Exploring public opinions on climate change policy in "Big Data Era"—A case study of the European Union Emission Trading System (EU-ETS) based on Twitter

Highlights

• This study analyses public opinions on the EU emission trading system.

• Public cognition of EU ETS is examined with social network and text topic analysis.

• The subjective assumption and sample size limitation of survey method was overcomed.

• Proposed suggestions for promoting public awareness on emission trading policies.

Abstract

Public awareness has an important effect on the legislation and implementation of climate change policies. Against the backdrop of the "Big Data Era," social media is an appealing and promising tool for a timely and complete understanding of public perception and attitudes towards climate policies. This paper examines the public's spontaneous attention and awareness about carbon emissions trading (ETS). Tweets related to the EU-ETS, published between 2008 and 2019, were collected for multi-dimensional analysis. Empirical results show several important findings. First, public attention on the EU-ETS has increased significantly since 2011. Second, government officials and industry practitioners have a stronger influence in the discussions than the public and industrial enterprises. Third, topic followers mostly gathered in Belgium (16.65%), the UK (11.6%), and some non-regulated countries like the US and Australia. Fourth, the public mainly focused on the policies and legislation, allowance price, and allocation. The innovation of this study rests in the development of a social media data-based research framework to examine the public's cognition of climate policies, which integrates the advantages of public social media, social network analysis, and text topic analysis. This study provides comprehensive analysis and support for climate policy implementation and public acceptance improvement.

Introduction

Under the existential and ubiquitous influence of climate change, countries all over the world have agreed to reduce their carbon emissions to address such environmental challenges (Field et al., 2014). The government's mandatory emission restrictions, pollution tax, and emission trading policies are the three main tools used by global societies to curb greenhouse gas (GHG) emissions (Li and Wei, 2021; Stuhlmacher et al., 2019). Amongst the available policies, market-based carbon emission trading scheme (ETS) is considered the most cost-effective and flexible option (Schmalensee and Stavins, 2017). According to the architects of the ETS, supervisory organisations have set limits on the emissions of polluting entities and have given each of these entities a certain number of emission permits (Wei et al., 2020). To comply with such regulations, these emitters have three options: the purchase of additional permits, the adjustment of production and outputs to ensure their emissions below the prescribed limits, and the sale of their excess permits (Egenhofer, 2007). Carbon emission permit trading is a combination per se of government intervention and market operation that encourages various emitters and market entities to participate in allocating carbon emission quota technological and capital resources to low-carbon development through flexible market mechanisms (Zhu et al., 2019).

The ETS mechanism is essentially an environmental policy designed to cope with climate change (Li et al., 2020). Generally, the formulation of a climate policy results from the interaction amongst multiple interest groups, including policymakers (i.e., governments and international organisations), advocacy groups representing specific industry interests, and civic activists (Guo and Saxton, 2014). Public sentiment fundamentally impacts the legalization and responsive implementation of a policy (Eun Kim and Urpelainen, 2018). Growing ex-post evidence shows that policies for addressing climate change require strong public participation and support (Bernauer and McGrath, 2016). An increase in public support positively influences public policies, particularly those that tackle prominent issues such as carbon emissions and climate change (Stroik et al., 2019), for several reasons. Firstly, climate mitigation measures significantly affect public life, production, and social welfare. Public engagement allows people to express their environmental interests, improves the transparency of policy governance, and promotes an in-depth understanding of the legitimacy of climate policies (Bernauer and Gampfer, 2013). Secondly, public support for climate policies contributes to climate cooperation strategies amongst countries (Tingley and Tomz, 2014). The efforts and contributions of different countries or regions to reduce GHG emissions can—through effective information dissemination and diffusion—stimulate support from more international communities and consequently drive environmental action (Victor, 2011; Wei et al., 2019). Therefore, the public awareness levels and participation in ETS are worth exploring.

Established in 2005, the EU carbon emission trading system is currently the largest and the only ETS operating across multiple countries and industries. This system covers all EU member states, Iceland, Liechtenstein, Norway, and accounts for 40% of all international trade in carbon emissions (Medina et al., 2014). As the earliest and most representative carbon market at the global and regional scales, the EU-ETS provides a flagship reference for the construction and development of ETSs in other parts of the world (Stuhlmacher et al., 2019). As the cornerstone of the EU climate policy portfolio, this trading system has demonstrated the effects of the gradual expansion of market regulation (Nava et al., 2018), the collapse in carbon prices due to the global economic crisis, and excessive quota supply (Eugenia Sanin et al., 2015) and certain institutional reforms (Mukherjee and Giest, 2019). However, with the public's lack of professional knowledge and participation, the importance of ETS for environment protection has often been ignored (Bernauer, 2013). Carbon pricing initiatives, which are considered to challenge longstanding practices and livelihoods of car-based and suburban lifestyles, may trigger strong political resistance among the broad public, such as the political disagreements in Canada, France, and Australia (Rosenbloom et al., 2020). The public's attitude towards the EU ETS is not always favorable because of disputes over the allowance allocation and price volatility (Genovese and Tvinnereim, 2019a).

ETSs have drawn growing concerns. Previous studies have explored a broad array of ETS-related topics, such as the status of major carbon markets (Zhu et al., 2019), the environmental, cost, and economic well-being of ETS (Haites, 2018), and assessment of ETS participants’ profitability (Makridou et al., 2019). However, few studies have examined public awareness of important climate policies, public opinions and attitudes towards carbon trading, and whether some interest groups are explicitly dedicated to carbon trading. This paper explores the online public attention and cognition about EU-ETS to provide concrete suggestions for the amelioration of the ETS policy.

Traditional methods, including online and face-to-face surveys, are often used to explore the perceptions and attitudes of the public towards environmental policies (Anderson and Bernauer, 2016; Kantenbacher et al., 2018; Stroik et al., 2019). With the rapid development of digital media, social media (including Twitter, Facebook, and YouTube) has gained extensive popularity among the public to share their views and experiences and participate in online discussions regarding climate- and environmental-related issues (Kirilenko and Stepchenkova, 2014). Twitter has become an effective information source of public opinions on environmental issues and climate policies for several reasons (Pearce et al., 2019). Firstly, Twitter users can independently publish and deliver diverse UGCs, including text, pictures, audio, and video. In this way, every user becomes a potential contributor and user of information, thereby greatly reducing the demands of users for traditional centralised authoritative news organisations (Cody et al., 2015). Secondly, the limited 140-character size of a tweet allows users to easily gauge public interest towards a topic, determine their attitudes, and understand the public narrative (Olofinlua, 2019). Thirdly, conversational replies, retweets, and relationships amongst Twitter users can reflect the social nature of information transmission (Russell et al., 2014). Fourthly, Twitter can gather real-time data on an unprecedented scale and dimension (i.e., time, location, user attributes) and reflect public awareness and response to social and environmental policies, facilitating discussions and spreading information (Jang and Hart, 2015). Therefore, taking advantage of new technologies developing an in-depth understanding of the public awareness of the carbon emissions trading scheme and discussing the effective diffusion of information and public opinion are of great value in promoting institutional changes and a low-carbon-living-related change in public behaviour.

Using text data collected from Twitter, this study performs social network analysis and natural language processing to analyse the public opinions and discussions regarding carbon emissions trading in the EU. Specifically, this study analyses the public attention towards this topic and the dynamic trends, build a network structure based on the tweeting and retweeting behaviors of users to identify those users who have the greatest influence on the discussion, determine the geographical distribution of the participants, and analyse the public concern on carbon trading.

This paper contributes to the existent literature in the following three aspects. Firstly, this study utilizes user-generated content (UGC) from web-based social media platforms to explore the public concern on ETS. Unlike traditional verification methods based on subjective assumptions, UGCs are widely recognized as useful data sources which can objectively reflect social media users' concerns and attitudes. Secondly, compared with traditional survey methods, this study, through an online real-time communication platform, collects extensive information about the social and demographic attributes of the research entities. Text analysis technology enables a large-scale data-driven exploration of public recognition's temporal, spatial, and content dimensions. Thirdly, examining the public's attention and recognition of climate policies can provide quantification-based references for the consummation of climate policies.

This paper is organised as below. The second part reviews the relevant literature. The third part introduces the research methodology. The fourth part presents the analysis. The fifth part summarises the findings.