Consumption and Biodiversity Conservation: Insights from Behavioral Science Using the MINDSPACE Approach
- 17 Downloads
Consumption is defined as purchase and use of all goods and services by individuals, households, and social groups.
MINDSPACE is a mnemonic that is used to set out nine robust influences on human behavior (messenger, incentives, norms, defaults, salience, priming, affect, commitment, ego).
Nudges are easy and cheap changes to some aspect of the choice architecture to alter human behavior in a predictable way without forbidding any options or significantly changing their economic incentives.
Human behavior is the key driver of all major threats to all life on land and biodiversity from habitat loss to climate change. Consumption of all goods and services requires the transformation of many natural resources via use as sources and sinks, which in turn impacts biodiversity. Individual’s lifestyle choices, in areas like food, energy, and water, can have indirect impacts by reducing emissions and demand-side pressures for wildlife and land. Individuals and communities can aid conservation efforts by directly giving time and money to conservation organizations and programs and reducing extraction of natural resources. But unsustainable consumption patterns can undermine the ability of ecosystems to provide services for industries and communities that rely upon them.
To tackle the underlying human drivers of biodiversity loss and implement solutions, conservation efforts must also focus on lifestyle choices and consumption habits. Academics and policymakers have increasingly recognized this need and attempted to incorporate behavior change into biodiversity conservation policies. For instance, ensuring Responsible Consumption is UN Sustainable Development Goal 12, along with Goal 7 on Sustainable Energy, Goal 11 on Sustainable cities and communities, and Goal 13 on Climate Action. The IPCC report also explicitly calls for demand-side solutions including strategies targeting technology choices, consumption, lifestyles, coupled production–consumption infrastructures, and systems to enable a climate transition (Creutzig et al. 2018). Changing behavior can entail taking up cleaner and more resource-efficient goods and services that minimize material footprint and pollution, and even participating in community-based natural resource management (CBNRM), ecosystem, and land management programs, which can benefit people and at the same time biodiversity.
Most biodiversity conservation approaches tend to rely on standard policy instruments to change human behavior. These typically include moral suasion via fact-based informational messages and appeals to raise awareness and reduce consumption, command, and control policies like mandates and bans on the consumption and regulation of wildlife products and market-based conservation instruments like payment for ecosystem services and carbon taxes. Yet evaluations of education campaigns focusing on facts and information-dissemination have been found to be relatively ineffective in producing behavior change (e.g., Schultz 2014). Bans can also backfire in particular contexts, for example, by unintentionally increasing poaching (Challender and MacMillan 2014). Price incentives can also have limited success; they can be difficult to introduce (e.g., carbon taxes) and then difficult to remove (e.g., subsidies) or even backfire in the long run (Bolderdijk and Steg 2015). These standard policy instruments tend to assume that people act in ways that reflects their best interests and therefore aim to change behavior by changing people’s minds. This has been variously referred to as the “cognitive” model (Dolan et al. 2012).
However, over the last two decades an important insight from the behavioral sciences, including behavioral economics and psychology, is that human behavior is significantly influenced by factors associated with the situation we find ourselves in (Thaler and Sunstein 2008; DellaVigna 2009; Steg et al. 2014). It recognizes that people sometimes make seemingly irrational and make inconsistent choices, go against what may be perceived to be in their best interest (even as acknowledged by themselves), and are influenced by automatic processes of judgment including emotions, biases, heuristics, and mental shortcuts, when responding to the context that we find ourselves in. In contrast with the cognitive model, behavioral science research shows that changing the decision-making context or the choice architecture can change human behavior, leading some to term this the “context” model of behavior.
Recognizing the automatic and irrational nature of human decision-making and the systematic role of contextual influences has led to the emergence of an increasingly popular set of behavioral public policy tools, of which the most well-known are nudges. Nudges can take the form of any easy and cheap changes to some aspect of the choice architecture to alter people’s behavior in a predictable way without forbidding any options or significantly changing their economic incentives or imposing mandates (Sunstein 2014). The implication is that in order to increase more responsible consumption for biodiversity conservation, we can change the context in which individuals and communities take decisions as well. There is much potential to leverage insights from behavioral science insights into enhancing the design of existing and new biodiversity conservation policies to accelerate and improve the effectiveness of existing approaches in cost-effective ways, while acknowledging that information campaigns, mandates and bans, and incentives can be crucial elements of the biodiversity conservation toolkit depending on the context.
This article summarizes insights from behavioral science pertaining to how contextual factors can increase responsible human consumption. These insights are organized using the ‘MINDSPACE’ framework, a simple mnenomic used to categorize nine behavioral effects that influence decision-making: messenger, incentives, norms, defaults, salience, priming, affect, commitment, and ego. This framework, set out by Dolan et al. (2012), is useful to understand how behavioral influences using the context model of behavior and has been applied to other agri-environmental contexts (Palm-Forster et al. 2019). The following section highlights recent research testing these effects in sectors linked to biodiversity conservation, primarily from wildlife, natural resource related domains especially waste, food, water, and energy consumption. Lessons and challenges are also discussed.
MINDSPACE to Change Human Consumption for Biodiversity Conservation
The weight we give to information depends greatly on the automatic reactions we have to the perceived authority of the source of that information – the “messenger” (Dolan et al. 2012). Rather than always listening to people (or the information source) by evaluating the content or accuracy of what they are saying, we tend to listen to those perceived to possess particular traits or attributes which signal that their messages are likely to be worth listening to. This implies that behavioral responses to informational interventions can be influenced by the messenger who delivers information. Martin and Marks (2019) outline two important characteristics of effective messengers: “soft” messengers are listened to because people feel connected to them (often emotionally) and “hard” messengers are listened to because they are perceived to possess some form of competence and status (often authority) over their audience.
Famous and culturally iconic charismatic animals, like pandas, elephants, and lions, are examples of soft nonhuman messengers that have been used extensively in biodiversity conservation charitable appeals and information campaigns. People tend to show a higher willingness to pay for natural park conservation and donations when information appeals feature charismatic megafauna, especially among those who less about the issue (Thomas-Walters and Raihani 2017; Shreedhar and Mourato 2019). Another example of soft messengers are mascots, which are used widely by sports teams, advertisers, and even environmental agencies, to reinforce a common group identity and belonging. For instance, Butler et al. (2020) found that participants in a lab experiment reduced pollution when a community mascot (an anthropomorphized wild animal) expresses negative emotions in response to poor water quality outcomes. Others have shown that anthropomorphizing both natural and manmade objects can increase charitable donations and pro-nature behavioral intentions (Ahn et al. 2014).
Hard messengers have been found to be important when communicating health risk information. For instance, expert interventionists produced greater behavior change than lay community members; but the demographic and behavioral similarity between the interventionist and the recipients are more effective at facilitating health behavioral change, especially among populations of low status and/or power (Durantini et al. 2006). This suggests that people, especially unempowered populations, are more likely to act on information when the hard messenger has similar characteristics and are more sensitive to characteristics of the interventionists. Other studies note that “similarity” and “liking” are important factors that can explain the success of the “block leader” approach, i.e., volunteers who help inform other people about a certain issue from the same social network, which have been important in local recycling and energy conservation drives (Abrahamse and Steg 2013). Evidence from climate change communication suggests reliance on hard messengers alone may be not sufficient. Studies have pointed out that “expert-led” in climate change communication can be less engaging for lay audiences than “non-expert” modes of communication, that use familiar, local, and culturally relevant icons instead (O’Neill and Hulme 2009).
Price incentives are central to economic and market-based interventions in biodiversity conservation contexts as they can impact the perceived benefits and the costs of taking an action. Dolan et al. (2012) offer useful insights about the role of incentives in changing behavior: (1) the value of a change appears from the status quo or reference point; (2) losses loom larger than gains (loss aversion); (3) small probabilities are over weighted; (4) money is allocated mentally to discrete accounts (mental accounting); and (5) choices consistently reflect living for today at the expense of tomorrow (present bias). Although economists have long studied the effects of incentives, they typically have ignored how the incentives are framed; how they are complemented by behavioral nudges; and how nonprice aspects of program structure and delivery can act as incentives to affect behavior.
Many benefits from biodiversity conservation programmers are mispriced (if at all valued in markets), collectively shared, and conservation efforts and technologies may not provide quick returns. Thus, incentive design can be made more effective, for instance, regarding the incentive size (how much), timing (when they are offered), and the targeting (to whom they are offered), by leveraging behavioral insights. In relation to point (1), Le Coent et al. (2017) found that farmers are willing to accept a higher payment to enroll when incentive-based contract which is framed as a biodiversity conservation rather than offset program, especially among those already adopting organic practices. The conservation framing presents the program goal as positive (it provides a gain from the reference point) rather than as a negative outcome as suggested by offset framing (it avoids a loss). In relation to point (5), studies have shown that being present biased is associated with late or low adoption of agricultural technologies and conservation measures (Duflo et al. 2011; Duquette et al. 2012). Farmers may procrastinate and postpone purchases until later periods, when they may be too impatient to purchase them even if they intended to in the past. Duflo et al. (2011) leverage the tendency to be present-biased into their program design by attempting to bring future benefits forward – they offer Kenyan farmers up front small, time-limited discounts on the cost of acquiring fertilizer, through free delivery, just after harvest.
An important unintended consequence of price-incentives is the “crowding-out” of local norms, values, and the “intrinsic” motivations to conserve biodiversity and to act prosocially more broadly. While much depends on the local context, crowding-out may be more likely with top-down schemes that offer small incentives and are poorly targeted (Akers and Yasué 2019). For instance, Chervier et al. (2017) found that a payment for ecosystems (PES) scheme negatively impacted on the perceived values of forest-based communities in Cambodia, from subsistence-related to money-related values, and that individuals emphasizing money-related values reported significantly more frequently that they would break conservation rules after an eventual end of payments, raising questions for the program effectiveness in the long term. Others find crowd-in may also be possible, i.e., increase in intrinsic motivation and values due to price incentives; this has been more likely when schemes deeply engage with informal community networks and leverage cultural values (Akers and Yasué 2019).
The effectiveness of incentive-based schemes can also depend on how people communicate with each other to set social norms out clearly. For example, the effectiveness of using fines to deter common pool resource extraction can depend on informal enforcement mechanisms, including communication between resource users and social norms (Balliet 2010). Whether incentive-based conservation programs are voluntarily accepted also hinges on the perceived participation of others in the community provisioning biodiversity as well. Chen et al. (2009), for instance, found conservation payments positively impacted participants intentions of maintaining forest on their Grain-to-Green Program (GTGP) land plots if the program ends, but negative effects if they perceived a lower share of their neighbors would participate.
In some situations, individuals take their cues from what others do; therefore, providing information on social norms can strongly influence a person’s behavior. In the focus theory of normative conduct, norms reflect the idea that sometimes group approval encourages people to engage in the “usual” behaviors (“injunctive norms”), and sometimes people engage in behavior because they believe that “most people do this” (whether that is objectively true or not; “descriptive norms”). This reflects the fact that people also have the ubiquitous tendency to engage in social comparison, i.e., we often look to others as appraisal standards for how to behave. Classic empirical investigations show how leveraging social norms can reduce littering in public spaces and stimulate more responsible lifestyle choices like reusing towels, but also many other pro-environmental intentions to use sustainable clothing and avoid pesticides (Farrow et al. 2017).
Utilities companies have been successful in leveraging social norms and social comparison in monthly home reports to reduce energy and water consumption, by telling people how much more they use more than their neighbors. Studies show that households are more likely to conserve energy when their utility bills contain both a descriptive normative message detailing average neighborhood usage and an injunctive message (conveying social approval or disapproval) (Allcott and Rogers 2014; Schultz et al. 2016). The inclusion of injunctive norms mitigates “backfire effects,” i.e., the tendency of households consuming less than the average to reduce their water consumption when hearing that they perform better. Moreover, follow-up studies show that effects can be persistent, even though they fall over time (Bernedo et al. 2014; Allcott and Rogers 2014). Most desirable behaviors can be counter-normative, for example, the meat-intensive diets in Europe and North America. Encouragingly, emerging evidence shows that people also confirm to dynamic norms, i.e., information about how other people’s behavior is changing over time. Highlighting that more people are starting to eat less meat led individuals to order more meatless entrees (Sparkman and Walton 2017).
Individuals regularly accept the default option, so changing the default to one that is more pro-conservation can help decide which of the available options to impose on individuals who fail to make a decision. Defaults or “status-quo” options are preset courses of action when no active choices are made. It can be useful when there is inertia or uncertainty in making decision. Defaults can provide a useful informational signal (people will believe that they have been given an implicit recommendation) and also establish the status quo or the reference point for counting changes as losses or instead as gains.
Changing the default in consumer choices is one of the most robust behavioral science insights. For instance, Ebeling and Lotz (2015) found that setting the default choice to more expensive “green” renewable energy (that is, where consumers have to actively opt out if they do not want it) increased purchases of such nearly tenfold in Germany. Egebark and Ekström (2016) cut paper use more than 6 months after the change, suggesting defaults influence behavior also in the longer run. However, the effects of defaults may differ by different groups and also by consumer behavior domain. For instance, Fowlie et al. (2017) found that most households only enroll in time-based pricing if assigned to the opt-out treatment, and these households reduced electricity use during higher priced peak periods. However, their reduction was lower on average compared to customers who actively opt in. Groups with strong antecedent preferences and experience may also show attenuated impacts; Löfgren et al. (2012) for instance, found that no significant effects of opt in or out default options among environmental economists at a large international conference.
Our behavior is greatly influenced by what our attention is drawn to and it can either be voluntarily controlled or it can be captured by some external event (Dolan et al. 2012). Since we are exposed to many different types of competing stimuli in our everyday lives, we may be more attentive to surprising, easily visible, simple, and comprehensible stimuli (e.g., items on sale next to checkouts, at the top of the menu, or at eye level in the aisle) (Thaler and Sunstein 2008). Eco-labeling is an example of how certain environmentally relevant characteristics (e.g., emissions or corporate sustainability practices) can be made salient and also convey social norms. Since consumers can underestimate emissions associated with food, researchers found that people purchased fewer high carbon food items when emissions were expressed using a colored rating scale ranging from “Lower” to “Higher Carbon Footprint” (green to red zone) (Camilleri et al. 2019).
People are also more likely to respond to personally relevant rather than abstract messages. This insight can be useful in designing information disclosure interventions. For instance, real-time, appliance-level energy data was feedback to university hall residents either only privately (along with the average usage in an US university residence halls) or also publicly disclosed on a poster visible to all residents, and the latter motivated a 20% reduction in electricity consumption achieved through reduced use of heating and cooling (Asensio and Delmas 2016; Delmas and Aragon-Correa 2016). In another set of experiments, households received feedback about their consumption in the metric of air pollution emissions. The environment information treatments motivated similar more energy savings, but were particularly effective on families with children. They also found that a health-based frame, in which households consider the human health effects of their marginal electricity use, induced persistent energy savings for longer than a cost savings frame (Asensio and Delmas 2016).
However, while it may be useful to increase the salience of the environmental consequences from different consumption choices, it works to change behavior only if this information is perceived to be credible and comes from a trustworthy source (e.g., a company with verifiable sustainability credentials backed by action). This is because there is extensive evidence and perceptions of companies engaging in “greenwashing” through the strategic labeling products to advertise companies’ eco-credentials among consumers. Another reason making eco-information salient may not work is because people can be actively counter-nudged or “sludged” if facing multiple competing advertisements and nudges to move towards less responsible options that are not in their best interest. The “Beyond Petroleum” advertisement campaign is a well-known example of this, and emerging evidence shows this pre-spill advertising significantly dampened the price response from the catastrophic oil spill, and may have reduced brand switching by BP stations (Barrage et al. 2020).
People behave differently if they have been “primed” by certain cues beforehand, seemingly outside of conscious awareness, through exposure to subconscious cues, like words, sights, and sounds which can activate knowledge in memory (Dolan et al. 2012). A commonly cited finding is that honesty primes, which ask respondents to give an answer that they would honestly pay (sometimes even going as far as having them sign oaths), affect an individual’s stated demand via willingness to pay for (or accept) an increase in environmental goods (bads) (Jacquemet et al. 2013).
Another useful insight is that while pro-environmental goals that people may have tend to be overruled by salient short-term goals and visceral impulses, they may be reactivated by exposure to relevant context-specific cues that can push behavior in line with the goal. For instance, priming environmental values increased hypothetical choices of environmentally friendly televisions, especially for those for whom environmental values were central to their idea of themselves (Verplanken and Holland 2002). More recently, Tate et al. (2014) found that exposure to environmental messages increased people’s choice of selecting loose rather than packaged products in a hypothetical choice task.
An important consideration is that much evidence from hypothetical choices and intentions may not translate into behavior in the field, especially since the effectiveness of the prime may be sensitive to the local context, as well as the personal cost of the action and the strength of prior preferences. Well-established results, like signing a veracity statement at the beginning instead of at the end of a tax or insurance audit form, have not been shown to replicate recently (Kristal et al. 2020). Similarly, others point to a “slippery slope effect,” where individuals more readily justify small indiscretions (as opposed to major ethical, moral disengagement) when unethical behavior develops gradually over time rather than abruptly (Welsh et al. 2015). Thus, more testing and validation across longer time frames and in revealed behaviors in natural settings will be necessary to verify when primes are effective in across different environmental context.
Affect is defined as any experience of feeling, such as positive to negative affect, mood, as well as specific emotions like sadness and joy. Affect can be a powerful force in decision-making, as emotional responses to other people, words, images, and events can be rapid and automatic often leading to behavioral reactions (Dolan et al. 2012). Research shows that integral affect, affect that is part of the perceiver’s internal representation of the option or target under consideration, has an influence on conservation-relevant behaviors (Västfjäll et al. 2016).
One of the most common spaces where integral affect has been used is to elicit charitable donations, since the “warm glow” or the positive emotional benefit that people get from giving is an important motivator of acting prosocially by helping others and acting pro-environmentally. Thus, charitable emotional appeals, especially those with attractive animals or those invoking empathy have been used extensively as previously noted. Pride is another important positive integral emotion. Rare, a conservation charity, leverages emotions of pride by using a variety of mascots, which are often native wildlife characters, in “Pride campaigns” to build a sense of collective identity associated with environmental protection in communities and recent research suggests that Rare’s mascots have been successful (Butler et al. 2020). Invoking empathy has been found to increase other types of cooperative behaviors too. Czap et al. (2015) finds that information about how upstream producers’ actions influenced downstream water quality using an empathy frame increased pollution abatement compared to a profit maximization frame in a lab experiment.
Similarly, others have found that pro-environmental consumption choices are positively related to also negative emotions like guilt for norm violation (Onwezen et al. 2013) as well as anger and moral outrage (Batson et al. 2007). However, a note of caution is that appeals to fear-inducing messages may not be effective, especially in the long run where people may feel the need to avoid bad news (O’Neill and Nicholson-Cole 2009). Apart from positive and negative integral emotions, incidental affect which is unrelated to a judgment or decision such as a mood has also been shown to impact consumer’s evaluative judgments and behavior (Västfjäll et al. 2016). But there is less systematic evidence about how it affects pro-environmental consumer preferences and behavior, with existing studies showing little effect on willingness to donate as well as pay for improved environmental quality (Hanley et al. 2017).
Although people may have good intentions and regularly set goals, like following low carbon diets or buying more environmentally friendly products, they also fail to follow through. We tend to procrastinate and delay taking decisions that are likely to be in our long-term interest due to self-control failures, by giving into temptations (such as a tendency to overspend or overeat) or automatically responding to contextual cues (big plate sizes, delicious smells, shop floor). In such cases, making commitment, either to oneself or to others, is commonly regarded as an effective tool to promote behavior change. Commitments can address self-control failures through imposing costs on the individual or by changing the choice architecture. More broadly, “commitment devices” are arrangements entered into by an individual with the aim of helping fulfill a plan for future behavior that would otherwise be difficult.
Bryan et al. (2010) distinguish between hard commitments that call for real economic penalties for failure or rewards for success and soft commitments for those devices that have primarily psychological consequences. Individuals may choose to bind their hands by privately pre-committing to particular choices food box vegetables and fruit subscriptions), which in itself might be a rational reflective action. For instance, studies find that consumers are less likely to buy “vice” foods like ice cream and more likely to buy fruits and vegetables when shopping for groceries online (Huyghe et al. 2017). Pledges to perform particular acts are common soft commitment device tested in research in relation to pro-environmental consumer choices like recycling and energy use. Contrary to perceptions that they are very effective in isolation, meta-analyses reveal that commitment effects can be much stronger and last longer when combined with other interventions like information feedback, incentives, or persuasive messages (Lokhorst et al. 2013).
Private commitments are also most effective if they are voluntarily entered into or if there is a prior intention to change behavior. This is because they will strengthen personal norms and be more aligned with one’s intentions. Intentions in turn sets, both the level of the goal or behavior (e.g., the number of days that the person intends to go vegetarian), and the person’s level of commitment (e.g., how determined they are). Indeed, there is much research that shows that people are more likely to engage in activity if someone elicits their “implementation intentions.” An implementation intention takes the form of if-then action plan that spells out the when, where, and how of goal striving in advance (“If situation Y is encountered, then I will initiate goal-directed pro-conservation behavior X!”) Initial evidence suggests implementation intentions have been effective in increasing recycling and self-reported fruits and veg consumption and energy savings (Holland et al. 2006; Adriaanse et al. 2011; Bell et al. 2016), but there is a need for more field-based evidence revealed rather than stated pro-conservation behaviors.
We tend to behave in a way that supports the impression of a positive and consistent self and social image (Dolan et al. 2012). This can be important motivations to undertake pro-environmental behaviors, since in many situations acting green is seen as the morally and socially desirable thing to do (Griskevicius et al. 2012). Importantly, people may be signaling their green virtues via social status signaling, only if the actions taken are sufficiently visible. Indeed, making the contributions of donors publicly visible through donor newsletters and funding circles, plaques, and other forms of public recognition is a common way for conservation and environmental organizations to both increase giving and express gratitude by appealing to social and self-image motivations (Ariely et al. 2009). Social status signaling can take the form of “conspicuous conservation” (Sexton and Sexton 2014). A classic example is the purchase and use of Toyota’s hybrid car, the Prius, and stickers that tell others about the fuel efficiency of one’s own car (Thaler and Sunstein 2008; Sexton and Sexton 2014).
People also like to stay consistent with their past behavior since it presents a more coherent self-image and mitigates cognitive dissonance. This suggests that there may be behavioral spillovers, i.e., when people engage in a first pro-environmental behavior (e.g., conserving energy at home), they can be more or less likely (positive and negative spillover, respectively) to engage in other pro-environmental behaviors (e.g., conserving water at home). Maki et al. (2019) undertake a meta-analysis of pro-environmental spillover studies and find that positive spillovers are most likely when considering self-reported behaviors and behavioral intentions, especially from interventions targeting intrinsic motivations and when behaviors were similar. However, they also find that the spillover effect was negative and small for actual behavior and policy support for the second behavior.
We may expect negative behavioral spillovers because people may also have the tendency to morally license themselves, i.e., when people initially behave in a moral way, they are later more likely to display behaviors that are immoral, unethical, or otherwise problematic. For instance, Mazar and Zhong (2010) found, in line with the halo effect associated with green consumerism, that people act more altruistically after mere exposure to green products than after mere exposure to conventional products. However, people act less altruistically and are more likely to cheat and steal after purchasing green products than after purchasing conventional products, as they feel morally licensed to do so. More broadly, this tendency highlights that individual’s choices in one context (e.g., travel to work) can impact their subsequent choices in other consumer domains and places (e.g., eating at home) (Nash et al. 2017).
In summary, biodiversity conservation requires moving human consumption and lifestyle choices toward more sustainable and responsible alternatives. While information campaigns, mandates and bans, and price incentives are important types of biodiversity conservation policies, there is much potential to leverage insights from behavioral science insights into enhancing the design of existing and new policies in order to accelerate and improve the effectiveness of existing approaches in cost-effective ways. A growing body of evidence shows that changing the decision-making context or “choice architecture” can open up new possibilities for conservation by the use of “nudges” to change human behavior.
The MINDSPACE framework is a helpful mnemonic which categorizes nine behavioral effects that influences consumption in sectors like food, waste, and natural resource conservation, namely messenger, incentives, norms, defaults, salience, priming, affect, commitment, and ego. It can be used to aid the design of conservation policies and projects. That said, there is a need for much more research in how to foster responsible consumption for biodiversity consumption, whether there are unintended behavioral spillovers across behaviors and contexts, short- versus long-run effects of behavioral and standard inventions, and how behavioral change this can be linked to measurable environmental outcomes to evaluate progress. Indeed, there is much potential to incorporate and test the impact of nudges and behavioral insights in existing policies in order to meet, protect, and restore life on land and promote sustainable use of natural resources and ecosystems, and mitigate biodiversity loss.
- Bolderdijk JW, Steg L (2015) Promoting sustainable consumption: the risks of using financial incentives. In: Thøgersen J, Reisch L (eds) Handbook of research on sustainable consumption. Edward Elgar Publishing, Cheltenham, pp 328–341Google Scholar
- Bryan G, Karlan D, Nelson S (2010) Commitment devices. Annu Rev Econ 2:671–698. https://doi.org/10.1146/annurev.economics.102308.124324CrossRefGoogle Scholar
- Creutzig F, Roy J, Lamb WF, Azevedo IML, Bruine de Bruin W, Dalkmann H, Edelenbosch OY, Geels FW, Grubler A, Hepburn C, Hertwich EG, Khosla R, Mattauch L, Minx JC, Ramakrishnan A, Rao ND, Steinberger JK, Tavoni M, Ürge-Vorsatz D, Weber EU (2018) Towards demand-side solutions for mitigating climate change. Nat Clim Chang 8:260–263. https://doi.org/10.1038/s41558-018-0121-1CrossRefGoogle Scholar
- Durantini MR, Albarracin D, Mitchell AL, Earl AN, Gillette JC (2006) Conceptualizing the influence of social agents of behavior change: a meta-analysis of the effectiveness of HIV-prevention interventionists for different groups. Psychol Bull 132:212–248. https://doi.org/10.1037/0033-2909.132.2.212CrossRefGoogle Scholar
- Fowlie M, Wolfram C, Spurlock CA, Todd A, Baylis P, Cappers P (2017) Default effects and follow-on behavior: evidence from an electricity pricing program. NBER working paper 23553. https://doi.org/10.3386/w23553
- Martin S, Marks J (2019) Messengers: who we listen to, who we don’t, and why. Random House, UKGoogle Scholar
- Thaler RH, Sunstein CR (2008) Nudge: improving decisions about health, wealth, and happiness. Yale University Press, New Haven/LondonGoogle Scholar