Keywords

4.1 Introduction

This final empirical chapter provides an intentionally different perspective. Specifically, we leap into an empirical study that is markedly different from Chap. 2 (on the funder-driven whole systems agenda) and Chap. 3 (on the establishment of long-term, large-scale, multi-stakeholder energy research centres). Herein, we investigate how many of the questions and issues that were generated by these two previous chapters—around high-level commitments/investments to particular visions of interdisciplinary working—can also point us to better understandings of more conventional projects that, whilst addressing inherently multifaceted problems, still use monodisciplinary research tools. Indeed, we strongly assert that there is much to be learnt around these issues, including the different implied epistemic cultures within one discipline (in this case Power Systems Engineering), the impacts of funding to applied projects that serve regulatory policymaking and are hence cross-professional, the ways in which specific boundary objects (in this case, ‘prices’) mediate social worlds in research project settings, and ultimately how all such issues may shape or be shaped by surrounding disciplines and epistemics, even when not consciously considered within a relatively rigid monodisciplinary stance. This all matters, not least because the calls for and touted merits of interdisciplinarity are inevitably relative to its dominant monodisciplinarity cousin.

This chapter investigates such issues by exploring the outputs and experiences of a large, monodisciplinary, Finnish energy research project that was set up in the mid-2000s. The project sought out to study how much reliable electricity supply is ‘worth’ to the energy end-users, by assigning this reliability a financial price. The scholars, mainly from the Engineering Sciences, were given this task by the public authorities (then Finnish Ministry of Trade and Industry) and power companies that wanted to use the ‘price’ in the new regulation model of electricity utilities in Finland. The project was called KAH, short for “Keskeytyksestä aiheutuva haitta” and Finnish for “harm caused by an interruption (of electricity supply)”. We henceforth refer to the project by its Finnish acronym, KAH.

The materials in this chapter stem from several sources, which indicate a double role as both project participant and its analyst—this is typical for Social Scientists in interdisciplinary projects, as we have outlined before in this book. First, one of the authors worked as a Research Assistant in KAH, gathering its data, conducting its surveys electronically and by phone, presenting the results to the funding body, and drafting the final report. Many of these reports and sources relevant to the KAH projects are also our materials in this chapter. Second, the author has examined this same project in a much larger study (Silvast 2017), whose related subsets have been published earlier (Silvast and Virtanen 2019). This larger study involved some 30 interviews with Finnish electricity professionals and laypersons, and participant observation (Silvast 2018). Herein, though, we take this study in a new direction by interrogating the calculations of ‘price’ for infrastructure reliability as disciplinarily distinct research on a public energy policy problem. However, for the purposes of this chapter, only selected parts of these larger datasets are used. We use these interview and personal experience data to enable a richer, deeper consideration of the dynamics in play.

The structure of this chapter is as follows: we begin by presenting background context with regard to the economic-theoretical rationale for market regulation, which provides the foundations for the Finish project (Sect. 4.2). Further context is then offered on the importance of minimising blackout disruptions within the Finnish energy policy landscape (Sect. 4.3). Following this, we discuss the KAH project in more detail, with particular attention given to its production and use of cost estimates, in the context of, for example, objectivities, power dynamics, science-policy translations, and interdisciplinarity roadblocks (Sect. 4.4). We conclude by directly discussing this chapter’s relevance for the pertinent issues being collectively generated by the three core empirical chapters (Chaps. 2, 3, and 4)—this cumulative progression in argument will, we hope, act as an appropriate stepping stone on route to Chap. 5, where we conclude the book with our presentation of a Sociology of Interdisciplinarity (Sect. 4.5).

4.2 What Is Market Regulation and How Does It Address Public Interest?

To understand the project context that sets the stage for this chapter, we must first visit the economic theory about energy networks and its market regulation in particular. Market regulation can be generally understood as a “negative feedback system in which asymmetries, whether due to social, economic or technological factors are balanced by rules which may (or may not) be codified in legal statutes” (Boyd 2001, p. 4). Separate from ministries and legislators in many countries, market regulators have “quasi-legislative power by allowing them to make rules having the force of law” (Hirsh 1999, p. 31). Hirsh (2004) explains how regulators emerged in the US electricity industry in the early twentieth century, first and foremost to guarantee that electricity networks have adequate financial integrity. However, they also therefore enacted the notion that these networks are natural monopolies (i.e. services for whom there are no competitors):

[T]he laws required regulatory commissions to maintain the financial integrity of utilities so they could expand their networks. To accomplish this goal, regulators ensured that utilities earned enough money to be able to pay investors attractive dividends and bond yields. At the same time, regulators guaranteed that competitors would not infringe on the franchise areas of utilities, thus explicitly codifying the notion that power companies constituted natural monopolies. (Hirsh 2004, p. 113)

In practice, hence, regulators are meant to ensure that public utilities (e.g. electricity networks) provide desired quality to their customers and are financially able to deliver that quality, especially when they will have no competitors. To address this ‘natural monopoly problem’, regulators have different regulatory instruments that they can use. Indeed, one previous piece of work (Boyd 2001, pp. 62–65) summarises a whole range of such instruments, some of which include persuasion and appeals to the public through the media, public operation, and ownership (instead of managing utilities privately); franchise bidding (running utilities as protected franchise for a set time); and whole deregulation (based on an assumption that gains from regulation are small compared to the costs of doing it). This said, the most common regulatory instruments are those focused on profits and pricing: namely, profit regulation, where utilities are allowed a specific rate of return, and price control regulation, where the prices to utility customers are controlled by the regulator.

As these various mechanisms show, regulation concerns far more extensive problems than merely efficiency and asset values. Regulation itself opens a set of issues around what is essentially a social contract between publics and utilities: where the utilities provide services that the public needs and the regulatory framework allows those services to be accomplished financially (Boyd 2001, p. 64). Questions around how exactly to govern these public interests, as well as what the public may need, have shifted over time, thereby introducing (typically monodisciplinary-framed) problems that are apt for study by interdisciplinary energy Social Sciences and Humanities (SSH) scholars—nevertheless, this topic has remained almost entirely underexamined, as we now go onto to discuss in Sect. 4.3.

4.3 Long Blackouts and Their Regulatory Impacts in Finland

In the previous section, we made clear how traditional economic theory emphasises how market regulation is meant to ensure a quality service provision. In building on this, for the rest of this chapter, we use the problem of electricity quality of supply as an exemplar of public interests in reliable infrastructures, and we focus our enquiry on Finland. This section provides the supporting policy context for Finland.

Finland is a northern European state that has been facing difficult and long electric power failures for the past several decades (Silvast 2017). These have generated an active political and public debate on protecting the electricity infrastructure and making it as riskless as possible for blackouts. We assert that, in practice, it has been the Finnish public market regulator (paired with research-based insight) that has sought to translate these public interests into the economic values that the utility companies can act upon.

Difficult electric power failures have been a long-standing policy issue in Finland. Indeed, we can point to five past examples where they arose in public discourse and led to specific calls for market interventions. First, back in 2001, two exceptionally strong storms—Pyry and Janika—struck Finland and doubled the yearly number of electricity supply interruptions compared to the whole previous decade (Kauppa-ja teollisuusministeriö 2006, p. 29). In 2002, a publicly commissioned report on these blackouts (Forstén 2002) recommended that Finnish energy end-users receive compensations from all blackouts that last longer than 12 hours, and this became operational in the Electricity Market Act in 2003. The aim of this compensation entitlement was to “motivate electricity distribution owners to act in a manner that shortens the duration of interruptions” (Forstén 2002, pp. 31–32), and to this end, the report also suggested a maximum duration of six hours for an electricity interruption “even in exceptional conditions” (Forstén 2002, p. 2).

Second, four years later, the Finnish Ministry of Trade and Industry suggested that customers, or “entrepreneurs” critically dependent on electricity, purchase their own private emergency power generators (Kauppa-ja teollisuusministeriö 2006, p. 56), although a similar idea had been afloat since 2002:

Uninterruptable electricity distribution cannot be guaranteed. If the customer’s production or other activities do not tolerate reasonable electricity distribution interruptions, then the customer should personally secure the electricity supply. (Forstén 2002, p. 35)

Third, as unattainable as risk-free electricity distribution may be, since 2008 Finnish electricity network companies became additionally penalised financially for each electricity blackout, according to market regulation (Energiamarkkinavirasto 2007). This penalisation was done by linking power supply failures to the allowed yearly profits of the utilities. This hence supposedly gave them a financial incentive to improve quality levels.

Fourth, a major Finnish storm on Boxing Day 2011 initiated a blackout that momentarily affected 570,000 customers and lasted for days for tens of thousands of customers (Energiateollisuus 2012). Immediately afterwards, the power failure led to an untypically wide public debate concerning the crisis communication and crisis preparedness among private energy companies, the impacts of the outsourcing of their maintenance, and the necessity of preventing similar storm damages in the future by burying electrical cables—thereby considerably increasing monetary compensations for customer damages from blackouts.

Fifth, a new Electricity Network Law was enacted in 2013 (Electricity Market Act [9.8.2013/588]), stipulating that electricity network companies must make preparedness plans and set maximum durations for electric power failures. In order to meet these requirements, the network companies would have to invest over three billion euros into their networks (according to the Finnish national broadcasting company Yle 2020), most often by burying their distribution cables underground.

In all five of these examples, the public common effect of a blackout is transformed into a calculable risk in order to create a fair, transparent, and market-based way of distributing harms across all energy consumers.

The Energy Authority that regulates Finnish electricity companies rose to these challenges from a particular viewpoint. It was concerned with how all electricity network companies in Finland, especially the comparatively smaller companies operating in rural areas, could produce the necessary investments. Hence, its new regulatory model (from 2016 onwards) raised the allowed profits of electricity network companies. The result—scrutinised in a study by the Finnish national broadcasting company (Yle 2020)—was an increase in electricity distribution tariffs among many electricity end-customers; including those served by urban electricity companies, where power failures were less common and investment needs smaller than for rural areas. These higher tariffs were the subject of a public debate for several months.

The Energy Authority has defended the new model by stating that the rules should be equal to all network companies in Finland, especially since the model may have been challenged in market courts had this not been the case (Yle 2020). Thus, what is at stake is not only the regulatory formula—which has remained almost entirely opaque in the public debate—but also overarching questions concerning how to incentivise infrastructure providers to ensure service reliability, how to deliver fair profitability among natural monopolies, and how much electricity customers are willing to pay for improved reliability of infrastructures that they critically depend upon.

Yet to paraphrase questions raised several times in this book before: how can the regulatory experts know what the public needs and what their energy demand is like, or which parts of their activities most critically depend on functioning infrastructure? This is clearly a research problem where the SSH disciplines should have much to offer. A revealing quotation was given by Professor of Electric Power Technology, Pertti Järventausta, who was interviewed for the Yle (2020) report. Järventausta argued that the market regulator had concentrated on the network companies but had not taken a whole systems perspective that would have included impacts on the customers:

The authority viewed this issue strongly from the perspective of network companies, so that they one would enable network investments and fulfill security of supply requirements. But now one forgot to conduct a holistic examination, what will this lead to in the customer end? (Järventausta, quoted in Yle 2020, no pagination)

This observation was clearly based on commissioned research Järventausta had co-conducted, where the first author was involved as a Research Assistant. This research study set to find out what indeed the customers need in terms of reliability, rather than taken that as a given.

4.4 Pricing Reliability in the Regulatory Model

This section briefly presents details of the original research study, KAH, which used monodisciplinary tools to examine what is a wide public matter: how members of the public value the importance of a critical infrastructure and how those valuations can be turned into prices that can be subsequently used in regulatory models. While such issues are often reported in project reports and publications, we offer unique access to them by drawing on the first author’s own experiences in the project.

4.4.1 Connecting the Original Research Study to the Themes of This Book

Between 2004 and 2005, the first author was the Research Assistant for a large-scale research study that set to cover how laypersons perceive electricity blackouts (Silvast et al. 2006). It was commissioned by the Finnish Ministry of Trade and Industry and several power companies operating in Finland.Footnote 1 It was conducted by Electrical Technology Departments of two technical research universities: the Helsinki University of Technology (today merged with Aalto University) and the Tampere University of Technology (today merged with Tampere University).

The way to address power failure risk is commonly called Value of Lost Load, which is essentially a monetary estimate of the damage caused by power interruptions. In Finland, the monitoring of this damage in turn depends on information from energy users to uncover the measured economic worth of reliable energy supply. Such values are assessed in surveys like the one in Fig. 4.1, which we conducted in the study. Filled with questions about multiple blackouts and their economic effects, the survey assumes that all energy users are rational economic actors that calculate the value of energy use and the financial risk of electricity blackouts. The survey received some 1500 responses and its outcome was a complex set of averages of blackout values, across different kinds of customers (including households, agriculture, public sector, and industries, and including summer cottages as a typical Scandinavian category).

Fig. 4.1
figure 1

A Finnish customer survey, 2004, asks what power cuts cost. More than a dozen similar questions are given in the survey. (Source: Silvast et al. [2006, p. 104])

Full size image

This survey (Fig. 4.1), as such, was not an exercise in interdisciplinary energy research. For example, it draws upon notions of rational consumers—common also today, for instance, in aspirations to make the electricity consumption more flexible by introducing dynamic time-of-use tariffs—that have been shown to be widely inadequate representations in social scientific research (Christensen et al. 2020). Further, this simplified survey did not always yield useful results, which may have been explained by the embedding of energy in everyday life. The household consumers, in particular, were not always able to estimate the monetary harms, especially in terms of short-lived interruptions. A report on the survey findings even notes this issue and, rather than including diverse viewpoints, ends up justifying the removal of ‘outliers’ that could not be explained by statistical averages:

79 % of respondents estimates the damage of one-hour unexpected power cut to be zero or did not respond at all. At the same time, 10% of responses in the same part was more than 100 euros. The largest response to this question as 1,600 euros … To find representative averages, the material had to be trimmed. We removed 10% of the biggest and the smallest responses, so that the average would represent the majority’s responses. (Silvast et al. 2006, p. 47)

During project discussions, these outlier responses were commonly referred to as “subjective”, as opposed to the seemingly “objective” answers that could be given by businesses, agriculture, and the industries on what damage power interruptions would cause. In fact, in the research group’s scientific reporting on the study, the “subjective” answers remained mostly unaddressed. The publications and conference papers (Kivikko et al. 2007, 2008) turned to detailed statistical analyses and reporting of averages. An example is from the study results for the residential sector (Kivikko et al. 2007, p. 3): an unexpected electric power failure of 1 second would ‘cost’ 0.23 €/kW, for 2 minutes 0.84 €/kW, 1 hour 5.8 €/kW, and so forth up to 36 hours (costing 147.60 €/kW). The conference paper that includes these figures does not comment on the numbers and how they were formed in any manner, other than one noteworthy observation: that as can be seen, the householders’ Willingness to Pay (WTP, i.e. how much more they would pay for reliable electricity at 1.10 €/kW for an hour’s blackout) is only a fraction of the Willingness to Accept (WTA, i.e. how much more risks they would accept for a cheaper tariff at 8.30 €/kW for a similar blackout). In simple terms: if householders were perfectly rational economic agents as envisioned by economics, their WTP and WTA should be identical or very close to one another. The paper does not take on this difference and discuss it in more detail, although it does seem to indicate that people expect reliability to be higher than the actual costs of electricity that they pay for.

The first author did conduct some research on his own, interviewing and surveying householders (reported in Silvast 2017), which showed there were various kinds of blackouts and different people had a variety of responses to them. The acceptance of a blackout varied according to gender, to age, to region, and especially to memory about past blackouts. To be acceptable, a blackout also had to feel ‘voluntary’, rather than imposed from above. Such an acceptable electricity supply interruption, even though anticipated, should not halt those household practices that were perceived as important—and it also did not prevent less significant practices regularly or permanently.

In fact, temporality explained the seriousness of the blackout in at least three senses. First, a blackout should not interrupt everyday routines on a regular basis. Second, a blackout should not occur at a time when people have planned to do something else that requires functioning electricity. Finally, a blackout should not impact on tangible objects which are the result of time and investment (e.g. contents of a freezer and computer’s hard disk drive).

Nonetheless, it was not possible to include these kinds of qualitative accounts as part of the scientific content of the report. In an interesting indication of interdisciplinary working and hierarchies of disciplines, there was an allowance to categorise the open qualitative responses that were given at the end of the survey study. However, even then, these categorised findings only became an Appendix (Silvast et al. 2006) in the final output, where they received only one table on one page of the 175-page report. The difficulty of qualifying the quantities of power cut damage persists in this research problem and also so in the regulatory domain. Undoubtedly, the monodisciplinary framing of this study held itself stubbornly strong throughout the project journey, including during the writing-up process discussed here.

4.4.2 How Did the Regulator Use the Cost Estimates?

The Finnish KAH study from 2006 (Silvast et al. 2006) was commissioned amid a change in the market regulatory model, which concerned not only Finland but European energy regulators at large. In digging deeper into the relevance and implications of this change for the Finnish economic regulation study that we have been discussing thus far in this chapter, we now draw upon CEER’s (The Council of European Energy Regulators’) published overviews of the European regulatory practices. CEER is a co-operation body of European national electricity and gas regulators.

According to CEER, many European countries’ electricity regulation shared a common starting point until around 2000. This is because regulators operated through assigning price caps for the electricity network service that is billed from customers (CEER 2005, p. 31). Soon, however, the regulators noted that while managing one risk (overpricing), this mechanism created another risk. Specifically, even if prices are capped, electricity network companies might reduce their maintenance and investments to make a profit. And according to a popular line of thinking by economists (Gramlich 1994), lack of investment directly influences the quality of infrastructure provision: “Price-cap regulation without any quality standards or incentive/penalty regimes for quality may provide unintended and misleading incentives to reduce quality levels” (CEER 2005, p. 31).

New electricity regulation models, which are increasingly popular in Europe since 2005, strive to monitor and motivate improvements in this quality (CEER 2005, pp. 31–32). In practice, the theory would say that this means: statistics of quality are made public; “incentive” and “penalty” schemes are enforced so that utility companies control their profits in terms of their quality of supply; and a growing number of arrangements emerge that fix maximum durations for electricity blackouts and customer compensations for cases when the durations are not met. Along with compensations, however, the matter has also been about making customers aware of the costs of quality. Thus, specific emphasis has been paid to electricity customers’ “expectations” and “their willingness to pay” for good-quality electricity (CEER 2011, p. 4). As has been summarised previously, “[r]esults from cost-estimation studies on customer costs due to electricity interruptions are of key importance in order to be able to set proper incentives for continuity of supply” (CEER 2010, p. 9). This is exactly the research problem that the original commissioned study (Silvast et al. 2006) set to address.

We now move to an example of concrete regulatory formula, to demonstrate how these issues are turned to activities in the regulatory profession. Specifically, between 2008 and 2011, the Finnish energy network quality regulation depended to a large part (although other measures were also deployed) on a method called Data Envelopment Analysis (DEA).Footnote 2 Developed in the US and situated in a scientific tradition called operational research, the method calculates the technical efficiency of multi-output, multi-input production units or decision-making units (Charnes et al. 1978). In so doing, it compares different decision-making units with one another, identifying the most efficient unit relatively and, in most contemporary applications, prescribing how the other units may improve their efficiency by altering their input, output, or both. While these aims may sound “neoliberal”, the method’s original intention was different: the DEA was developed to study “public programs” and “decision making by not-for-profit entities rather than the more customary ‘firms’ and ‘industries’”, and it depended on data “not readily weighted by reference to (actual) market prices and/or other economic desiderata” (Charnes et al. 1978, p. 429). Undoubtedly, such traits also made the method appealing to measure public electricity utilities and their efficiency.

This Finnish electricity regulation DEA model has the following formula for technical DEA ‘efficiency’, by which we mean: a utility’s yearly outputs divided by inputs (Energiamarkkinavirasto 2007, p. 53):

$$ \mathrm{Data}\ \mathrm{Envelopment}\ \mathrm{Analysis}\ (DEA)=\frac{\begin{array}{l}{u}_1\ast \mathrm{Energy}+{u}_2\ast \mathrm{Network}\\ {}\kern2em +{u}_3\ast \mathrm{Customers}\end{array}}{v_1\ast \left(\mathrm{OPEX}+ TP+ KAH\right)} $$

Most of the inputs at this bottom of equation and the outputs at top of this equation are relatively common sense. Factors like a utility’s operational expenses (OPEX, in the above formula) and property value depreciation (TP) are obviously a ‘cost’ from the quality point-of-view. A utility’s ‘productions’ include the financial value of distributed electricity during a year (Energy), as well as the length of the utility’s electricity network (Network) and the number of customers served by the utility (Customers) to normalise the utility’s size. The variables—u1, u2, u3, and v1—are altered during a linear optimisation to maximise ‘efficiency’ relative to other utilities.

Along with these parameters, however, the customer’s costs from blackouts (KAH ) are an input. What does this mean in practice? Such costs have been first gathered by means of the surveys described in the previous sub-section (Fig. 4.1). Based on these, the Finnish regulation model then concluded on the ‘pricing’ for blackouts (Table 4.1). For example, an unexpected electricity blackout would cost €1.10/kW of lost customer electric power and €11.00/kWh of lost customer electric energy. Other costs were assigned to planned interruptions and reclosing operations used by utilities to protect their systems that cause short-lived blackouts.

Table 4.1 The regulatory ‘pricing’ of electricity blackouts in Finland between 2008 and 2011
Full size table

It is worth noting that the figures are not using entirely the same units and are not in the magnitude of the figures reported by Kivikko et al. (2007). There was a process of translation between the scientific work and regulation, where the regulator needed figures for whole of Finland and did not disaggregate them to different kinds of customers. It is very difficult to find out how one cost became the other, and we assume that complex negotiations among electricity stakeholders took place although have no direct evidence of them.

Nevertheless, in the DEA input-output framework, such partly-researched, partly-constructed blackout ‘costs’ are then combined with managing electricity risk: the more costly the blackouts the customers have had, the more electricity the utility now has to distribute, or the less expense and property it has to have in order to appear ‘efficient’. What emerges is a loop between customers’ risk perceptions and potential for profit. This loop is furthermore performative: between 2008 and 2011, the Finnish Energy Market Authority set each electric utility an efficiency target to a large part based on a DEA formula (Energiamarkkinavirasto 2007, p. 49).

The “harm” (as per the project team) of a blackout and the techniques of its measurement have, we discovered, their own history. Often the “harm” it referred to was as the value of non-delivered electricity, not customer interruption harm like in Finland. According to an infrastructure and electricity expert, who was familiar with the first Finnish studies that concerned these harms decades ago:

the term interruption harm indicates that the customers experience blackouts as a harm and they should assess it. The perspective has not been similar elsewhere in the world, and one still hears talk about NDE [Non-Delivered Electricity] or such. Previously, in Finland, such NDE values were calculated without asking customers. (Man, 60–69 years, National authority)

But the assumption in the DEA model is the opposite to NDE: customers are asked to calculate the level of risk and these are factored in as an input variable. All answers, or their averages, play a part in minimising electricity risk and distributing harms.

4.5 Conclusions

In this chapter, we examined the workings of translating values of public interest in electricity distribution between scientific research and market regulation. In doing this, we drew on a past Finnish energy research project—KAH, short for “Keskeytyksestä aiheutuva haitta” and Finnish for “harm caused by an interruption (of electricity supply)”—which was led by Power Systems Engineers who held an interest in economics (but were definitely not economists).

This chapter was intentionally different from the ones that preceded it, given that the KAH project was not interdisciplinary in its remit. To repeat, though, we would emphasise that monodisciplinary projects are worthy of exploration from an interdisciplinarity perspective. This is primarily because such projects do not exist in isolation: other disciplines, methods, and academic communities connect, especially when wide matters of public importance (e.g. functioning of energy infrastructure) are at stake. Acknowledging this, immediately allows for relevant cross-disciplinary questions to be posed, even to conventional and (some might say) narrow monodisciplinary approaches.

We want to now tie in this purposively different case to themes that have been emerging from this book thus far. First, we have demonstrated the obvious importance of disciplines. Indeed, the whole KAH exercise of valuing public interest in infrastructure reliability was seemingly conducted within one discipline: that of Power Systems Engineering. Yet, whilst it did not mention other disciplines by name, it was clear that during its analytical phase some type of applied economics (i.e. the study of prices and costs) had assumed the place of critical-SSH. That is to say, the notion of there being a ‘cost’ for reliability was a proxy for the public importance of functioning infrastructure, and this cost was not merely postulated in theory but became the topic of KAH’s detailed empirical enquiry. This finding is not unique to the relatively esoteric topic of costs of reliability; it also appears, for example, when the national potential for energy efficiency and energy saving is translated to energy intensity (i.e. the ratio between energy output and Gross National Product). In all of these cases, many details (e.g. everyday practices of energy use and social norms) could not be acknowledged by the disciplinary scientific tools being used. We argue that it very much matters that economics was parachuted in to cover the societal elements of KAH, without due thought for its implications—as is emphasised by the common belief of SSH researchers that economics is ill-equipped to conceptualise and investigate matters of social order, and thus is why economics is fundamentally regarded as not being an SSH discipline (Foulds et al. 2017).

Second, however, while only one discipline worked on this topic, it did so by using different research tools and approaches, which were not always compatible. That is to say, we suggest different ‘epistemic cultures’ were at play, although this finding is not based on conventional ethnography (Knorr Cetina 1999). As has been seen, the scientific version of statistics in the universities was different from the more applied economic regulatory models that the regulator wanted to create. For instance, the DEA regulatory model mentioned was driven forward by very different requirements (e.g. aggregated prices at the national level), whereas the Engineers in the project wanted to subject prices to detailed empirical enquiries. Even within KAH, they did not seem to agree on the level of detail and statistical sophistication necessary, with viewpoints diverging from the more pragmatic (what was needed to, for instance, complete the research) to the more explicitly scientific (what was needed to, for instance, publish in conference papers).

Third, the disciplinary concerns of the SSH—such as consumer research of electricity use—were wholly taken on board by a statistical style of reasoning, suggesting the dynamics of ‘appropriation’ (Forsythe 1999). Applied statistics also ended up essentially eliminating the putative ‘subjective’ answers that did not fit into the model of the rational consumer. This finding may relate to paradigms in scientific research; for example, the paradigm of Power Engineering cannot account for subjectivity; therefore, the more subjective observations can only appear as anomalies to the scientific method. This is interesting insofar it was not permitted to use these subjective answers within the KAH project, other than to a comment on methodological weaknesses, and there was no person in the KAH project dealing with the issue (including from a conventional economics, let alone SSH, perspective). Whilst it is true that the first author was allowed to work on the topic, this seemed to be under the normal arrangements of project work: the instrumental project work was prioritised and had to be completed first, and it was clearly the case that the study of ‘subjective’ answers was something he could write an academic thesis on (he was an undergraduate at the time). Even this arrangement, though, was not wholly clear-cut, as the author did, for example, present subjective findings in project board meetings and was invited to give workshops also in other contexts. The intriguing finding is that whilst SSH work was at one time seen as important, it was still nevertheless underprioritised.

Fourth, the cost features as a ‘boundary object’ (Star and Griesemer 1989): it is what mediates between the social worlds of public interest, market regulation models, and scientific research studies. It is also what is used to ‘scope’ out everyday life experience. As is often the case, and almost expected, the readers in key policymaking positions cannot account for the complexities of scientific methodological details, whether it be (in our speculation) due to, for example, lack of time, expertise, or simply an inability to use the methodologies in what they do. It is also important to note that while the costs were a boundary object for the public interest, the public was not allowed to engage in any manner in the issue within the project. The public interest was represented only by the constructed reality of economic models, and by offering economic replies that were quite trivially insensitive to everyday energy demands (i.e. the vital infrastructure of energy and social practices it sustains was visible only as abstract cost calculations). Thus, even while boundary objects do mediate between social worlds, they are a distorted proxy for them.

Fifth, we observed ‘interpretative flexibility’ (Pinch and Bijker 1984): the notion of ‘cost’ is itself quite flexible in how it can be interpreted. One important matter to notice is that while we have talked to ‘cost’ in this chapter, these were perhaps never meant to be ‘actual costs’—although some research project participants probably still thought so. In contrast, a critical-SSH analyst would see the costs as constructed in the research process. For the power companies, the costs were possibly a proxy for their customer interests. For the regulator, we would argue that the costs were ‘performative’, in that they were meant to incentivise the power companies, not (necessarily) be based on real costs from actual consumers, whatever those may be.

Finally, we point to the key role of research funding. The KAH project was commissioned research by the Finnish Ministry of Trade and Industry and several Finnish electric power distribution companies. In the KAH project report (Silvast et al. 2006, p. 33), these were simply referred to as the “research funders”. The funders did not generate an explicit interdisciplinary agenda, but it would be important to study what kind of an agenda they did create and how that might have impacted the project content. For instance, the KAH project had a steering board with representatives from the Ministry and the power companies that would actively steer the research work. We argue that the funders actively co-created the project, which is indeed more generally the case in applied technical settings, such as this. This situation furthermore made the inclusion of SSH highly difficult as a practical matter. Those engaging in interdisciplinary research should pay close attention to the restrictions imposed by (the required process of constructing) the project’s design, as that can have tangible impacts on what kind of project work is considered to be relevant and possible, even if academically we know the issues demand interdisciplinary interrogation.