Abstract
Big data have become a driver of innovation in multiple sectors, including the management of infrastructures employed for the provision of essential goods and services, such as drinking water. As technology enables new possibilities of action for infrastructure managers, it could be questioned whether the regulations in place still deal adequately with such possibilities or if certain adjustments are necessary, especially considering that infrastructure managers usually operate in highly regulated environments. This study explores the regulatory challenges of introducing smart water meters (SWM) in the Netherlands. In particular, it discusses whether the introduction of SWM will require adjusting the regulations of the sector, to deal with the new possibilities of action enabled by this technology.
Introduction
Big data have become an important driver for innovation with the potential of having a positive impact on economic and social challenges (European Commission, 2018; OECD, 2015). This so-called rise of big data might be explained by two principal factors: on the one hand, the increasing availability of large volumes of data at a low cost, facilitated by information and communication technologies (ICTs); and, perhaps more importantly, the increasing ability of firms and governments to analyze and extract value from the generated data (Mayer-Schönberger & Cukier, 2013, as cited in Prufer & Schottmüller, 2017). Against this background, the OECD issued a report in 2015 in which the term data-driven innovation (DDI) was introduced to refer to “[t]he use of data and analytics to improve or foster new products, processes, organisational methods and markets” (OECD, 2015, p. 17).
Infrastructure is one of the multiple sectors that may benefit from DDI. Employing DDI for the management of infrastructure “enables analysis at unprecedented depth and granularity, as well as targeted interventions in and better management of urban systems” (OECD, 2015, p. 382). With the help of technological developments like smart metering, smart grids, sensors, and data analytics techniques, infrastructure operators can obtain improved and (near-to) real-time information about the condition and operation of the networks they manage. This “smartification” has the potential to facilitate more targeted interventions, reducing expenditures in time and costs, ensuring safer, more resilient and sustainable infrastructure, and delivering better quality service to the general public (Ogie, Perez, & Dignum, 2017).
For example, smart sensors and big data infrastructure enhance the use of maintenance techniques such as data-driven condition-based maintenance or risk-based maintenance (Akkermans, Besselink, van Dongen, & Schouten, 2016) and might even enable predictive maintenance (Mainnovation & PwC, 2017). The result is a more targeted and timelier (“not too early, not too late”) infrastructure maintenance that can lead to an “improved availability of installations, reduction of failure costs and lower costs over the entire life cycle” (Fang, van de Kerkhof, & Lamper, 2018, p. 9).
As infrastructure managers increasingly rely on data to obtain better information about networks, certain changes may arise in the way they perform their tasks. Firstly, implementing DDI requires that infrastructure managers have sufficient resources to collect, transmit, store, and analyze large data sets. That means that in addition to investments in traditional physical infrastructure, or “burying copper in the ground” (Edens, 2017), infrastructure managers must now invest in ICT and other less conventional resources, including “smart devices” and specialized expertise in ICT and data science (see e.g. Rijkswaterstaat, 2018; Vitens, 2017). Secondly, to improve their processes through DDI, infrastructure managers require data from an array of different sources, which include their own systems but also government data and data generated by the consumers 1 of services provided through infrastructure (see OECD, 2015, Chapter 9). Considering that infrastructure managers usually operate in highly regulated sectors (section “Managing infrastructures in the context of heavily regulated sectors” of this article), it might be questioned whether the regulatory frameworks 2 that govern their tasks are still able to cope with the new possibilities of action enabled by the increasing use of DDI or whether eventually they should be revised.
Smart meters are an example of technology that can facilitate the implementation of DDI in the management of utility networks. These meters not only provide more accurate information regarding the consumption of water, electricity, gas, or heating, they also generate information about the functioning of the networks. For example, smart meter data can help to pinpoint failures such as outages or leakages and can also give insights regarding the quality of the service in question (Arniella, 2017; Cervigni & Larouche, 2014; He, Jenkins, & Wu, 2016).
In the European context, a great deal of attention has been paid to the rollout of smart meters for small consumers in the energy sector (electricity and gas), but considerably less attention has been given to smart metering in the drinking water sector. While energy meters are explicitly mentioned in European Union legislation aiming to promote energy efficiency, 3 European legislation applicable to drinking water 4 contains no mentions of any comparable technology. The same situation occurs at the national level in the Netherlands: While a large-scale rollout is taking place with specific regulations for energy smart meters, 5 no national rollout of smart water meters (SWM) is being carried out and no rules have yet been devised regarding the use of this technology.
Nevertheless, some Dutch drinking water companies are already conducting pilots to explore the viability of implementing SWM for household consumers. This provides a good opportunity to start exploring what kinds of regulatory challenges would arise if a rollout of SWM is to be carried out in the Netherlands. To the best of our knowledge, there are no previous studies on this subject. Therefore, this article aims to start a discussion on the implications of increasing digitalization of infrastructures, vis-à-vis the regulatory framework applicable to the Dutch drinking water sector.
In terms of approach, this article looks at the introduction of SWM as a technological development that brings changes to the management of drinking water networks in the Netherlands, in the sense that it enables new possibilities compared to traditional water meters. From that perspective, the article explores whether such changes may require revisiting existing regulations that are in place, both at the microlevel (specific rules applicable to the metering activity) and at the macrolevel (general rules applicable to the drinking water sector).
The method employed for this research combines a review of literature regarding smart (water) meters, DDI, law, regulation and technology, and regulation of network industries, together with an analysis of Dutch and European Union regulations, policy documents, and reports related to the drinking water sector.
The rest of this article is structured as follows: The second section will be dedicated to a review of relevant literature; the third section includes the aspects of the current Dutch regulatory framework considered in this study; the fourth section contains the specific analysis of the regulatory challenges of implementing smart metering for drinking water in the Netherlands. The final section presents the main conclusions of this research.
Literature review
This chapter will introduce the main concepts, as well as the theoretical and contextual background that are employed in this study. It will start by describing the main features of smart meters, their possible uses in the management of water infrastructure, and the current situation in the Netherlands. It will then refer to the possible need to revise regulatory frameworks as a result of technological change as explained by literature on law, regulation, and technology. Finally, it concludes with a discussion of some particular features of the network industries, like the provision of drinking water, which characterize them as highly regulated sectors.
Smart water meters
Smart meters in general can be defined as “a component of the smart grid that allows a utility to obtain meter readings on demand (daily, hourly or more frequently) without the need of manual meter readers to transmit information” (Arniella, 2017, p. 15). They are commonly used for utilities such as electricity, gas, and drinking water.
Although the specific functionalities of SWM vary according to the particular configuration of the system, they can be summarized in three main aspects. Firstly, SWM provide high resolution data that allow sampling water consumption on sub-daily basis (Cominola, Giuliani, Piga, Castelletti, & Rizzoli, 2015). Secondly, SWM are connected to a communication system that allows meter data to be remotely accessible. Thirdly, SWM can provide information about the functioning of the system and eventually the quality of the water itself. These features enable a number of possibilities, such as
6
precise consumption measurement, reducing billing errors and disputes with consumers; monitoring the water system in a timely manner; easing and lowering the cost of meter reading (avoiding manual reading); providing precise data to balance water demand; facilitating prompt leak detection in consumer premises or other parts of the network (e.g. analyzing information generated from a building or a block); prompt detection of theft or other causes of water loss; creating awareness about water conservation and facilitating enforcement of local water restrictions; applying dynamic prices; and additional features may also enable to measure water quality parameters, such as temperature or pressure.
In addition to the opportunities that SWM enable, it is also relevant to mention the main concerns related to this technology. From a legal perspective, the most recurrent concerns related to smart meters are the challenges that they may create for the protection of privacy and personal data (see Baumgart, 2017; Cuijpers and Koops, 2013; Papakonstantinou & Kloza, 2015, regarding energy smart meters). This can be the case because smart meters are installed at the homes of consumers and generate information which can give insights about their private and family life, including behavior, habits, or preferences, which in turn might result in unintended consequences such as profiling or tracking.
Current status in the Netherlands
Unlike network operators in the Dutch energy sector (gas and electricity), Dutch drinking water companies are not obliged to carry out a national rollout of smart meters, and at the moment there are no government-lead initiatives, policies, or studies conducive to a nation-wide implementation of SWM in the Netherlands. Therefore, there is no particular framework that provides guidance to the drinking water companies regarding the implementation of this technology.
Regarding the reasons why there is no large rollout of SWM in the Netherlands yet, it seems that the cost–benefit analysis is not positive at the moment. Firstly, it appears that the cost of implementing this technology national-wide is too high for the time being (Quist, 2013). Secondly, as previously mentioned, there is no national or European Union policy mandating or at least facilitating the implementation of SWM, in contrast with the situation of the same technology in the energy sector. In addition, the price of water in the Netherlands is rather cheap compared, for instance, to electricity, and this may also decrease the interest of consumers in obtaining better insights of their water consumption (Quist, 2013). Finally, the country is not facing serious threats of water scarcity at the moment (Geudens & van Grootveld, 2017; OECD, 2017; Waterbeschikbaarheid en de waterketen, 2018).
However, the availability of fresh water sources may be endangered by extreme climate, like the climate experienced during the spring and summer of 2018, which were registered as the driest ever, breaking a record from 1976 (Voorjaar en zomer van 2018 tot nu toe de droogste ooit gemeten, 2018; Waterbeschikbaarheid en de waterketen, 2018). This could be a game changer which, together with the additional opportunities that SWM brings for a smarter management of the drinking water network, may eventually lead to a systematic rollout, especially if the technology becomes more affordable overtime. According to publicly available information, at least 3 of the 10 Dutch drinking water companies, namely Vitens, 7 Oasen, 8 and Brabant Water, 9 have each started pilots to explore the viability of such technology, mainly for the following purposes: to improve the monitoring of the network, especially leak detection; 10 to increase consumption awareness and stimulate rational use; and to measure factors affecting the quality of the water, such as temperature or pressure.
Technological change and regulation
The introduction of new possibilities of action as a result of technological change 11 can reveal the need to revisit regulations in place and eventually adjust them to deal with new practices and their positive and negative consequences. As explained by Bennett Moses (2013), regulations are designed to operate in an assumed (explicit or implicit) socio-technical landscape. In that sense, to the extent that technological change may enable practices that were previously unknown or irrelevant, such change is at least potentially problematic vis-à-vis the regulations in place (Bennett Moses, 2013, p. 18).
The same author introduced a classification of the main legal issues that may follow technological change, namely: (1) the potential need for laws to ban, restrict, or, alternatively, encourage a new technology; (2) uncertainty in the application of existing legal rules to new practices; (3) the possible over-inclusiveness or under-inclusiveness of existing legal rules as applied to new practices; and (4) alleged obsolescence of existing legal rules. (Bennet Moses, 2007, p. 243)
The first issue refers to the fact that new practices may enable new social dynamics with associated benefits and risks that are not adequately addressed in already existing regulations, revealing a regulatory gap. In this case, the creation of new rules is deemed necessary to limit possible harms and/or to take advantage of technologies that are perceived as beneficial for society (Bennett Moses, 2007; Butenko and Larouche, 2015; Marchant, 2011a).
In the second scenario, there is no immediate need for new rules, to the extent that there are certain rules that should be in principle applicable, but the new context gives rise to uncertainty regarding the application of such rules. The core of the problem lies in the fact that the permissibility of the newly enabled conducts is generally determined by legal categories and concepts which determine what kinds of actions fall within the scope of a certain rule (Bennett Moses, 2007; Mandel, 2017). Thus, the emergence of new possibilities of action enabled by technology may create challenges regarding preexisting categories, in some cases because new practices may not fit neatly in already existing “legal boxes,” 12 and in some other cases because the category in itself becomes contestable. 13
A third issue that might become evident as the result of technological change is a mismatch between the content and scope of a rule and the goal that such rule intended to achieve (Bennett Moses, 2007). The mismatch will result in over-inclusiveness when a certain rule regulates behavior that should not be subject to control (Baldwin, Cave, & Lodge, 2011, p. 70) or will result in under-inclusiveness when a conduct or situation that should be controlled escapes constraint (Baldwin et al., 2011, p. 70). 14
Finally, technological change may render existing rules obsolete in different ways: Technological change may reduce or eliminate the importance of the regulated conduct; the reasons to regulate a certain conduct may disappear as a result of technological change; and technological change may reduce the cost-effectiveness of a certain rule and therefore its enforcement turns into prohibitively expensive (Bennett Moses, 2007, p. 265-269).
The four regulatory issues just described illustrate that technological change can reveal the need to revisit the regulations in place, either because there are no rules to deal with the new possibilities of action (with their associated benefits and concerns) or because the existent rules do not operate as effectively as in the past given the new reality. The importance of addressing such challenges and trying to keep a good connection between practice and the relevant regulatory frameworks relates to the roles of regulation in respect of innovation, which can be summarized in maximizing its benefits (ensuring that there is an adequate regulatory environment for innovation to flourish and steering it in the “right” direction) while reducing possible or actual negative consequences (by prohibiting undesirable innovations or stablishing limitations to the use of innovation) (see among others, Bennett Moses, 2007; Brownsword & Somsen 2009; Butenko & Larouche, 2015). 15
Managing infrastructures in the context of heavily regulated sectors
Water, energy, transport, or telecommunications are considered network sectors characterized by the use of infrastructure such as wires, cables, pipes, roads or railways, or junctions through which such services are offered (Dumaij, van Heezik, & Felsö, 2014, p. 15). According to Bauer (as cited in Knops, 2008, p. 2), some of the most salient features of network industries are high capital intensity and high investment requirements; the vital importance of the services provided by the network industries for individuals and businesses; their strategic value in terms of economic growth and national security; and the indispensability and non-substitutability of the services provided by network industries.
The key value of network industries for society explains why they were initially directly controlled by the state (Baldwin et al., 2011, p. 443; Bruijn & Dicke, 2006, p. 718; Dumaij et al., 2014, p. 15). However, early experiences in the 1980s in the United States and the United Kingdom led many to embrace the idea that dismantling the state monopoly in the provision of network-based services and opening such services to the market would increase efficiency, which in turn gave rise to processes of liberalization and privatization (de Bruijne, 2006, p. 42; Dumaij et al., 2014, p. 15). The loss of direct control by the state suggested, in principle, that there would be more freedom in the market and less subjection to regulations. However, as explained by Vogel, the introduction of more competition is not necessarily accompanied by deregulation, indeed on the contrary, in many cases, the result has been “freer markets and more rules” (Vogel, 1996, p. 3).
Adding to this, while certain parts of the value chain of network industries might be open to competition (e.g. supplying the end customer in the energy sector), the management of distribution networks is usually carried out by natural monopolies because it would be extremely costly if every new market entrant would have to stablish an entire network from scratch (Baldwin et al., 2011, p. 443–444; Dumaij et al., 2014, p. 15). 16
To control possible monopolistic behavior (e.g. limiting output or quality and setting profit maximizing prices) and/or to simulate actual competition at the network management level, different sector specific regulatory instruments, such as economic regulation (e.g. rate of return, price-cap or yardstick regulation) or rules of third-party access, are used (Dumaij et al., 2014, pp. 20–22). In addition, sector-specific regulations require that infrastructure managers ensure that public values such as universal access, safety, reliability, affordability, and sustainability can be achieved. 17
In addition, a complex network of public authorities, including, among others, legislators, ministries, regulatory agencies, regional, and local authorities, is in charge of the creation and enforcement of rules governing the management of infrastructure (see Steenhuisen, 2009, pp. 2–3). In some cases, such a network can be expanded in scope as a result of the internationalization of its governance, as it is the case within the European Union (Groenleer, 2016; Kohlbacher & Lavrijssen, 2018). Furthermore, other stakeholders can play a role in terms of oversight of infrastructure sectors, such as the public or private shareholders of infrastructure managers (when the latter are organized as companies) and private actors like consumer organizations. The result of this complex system is “an increasingly packed oversight environment surrounding the network-based utility industries” (Steenhuisen, 2009, p. 3).
In sum, infrastructure managers operate in highly regulated sectors. This suggests that infrastructure managers may face more constraints than other actors in less regulated markets. For example, while less regulated organizations may freely decide whether and how to invest in innovative solutions, network operators are compelled to follow the priorities and policy goals established by national (and supranational) rule-makers. While less regulated market actors may diversify their businesses and even try to tap new markets (see e.g. Prufer & Schottmüller, 2017, regarding digital platforms), network operators are limited to developing the activities assigned to them by the regulations that govern their sector. While less regulated organizations have more freedom to stablish the prices they charge for their products and services to maximize their profits, network managers are usually subject to different types of economic regulation and/or regulated network tariffs, as set by independent authorities intending to ensure that the public values related to the service in question will not be sacrificed to profitability (Lavrijssen & Ottow, 2012). This is usually accompanied by regulatory requirements to apply tariffs that are cost-oriented, reasonable, transparent, and nondiscriminatory (Lavrijssen & Vitez, 2015; Edens & Lavrijssen, 2018, p. 9).
The previous examples illustrate that, given the highly regulated context in which infrastructure managers operate, they are likely to face more legal constraints in their possibilities of using innovative solutions than less regulated organizations. In this sense, the need to maintain a good connection between what is possible in practice and the existing regulatory frameworks (section “Technological change and regulation”) becomes even more relevant.
The relevant Dutch regulatory framework
This section will outline the aspects of the Dutch regulatory framework for drinking water that are relevant for the analysis in section “Microlevel.” It will start with a reference to the main legislative and regulatory provisions of the sector and will continue with the current regulations applicable specifically to the metering.
General regulatory framework
The general regulatory framework for the drinking water sector 18 in the Netherlands is mainly provided by the Drinking Water Act (Drinkwaterwet in Dutch) of July 18, 2009; the Drinking Water Regulation (Drinkwaterregeling) of June 14, 2011; and the Drinking Water Decree (Drinkwaterbesluit) of May 23, 2011. Of relevance is also the Policy Note (Beleidsnota in Dutch), a document adopted at least every 6 years by the Minister of Infrastructure and Water Management, which contains the guidelines and principles of the drinking water supply policy that the Dutch government will follow (Article 6, Drinking Water Act). The most recent Policy Note was issued in 2014. As it will be shown in this section, the Dutch drinking water sector is subject to tight regulations.
According to this framework, 10 drinking water companies (drinkwaterbedrijven) 19 are in charge of producing, distributing, and supplying drinking water in the Netherlands. As part of their legal obligations, these companies are tasked with establishing and maintaining the infrastructure necessary for the production and distribution of drinking water (Article 7, Drinking Water Act). As such, these companies are the network operators in this sector. They are incorporated as public limited companies with municipalities and provinces as their shareholders, with the exception of one organization, which has the legal form of a foundation. Only public entities or companies controlled by legal entities subject to public law can be the shareholders of drinking water companies (Articles 1 and 15, Drinking Water Act).
These drinking water companies operate as monopolies within the distribution areas assigned to each of them by the Ministry of Infrastructure and Water Management (Drinking Water Regulation, Appendix 1, belonging to Article 4). As a consequence, consumers cannot freely choose or change their water supplier, and they are “tied” to the drinking water company that operates in their area of residence. As such, drinking water consumers are captive consumers (Lavrijssen & Vitez, 2015, p. 12–13).
The oversight of the drinking water sector in the Netherlands is mainly the responsibility of the Inspectorate of Human Environment and Transport (Inspectie Leefomgeving en Transport—ILT), an authority belonging to the Ministry of Infrastructure and Water Management (Article 48, Drinking Water Act), which also takes part in the oversight of the sector.
In addition, the Authority for Consumers and Markets (ACM) fulfils an advisory role, mainly assisting the ILT in its tariff monitoring task (Article 8a, paragraph 2 of the Drinking Water Decree), as well as the Ministry in determining the weighted average cost of capital (WACC) that the drinking water companies must observe (Article 8a, paragraph 1(b) of the Drinking Water Decree), as it will be discussed below.
There are three major regulatory instruments in place in the Dutch drinking water sector (Andersson Elffer Felix, 2017). The first one is the performance comparison (prestatievergelijking), also known as “benchmark” (Lavrijssen & Vitez, 2015). This benchmark is performed triennially by the ILT and compares the 10 drinking water companies on 5 main points: the quality of supplied water, the environmental aspects of the drinking water supply, customer service, cost efficiency, and research and development (Article 39, Dutch Water Act). Within 6 months of the delivery of the ILT report, the drinking water companies must inform the Ministry in writing how they will improve their performance, information which will be also shared with both Houses of the States General (Article 44, Drinking Water Act).
The second regulatory instrument entails financial oversight coupled with tariff monitoring. In the Netherlands, the costs involved in the supply of drinking water are exclusively covered by the tariff charged by the drinking water companies (Blank & van Heezik, 2017, p. 42). However, the drinking water companies are not entirely free to decide what tariffs they will apply. Any costs of capital that they want to pass on to the tariff are limited by the WACC that is predetermined by the Ministry of Infrastructure and Water Management with the advice of ACM (Article 10, Drinking Water Act; Article 6, Drinking Water Regulation). If the profit earned by the drinking water companies exceeds the WACC, they must compensate this excess in the tariff for the following calendar year (Article 12(3) Drinking Water Act; Lavrijssen & Vitez, 2015). In addition, the ILT (advised by ACM) monitors compliance with the requirements laid down in Article 11(1) of the Drinking Water Act, according to which the tariffs must be cost-effective, transparent, and nondiscriminatory.
The third way of exercising control over drinking water companies relates to their task of investing in infrastructure improvements. As part of their legal obligations, drinking water companies must submit to the ILT a “delivery plan” stating how they will ensure the adequate and sufficient supply of drinking water and how they will address any possible disruptions (Article 37, Drinking Water Act). Such a plan must consider the Policy Note prepared by the Ministry of Infrastructure and Water Management (Article 37, paragraph 2 of the Drinking Water Act). Furthermore, the plan must contain multiyear investment plans to improve the drinking water infrastructure, which in turn require the approval of the licensing department of the ILT, thereby allowing authorities to monitor the planned investments of the drinking water companies (Andersson Elffer Felix, 2017). In addition, these infrastructure improvement investments are also used as an indicator assessed by the ILT when conducting the abovementioned benchmark (Andersson Elffer Felix, 2017).
Specific regulations for metering
In addition to the general regulatory framework, it is relevant to explain how metering is currently regulated in the Dutch drinking water sector. 20 The first observation is that this activity is not explicitly mentioned in any of the legal or regulatory texts above mentioned. In fact, in the Netherlands, the metering is governed by the general conditions of drinking water supply published and applied by the drinking water companies, which form part of their contracts with consumers. In theory, each of the companies may draft its own general conditions, but in practice they replicate the conditions included in a model prepared by Vewin (the Dutch association of water companies), in consultation with the Dutch Consumers Association (Consumentenbond), a practice established since 1979 (Vewin, 2012). For this reason, the latest model published by Vewin in 2012 (hereinafter “the Vewin model”) will be used in this contribution as the specific regulatory framework for the metering activity.
The measuring device (water meter) is defined in the Vewin model as the equipment used by the drinking water company to determine the quantity of the drinking water supplied, which generates the data deemed necessary for invoicing and monitoring consumption (Article 1, Vewin model). The specific provisions related to metering are mainly found in Articles 10 to 13 of the Vewin model. The relevant aspects of these provisions are The drinking water company decides how the consumption is measured. It is the choice of the company whether to employ a measuring device (water meter) or not. When this is the case, the readings of the water meter are (as a general rule) binding for the company and the consumer. The use of the water meter is only oriented to determine the consumption of drinking water. The measuring device is installed and maintained by or on behalf of the company and at its own expense, unless the device gets lost or damaged by the consumer. At least once a year the consumer must record the position of the measuring device and inform it to the drinking water company in the way and term specified by the latter. However, the companies have the right to take the readings themselves. If the consumer does not inform the meter readings to the drinking water company, the company may make an estimation of the water consumed for billing purposes. The consumer must ensure that the measuring device is always accessible and can be properly read. In addition, the consumer must protect the device against damage, breaking of the seal, and must prevent frost damage.
Analysis
This section brings together the aspects discussed in sections “Literature review” and “The relevant Dutch regulatory framework” to explore what kinds of regulatory challenges may become evident if SWM is to be introduced in the Netherlands. The analysis is divided in two parts: The first part refers to the specific rules concerning the metering activity (microlevel). The second part refers to possible implications of SWM in the context of the general regulatory framework (macrolevel).
Microlevel
At the microlevel, the specific rules applicable to the metering in the Dutch drinking water sector will clearly require adaptations or additions to deal with both the opportunities and concerns of SWM. Under the current framework (see section “Specific regulations for metering”), the only use of water meters is to determine the consumption of water, mainly for billing purposes. The additional possibilities enabled by SWM which can improve the managing of the drinking water network (section “Smart water meters”) are not yet accounted for in the current rules applicable to the use of meter devices.
The adaptations will entail among others: introducing a new definition of the measuring device that includes the additional purposes for which SWM can be used; adjusting the frequency of the meter readings to attain the proposed goals; and eliminating the consumer obligation of manually sending the readings to the drinking water company.
In addition, since the current rules were drafted for traditional meters, they do not contemplate possible concerns regarding privacy that SWM may introduce as a result of more frequent measurements of water consumption (section “Smart water meters”). 21 Therefore, introducing SWM would require additional rules that deal properly with the challenges that such technology creates vis-à-vis the protection of consumer privacy. As explained by Cuijpers and Koops (2013), a major aspect to consider is that more intrusive approaches (e.g. mandatory installation of SWM and high-frequency measurements and readings by the utility company) require more substantiation and empirical evidence to justify the interference with the right to privacy. 22
An additional issue relates to the protection of personal data. Since the data generated by SWM can be traced back to an identifiable natural person (the consumer), it qualifies in principle as personal data under the meaning of the General Data Protection Regulation (Regulation (EU) 2016/679, known as the “GDPR”) in force in the European Union since May 2018. Among other aspects, the GDPR requires that all processing of personal data should be based on a lawful ground (see Article 5, paragraph 1(a); and Article 6 of the GDPR). Since the metering currently serves the purposes of performing the contract of supply of drinking water concluded between the drinking water company and the consumer, it is likely that the current legal basis for the processing of such personal data is the performance of a contract (Article 6, 1(b), GDPR). However, some of the additional possibilities enabled by SWM may not have a direct link with the performance of the contract but serve other purposes (e.g. leak detection, better monitoring of the network). Therefore, a different basis for the processing of such data will probably be required. 23
Macrolevel
As noted earlier (section “Specific regulations for metering”) at the moment, there are no specific legal or regulatory provisions that instruct the drinking water companies on how they should carry out the metering, even with traditional meters. Again, a reasonable explanation for this seems to be that the metering is nowadays used mainly for contractual purposes, thus, the metering is not seen as a means to achieve policy goals or public values of the drinking water sector. 24 Although the lack of such references does not automatically mean that drinking water companies cannot rollout SWM, their existence could help to provide certainty to the drinking water companies that implementing this technology is allowed under the regulations of the sector (sections “Managing infrastructures in the context of heavily regulated sectors” and “General regulatory framework”). 25
In addition to connecting the possibilities of SWM with the policy goals of the sector, adjusting the national regulatory framework might be also relevant to provide certainty regarding the scope of the responsibilities and rights of the different stakeholders involved. Firstly, to give some guidance regarding the goals that are intended to be achieved via the SWM (e.g. prompt leak detection, better monitoring of the network, consumption awareness, water conservation, etc.) and how these goals should be reflected in the technical aspects of the SWM. As explained by Hoenkamp (2015) in the context of electricity smart meters, the functionalities of such devices (determined via standards) ought to be coherent with the goals they are intended to serve. In her analysis of the rollout of electricity smart meters in the Netherlands, she explains that the functionalities initially established for the electricity smart meters did not allow to materialize the European Union policy aim of encouraging end-user energy efficiency, which was supposed to be the main motivation behind the rollout of the smart meters. This was the case because the adopted technical standard did not include functionalities (e.g. a display) to increase the consumption awareness of consumers (Hoenkamp, 2015, p. 31). In addition, a clear indication of the goals to be pursued with SWM is necessary to exclude unwanted functionalities (Vlaamse Milieumaatschappij, 2017). In this sense, additional rules at the macrolevel could contribute to steer the rollout of SWM in a way that the benefits are maximized (section “Technological change and regulation”).
Second, macrolevel rules could also provide guidance regarding the way the goals of SWM ought to be balanced against the rights to privacy and protection of personal data of the consumers. As explained earlier, certain approaches (e.g. mandatory rollout, sub-hour reading frequency) might be more intrusive for the private life of the consumers and therefore would require more justification of their proportionality and necessity in a democratic society (Article 8 ECHR; Cuijpers & Koops, 2008). Given that the Dutch state has the positive obligation of ensuring effective respect of private and family life laid down in Article 8 of the ECHR (see Council of Europe/European Court of Human Rights, 2018, p. 8), some guidance might be necessary to ensure that the rollout of SWM does not constitute an illegitimate interference with the right to privacy of the consumers, especially considering that they are “captive” consumers.
Similar considerations apply for the protection of personal data: Some guidelines regarding the legal grounds for processing and what kind of purposes are considered necessary to attain the goals of using SWM could provide certainty to the drinking water companies regarding their possibilities of action and at the same time contribute to the protection of consumers. In this respect, additions to the national regulatory framework might be necessary to limit the new harms or risks introduced with SWM (section “Technological change and regulation”).
Finally, macrolevel rules could also help to adequately plan the rollout of SWM, bearing in mind that implementing SWM may involve high investments (see Arniella, 2017; March, Morote, Rico, & Saurí, 2017) and the costs of the drinking water sector are currently covered by the tariff charged to the consumers (sections “Managing infrastructures in the context of heavily regulated sectors” and “General regulatory framework”). Macrolevel rules may contribute to structure the rollout in a way that the affordability of drinking water in the Netherlands is not endangered, while giving the drinking water companies enough room to make the necessary investments to implement SWM. The latter may require aligning or giving certain flexibility to the current regulatory instruments in place, described in section “General regulatory framework.” 26
Conclusions
This article is a first exploration of possible legal challenges of introducing SWM in the Netherlands. The most important observation resulting from this study is that the current regulations are still based on the assumption that metering only serves for the purposes of determining the consumption of water, as part of the performance of the supply contracts between drinking water companies and consumers. The additional uses of metering as a valuable source of data for the management of the network are not yet present in the current rules, nor the challenges that the use of SWM can create for the protection of privacy and personal data of consumers.
The need to revisit the existing rules to implement SWM is easier to pinpoint at the microlevel (rules governing the metering) than at the macrolevel (general rules of the drinking water sector), but that does not mean that the latter requires less attention. Dutch drinking water companies operate in a highly regulated sector and as such their possibilities of action are largely determined by the regulations in place, including the supervisory instruments. These circumstances may create challenges for the implementation of SWM that are less explicit but nevertheless relevant. More empirical research is necessary to assess to which extent the national regulatory framework in place in the Dutch drinking water sector facilitates or hinders the implementation of SWM.
Further research would be also required to investigate more specific regulatory challenges of implementing SWM, such as the governance of the system, in terms of who ought to be responsible for the installation and operation of the SWM, and the processing of the collected data. In this regard, some questions to consider are: should this be a task carried out by the drinking water companies or should there be a separate organization in charge of this? Who should be entitled to access SWM data and for which purposes? Will there be room for the creation of additional services for the water consumers based on the data generated by SWM? If so, who should be allowed to provide such services and under what conditions?
With this exploratory study, we also aim to start a discussion at a broader level, regarding the legal challenges that may become apparent as a result of the increasing digitalization of infrastructure in the drinking water sector.
Footnotes
Acknowledgements
We would like to express our gratitude to Dr Inge Graef, Prof. Dr Martijn Groenleer, Dr Leonie Reins (Tilburg University); Ben Puhl, Eugenie Westhuis, Ed Bakker, Jorik van Vilsteren, Rian Kloosterman, Rik Thijssen (Vitens); Daan Rutten (Alliander); Simon Porcher and Maria Salvetti (Sorbonne Business School); for commenting on our drafts and/or providing valuable insights to enrich our research. Any views expressed in this article and any errors or omissions remain responsibility of the authors.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research is supported by the Responsive Innovations program of Next Generation Infrastructures (NGInfra) and The Netherlands Organisation for Scientific Research (Nederlandse Organisatie voor Wetenschappelijk Onderzoek – NWO) project number: 439.16.807.
