Scholarly article on topic 'Uptake and usage of cost-reflective electricity pricing: Insights from psychology and behavioural economics'

Uptake and usage of cost-reflective electricity pricing: Insights from psychology and behavioural economics Academic research paper on "Educational sciences"

Share paper
OECD Field of science
{"Cost-reflective pricing" / "Electricity tariffs" / "Behavioural economics" / "Consumer behaviour" / "Electricity consumption" / "Energy demand"}

Abstract of research paper on Educational sciences, author of scientific article — Elizabeth V. Hobman, Elisha R. Frederiks, Karen Stenner, Sarah Meikle

Abstract The Australian electricity industry – like many other countries globally – is currently facing the complex challenge of reforming electricity tariffs. Momentum is growing for transitioning residential consumers toward more ‘cost-reflective’ pricing that better reflects the true costs of generation and supply, and sends a ‘price signal’ that presumably incentivises reduced consumption during peak periods. Under such tariffs, customers pay more for electricity used during times of peak demand – unlike traditional ‘flat-rate’ tariffs where the price remains stable regardless of time or demand. Pilot trials indicate that cost-reflective tariffs might succeed in reducing peak demand, but often only for a small minority of customers, such that population-wide demand response is minimal or insignificant. In this paper, we apply insights from psychology and behavioural economics to identify how cost-reflective pricing can be designed, depicted and delivered to enhance customer uptake and optimal usage –thereby facilitating ‘appropriate’ demand response – for a larger cross-section of the population. By carefully considering the likely impact of relevant cognitive biases and psychological factors –which routinely shape human decision making and behaviour – we are able to propose practical strategies that industry can adopt to maximise the prospect of cost-reflective pricing achieving significant population-level peak demand reductions, while providing shared benefits for customers, retailers, networks and regulators alike.

Academic research paper on topic "Uptake and usage of cost-reflective electricity pricing: Insights from psychology and behavioural economics"

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

journal homepage:

Uptake and usage of cost-reflective electricity pricing: Insights from psychology and behavioural economics

Elizabeth V. Hobman*, Elisha R. Frederiks1, Karen Stenner2, Sarah Meikle3

CSIRO Adaptive Social and Economic Sciences, Ecosciences Precinct, 41 Boggo Road, Dutton Park Qld 4102 Australia



Article history: Received 4 August 2014 Received in revised form 27 October 2015 Accepted 17 December 2015 Available online 5 January 2016


Cost-reflective pricing Electricity tariffs Behavioural economics Consumer behaviour Electricity consumption Energy demand


The Australian electricity industry - like many other countries globally - is currently facing the complex challenge of reforming electricity tariffs. Momentum is growing for transitioning residential consumers toward more 'cost-reflective' pricing that better reflects the true costs of generation and supply, and sends a 'price signal' that presumably incentivises reduced consumption during peak periods. Under such tariffs, customers pay more for electricity used during times of peak demand - unlike traditional 'flatrate' tariffs where the price remains stable regardless of time or demand. Pilot trials indicate that cost-reflective tariffs might succeed in reducing peak demand, but often only for a small minority of customers, such that population-wide demand response is minimal or insignificant. In this paper, we apply insights from psychology and behavioural economics to identify how cost-reflective pricing can be designed, depicted and delivered to enhance customer uptake and optimal usage -thereby facilitating 'appropriate' demand response - for a larger cross-section of the population. By carefully considering the likely impact of relevant cognitive biases and psychological factors -which routinely shape human decision making and behaviour - we are able to propose practical strategies that industry can adopt to maximise the prospect of cost-reflective pricing achieving significant population-level peak demand reductions, while providing shared benefits for customers, retailers, networks and regulators alike.

© 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND

license (


1. Introduction ........................................................................................................ 456

2. Prior research and evidence: rates of customer uptake and usage.............................................................456

3. The current review: adopting a behavioural approach......................................................................458

4. Human decision making and behavioural biases...........................................................................458

4.1. Simplicity: an essential, overarching principle.......................................................................459

4.2. Trust as a decision heuristic: first establish credibility and create positive customer relationships ............................. 460

4.3. Status quo bias: present new tariff offers in ways that encourage active consideration by customers...........................460

4.4. Loss aversion: reduce all costs associated with the shift to dynamic pricing............................................... 461

4.5. Risk aversion: provide assurances that customers do not risk higher electricity bills under cost-reflective pricing ................ 462

4.6. Temporal and spatial discounting: Reduce immediate costs and increase the salience of immediate benefits from cost-reflective pricing 463

4.7. Normative social influence: describe how other customers have experienced cost-reflective pricing ........................... 463

4.8. Perceived fairness: explain inequity in flat-rate pricing, and how cost-reflective pricing restores fairness ....................... 464

5. Conclusions and recommendations ..................................................................................... 465

References ............................................................................................................. 465

* Corresponding author. Tel.: + 61 7 3833 5744; fax: + 61 7 3833 5504 E-mail addresses: (E.V. Hobman), (E.R. Frederiks), (K. Stenner), (S. Meikle).

1 Tel.: +61 7 3833 5753; fax: + 61 7 3833 5504

2 Tel.: +61 7 3833 5745; fax: + 61 7 3833 5504

3 Tel.: +61 7 3327 4182; fax: +61 7 3327 4455

1364-0321/© 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (

1. Introduction

Overinvestment in network infrastructure, falling electricity consumption and (up until recent years) growing peak demand are contributing to a challenging problem of network inefficiency for Australian energy distribution and network service providers (DNSPs). One of the significant factors thought to perpetuate this inefficiency problem is an imbalance between the financial costs of producing and supplying electricity, and the price actually paid by customers. Regardless of demand on the electricity grid, many residential customers are still on traditional 'flat-rate' tariffs where the price (per kilowatt hour of electricity used) remains stable over time - essentially insulating them from moment-by-moment fluctuations in wholesale prices on the electricity market. One potential solution to this disparity is to simply price electricity in closer accordance with the actual costs incurred by DNSPs. By doing so, customers are presented with a 'price signal' that more accurately conveys the true costs of electricity generation and supply and (at least in theory) incentivises them to reduce or shift electricity usage to different (i.e., off-peak) times, thereby flattening peak demand. This pricing arrangement is commonly termed cost-reflective pricing or dynamic pricing4.

The potential benefits of cost-reflective pricing for reducing the adverse impacts of peak demand have been promulgated for decades [2-5]. Numerous pilot trials and field experiments have appeared to suggest that pricing electricity dynamically (rather than statically) could yield a range of payoffs, including substantial improvements in network efficiency, reduced infrastructure costs, and lower average market prices [6-8]. Cost-reflective tariffs also are seen as a means of ensuring greater social equity in the mass market, by reducing the largely invisible cross-subsidies embodied in flat-rate tariffs. On the face of it then, it would seem that tariff reform has the potential to yield significant benefits for a range of stakeholders. However, we are at pains to stress that this potential is heavily contingent on the optimal uptake and usage of such tariffs by a large proportion of the population. Indeed, some critics have questioned whether a positive case even exists for the large-scale introduction of cost-reflective pricing, with doubts remaining over customer acceptance, adverse impacts on vulnerable groups (e.g., low-income households), and its overall efficacy in producing the purported benefits [9,10]. We argue that, for cost-reflective pricing to yield the desired outcomes for utilities and end-users alike, there are at least two critically important and inextricably linked requirements: first, there must be sizeable and widespread uptake among customers; and second, there must be optimal usage (i.e., appropriate demand response) that is sustained over time. Identifying practical strategies to achieve both of these requirements is the core focus of our paper.

The design of strategies to enhance uptake and optimal usage of cost-reflective pricing naturally involves a deep understanding of consumer behaviour and the underlying psychology of human decision making. There are numerous theories of consumer behaviour and decision making available to guide our thinking and practice in this regard, the best of which endeavour to integrate insights from multiple disciplines, including psychology, sociology, marketing and economics5. Yet in

4 Several different types of cost-reflective pricing have been cited in the literature, ranging from the simplest form of time-of-use (TOU), to more dynamic forms of critical peak pricing (CPP), peak time rebate (PTR), variable peak pricing (VPP) rates and real-time pricing (RTP) [for a recent overview, see [1]].

5 Most early approaches, such as classical and neoclassical theories of human

behaviour, drew on rational choice models [for overviews, see [11 -14]] that assume that consumers engage in economically rational information-processing and

practice, industry stakeholders often pay insufficient attention to (even entirely overlook) these 'behavioural insights', and instead tend to pursue tariff reform guided by little other than pure economic modelling. We propose that non-economic factors are likely to also be influential - especially in this complex and unfamiliar decision making scenario of choosing among alternative electricity pricing offers. By considering a more complete set of factors with the potential to shape customer choices and actions - particularly psychological and motivational influences - stakeholders would be better equipped to understand, influence and predict how customers will react to, and utilise new pricing schemes.

2. Prior research and evidence: rates of customer uptake and usage

Empirical evidence from real-world pricing trials shows that the outcomes of cost-reflective pricing do not always meet expectations. For example, if we confine our assessment only to empirical results from scientifically robust research (e.g., randomised controlled trials, economic modelling of historical data), it becomes clear that voluntary uptake rates are often low [29] and in some cases only a small minority of customers exhibit demand responsiveness [see 30,31]. Ultimately, these findings imply that the benefits observed in small-scale (often highly resource-intensive) trials might not be obtained in large-scale roll-outs to the broader population, because the latter invariably involves diverse customer segments - each of which has unique needs, wants, interests and motivations, and faces varying constraints and supports when it comes to using electricity. A convincing argument for the widespread introduction of cost-reflective tariffs needs to offer something of value to the majority, if not totality of customers. To this end, our paper endeavours to provide a suite of practical strategies that policymakers and industry stakeholders could employ to ensure that tariff reform is undertaken in a way that minimises social, economic and political risks across the board.

In terms of what we currently know about the initial uptake of cost-reflective pricing, while there is surprisingly little data available6, a recent review of trials conducted by nine U.S. utilities indicates that voluntary recruitment rates for opt-in trials vary widely across the lower end of the scale (5 to 28% for opt-in studies, with an unweighted average rate of 14%) and are often significantly less than expected (about 7% to 22% less than utilities' predictions) [29]. Findings from recent large-scale experimental work conducted in Australia [32] and the U.K. [33,34] also tend to confirm that the voluntary uptake of cost-reflective tariffs is likely

(footnote continued)

decision-making [15,16]. However, extensive research shows that consumer choices and actions often deviate systematically from such assumptions [17-19]. There are various psychological anomalies and biases in human decision-making that routinely lead consumers to behave in ways that traditional economic models cannot explain, and this extends to the domain of residential electricity usage [2022]. More contemporary perspectives are increasingly integrative, taking into account a range of socio-demographic, psychological and situational factors [for a review, see 23]. An exhaustive list of all relevant theories and models of energy consumption behaviour is beyond the scope of this paper; however, some noteworthy approaches include Van Raaij and Verhallens [24] behavioural model of residential energy use; Costanzo et al. [25] socio-psychological model of energy conservation behaviour; Stern and Oskamps [26] causal model of resource use, and more recent approaches put forward by Abrahamse et al. [27] and Stern [28].

6 Very little information is available on uptake rates and the factors that might influence uptake, because utilities are not actually required to report such data [For more information, see [29]].

Table 1

Summary of decision-making principles and behavioural biases associated with cost-reflective pricing.



Behavioural economics-inspired solutions

Key sources/ references

Aversion to complexity/ Cognitive and choice overload

Trust as a decision heuristic

Status quo bias

Loss aversion

Risk aversion

Tendency to adopt simpler decision rules (and therefore potentially make worse decisions/choices) as the information and stimuli in one's environment becomes more complex.

Tendency to use perceptions of trust as a decisionmaking 'heuristic' - i.e., a mental shortcut or rule-of-thumb to speed up information processing, problemsolving and decision-making in complex and cogni-tively demanding situations.

Tendency to resist change and instead favour the status quo or 'default' setting, which oftentimes means not acting (i.e., inertia) or avoiding making a decision altogether.

Tendency to focus more heavily on losses than on comparable gains, and to exert far greater effort in order to avoid a loss as compared to an equivalent-sized gain.

Time inconsistency and spatial and temporal discounting

Tendency to prefer certainty over risk, especially when the stakes are high - i.e., people are more willing to choose a certain or guaranteed gain as compared to gamble for an uncertain pay-out. While risk-taking behaviour is more likely in the context of losses, when the stakes are high, again, people are far less likely to take a gamble and will instead prefer a certain or guaranteed loss.

Tendency to be short-sighted on nearby or immediate costs/benefits and farsighted on costs/benefits that are further away or in the future - i.e., people 'discount the future' and prefer smaller immediate rewards (e.g., $5 now) rather than larger future rewards (e.g., $100 next year).

Ensure simplicity ('keep it short and simple') in all cus- [45-49] tomer communication by avoiding unnecessary complexity (i.e., present smaller amounts of the most important information in a clear, concise and understandable format).

Simplify the registration process; design tariffs that are more structured and less dynamic; and describe pricing offers using straightforward, jargon-free language. Avoid inundating customers with too many different choices. Instead, reduce the number of pricing offers to a handful (i.e., fewer choices rather than many) and present them side-by-side to facilitate relative comparisons. Ensure messages stem from a source that customers [50-55] perceive as credible, trustworthy, competent, accountable, and genuinely acting in good faith. Build customer trust in the specific entity (network provider, energy utility/retailer) that is seeking to introduce cost-reflective pricing.

Provide services that assist customers in managing their electricity consumption, such as text/email alerts when they may be approaching a certain percentage of their previous bill value, and simplifying the content and presentation of bill statements.

Provide a 'recommended' or 'popular' pricing offer that is [56-62] framed as the default.

Present traditional flat-rate tariff in a neutral way by presenting and depicting it in the same way as all other options. It is less likely that customers will naturally default to it.

Capitalise on salient lifecycle 'trigger points' (e.g., moving homes, contract end) when people may be more amenable to changing electricity plans. For example, customers may be more willing to switch to cost-reflective pricing when moving homes and/or changing electricity providers; when they have received a higher than normal electricity bill and/or an electricity price rise has recently been announced.

Emphasize the costs/losses incurred by sticking with the [18,60,63,64] status quo (flat-rate pricing), and how such losses could be avoided by switching to cost-reflective pricing. Reduce the perceived and actual 'costs' (e.g., loss of time, effort, money, functionality, comfort, etc.) of switching pricing plans by simplifying the registration/application process, providing automated technology free of charge. To maximise the salience of a cost/loss, present it in isolation by itself (e.g., "You are currently losing $50 per quarter by staying on flat-rate pricing") rather than integrating or embedding it within a larger amount. Design tariffs where the off-peak reduction in price (i.e., potential benefit/gain) is greater than the peak increment in price (i.e., potential cost/loss).

Design 'risk-free' tariffs that provide an incentive (e.g., bill [64-69] rebate) for reduced demand, rather than relying solely on disincentives (e.g., penalty charges) to encourage the same behaviour.

Offer an obligation and risk-free 'introductory offer' that includes safeguards against higher bills for an initial period of time - e.g., offer customers money-back guarantees or free bill protection insurance for a trial period. Offer an obligation- and risk-free 'try before you buy' period, so customers can trial cost-reflective pricing without incurring any costs.

Repackage the 'costs' of cost-reflective pricing so that the [70-74] benefits are gained upfront. For example, make the immediate benefits (i.e., potential cost-savings) more salient by communicating dollar savings in, or as close to, real-time.

Provide an initial period of 'immediate feedback' where new customers receive real-time or next-day feedback on the actual cost savings attained under different tariffs (e.g., a side-by-side graph comparing bill costs under flatrate vs. cost-reflective pricing).

Minimise the time, effort, hassle and inconvenience of switching pricing plans.

Table 1 (continued)

Barrier Description Behavioural economics-inspired solutions Key sources/


• Provide hints and tips, as well as automated technology to assist the customer in changing their energy-consuming practices, especially long-standing habits.

• Emphasize the non-monetary benefits of cost-reflective pricing - e.g., improved supply/reliability of electricity, reductions in peak demand, network optimization, etc.

Normative social influence Tendency to follow the behaviour of others, make • Convey messages that encourage positive social norms [75-85] social comparisons, and adhere to social norms - i.e. (i.e., provide examples of other households adopting the people are heavily influenced by how others think, desirable behaviour) rather than messages that encourage feel and act; they often care about performance, negative social norms (i.e., don't use examples of other

possessions and wellbeing relative to other people, households engaging in undesirable behaviour) rather than in absolute terms. • Where possible, appropriate and genuine (i.e., ethically

and professionally sound), frame cost-reflective pricing as something that is common, valued, accepted and desired among one's local peer network/group, community or society.

• Present a realistic image of the types of people who are likely to benefit from cost-reflective pricing.

Perceived fairness /inequality Tendency to be averse to inequalities and unjust • Provide hard evidence on the social equity benefits of [86-88] outcomes, and to seek fairness in one's transactions. cost-reflective pricing.

• Ensure vulnerable groups (e.g., working families on low-income) are not unfairly disadvantaged by providing extra supports - e.g. maintain the standard two-part tariff and/ or design a low-capacity flat-rate tariff that is extremely cheap, yet still sufficient to maintain basic living standards; provide government concessions and assistance programmes.

to be relatively low across the population7. Related research examining whether customer participation depends on favourable load patterns also seems to indicate that those customers who self-select cost-reflective pricing are a unique subset of the population. For example, some research indicates that willing customers tend already to possess favourable demand profiles (i.e., they already consume less electricity during peak times) [35-39] and/or demand flexibility (i.e., they have the flexibility to modify demand by using energy management systems or alternative sources of electricity) [40]. However, some contrary evidence also exists suggesting, alternately, that customers who willingly choose cost-reflective pricing do not tend to have different patterns of consumption [40-42].

In terms of what we know about subsequent usage of cost-reflective tariffs, recent empirical evidence (from two opt-out trials with valid control group comparisons) indicates that demand response is typically confined to a small sub-sample of the population [30,31]. The modelling of historical data to determine the price elasticity of demand for electricity also indicates that the majority of the population are relatively price inelastic (i.e., their electricity consumption is largely unresponsive to price changes), with only a small proportion substantially price elastic [43]. The fact that household electricity demand, in the aggregate, appears to be relatively price inelastic is unsurprising, given that the magnitude of elasticity is known to have fallen across the last three decades [44]. This trend suggests that, overall, householders have already made the bulk of possible reductions and substitutions in electricity consumption (whether by behavioural means or via the adoption of energy-efficient technology/appliances).

7 The first survey of a nationally representative sample of people across the U. K. found that 25% to 30% of people said that they were either strongly or moder-

ately in favour of switching to a cost-reflective tariff (either dynamic or static TOU, respectively) if it was offered. The second experimental survey of a large sample of Australian households found that likely rates of accepting a hypothetical cost-reflective tariff offer ranged from ~ 37% (for RTP) to -53-54% (for TOU, PTR and CPP), or just 2.5% to 3.7% when adjusted for predicted response rate, in the absence of additional risk relief mechanisms. It must be emphasised, however, that these estimates are based on self-reported data only and do not reflect actual uptake rates.

3. The current review: adopting a behavioural approach

On balance then, the best empirical evidence currently available suggests that initial uptake tends to be low, and subsequent demand response tends to be confined to a small and unique segment of customers. Accordingly, we anticipate that from a whole-of-population point of view, the benefits of cost-reflective pricing may often fall short of expectations. For new pricing schemes to successfully achieve the desired peak demand reductions across the population, much more work needs to be undertaken. Our paper takes an important step forward in this regard, by introducing a different way of viewing the problem of cost-reflective pricing. We draw on evidence-based insights from the behavioural sciences that illustrate how consumers actually make decisions and behavioural choices, and integrate these insights with the limited empirical evidence available - confining ourselves only to scientifically robust research in the field - to examine how the uptake and usage of cost-reflective pricing might be improved.

Specifically, we (1) discuss a set of powerful cognitive biases and psychological anomalies that are likely to influence customer responses to electricity pricing, and (2) propose practical and cost-effective solutions that address and/or leverage these influences to yield optimal outcomes for both customers and the energy industry. We will demonstrate that by understanding the relevant behavioural forces and how they interact to shape consumer choices and actions, we open up opportunities to effectively capitalise on these forces to achieve outcomes beneficial to all stakeholders, consumers and industry alike. Table 1 summarises the key psychological forces relevant to cost-reflective pricing, alongside some behavioural economics-inspired solutions for addressing and/or leveraging these forces.

4. Human decision making and behavioural biases

In terms of strategies that might be effective for encouraging uptake and optimal usage of cost-reflective pricing, it is often

assumed that people simply need to be given more information and choice, and to become more actively involved and 'engaged' with the issue. However, this commonly-held assumption (that greater information, knowledge and awareness will induce behaviour change) is inconsistent with several fundamental, well-established principles of human decision making and behaviour. These principles explain how people generally make behavioural choices under real-world conditions (vs. the assumptions of classic economic models), and more specifically, how people tend to make decisions and act in ways that are economically 'irrational' (from the perspective of maximising personal gain) [19,89,90]. These 'irrationalities' -known as 'cognitive and decision-making biases' -have been the subject of intensive scholarly inquiry for decades, with research confirming their existence across a range of behavioural domains [91-93], including electricity usage [2022,94]. Many of these biases are simple mental 'short-cuts' or 'heuristics' that alleviate the need for intensive cognitive processing, thereby hastening the speed of decision making [95-98]. In a world where information, choices and opportunities abound and overwhelm, people come to rely on these short-cuts to guide their decision making. It simply saves time, energy and thought, even though it often leads to (economically) sub-optimal outcomes for the individual.

4.1. Simplicity: an essential, overarching principle

Before discussing the various biases that might be implicated in consumers' responses to cost-reflective pricing, it is essential to understand the overarching principle of simplicity that, when properly attended to (for example, in programme design), can in itself greatly reduce the expression of many common behavioural biases. As intimated above, there are inescapable limits to human capacity to process information, such that people tend to rely -especially in environments of high complexity, uncertainty and risk -on simple heuristics or other decision-making short-cuts [95,97,98]. For example, when the amount or complexity of information overwhelms, people often 'satisfice' rather than 'optimize' [15,16], i.e., they process only enough information to reach a satisfactory decision rather than exhaustively weighing all available information to achieve the optimal outcome. Further, as the amount and/or complexity of information increases, people tend to adopt increasingly simple rules-of-thumb, such as sticking to the status quo [60,61], choosing the default option [48,56], or avoiding decision making altogether [99]. Overall, there is an abundance of studies confirming that people tend to reach worse decisions when given more information and/or greater choice, and conversely, are better served by 'keeping things simple'.

Thus, our overarching premise in any sensible strategy to enhance customer decision making around cost-reflective pricing is that customer behaviour will be heavily determined by the simplicity of incoming information. In terms of cost-reflective tariffs, new pricing schemes - as well as their enrolment processes and the various ways in which the scheme is supported -should ideally be designed, depicted and delivered in an exceedingly simple manner. If overloaded with too much information, too many choices and/or too much complexity, customers are likely to find it difficult to select among options and instead, will tend to satisfice and rely on heuristics (e.g., stick with the default option) to guide their decision making. Thus, the number of pricing options should be reduced to just a few, and presented in a rudimentary fashion side-by-side to facilitate decision making8.

8 Research suggests that decision making is improved when individuals are presented with multiple options simultaneously, rather than separately [100-104]. A lack of comparison information in the latter scenario is thought to induce greater reliance on decision making heuristics, which then leads to sub-optimal decisions.

Additionally, if the pricing structure itself is too complicated, customers are not only unlikely to choose it in the first place, but may also find it difficult to utilise effectively on a daily basis. They may struggle to keep track of the changing schedule of fees in order to know precisely when (and for how long) to reduce demand9. Additional strategies such as prompts (e.g., text or email reminders; energy orbs that glow green when energy use/prices are modest and pulse red when energy use/prices are high [106]) and automated demand management technology are likely to be needed to effectively support these types of schemes.

Simpler tariff configurations are bound to be more manageable in everyday life10. Customers need only change a few electricity-intensive practices, which may eventually create new habits and merge into a new routine, further reducing effort. Both qualitative and quantitative research indicate that customers often find highly complex, variable pricing schemes less appealing than simpler, more static ones [32-34,107,108]. For example, an experimental online survey in the U.K. revealed greater consumer willingness to switch to static time-of-use pricing than dynamic pricing (where prices fluctuate based on the predicted demand and supply on the grid), and greatest willingness to switch to a direct load control tariff (where in return for a lower flat rate tariff, the consumer allows the electricity supplier to cycle their heating system off/on at certain times) -which essentially does not require any behavioural response at all [33,34]. Similarly in Australia, an experimental mail-out survey of different (randomly assigned, hypothetical) tariff offers found that self-reported consumer acceptance was greater for more simple, structured and straightforward pricing plans (such as time-of-use, critical peak pricing and peak time rebate), and lower for those that were more complex, highly dynamic and/or unfamiliar (such as real-time or capacity-based pricing) [32]. According to the researchers, these relative tariff preferences can be understood as "roughly reflecting public perceptions both of how difficult a proposed pricing structure is to comprehend, and how hard it might be for households to behave in ways that would maximise its benefits" (p. 8).

While it is essential to always attend to this core principle of simplicity, there are a number of additional strategies that can either ameliorate or exploit those cognitive biases and other psychological anomalies that tend to surface in all complex decision making such as this. In the remainder of this paper, we explain each of these potential influences in turn, specifically highlighting how they might be leveraged to maximise both the uptake and effective usage of cost-reflective pricing.

9 This notion of managing the schedule of charges has also been reflected in some qualitative research, which found that consumers typically do not fully familiarise themselves with the precise implications of a tariff, but rather only note that they are financially better off using electricity during off-peak periods [105]. This suggests that consumers may rely on a rough time-based 'rule of thumb' (rather than conducting a fully 'rational' calculation of the price differential) when endeavouring to understand and respond appropriately to a tariff's 'price signal'.

10 Striving for simplicity may not seem feasible for highly dynamic tariffs such

as real-time pricing. Such pricing is inherently complex and, frankly, not usually designed with customer needs at the forefront. Yet even here, we think it is pos-

sible to design relatively simple, user-friendly cost-reflective pricing packages that can achieve mutual benefits for both utilities and end-users (e.g., imagine a cost-reflective tariff combined with a complementary automated device that can manage demand for the consumer, along with some simple visual aid that conveys information on shifting prices without requiring much cognitive effort). Consider also that by offering simple solutions that better satisfy customers' needs, utilities might enjoy other indirect payoffs - greater trust and loyalty, improved reputation and credibility, perhaps enhanced status as the 'provider of choice' - all of which might presumably lead to greater customer retention and expansion of market share.

4.2. Trust as a decision heuristic: first establish credibility and create positive customer relationships

One overarching decision-making heuristic that is commonly used to curtail cognitive processing - especially in the face of high uncertainty, risk and complexity - is the perception of trust [5052,54,55]. In these kinds of situations, consumers often defer to what their feelings 'say' regarding whether an entity (e.g., another person, organisation, etc.) is trustworthy, as an indicator of whether the benefits of a particular interaction/service/product are likely to exceed the costs. In essence, perceptions of trust and credibility become a simple barometer of 'risk'. Given the strong and pervasive influence of trust, we recommend that utilities make genuine and substantial efforts to build customer trust and confidence well before introducing cost-reflective pricing. This endeavour is not expected to be easy, given that consumers around the globe tend to share the conviction that electricity utilities are more interested in revenue and profits than customer welfare [109,110]. Nonetheless, it is absolutely essential that utilities take up this challenge and engage in early, repeated efforts to repair customer relationships and build a strong sense of customer trust and loyalty.

For example, utilities could make serious efforts to provide goods and/or services that help customers better manage their electricity bills11, to always communicate in an open and transparent manner, to partner with welfare organisations to assist customers in need, and/or to participate in 'public good' type activities that 'give back' to the community. Additionally, given that price (and particularly perceived price unfairness) and billing errors may have the potential to influence customer loyalty and retention [87,111], and presumably trust, other potentially useful strategies include ensuring that electricity bills are always factually correct, delivered on time, and free from 'nasty surprises' (e.g., higher-than-expected charges). For example, customers could be alerted when they are approaching a certain self-selected limit (e.g., dollar value, or proportion of previous bill value) to ensure they have forewarning sufficient to enable shifts in usage behaviour before excess charges apply12. Billing information should also be simple and transparent (e.g., providing clear rules on how fees are calculated; date/s of metre readings at the property). And in cases where over-charging and/or service failures occur, the utility should issue a timely apology (ideally outlining what caused the problem, and the steps taken to remedy the problem and prevent it happening again [113]) and where appropriate, provide fair compensation. Together, these trustbuilding activities would help assure customers that the retailer does care about their welfare, and the customer might eventually

11 Some examples include: offering automated demand management devices for high demand household appliances (e.g., air conditioning, swimming pool filtration, and electric hot water) to help consumers reduce or shift electricity consumption during peak periods; and providing early-warning notifications (via email, phone or in-home visual aid) when they are approaching a certain predefined level of usage.

12 It is important for new strategies such as this to be thoroughly piloted and systematically evaluated first to ensure no unintended consequences arise. To illustrate, pre-payment of electricity is increasingly popular among consumers, yet some evidence suggests householders who pre-pay (rather than post-pay), and who pre-pay with smaller (rather than larger) top-up amounts, tend to consume more electricity [112]. While this finding may be contrary to what one might expect (given that such plans are designed to offer consumers more flexibility in managing their bills), it may be explained by the ways in which people categorise and perceive costs of different values and at different levels of aggregation - smaller, disaggregated expenditures might be seen as more trivial and less salient than larger, aggregated ones (see Brutscher [112]).Whatever the underlying explanation, this example illustrates the importance of pilot-testing new strategies prior to broader scale roll-out, in order to properly assess all impacts, whether positive or not.

start to feel that their retailer has integrity and can indeed be trusted.

4.3. Status quo bias: present new tariff offers in ways that encourage active consideration by customers

Another common anomaly in human decision making is the status quo bias - a tendency to stick with one's current position (or the default option), which may often mean avoiding decision making altogether and simply failing to act [60,61]. It is well-known that people are inherently inertial and tend to resist change. They usually prefer to retain the current state of affairs, partly because any deviation from normality seems to pose risk and potential loss, partly because it appears effortful and inconvenient. In the energy domain, research has shown that customers are generally reluctant to switch to new electricity suppliers, even when there are material benefits to be gained and/or they are well aware of product and service characteristics [57,62]. Instead, people tend to stick to the status quo. However, note that this pervasive inertial tendency can sometimes be turned in favour of change13 when the new pricing scheme is set as the default option (because then the decision to explicitly reject the default is what seems effortful and risky). Unsurprisingly, a recent review of cost-reflective pricing programmes revealed relatively low rates of recruitment (— 5 to 28%) for 'opt-in' pricing schemes (where the default is non-enrolment in the programme), but substantially higher rates (— 78% to 87%) for 'opt-out' schemes (where the default is automatic programme enrolment) [29].

Collectively, these insights suggest that status quo bias is likely to exert a powerful impact on customers' response to the introduction of cost-reflective pricing. It is reasonable to expect, for instance, that many customers will demonstrate a deeply entrenched preference for the status quo (traditional flat-rate tariffs) because they perceive the potential risks/losses of abandoning their familiar state and switching to a new pricing method (including the potential inconvenience and cognitive effort involved both in choosing and using the new tariff) as greater than the likely benefits/gains. Notwithstanding the potency of status quo bias, there are several strategies that could be deployed to lessen (or else exploit) its impact. As mentioned at the outset, it is critically important first to simplify information so that the average consumer is able to understand and act upon it. Beyond those fundamentals, other strategies include: (1) providing a 'recommended' offer that is framed as the default14, (2) avoiding indicating that the traditional (flat-rate) tariff is favoured by the majority of customers and instead endeavouring to present it on an 'even footing' i.e., in neutral terms that should increase the prospect that other kinds of pricing schemes are at least given proper consideration, and (3) introducing cost-reflective pricing offers at those times in a person's life when they might naturally be more open and amenable to considering alternative options

13 However, it is crucial to note that this only means increased probability of tariff acceptance, which does not necessarily translate into any increase in effective usage of the new pricing scheme: a critical difference. Both logically and empirically, an opt-in (vs opt-out) scheme will see a greater proportion of its subscribers willing and able to use the tariff effectively.

14 In regard to providing a 'recommended' pricing offer, Toh and Low [114] reported on a smart metering trial with 400 customers in which 95% of participants opted for time-of-use pricing when it was labelled as a 'recommended offer'. The recommendation was made possible by installing smart metres for a one-month period in order to assess the customer's consumption profile and determine the 'recommended offer'. It is unclear whether Toh and Low [114] included a control group in this particular trial. Thus, it remains uncertain what the participation rate would have been in the absence of labelling the pricing plan as a 'recommended offer'.

(e.g., when moving homes, ending a contract, making a complaint about a high bill, or wanting to change plans for any other reason).

As noted earlier, status quo bias could also be exploited by setting an 'opt-out' default pricing offer, rather than 'opt-in', with the former likely to yield a higher 'participation' rate [see 29]. However, we caution that this strategy might create significant customer backlash, especially among those segments that struggle to reduce or shift their electricity consumption appropriately. If customers subjected in this way to an 'opt-out' cost-reflective tariff failed to actively switch to a more familiar, less dynamic pricing scheme, they might unwittingly run up markedly higher electricity bills. Further, any change to pricing that impacts electricity affordability may pose health and safety risks for vulnerable groups that might have limited capacity to respond 'appropriately' e.g., low-income households, the elderly, those with a disability. Vulnerable customers may be induced to make choices that compromise thermal comfort and wellbeing, e.g., not using fans or air-conditioners during a heat wave in their attempts to avoid a high bill [10]. Some researchers claim there is little empirical evidence that cost-reflective pricing will 'hurt' low-income households, e.g., Faruqui and Palmer [115] cite evidence they say suggests most low-income customers would immediately save money by switching to dynamic pricing. But the potential risks for vulnerable groups cannot be easily dismissed. Thus, rather than introducing cost-reflective pricing as an opt-out or mandatory programme, we contend that a more socially responsible approach would limit itself only to opt-in programmes, and direct considerable efforts toward improving the design, promotion and ongoing support of such programmes. At the very least, customers should be explicitly informed about any impending opt-out tariff changes before they are automatically enroled in them - and provided with very prominent and exceedingly simple means to un-enrol, if they so choose.

4.4. Loss aversion: reduce all costs associated with the shift to dynamic pricing

Another prominent cognitive bias that may impact customer acceptance of cost-reflective pricing is loss aversion, which describes a tendency to place substantially greater weight on losses than comparably-sized gains [60,64]. This persistent tendency is manifest whenever people are weighing up the losses/ costs15 and gains/benefits of different choices, and ultimately reflects the fact that humans tend to experience or 'feel the pain' of losses far more than the pleasure of gains. This strong aversion to loss has important implications for designing and delivering policies and programmes intended to induce behaviour change, including the framing of messages, the provision of options, and the manner in which incentives are described and administered. For example, many studies have found that loss-framed messages tend to have greater behavioural impact than gain-framed messages, particularly when a self-referencing frame (emphasising losses to oneself or the current generation) is used [116-119] 16.

The principle of loss aversion would suggest that messages around cost-reflective pricing should ideally be framed in ways

15 This may include financial costs (how much does it cost?), physical risks (is it safe?), social costs (what do others think?), ecological risks (is it environmentally friendly?), time (will it take me longer?), functional risks (will it suit my lifestyle/ routine?) and even psychological costs (how will it make me feel?).

16 However, the motivational impact of a particular message frame may

depend on a number of factors - for example, the perceived risks of a target

behaviour, who/what is the reference point, and attributes of the target audience. To achieve optimal effectiveness, factors such as these should be carefully con-

sidered when designing and delivering consumer-focused communications, particularly when tailoring messages to different customer segments [for a comprehensive summary, see [118]].

that specifically focus on the losses, costs or risks (in terms of time, effort, money, etc.) associated with retaining the status quo of flatrate pricing or, conversely, convey how cost-reflective pricing may help customers avoid losses, reduce costs, or minimise risks. This is as opposed to simply focusing on the payoffs or benefits to be gained by switching from the status quo. For example, a statement such as "You would lose $15 per quarter by not switching plans" should be more motivating than "You would gain $15 per quarter by switching plans".

While these would be our general expectations - derived from the convergence of relevant findings across many diverse domains - in the specific domain of cost-reflective pricing, the evidence seems less certain. To our knowledge, there has been only one empirical examination of the relative effectiveness of loss- versus gain-framing in the context of cost-reflective pricing. This work found no significant differences in the impact of advertisements framed in terms of losses (e.g., "Not switching could cost you money on your bill and harm the environment") versus gains (e.g., "Switching could save you money on your bill and help the environment") [33]. These scholars concluded that their counterintuitive result could be due to the loss-framed marketing messages not being strong enough to stimulate feelings of loss-aversion17 in the context of such an important decision as whether to switch to a cost-reflective tariff. To ascertain whether tariff design 'trumps framing' (p. 32-33) as suggested by the authors, we would want to conduct further experimentation. These additional studies would pointedly test the impact of advertisements free of both loss- and gain-frame messages against those with loss- and those with gain-framing. Only this would allow confident evaluation of whether tariff design truly overshadows the effects of framing.

In addition to designing motivational messages that highlight the avoidance of losses/costs, tangible efforts should also be made to actually reduce the transaction costs of switching to, and using cost-reflective pricing, particularly in terms of time, effort, functionality and comfort. This might entail, for example, making the registration process as quick, easy and hassle-free as possible; offering automated demand management technology free of charge; and ensuring customers do not have to sacrifice thermal comfort or forgo valued lifestyle choices in order to switch pricing plans.

Loss aversion also has important implications for the specific configuration of tariff charges. For example, Dutschke and Paetz [107] found that participants typically preferred pricing programmes with a narrower spread of charges, which minimised the risk of incurring the financial loss of a high bill. It has also been suggested that any price rises during peak periods (potential losses) should ideally be compensated with much larger cost savings during non-peak periods (potential gains) [120], which may help alleviate any avoidance of new tariff offers for fear of incurring higher fees. Furthermore, we expect that many customers are aware that any increase in fixed charges significantly constrains their ability to reduce bill costs, which contradicts the very rationale for trying out a novel tariff in the first place. In fact, given consumer awareness of those 'uncontrollable' fixed charges, it may even be the case they might be more accepting of, and compliant with a cost-reflective pricing scheme that offers a reduction in fixed charges. Of course, we need further research to test this empirically, with varying price differentials. But the general principle in designing the schedule of charges in cost-reflective pricing schemes is that any loss (i.e., increased costs, especially around

17 Which, in this sample, were almost universal, with 95% of consumers being classified as loss averse (and almost a third of the sample highly loss averse), as measured via a standard battery of 'loss aversion' financial decision-making questions.

fixed charges) must ideally be offset by much larger gains (i.e., decreased costs during off-peak times)18.

4.5. Risk aversion: provide assurances that customers do not risk higher electricity bills under cost-reflective pricing

Related to loss aversion is the tendency for people to be risk averse. For example, people tend to over-weight certainty and sure gains, favouring a smaller but guaranteed payoff over the gamble of less certain gains of markedly higher value [64,65,69]. Interestingly - and this may be of particular relevance to the content and framing of different tariff offers - this risk aversion is more prevalent in the domain of gains (e.g., people will over-value a smaller but more certain gain compared to a larger yet uncertain gain), whereas people tend to be somewhat more tolerant of risk in the domain of losses (e.g., they might take a chance on an uncertain loss in order to avoid a certain loss) [65,66].

These regularities in how decision-makers typically choose between probabilistic alternatives involving a certain degree of risk have important implications for understanding customer responses to electricity tariffs, because different pricing options appear to invoke different perceptions of risk and reward. Compared to traditional flat-rate pricing, for example, cost-reflective pricing holds out the prospect of higher reward, but also higher risk [115]. Especially if the pricing scheme is highly variable, it may be difficult to predict bill savings, and there is a (real or perceived) risk that nothing will be gained. It seems the advantages of cost-reflective pricing are not obvious to customers, with many either sceptical there are gains to be had or simply not valuing the purported benefits in the first place [107,121 ]. While the most obvious risk faced by customers when switching electricity tariffs is financial (i.e., the chance of higher bills), there may also be physical risks (e.g., sacrificing health/comfort to save on bills), psychological risks (e.g., feeling disappointment /regret over selecting the 'wrong' tariff) and even functional and lifestyle risks (e.g., around the time/effort involved in switching plans, the inconvenience and 'hassle' of trying to use tariffs effectively, lack of 'fit' with one's routine/lifestyle).

Choosing a new cost-reflective pricing plan might therefore be considered an 'uncertain loss' gamble for customers. They might avoid risk by opting for/sticking with a standard flat-rate plan where the loss is more predictable. Or they might be motivated to move off a standard flat-rate plan by tariffs that offer only the prospect of gains without the risk of loss, e.g., a peak time rebate schemes assuring customers that their bills will not increase. In this regard, it is quite likely that peak time rebates - even if the discounts are variable - would be much more popular than other types of cost-reflective pricing that do not involve such discounts. Note that the relative attractiveness of peak time rebates was seen in pilot trials run by Baltimore Gas & Electric (BGE), whose customers proved to be more satisfied with a specially-designed peak time rebate scheme than with critical peak pricing [122] 19. BGE has now rolled out the peak time rebate programme alone, and expects all of its residential customers to be enroled by 2015.

18 Of course, offsetting peak prices with much lower non-peak prices that deliver real cost savings to consumers willing and able to respond 'appropriately' might materially damage a utility's 'bottom-line', at least on the face of it. However, a proper assessment of benefits relative to costs must also take account of any longer-term, indirect outcomes, such as customer satisfaction and retention, and increased market share. Additionally, the utility would be expected ultimately to benefit from any resulting shift in demand that might consequently reduce, delay or circumvent future infrastructure investments.

19 The rebate scheme - BGE's 'Smart Energy Rewards' programme - in this case also incorporated behavioural solutions, in the form of personalised energy saving tips and timely feedback, delivered by OPower's Behavioural Demand Response.

There are other potentially effective means of reducing the perceived risk of cost-reflective pricing. Extensive research has assessed the impact of various risk-reduction strategies ('risk relievers') on consumers' purchasing decisions and behaviour, particularly in online retail environments where the perception of risk can often deter potential customers [123-125]. In this environment, strategies such as money-back guarantees, warranties and free trials/samples may prove effective [126,127]. In regard to cost-reflective pricing, it might likewise prove worthwhile offering obligation-free trials and bill protection insurance for some initial period, as a means of reducing perceived risk [6], and thereby enhancing uptake. Recent empirical work specifically examining the impact of money-back guarantees on intention to accept cost-reflective pricing appears to support this proposition [32].

Further, Faruqui and Palmer [115] have suggested that the perceived risks of cost-reflective pricing could be addressed by introducing a pricing platform enabling customers to select (or have recommended to them) a pricing plan based on their individual tolerance of risk relative to reward. We would also recommend including - as an integral part of such a scheme - at least one completely 'risk-free' tariff (e.g., peak time rebate), where customers face no risk of incurring a higher bill, only incentive payments for reducing consumption during peak periods. Ideally, such a scheme would also build in the provision of simple demand-management tips, and the offer of automated demand management devices and/or other aids to help householders reduce consumption at higher-priced times.

With regard to the potential of automated demand management devices as a risk-relieving mechanism, the limited field trials conducted to date do not seem to indicate that the offer of technology (e.g., in-home displays and programmable controllable thermostats) significantly increases uptake rates [29]. But other consumer-based research offers some hope that automated devices do have the potential to boost uptake. These kinds of discrepancies in results might be explained by differences in the technology on offer. For example, Stenner and colleagues [32] observed elevated uptake20 rates when the offer of a free load control device was bundled with a tariff. Fell and colleagues [33,34] also found that a smart thermostat made dynamic pricing more attractive21 and that the best performing scheme was a direct load tariff, which is essentially an entirely automated programme: in return for a lower flat rate tariff, the consumer allows the electricity supplier to cycle their heating system on/off at certain times. Qualitative research has shown that consumers talk favourably about home energy automation systems and smart appliances, in regard to how such technologies might ease the burden of changing routines, and maintain householder comfort [108]. Further, when already on a cost-reflective pricing plan, consumers show a preference for enabling technology [107] 22,23.

20 Although the outcome measure in this case rested on self-reported intentions to take up a hypothetical tariff.

21 Note that this difference was only significant for attitudes towards the tariff, not for self-reported willingness to switch.

22 It is important to note that such preferences cannot be readily generalised, because acceptance of automated devices appears to be highly appliance-specific. It seems to depend on how long a consumer is willing to delay use of the specific appliance, something that varies considerably across appliances [128].

23 In addition to examining the likely uptake and usage of automated demand management solutions, it is also critical to consider their overall economic viability. For example, research examining the value of smart appliances (per se) has indicated that their value is highly system-specific. And once costs are factored into the equation, it appears that the profit margin is very low, especially as the demand side management capacity increases (because cost reductions do not increase proportionally) [128]. While such research suggests diminishing marginal returns from the mass-scale introduction of smart appliances, it is possible that their value could be enhanced through future technological efficiencies (which might reduce the cost of installation and replacement) and the inclusion of 'intentional' demand

4.6. Temporal and spatial discounting: Reduce immediate costs and increase the salience of immediate benefits from cost-reflective pricing

Another cognitive bias to consider in the context of customer uptake and usage of cost-reflective pricing is time inconsistency, i.e., the tendency for people to be short-sighted when some costs or benefits are immediate, but farsighted when all costs and benefits are in the future [70,71]. People tend to discount things that are further away in time (temporal discounting) and/or space (spatial discounting), and conversely, place disproportionate weight on things that are immediate and/or nearby. For example, people often procrastinate over, delay or avoid actions that are costly to perform in the short-term (e.g., outlaying time, effort and money to purchase new energy-efficient devices), even if they might lead to significantly larger payoffs in the long-term (e.g., markedly reduced electricity bills in the next few years). This systematic bias is extensively documented in the literature. Far in excess of reasonable discounting for the greater uncertainty around future outcomes, people tend to value immediate rewards (and dislike immediate costs) far more than they value future rewards (and dislike future costs) [70-72,74].

In regard to cost-reflective pricing, there are several upfront costs that may be immediately salient to customers. These include 'learning costs' [129] and 'transaction' costs [130], in this case: the time and effort customers must invest to learn about new tariff offerings, and about how to modify household energy usage appropriately. There is also the immediate inconvenience incurred by switching electricity plans, often referred to as 'hassle costs' [131 ]. Modifying one's pattern of energy usage may, in turn, lead to other ongoing costs such as losses in convenience, comfort and cleanliness. In contrast, the main potential customer benefit from cost-reflective pricing (and this is by no means assured) is a lower electricity bill, which arrives well into the future and might only involve a small saving on each bill payment. As such, customers may have a tendency to discount these potential longer-term payoffs by placing a disproportionate emphasis on the upfront costs.

In the mind of typical consumers, cost-reflective pricing likely poses certain, salient and immediate costs (and the accrual of unknown future costs) as against uncertain, potentially negligible and delayed benefits. In these circumstances, it is likely that cost-benefit appraisals (to the extent the typical consumer even undertakes such calculations) will be heavily biased towards the upfront costs, and customers may well judge that the costs are not worth the benefits. To achieve a more balanced appraisal, the industry should consider more explicitly highlighting and making more tangible the potential benefits of cost-reflective pricing, and reducing the (real and perceived) costs of participating. Costs and benefits could be restructured, so that the benefits heavily outweigh the costs and/or are gained much sooner, e.g., increase the salience of immediate benefits by communicating dollar-value savings in (or close to) real-time. For example, BGE's 'Smart Energy Rewards' programme provides same- or next-day communication advising customers how much money they have saved, compared to either their baseline, or peers' consumption. At the very least, a simple side-by-side graphical comparison of the quarterly electricity bill under the alternative tariff plans (e.g., traditional flat-rate versus new cost-reflective pricing structure)

(footnote continued)

response (manual demand response by the consumer), the latter perhaps stimulated by such things as cost-reflective pricing and non-pecuniary, behavioural interventions. Research appears to indicate that optimal demand response is achieved when automated demand management technology is coupled with cost-reflective pricing [1].

should be provided to customers, showing the exact amount of money saved (or lost) over time. Customers can then evaluate for themselves whether they wish to continue with the new pricing scheme.

Finally, it is important to consider the benefits that cost-reflective pricing might provide not only to customers themselves (potential savings), but also to the community (restoring fairness and equity), and to industry (reducing peak demand and improving electricity supply/reliability). In this regard, we note that Gyamfi and Krumdieck [132] found that when households were asked to rate the relative importance of price, environment (carbon reduction) and supply security (blackouts) as reasons for reducing electricity demand, price and supply were judged similarly important whereas environmental impacts (while also important) were of lesser concern.

4.7. Normative social influence: describe how other customers have experienced cost-reflective pricing

Another mental shortcut that people commonly use in decision making is following the behaviour of others in similar situations, i.e., conforming to social norms. Social norms, quite simply, convey the explicit and/or implicit guidelines or behavioural expectations within a group or society concerning what is normal/common and/or desirable behaviour in that context [76,82,133]. Because people tend to do what is socially approved (because it yields social rewards) and/or popular (because it reflects effective or adaptive behaviour) [83], social norms can exert a powerful impact on attitudes and behaviour. Numerous studies in the energy domain have found that simply conveying descriptive normative information - e.g., personalised messages comparing one's electricity consumption to a neighbourhood norm - can significantly influence customers' energy usage [e.g., 80,134,135-138]. The behavioural impact of such messages appears to be greatest when norms are contextualised with personally relevant, meaningful or localised information (e.g., referring to the norms in one's immediate surrounds [78]). While the general trend of adhering to social norms is a well-observed phenomenon, other social influence research has demonstrated that in some circumstances people can actually be induced to deviate from the prevailing social norm. For example, when provided with a positive image of an alternative, favourable, yet non-normative behaviour, it appears that people are sometimes willing to 'stand out from the crowd' in order to be associated with the positive features of the 'rare' behaviour [139,140].

In regard to customer response to electricity pricing, the powerful impact of social norms plus the moderating influence of positive images together suggest that likely barriers to uptake of cost-reflective pricing may include (1) a general customer perception that few other people approve of and/or have accepted this form of pricing themselves (i.e., if customers think that many other people are not shifting to cost-reflective tariffs, for whatever reason, they may be unlikely to do so too); and (2) a general perception that customers who choose cost-reflective pricing must be (for example) gullible, unintelligent, poor, risqué, 'greenies'. Conversely, if cost-reflective tariffs are seen as socially desirable and popular, or (alternatively), as uncommon but associated with positive features, then customers might be more likely to test out this new method of pricing themselves. In general, the literature suggests that peer reviews and customer testimonials from credible, trustworthy sources (e.g., positive word-of-mouth from friends and family, celebrity endorsements, etc.) can serve as effective 'risk-relievers' for consumers facing purchasing decisions, as can making a product/service appear popular and/or something that is socially approved [126,127,141].

These insights have important implications for how customer-focused communication should be designed and delivered. To increase the uptake of cost-reflective pricing, industry stakeholders and/or policymakers should consider whether it is possible and appropriate (i.e., ethically and professionally sound) to frame cost-reflective pricing as something that is valued and accepted within the local peer group, community or society, and to associate it with positive images of the types of people who choose cost-reflective pricing. At the very least, information about the experiences - good or bad - of real customers is something many will use to make their own decisions about whether try cost-reflective pricing themselves.

4.8. Perceived fairness: explain inequity in flat-rate pricing, and how cost-reflective pricing restores fairness

One final factor that conceivably might impact customer response to cost-reflective pricing is its perceived fairness and equity compared to traditional flat-rate tariffs. While industry support for cost-reflective pricing is generally strong, there remains a common public perception it is harmful and unfair24. In particular, variable pricing is thought to harm 'vulnerable groups', specifically, those segments of the community with limited capacity to reduce energy usage during peak times, e.g., low-income households, those with disabilities or medical/health issues, shift workers, and young families with many children. Yet advocates of cost-reflective pricing counterargue that it is flat -rate pricing that is actually unfair, since customers that do not use much electricity at peak times ('flat users') essentially cross-subsidize those that do ('peaky users'). Considered in this light, pricing electricity in a more cost-reflective way is said to be more equitable, with 'peaky users' paying the appropriate premium for using electricity during high-demand times. In support of this notion, Faruqui [6] has conducted a modelling exercise to illustrate that flat-rate charges may actually be unfair for 'flat users'. Using a population of 10 million customers, he demonstrated that under the current (flatrate) pricing, 'flat users' incur a $10 loss per month whereas 'peaky users' gain a $10 benefit. His analysis showed, further, that the benefits of dynamic pricing for low-income households actually outweigh the negative impacts, with the vast majority of low-income households actually better off under dynamic pricing (e.g., he reports that 80% would benefit even without any demand response, while 92% would benefit with modest demand response).

It would appear that the controversy surrounding the relative fairness (or otherwise) of cost-reflective pricing stems from viewing the potential impacts at different levels of analysis. In the aggregate, it would appear that there are more winners than losers. But at lower levels of analysis, it is likely that distinct customer cohorts could well be worse off: not just low-income

24 In regard to the perception that cost-reflective pricing is unfair, we acknowledge that 'electricity usage' has multiple dimensions beyond mere time (i.e., apart from just when it is used). For example, there is also variability in the geographical location (where the electricity is used), quantity (how much is used overall), momentary demand (how much is used at any one time) and duration (how long it is used for). Pricing electricity according to only one of these dimensions could arguably be considered 'unfair'. The fairest system might be one that takes into account all these different facets of electricity usage that contribute to the final cost of generating and supplying electricity to the end user. In practice, however, designing such a system could prove a difficult feat. Current models of cost-reflective pricing get part of the way there, by charging consumers for how much electricity they use over a set period (quantity), as well as when they use it (time). However, they obviously do not account for all dimensions. Integrating elements of capacity pricing with traditional time-based tariffs (i.e., charging consumers for momentary demand) may be seen as an even 'fairer' approach, although the challenge lies in designing a tariff of this nature that is still relatively simple, easy and manageable for consumers.

households, but groups of customers with particular attributes that limit their capacity to shift demand, such as working families and those with certain health concerns. Considering the real potential for these stratified negative impacts, cost-reflective pricing should be designed and introduced in ways that clearly prioritise the welfare of at-risk customers, for example, incorporating government concessions and assistance programmes, specially-designed and targeted tariffs coupled with supporting mechanisms, a highly subsidized retrofit programme to improve household energy efficiency, and even the option to remain on a traditional flat-rate tariff25. It would also be socially responsible for utilities to offer their customers basic energy-efficiency and demand-management support (e.g., simple energy-saving tips, access to automated devices and direct load control), with this assistance perhaps bundled within special cost-reflective pricing offers tailored to particularly vulnerable groups.

Beyond these additional mechanisms to support vulnerable customers, it might also help to communicate simply and openly the relative (un)fairness (at the aggregate level) of different methods of pricing (although as for most communication/community education campaigns, we expect that only a modest portion of the public would ultimately be interested in engaging with this). The first step in such an enterprise might entail simply explaining how electricity is generated and supplied; the variable demand for electricity across different customers, at different times of the day and year; the inherent (and expensive) constraints in supply that are stressed at times of high demand, and (consequently) why utilities are seeking to price electricity differently. As prior research on retail price rises makes clear, it is critically important to convey a positive (rather than negative, self-interested) reason why prices need to change, since this underlying rationale is linked to customers' perceptions of fairness, and subsequent intentions to continue their patronage [142].

In terms of tariff configurations, consideration should be given to the fairness of 'capacity' pricing schemes, which include a charge for the actual volume of electricity being drawn from the grid at any one point in time, instead of (or in addition to) a fixed network charge and/or usage charges. These novel pricing schemes are often recommended as being among the best mechanisms for redressing the purported unfairness inherent in the way electricity is currently priced. Since consumers would pay a set price per volume of electricity drawn from the grid at any one point in a given period, 'non-peaky' customers would usually pay less than 'peaky' ones. However, some might still consider this pricing method unfair, on the grounds that generally non-peaky users might end up being charged just as much as regularly peaky-users, due to some unusual instance in which they happened to consume a great deal of electricity at once, e.g., when entertaining or hosting visitors. Thus, the purported fairness of capacity-style schemes is not without dispute, and they would likely require the thoughtful addition of features that could mitigate some of these issues for a variety of household types.

It has also been suggested that capacity-based tariffs might be especially difficult to understand - being relatively complex and unfamiliar to consumers, who are mostly accustomed to a consumption-based method of pricing) [143] - and that a 'targeted

25 The Australian Energy Market Commission (AEMC) has recommended a review of energy concession schemes and assistance programmes to ensure these appropriately target and support customers who are at risk of increased financial stress due to cost-reflective pricing (Power of Choice review). It should be noted that as more and more customers switch to cost-reflective pricing, the cost incurred by those who remain on standard flat-rate pricing is likely to increase because the per capita cost of servicing this smaller number of customers is higher. Thus, it would be prudent to offer suitable support mechanisms to help vulnerable customers eventually transition to cost-reflective pricing.

and comprehensive education campaign' should be undertaken to help them understand demand charging [144]. While again, we have modest expectations regarding the likely impact of these kinds of community education campaigns, we imagine that any such education efforts might usefully capitalise on analogical reasoning, i.e., using a familiar example with a similar underlying principle to help people understand a new situation. Prior research has certainly shown that analogical reasoning can be an effective means of improving decision-making [145,146]. Demand-pricing is an unfamiliar concept in the context of electricity, but commonly used in many other domains, such as transport (airfares, car parking) and hospitality (cinema/theatre tickets, accommodation, dining on public holidays). It is possible that framing capacity-based tariffs in analogous ways - with reference to other industries where people pay a premium when demand is greater and/or operating costs are higher - could potentially lead to greater consumer understanding and acceptance of this type of pricing regime for electricity.

The power of analogy could also be used to design and depict capacity-style tariffs in ways that imitate internet or mobile service plans that cater to different lifestyles (e.g., $50/month low user, couple-household, heavy entertainment packages). Indeed, capacity pricing of electricity might even lend itself well to the design of an extremely cheap flat-rate tariff, with a very low capacity just sufficient to run a small number of essential appliances (hot water system, lights, stove, etc.), e.g., $40/month for 1.5 kW capped demand, with unlimited consumption. This type of tariff might appeal to low-income households that are already using very little electricity at all times. In fact, it might even help raise the living standards of those that are currently using even less electricity than the capacity limit, as they would suddenly have access to extra electricity for fairly rudimentary activities that improve health, wellbeing and comfort (e.g., reading, bathing, cooking fresh food). While such a tariff would need to be carefully designed to ensure it remains fair and economically feasible, the potential benefits extend also to the utility. For instance, if this type of pricing is as appealing as it is socially responsible, multiple benefits might accrue for the utility in terms of a larger customer base, higher levels of trust, and less customer 'churn'. We note here that customers in general may be especially motivated to learn about capacity pricing plans, given 'flat-rate bias' [147]: their well-known tendency (particularly for hedonistic services [148]) to prefer unlimited consumption with a flat fee for service over per unit pricing, and similarly, to favour tariffs with a relatively high allowance independent of actual usage [147].

5. Conclusions and recommendations

This article has examined a number of common cognitive biases and psychological influences that may shape consumer decision making and behaviour in regard to cost-reflective electricity pricing. These 'behavioural' aspects are likely to be critically important when designing and introducing new tariffs. Not only do they explain why consumer uptake and usage of cost-reflective pricing often falls short of expectations, they provide guidance on how cost-reflective pricing can be designed, depicted and delivered to enhance customer uptake and promote optimal usage.

In this paper, we have discussed a number of behavioural interventions that might usefully harness and leverage these well-established biases in decision making. However, before such interventions can be implemented effectively at scale, it is critical that we first empirically determine which of these solutions (or what suite of solutions) would likely yield the greatest peak demand reductions, at lowest cost, and with minimal unintended, negative effects. To derive compelling evidence in this regard, we

strongly recommend that such research employs a randomised controlled trial (RCT) design: generally the optimal method for any programme one intends ultimately to validate [149-152]. RCTs are a scientifically robust and empirically defensible way of determining the precise extent and nature of the cause-and-effect relationship existing between a particular intervention and its intended outcome(s). This methodology allows one definitively to determine the utility and overall cost-effectiveness of an intervention (or policy choice) relative to business-as-usual and/or alternative interventions (programmes or schemes).

We close with a number of specific recommendations for effectively deploying RCTs to empirically test various strategies relating to the uptake and usage of cost-reflective pricing. We would suggest that stakeholders first perform a sequential series of smaller-scale RCTs in the field, with randomly-selected, representative samples of customers, all of whom should be equipped with advanced metering infrastructure (to provide objective outcome data on electricity demand). Each trial should be of sufficient duration to examine not just the immediate impact of an intervention, but also its effects as the intervention is continued over time ('durability') and once it is discontinued ('persistence') [for more information on these different temporal effects, see [153,154]. And while additional interaction with these customers (beyond that involved in the behavioural intervention itself) should be minimised throughout the experiment to reduce the influence of participant reactance and so-called 'demand characteristics' see [155-157], in some cases it might be useful to undertake a brief end-of-experiment survey to collect information on participants' subjective experiences, along with any ideas for improvement in further iterations of the behavioural intervention.

Ultimately, the RCT method could usefully be applied to test the cost-effectiveness and mass scalability of many, if not all of the solutions discussed in this paper. Additionally, the cyclical nature of this applied experimental method, i.e., test-learn-adapt see [158] would enable such strategies to be iteratively modified -and even interwoven with other approaches - leading ultimately to the identification of optimal pricing solutions that could then, with minimal risk, be rolled out en masse across the broader population.


[1] Faruqui A, Sergici S. Arcturus: international evidence on dynamic pricing. Electr J 2013;26:55-65.

[2] Vickrey WS. Responsive pricing of public utility services. Bell J Econ 1970;2:337-46.

[3] Faruqui A, Malko JR. The residential demand for electricity by time-of-use: a survey of twelve experiments with peak load pricing. Energy 1983;8:781-95.

[4] Caves DW, Herriges JA, Kuester KA. Load shifting under voluntary residential time-of-use rates. Energy J 1989;10:83-99.

[5] Caves DW, Christensen Lr, Herriges JA. Consistency of residential customer response in time-of-use electricity pricing experiments. J Econom 1984;26:179-203.

[6] Faruqui A. The ethics of dynamic pricing. Electr J 2010;23:13-27.

[7] Faruqui A, George SS. The value of dynamic pricing in mass markets. Electr J 2002:45-55.

[8] Simshauser P, Downer D. Dynamic pricing and the peak electricity load problem. Aust Econ Rev 2012;45:305-24.

[9] Brand SA. Dynamic pricing for residential electric customers: a ratepayer advocate's perspective. Electr J 2010;23:50-5.

[10] Alexander BR. Dynamic pricing? Not so fast! A residential consumer perspective Electr J 2010;23:39-49.

[11] Coleman JS, Fararo TS. Rational choice theory: advocacy and critique. New-bury Park: SAGE Publications; 1992.

[12] Goode WJ. Rational choice theory. Am Sociol 1997;28:22-41.

[13] Weintraub E.R. Neoclassical economics. In: the concise encyclopedia of economics. Library of Economics and Liberty; 1993.

[14] Henry JF. The Making of Neoclassical Economics. Oxon, UK: Routledge; 2011.

[15] Simon HA. A behavioral model of rational choice. QJ Econ 1955;69:99-118.

[16] Simon HA. Models of man: social and rational. New York: Wiley; 1957.

[17] DiClemente DF, Hantula DA. Applied behavioral economics and consumer choice. J Econ Psychol 2003;24:589-602.

[18 [19 [20

[21 [22 [24 [25 [26

[28 [29

[35 [36 [37 [38 [39 [40 [41

[42 [43 [44

[47 [48 [49 [50 [51 [52

Kahneman D. A psychological perspective on economics. Am Econ Rev [54 2003;93:162.

Kahneman D. Maps of bounded rationality: psychology for behavioral eco- [55

nomics. Am Econ Rev 2003;93:1449-75.

Frederiks ER, Stenner K, Hobman EV. Household energy use: applying behavioural economics to understand consumer decision-making and [56

behaviour. Renew Sustain Energy Rev 2015;41:1385-94. [57

Wilson C, Dowlatabadi H. Models of decision making and residential energy use. Ann Rev Environ Resour 2007;32:169-203. [58

Stern P. What psychology knows about energy conservation. Am Psychol 1992;47:1224.

Van Raaij WF, Verhallen TMM. A behavioral model of residential energy

usage. J Econ Psychol 1983;3:39-63. [59

Costanzo M, Archer D, Aronson E, Pettigrew T. Energy conservation behavior:

the difficult path from information to action. Am Psychol 1986;41:521-8. [60

Stern PC, Oskamp S. Managing scarce environmental resources. In: Stokols D,

Altman I, editors. Handbook of environmental psychology. New York: John [61

Wiley & Sons, Inc; 1987. p. 1043-88.

Abrahamse W, Steg L, Vlek C, Rothengatter T. A review of intervention stu- [62

dies aimed at household energy conservation. J Environ Psychol 2005;25:273-91.

Stern P. Toward a coherent theory of environmentally significant behavior. J [63 Soc Issues 2000;56:407-24.

Todd A, Cappers P, Goldman C. Residential customer enrollment in time- [64

based rate and enabling technology programs. Lawrence Berkeley National Laboratory; 2013. [65

Braithwait S., Hansen D., Armstong D. Impact evaluation of San Diego gas & electric's peak-time rebate pilot program. Part 1. Ex Post Report. Christensen [66 Associates Energy Consunlting; 2012.

Navigant Consulting. The effect on electricity consumption of the common- [67

wealth edison customer applications program: phase 2 Final Analysis. Palo Alto, CA; 2011. [68

Stenner K, Frederiks ER, Hobman EV, Meikle S. Australian consumers' likely response to cost-reflective electricity pricing. Brisbane, Australia: The Commonwealth Scientific and Industrial Research Organisation (CSIRO); 2015. [69 Fell MJ, Nicholson M, Heubner GM, Shipworth D. Is it time? Consumers and time of use tariffs. UCL Energy Institute, London. Available online at: «http:// [70 %20of%20use%20tariffs.pdf»; 2015. [71 Fell MJ, Shipworth D, Huebner GM, Elwell CA. Public acceptability of domestic demand-side response in Great Britain: the role of automation and [72 direct load control. Energy Res Soc Sci 2015;9:72-84.

Train KE, McFadden DL, Goett AA. Consumer attitudes and voluntary rate [73

schedules for public utilities. Rev Econ Stud 1987;69:383-91.

Aigner DA, Ghali K. Self-selection in the residential electricity time-of-use [74

pricing experiments. J Appl Econom 1989;4:S131-44.

Patrick RH. Rate structure effects and regression parameter instability across [75

time-of-use electricity pricing experiments. Resour Energy 1990;12:179-95. Matsukawa I. Household response to optional peak-load pricing of elec- [76

tricity. J Regul Econ 2001;20:249-67.

Train KE, Mehrez G. Optional time-of-use prices for electricity: econometric analysis of surplus and Pareto impacts. RAND J Econ 1994;25:263-83. [77

Ericson T. Households' self-selection of dynamic electricity tariffs. Appl Energy 2011;88:2541-7.

Mountain DC, Lawson EL. Some initial evidence of Canadian responsiveness [78

to time-of-use electricity rates: detailed daily and monthly analysis. Resour Energy Econ 1995;17:189-212.

Baladi SM, Herriges JA, Sweeney TJ. Residential response to voluntary time- [79

of-use electricity rates. Resour Energy Econ 1998;20:225-44. Reiss PC, White MW. Household electricity demand. Rev Econ Stud 2005;72:853-83. [80

Rai AM, Reedman L, Graham PW. Price and income elasticities of residential electricity demand: the Australian evidence. Australian Conference of [81 Economists (AEC) Hobart, Tasmania, Australia: Economic Society of Aus- [82

tralia; 2014.

Scheibehenne B, Greifeneder R, Todd PM. Can there ever be too many [83 options? A meta-analytic review of choice overload J Consum Res 2010;37:409-25. [84

Greifeneder R, Scheibehenne B, Kleber N. Less may be more when choosing is difficult: Choice complexity and too much choice. Acta Psychol [85


lygnar S, Lepper M. When choice is demotivating: can one desire too much of [86 a good thing? J Personal Soc Psychol 2000;79:995-1006. Johnson EJ. Man, my brain is tired: linking depletion and cognitive effort in choice. J Consum Psychol 2008;18:14-6. [87

Schwartz B. The paradox of choice: why more is less. New York: HarperCollins Publishers lnc.; 2004. [88 Earle TC, Cvetkovich GT. Social trust: toward a cosmopolitan society. Westport, CT: Praeger; 1995. [89 Poortinga W, Pidgeon NF. Exploring the dimensionality of trust in risk regulation. Risk Anal 2003;23:961-72. [90 Renn O, Levine D. Credibility and trust in risk communication. ln: Kasperson RE, Stallen PJM, editors. Communicating risks to the public. The Hague: [91 Kluwer; 1991. p. 175-218.

Siegrist M, Cvetkovich M, Roth C. Salient value similarity, social trust, and [92

risk/benefit perception. Risk Anal 2000;20:353-62.

Siegrist M, Cvetkovich G. Perception of hazards: the role of social trust and knowledge. Risk Anal 2000;20:713-9.

Poortinga W, Pidgeon NF. Prior attitudes, salient value similarity, and dimensionality: Toward an integrative model of trust in risk regulation. J Appl Soc Psychol 2006;36:1674-700.

Johnson EJ, Goldstein DG. Do defaults save lives? Science. 2003;302:1338-9. Pichert D, Katsikopoulos KV. Green defaults: information presentation and pro-environmental behaviour. J Environ Psychol 2008;28:63-73. Sunstein CR, Reisch LA. Automatically green: behavioral economics and environmental protection. Forthcoming, Harvard Environmental Law Review; 2013. Available at SSRN: «» or

Hartman RS, Doane MJ, Woo C. Consumer rationality and the status quo. Q J Econ 1991;106:141-62.

Kahneman D, Knetsch JL, Thaler RH. Anomalies: the endowment effect, loss aversion, and status quo bias. J Econ Perspect 1991;5:193-206. Samuelson W, Zeckhauser R. Status quo bias in decision making. J Risk Uncertain 1988;1:7-59.

Brennan T. Consumer preference not to choose: methodological and policy implications. Discussion Paper 05-51. Washington DC: Resources for the Future (RFF); 2006.

Tversky A, Kahneman D. Rational choice and the framing of decisions. J Bus 1986;59:251-78.

Kahneman D, Tversky A. Choices, values and frames. Am Psychol 1984;39:341-50.

Kahneman D, Tversky A. Prospect theory: an analysis of decision under risk. Econometrica 1979;47:263-91.

Tversky A, Kahneman D. Advances in prospect theory: cumulative representation of uncertainty. J Risk Uncertain 1992;5:297-323. Thaler RH. Toward a positive theory of consumer choice. J Economic Behavior and Organization 1980;1:39-60.

Weber BJ, Chapman GB. Playing for peanuts: why is risk seeking more common for low-stakes gambles. Organ Behav Hum Decis Process 2005;97:31-46.

Bernoulli D. Exposition of a new theory on the measurement of risk. Econ-ometrica 1954;22:23-36.

Loewenstein G, Thaler RH. Anomalies: intertemporal choice. J Econ Perspect 1989;3:181-93.

Thaler RH. Some empirical evidence on dynamic inconsistency. Econ Lett 1981;8:201-7.

Lynch Jr JG, Zauberman G. When do you want it? Time, decisions, and public policy J Public Policy Mark 2006;25:67-78.

Critchfield TS, Kollins SH. Temporal discounting: Basic research and the analysis of socially important behavior. J Appl Behav Anal 2001;34:101-22. Frederick S, Loewenstein G, O'Donoghue T. Time discounting and time preference: a critical review. J Econ Lit 2002;40:351-401. Cialdini RB, Demaine L, Sagarin BJ, Barrett DW, Rhaods K, Winter PL. Managing social norms for persuasive impact. Soc Influence 2006;1:3-15. Cialdini RB, Trost MR. Social influence: social norms, conformity and compliance. ln: Gilbert DT, Fiske ST, Lindzey G, editors. The handbook of social psychology. 4th ed.. New York, US: McGraw-Hill; 1998. p. 151-92. Cialdini RB, Trost MR. lnfluence: social norms, conformity and compliance. ln: Gilbert DT, Fiske ST, Lindsey G, editors. Handbook of social psychology. 4th ed.. Boston: McGraw-Hill; 1998. p. 151-92.

Goldstein NJ, Cialdini RB, Griskevicius V. A room with a viewpoint: using social norms to motivate environmental conservation in hotels. J Consum Res 2008;35:472-82.

Schultz PW, Nolan JM, Cialdini RB, Goldstein NJ, Griskevicius V. The constructive, destructive, and reconstructive power of social norms. Psychol Sci 2007;18:429-34.

Nolan JM, Schultz PW, Cialdini RB, Goldstein NJ, Griskevicius V. Normative social influence is underdetected. Personal Soci Psychol Bull 2008;34:913-23. Turner JC. Social influence. Pacific Grove, CA: Brooks/Cole; 1991. Feldman DC. The development and enforcement of group norms. Acad Manag Rev 1984;9:47-55.

Cialdini RB. lnfluence: science and practice. 5th ed.. United States: Pearson Educational lncorporated; 2003.

Corneo G, Snobs Jeanne O. bandwagons, and the origin of social customs in

consumer behavior. J Econ Behav Organ 1997;32:333-47.

Pratkanis AR. The science of social influence. New York: Psychology Press;

van den Bos K, Maas M, Waldring l, Semin G. Toward understanding the psychology of reactions to perceived fairness: the role of affect intensity. Soc Justice Res 2003;16:151-68.

Xia L, Monroe KB, Cox JL. The price is unfair! A conceptual framework of price fairness perceptions. J Mark 2004;68:1-15.

Kahneman D, Knetsch JL, Thaler RH. Fairness and the assumptions of economics. J Bus 1986;59:S285-300.

Pesendorfer W. Behavioral economics comes of age: A review essay on

advances in behavioral economics. J Econ Lit 2006;44:712-21.

Thaler RH, Sunstein CR. Nudge: improving decisions about health, wealth,

and happiness. New Haven: Yale University Press; 2008.

Rabin M. A perspective on psychology and economics. Eur Econ Rev


Camerer CF, Loewenstein G, Rabin M. Advances in behavioral economics. Princeton, NJ: Princeton University Press; 2004.

[93] Camerer CF, Loewenstein G. Behavioral economics: past, present, future. In: Camarero C, Loewenstein G, Rabin M, editors. Advances in behavioral economics. Princeton, NJ: Princeton University Press; 2004. p. 3-52.

[94] Allcott H, Mullainathan S. Behavior and energy policy. Science. 2010;327:1204-5.

[95] Gigerenzer G, Gaissmaier W. Heuristic decision making. Annu Rev Psychol 2011;62:451-82.

[96] Kahneman D, Slovic P, Tversky A. Judgment under uncertainty: heuristics and biases. New York: Cambridge University Press; 1982.

[97] Tversky A, Kahneman D. Availability: a heuristic for judging frequency and probability. Cogn Psychol 1973;5:207-32.

[98] Tversky A, Kahneman D. Judgment under uncertainty: heuristics and biases. Science. 1974;185:1124-31.

[99] Anderson CJ. The psychology of doing nothing: Forms of decision avoidance result from reason and emotion. Psychol Bull 2003;129:139-66.

[100] Bazerman MH, White SB, Loewenstein GF. Perceptions of fairness in interpersonal and individual choice situations. Curr Dir Psychol Sci 1995;4:39-43.

101] Bazerman MH, Loewenstein GF, White SB. Reversals of preference in allocation decisions: Judging an alternative versus choosing among alternatives. Administrative Sci Q 1992;37:220-40.

102] Nowlis SM, Itamar S. Attribute-task compatability as a determinant of consumer preference reversals. J Mark Res 1997;34:205-18.

103] Hsee CK, Blount S, Loewenstein GF, Bazerman MH. Preference reversals between joint and separate evaluations of options: A review and theoretical analysis. Psychol Bull 1999;125:576-90.

104] Bohnet I, van Geen A, Bazerman MH. When performance trumps gender bias: Joint versus separate evaluation. John F. Kennedy School of Government, Harvard University: HKS Faculty Research Working Paper Series RWP12-009; 2012.

105] Bartusch C, Wallin F, Odlare M, Vassileva I, Wester L. Introducing a demand-based electricity distribution tariff in the residential sector: demand response and customer perception. Energy Policy 2011;39:5008-25.

106] Snook J. Boomgard E. Driving sustainable behavior in the mainstream consumer: leveraging behavioral economics to minimize household energy consumption. Duke University; 2011.

107] Dutschke E, Paetz A-G. Dynamic electricity pricing-Which programs do consumers prefer? Energy Policy 2013;59:226-34.

108] Paetz A-G, Dutschke E, Fichtner W. Smart homes as a means to sustainable energy consumption: A study of consumer perceptions. J Consum Policy 2012;35:23-41.

109] EYGM Limited. Reconnecting with customers. Utilities unbundled: new perspectives on power and utilities. Ernst & Young; June 2013. p. 10-19.

110] EygM Limited. In utilities we trust? How can energy providers win back the trust of their customers? Ernst & Young Limited; January 2013.

111] Keaveney SM. Customer switching behavior in service industries: an exploratory study. Journal of Marketing. 1995;59:71-82.

112] Brutscher P-B. Payment matters? An exploratory study into pre-payment electricity metering. University of Cambridge; 2011. p. 1-49.

113] Choi S, Mattila A. Perceived controllability and service expectations: influences on customer reactions following service failure. J Bus Res 2008;61:24-30.

114] Toh E, Low V. Promoting competition in electricity retail: Insights from behavioural economics. In: College CS, editor. Behavioural economics and policy design: examples from Singapore; 2011. p. 87-99.

115] Faruqui A, Palmer J. Dynamic pricing and its discontents. Regulation 2011;34:16-22.

116] Davis JJ. The effects of message framing on response to environmental communications. Journal Mass Commun Q 1995;72:285-99.

117] Loroz S. The Interaction of message frames and reference points in prosocial persuasive appeals. Psychol Mark 2007;24:1001-23.

118] Cheng T, Woon DK, Lynes JK. The use of message framing in the promotion of environmentally sustainable behaviors. Soc Mark Q 2011;17:48-62.

119] Obermiller C. The baby is sick/the baby is well: a test of environmental communication appeals. J Advert 1995;24:55-70.

120] Pollitt MG, Shaorshadze I. The role of behavioural economics in energy and climate policy. In: Fouquet R, editor. Handbook on energy and climate change. Cheltenham, UK: Edward Elgar; 2013. p. 523-46.

121] Lineweber DC. Understanding residential customer support for - and opposition to - smart grid investments. Electr J 2011;24:92-100.

122] Lich J, Mercer T, Tinjum A. How Baltimore gas and electric is solving the dynamic pricing puzzle. OPower. Retrieved online from: «http://blog.opo»; 2014.

123] Cases A-S. Perceived risk and risk-reduction strategies in Internet shopping. Int Rev Retail, Distrib Consum Res 2002;12:375-94.

124] Pires G, Stanton J, Eckford A. Influences on the perceived risk of purchasing online. J Consum Behav 2004;4:118-31.

125] San Martin S, Camarero C. How perceived risk affects online buying. Online Inf Rev 2009;33:629-54.

126] Schiffman LG, Kanuk LL. Consumer behavior. 10th ed.. Englewood Cliffs, NJ: Prentice-Hall; 2009.

127] Lantos GP. Consumer behavior in action: real-life applications for marketing managers. New York: M. E. Sharpe, Inc; 2011.

[128] Silva V, Stanojevic V, Aunedi M, Pudijanto D, Strbac G. Smart domestic appliances as enabling technology for demand-side integration: Modelling, value and drivers. In: Jamasb T, Pollitt MG, editors. The future of electricity demand Customers, citizens and loads. New York: Cambridge University Press; 2011. p. 185-211.

[129] Salies E. Real-time pricing when some consumers resist in saving electricity. Energy Policy 2013;59:843-9.

[130] Faruqui A, Hajos A, Hledik RM, Newell SA. Fostering economic demand response in the Midwest ISO. Energy 2010;35:1544-52.

[131] Lambrecht A, Tucker C. Paying with money or effort: pricing when customers anticipate hassle. J Mark Res 2012:66-82.

[132] Gyamfi S, Krumdieck S. Price, environment and security: Exploring multimodal motivation in voluntary residential peak demand response. Energy Policy. 2011;39:2993-3004.

[133] Hackman JR. Group influences on individuals in organizations. In: Dunnette MD, Hough LM, editors. Handbook of industrial and organizational psychology. Palo Alto: Consulting Psychologists Press; 1992. p. 234-45.

[134] Allcott H. Social norms and energy conservation. J Public Econ 2011;95:1082-95.

[135] Allcott H, Mullainathan S. Behavioral Science and Energy Policy. Cambridge, MA, USA: Working Paper; 2010.

[136] Allcott H, Mullainathan S. External validity and partner selection. NBER working paper series no. 18373; 2012.

[137] Ayres I, Raseman S, Shih A. Evidence from two large field experiments that peer comparison feedback can reduce residential energy usage. NBER working paper 15386. Cambridge, MA, USA: National Bureau of Economic Research; 2009.

[138] Costa D.L., Kahn M.E. Energy conservation "nudges" and environmentalist ideology: Evidence from a randomized residential electricity field experiment. NBER working paper series no. 15939; 2010.

[139] Blanton H, Stuart AE, Van den Eijnden R. An introduction to deviance-regulation theory: The effect of behavioral norms on message framing. Personal Soc Psychol Bull 2001;27:848-58.

[140] Blanton H, Christie C. Deviance regulation: a theory of action and identity. Rev Gen Psychol 2003;7:115-49.

[141 ] Hu N, Liu L, Zhang J. Do online reviews affect product sales? The role of reviewer characteristics and temporal effects Inf Technol Manage 2008;9:201-14.

[142] Campbell MC. Perceptions of price unfairness: antecedents and consequences. J Mark Res 1999;36:187-99.

[143] Deloitte access economics. residential electricity tariff review. Report commissioned by the Energy Supply Association of Australia. Kingston, ACT: Deloitte Access Economics; 2014.

[144] AEMC. Power of choice review-giving consumers options in the way they use electricity. Sydney, Australia: Australian Energy Market Commission; 2012 Final Report.

[145] Thompson L, Gentner D, Loewenstein G. Avoiding missed opportunities in managerial life: Analogical training more powerful than individual case training. Organ Behav Hum Decis Process 2000;82:60-75.

[146] Idson LC, Chugh D, Bereby-Meyer Y, Moran S, Grosskopf B, Bazerman M. Overcoming focusing failures in competitive environments. J Behav Decis Mak 2004;17:159-72.

[147] Lambrecht A, Skiera B. Paying too much and being happy about it: existence, causes, and consequences of tariff-choice biases. J Mark Res 2006;43:212-23.

[148] Uhrich F, Schumann JH, Wagenheim F. The impact of consumption goals on flat-rate choice: can "hedonizing" a service increase customers' propensity to choose a flat rate? J Serv Res 2012;16:216-30.

[149] Schulz K, Altman D, Moher D. CONSORT 2010 statement: updated guidelines for reporting parallel group randomized trials. Ann Intern Med. 2010;152:726-32.

[150] Solomon P, Cavanaugh MM, Draine J. Randomized controlled trials: design and implementation for community-based psychosocial interventions. New York, US: Oxford University Press; 2009.

[151] Kendall JM. Designing a research project: randomised controlled trials and their principles. Emerg Med J 2003;20:164-8.

[152] Sibbald B, Roland M. Understanding controlled trials: why are randomised controlled trials important? Br Med J 1998;316:201.

[153] Allcott H, Rogers T. How long do treatment effects last? Persistence and durability of a descriptive norms intervention's effect on energy conservation. HKS Faculty Research Working Paper. RWP12-045; 2012. p. 1-35.

[154] Allcott H, Rogers T. The short-run and long-run effects of behavioral interventions: Experimental evidence from energy conservation. NBER working paper. Cambridge, MA: National Bureau of Economic Research; 2012. p.1 -63.

[155] Brewer MB. Research design and issues of validity. Handbook of research methods in social and personality psychology 2000:3-16.

[156] Nichols AL, Maner JK. The good-subject effect: investigating participant demand characteristics. J Gen Psychol 2008;135:151-65.

[157] Orne MT. On the social psychology of the psychological experiment: With particular reference to demand characteristics and their implications. Am Psychol 1962;17:776.

[158] Haynes L., Service O., Goldacre B., Torgerson D. Test, learn, adapt: developing public policy with randomised controlled trials. UK Cabinet Office, Behavioural Insights Team London, United Kingdom; 2012.