Higher Education, Ministry of Education and Culture of Indonesia with contract number: PDD-001/SKPP/III/2018, which is highly appreciated. The authors also thanked the students of Industrial Engineering Department, Ahmad Dahlan University who helped in collecting data for this study.

Conflicts of Interest
The authors declare no conflict of interest.

Appendix I
The results of the reliability and validity test of the research questionnaire.

A2-2. Multicollinearity Test Results
The results of the multicollinearity test are presented in Table A2-2. From the figure, it can be seen that the tolerance value for all variables is greater than 0.1 and the Variance Inflation Factor (VIF) value for all variables is less than 10, in all research areas. Thus, it can be concluded that there is no multicollinearity between the independent variables.