The concept of value in sustainable infrastructure systems: a literature review

Infrastructure choices and decisions widely employ the language of value, whether to articulate what is worthwhile or to debate which principles or approaches are most appropriate to specific contexts. As the world strives to achieve long-term sustainability goals, incorporating sustainability values into infrastructure decision-making becomes progressively more important. Yet, the term ‘value’ has been used under different meanings and implications throughout the infrastructure sustainability literature, obstructing the debate on which values are important and what is valuable to infrastructure decision-making processes. This paper reviews how the concept of value has been used to position different sustainability dimensions of large infrastructure systems. Specifically, a conceptual framework proposed by Tadaki et al is used to highlight different notions of infrastructure value under four general headings: value as a magnitude of preference, as a contribution to specified goals, as a means of communicating key priorities, and as a representation of historical relations. This review shows that the discussion of infrastructure value has often focussed on monetary measures to the exclusion of other relevant measures of value. However, if long-term sustainability goals are to be met, a transformation of the ways that value is understood and measured in the context of infrastructure systems is required. This review discusses key similarities, interdependencies, and disparities between published notions of infrastructure value in order to provide a conceptual reference guide that highlights the variety of perspectives that are both implicit and explicit among practitioners and academics.


Introduction
Infrastructure systems play crucial roles in the functioning of urban areas, are extensive in spatial scope, have immense financial costs as well as human implications, and are controversial in both form and function.How these systems are defined, used, and evaluated is thus crucial, and often contested within industrial societies.What is built-and how it is built-is a clear indication of the priorities and values of a society, and will be critical in determining the extent to which global societies live up to local and international sustainability needs and commitments.Infrastructure commissioning and decision-making widely uses the language of value: both to explore what is valuable (i.e., worthwhile), and to debate which values (i.e., principles) are most relevant in specific contexts.As the relationship between sustainability and infrastructure has received increasing attention in the public discourse and academic literature, a growing body of work has been written on the value of infrastructure in sustainability.Throughout this literature, the term 'value' has been used in different ways to express different things.This review paper explores how the concept of 'value' has been used to study different sustainability dimensions of large infrastructure systems in the academic literature.Using a conceptual framing value developed by Tadaki et al (2017) in the context of ecosystem services, we classify the reviewed literature under four separate concepts of value.
The term 'infrastructure' herein refers primarily to distributed horizontal systems such as roads, bridges, and both water supply and sewerage networks.Further, this paper focuses on large-scale systems, which are usually identified by their significant capital costs and their impact on large populations (Sheng 2018).Projects of this scale have long planning horizons, several layers of physical and institutional complexity, and are often owned or commissioned by public authorities (Chester et al 2019).Although the focus is thus primarily on civil infrastructure (e.g., transport, water), the discussion is generally applicable to several other infrastructure systems such as energy or telecommunication systems.Moreover, the paper also accesses in passing green (natural) infrastructure and ecosystem services.
While the notion of the value these infrastructure systems provide is often explicitly and implicitly discussed in the context of public infrastructure decision-making and academic research, the published conversations have often diverged due to inconsistencies in both meaning and implication of 'value' and 'values'.This divergence has sometimes interfered with interdisciplinary conversations, the very conversations that are so vital if complex socio-ecological challenges are to be tackled (UN Habitat 2020).As one example, consider the ongoing discussion of how the concept of 'sustainability' should be included in infrastructure decision-making, where different levels of abstraction, context dependence, and valuation methods have lead to diverging understandings of what is valuable across social, environmental, and economic dimensions (Rawluk et al 2019).This paper seeks to gather diverse conversations together to provide a discussion with the goal of clarifying the multidimensional, dynamic, and interdependent nature of 'value' in infrastructure.A conceptual framework proposed by Tadaki et al (2017) is used to frame different notions of infrastructure value under four general headings: value as a magnitude of preference, as a contribution to specified goals, to communicate key priorities, and as a representation of historical relations.This framework provides a holistic and complementary view of preference and value that facilitates exploration of complex notions such as sustainability, and focuses on the particularities of decision-making and the valuation related to natural (and built) environments.Although the focus of this discussion are different concepts of value and their implication, this paper also explores the methodologies used to measure or describe these values.The methods through which value is operationalized provide critical insights into the assumptions and implications behind different concepts of value.
Large infrastructure systems provide the foundation for most activities in modern societies and strongly influence their surrounding environment, both in the short-and long-term.As such, infrastructure plays a defining role in the quality of life and environmental sustainability of most individuals and communities (Kasraian et al 2016 and Rezvani et al 2015).Further, these systems both operate and evolve in a highly uncertain and dynamic environment.Infrastructure needs to adapt to changing pressures including climate change, technological change, rapid urban expansion, and population growth (Gillespie-Marthaler et al 2019 and Sánchez-Silva 2018).The development and operation of infrastructure systems thus inevitably involve a wide range of stakeholders (owners, community members, engineers) who often have conflicting interests (Vuorinen and Martinsuo 2019).Consequently, infrastructure development and implementation are frequently controversial, competing over priority, resources, mechanism, and participation (Barclay and Klotz 2019, Pekkanen and Pearson 2018and Thomas Ng et al 2012).Conflicting interests arising from fragmented responsibilities across parties can also hinder systemic management and innovation of infrastructure systems.
The development of infrastructure is closely related to the collective objectives of societies (Alford andO'Flynn 2009 andMu et al 2015).Yet, converging on these objectives is often challenging due to lack of agreement between stakeholders, and the mechanisms used to aggregate their preferences (O'Flynn 2007).Being the physical representation of collective objectives, infrastructure plays a central role in the way priorities materialize in the built environment.For instance, the choice of the scale of a system is usually motivated by a balance of priorities including ownership structure, reliability, and impacts on sustainability (Alanne and Saari 2006).An example of collective objectives that reflect public values are the sustainable development goals (SDGs) adopted by the United Nations (United Nations 2015), which provide a comprehensive framework for sustainable development.The SDGs explicitly highlight the importance of infrastructure in achieving significant progress towards social, environmental, and economic objectives (Thacker et al 2019).Some of these goals require significant infrastructure investment, such as the provision of universal access to drinking water, sanitation, and electricity by 2030 (Oxford Economics 2017).Infrastructure systems are expected to deliver adequate, reliable services (e.g., access to drinking water, safe modes of travel) while promoting other socio-economic objectives such as equity and accessibility (Koppenjan et al 2008).
Most often the value or preference towards infrastructure projects is assessed in monetary terms.Further, infrastructure development has often been driven by subjective motivations such as biases towards technological innovation or the search for increased political visibility (Flyvbjerg 2014, Frick 2008and Pearce 1983).More recently, methodologies such as life cycle assessment (LCA) and multi-criteria methods (MCM) have been used to account for other aspects relevant to overall preference, such as social and environmental performance.However, decision makers often disagree on the evaluation of social and environmental dimensions, and their role in overall preference (Mouter et al 2013).Moreover, these dimensions often have largely different scopes, limitations, levels of detail, and assessment methodologies.For instance, it is common for infrastructure projects to give priority to cost-benefit analyses that focus on benefits and cost that can be readily monetized (e.g., travel time savings in road projects); in contrast, environmental and social impact assessments

Priorities
The structure/hierarchy of priorities within individuals/groups that guides their decisions.
Residents of city X prioritize the aesthetics of their built environment, followed by its safety.(Brown 1984, Tadaki et  usually receive less or delayed attention, perhaps only considered after the design is essentially completed (Ward and Skayannis 2019).Further, the results of environmental and social impact assessments are often limited due to high levels of uncertainty (Saxe et al 2020 andWard andSkayannis 2019).Monetary assessments of large projects have also consistently led to faulty estimations of risk and poor economic performance (Flyvbjerg 2014).This imbalance in the evaluation of infrastructure projects may lead to inaccuracies in the evaluation of their consequences for the public and the environment, hindering the development of infrastructure that manages to integrate the multiple facets of sustainable development.This article is structured as follows: the next section provides a conceptual framework for the paper by examining the concept of value and drawing a connection between value and sustainability.This is followed by a review of how value is represented in existing studies related to infrastructure systems.Section 4 uses this framework to discuss alignments, disparities, and interdependencies leading to an articulation of a more complete understanding of value for infrastructure.Finally, conclusions are drawn along with suggested pathways for future research.

Examining the concept of value
Value is a term widely used across disciplines under several interpretations.In the fields of philosophy, sociology, and psychology, it is used to describe the set of end-states or qualities that an individual or group could consider desirable (e.g., happiness, justice) (Brown 1984 andIves andKendal 2014).Other fields such as economics or engineering refer to 'value' in quantitative decision-making applications built on measurable notions: thus, value becomes the estimated worth of an object or place for an agent, or the means of estimating this worth (Brown 1984 andRawluk et al 2019).While this perspective has been useful in traditional monetary decision-making, it obscures the study of more abstract goals such as sustainability, which incorporate nonmaterial values.This clear limitation puts existing value engineering methods into question when designing decision-making processes for complex systems such as large infrastructure.
There are several published 'value' typologies that propose conceptual distinctions for the term.Some examples include the separation between 'held' and 'assigned' values (Brown 1984), or a social-value focussed typology of 'instrumental' and 'deliberative' values (Raymond et al 2014).Different typologies highlight specific mechanisms of evaluation, i.e., the mechanism of measurement and participation through which value is assessed.These typologies also shed light on the assumptions, intended applications, and limitations of different methodologies used for evaluation (Daily et al 2000, Rawluk et al 2019and Tadaki et al 2017).Brown (1984) and Tadaki et al (2017) have conceptualized values and preference over specific aspects of the environment and shared goods.In particular, the latter identifies four distinct concepts of value that are commonly present in the academic literature and applications from the context of environmental management: value as a magnitude of preference, value as contribution to a goal, value as a set of priorities, and value as historical relations (Tadaki et al 2017).While the authors emphasize the natural environment and ecosystem services, their concepts can be extended to the built environment and to infrastructure systems broadly.Further, this framework provides a holistic, and complementary view of preference and value useful to articulate complex notions such as the one of sustainability in infrastructure systems.Thus, this paper will focus on this framework and use it to explore existing discussions of value in the context of infrastructure systems.Summarized in table 1, this section explores Tadaki et al's conceptualization, its broad relevance to infrastructure, and draws a conceptual connection to the concept of sustainability.

Value as a magnitude of preference
The first concept of value presented by Tadaki et al (2017) is value as a magnitude of preference, which refers to a quantitative measure of the preference that an individual or a group has for an alternative or attribute relative to others.Value as a magnitude of direct preference is not an attribute of a system or alternative, but rather a representation of a stakeholder's preference towards it.A well-known application is the idea of market valuation and willingness-to-pay, which has been widely used to measure the worth of a specific component of the (built) environment (Daily et al 2000 andFreeman et al 2014).A typical infrastructure example is the calculation of the monetary value of providing additional water in a supply network based on water demand, supply and operating costs (Jenkins et al 2004).Other approaches such as mapping of stated preferences and multicriteria analysis use generic utility metrics to appraise the relative preferences of stakeholders over a defined set of alternatives (Kabir et al 2014).These analyses are used, for instance, to select a design alternative based on the mapping of experts' preferences using a multi-criteria approach that includes environmental, social, equity, and economic performance (Martin-Utrillas et al 2015).
Assuming that the value of an alternative relative to others is quantitatively measurable implies commensurability (Tadaki et al 2017), allowing for direct comparison and aggregation in collective contexts.For example, when quantifying the value of an infrastructure project for a group of people (e.g., users, owners, planners), one would add their individual measures of value for the project.Furthermore, commensurability also allows for value to be broken down and compared across components.For instance, monetary or utility value metrics might allow one to directly compare disparate environmental and social aspects of a proposal, even though they are different in nature.This further implies that the preference over different dimensions is fungible (i.e., interchangeable).For example, a reduction in stated preference due to poor social performance of a project could be offset by improvements in economic performance.
Recent discussions of 'value-as-magnitude' show that much of the literature on project management has used the term 'value management' to study cost reduction (Laursen and Svejvig 2016).These studies often focus on shorter term investments that do not explicitly acknowledge a project's long-term and holistic nature.Others have discussed the differences between individual, organizational, and societal value from a monetary perspective (Lepak et al 2007); or the importance of non-commercial values in project management, calling for a notion of preference that is broader than a purely monetary perspective (Martinsuo and Killen 2014).

Value as a contribution to goals
Value can also be understood as 'the contribution of an action or object to user-specified goals, objectives, or conditions' (Tadaki et al 2017).This second aspect arises from the idea that the overall preference of a collective may differ from the aggregation of individual preferences (Farber et al 2002).This notion of value captures the contribution of an object (e.g., an infrastructure product) to a set of goals or objectives (often defined by experts) measured through dedicated metrics.In contrast to value as magnitude of direct preference, this notion of value is not based in stakeholder preference but rather an attribute of a system in relation to specific goals.Collective objectives are often inadequately captured by individual utility measures such as willingnessto-pay.Further, individual evaluations may be difficult to compare and aggregate.An example of a collective goal that is inadequately captured by individual preferences is reducing greenhouse gas (GHG) emissions.This is known in economics as the 'tragedy of the commons', a market failure due to externalities where individual actors do not bear the collective cost of their actions (Fang 2018).The notion of value as contribution to a goal recognizes that shared goods such as infrastructure products are complex, which creates the need for an evaluation through attributes and goals separate from individual preference.
One of the main steps in measuring value as a contribution to a goal is the specification of appropriate metrics.However, the selection of these may be subject to disagreement and have implications to the outcomes of the assessment and the decisions they might lead to (Hubbard and Hubbard 2019, Huizar et al 2018and Tadaki et al 2017).For instance, using only GHG emissions to measure the environmental impact of a new road may induce it to be built through a vulnerable ecosystem to reduce construction material use; in contrast, using a measure of ecological fragmentation might protect vulnerable areas but increase material consumption and emissions (Cornet et al 2018).Moreover, expert-defined metrics may also embody the differing priorities between planners, operators, and users of infrastructure products (Chan et al 2012 andTadaki et al 2017).

Value as priorities
Value can also conceptualize individual priorities or held values (Brown 1984, Dietz et al 2005, Tadaki et al  2017, Webb 2013 and Wolters et al 2020).This notion refers to the structure of priorities within individuals, understood as the fundamental driver behind their actions and decisions.For example, some people make decisions trying to maximize their individual sense of freedom, while others might prioritize their sense of safety.This concept of value matches other relevant discussions in the public decision-making literature through the concept of 'public values', i.e., the collective aspirations that should guide public decisions and operations (Alford and O'Flynn 2009, Mu et al 2015and O'Flynn 2007).Studies of the priority hierarchy for different individuals often seek to demonstrate the differences in the objectives of policy makers and users, highlighting the challenges of agreeing upon a set of societal objectives for policymaking (Wallis and Gregory 2009).
In the context of infrastructure, the use of the concept of value-as-priorities has been used to assess the need for infrastructure development.For instance, some studies have explored the existing trade-offs between health impacts, service efficiency, and capital investment associated with infrastructure investment ( Álvarez et al 2007 andBurchell et al 2010).Further, infrastructure systems are expected to respond to public needs such as accessibility, reliability, and equity (Koppenjan et al 2008).In practice, physical infrastructure represents a materialization of the priorities used for its planning and design and has a long-lasting influence on the way societies develop (Bivens 2014, Eskerod and Ang 2017and Kasraian et al 2016).

Value as relations
The final concept of value described by Tadaki et al (2017) is the one of value as relations.This concept seeks to acknowledge the historical relationships between communities or individuals with their surrounding environment.As such, relational value is an emergent property from the unique relationships between people and their environment, instead of arising from the individual, community, or the environment itself (O'Neill et al 2007).In other words, the way environments bring satisfaction or value to people or communities may challenge traditional economic or social measures, and even collective goals.For instance, fishing communities may find value in preserving this traditional activity, which may not necessarily align with economic or larger societal objectives.This concept has been used to explore the different ways in which people relate to the environment when looking for a satisfying life, emphasizing that sometimes communities and individuals value their relationship to an environment beyond any social, economic, or intrinsic (e.g., value-as-magnitude and value-as-contribution) measures (Tadaki et al 2017).

Value and sustainability
While the Brundtland report (s) is known for its intergenerational definition of sustainable development, it also called for the development of a broader, multidimensional notion of sustainability.Afterwards, the concept of the triple bottom line (TBL) was developed to formally describe the multidimensional nature of sustainability (Elkington 1999).The very interpretation of what sustainability means, and how it translates to physical systems such as infrastructure, is dependent on the values and background of the engineers and developers designing and operating a system (Carew and Mitchell 2008).
The growing importance of sustainability and the prominent role of physical systems in achieving it has also led to expectations that infrastructure systems meet multidimensional performance thresholds (Ainger and Fenner 2014).However, the extent to which this is achieved in practice remains limited, as existing infrastructure systems are often assessed through the lens of economic and monetary measures of performance, and often struggle to fully integrate or even acknowledge social and environmental goals (Atkins et al Bishop 2017, Flyvbjerg 2014, Glasson and Therivel 2019, Hawkins and Jill 2006and Purwohedi and Gurd 2019).Therefore, it is relevant in this context to attempt to frame the concepts of value in the domain of infrastructure through an integral framework that includes broader social, environmental, and economic considerations (Glasson andTherivel 2019 andMarvin andGuy 1997).This relevance has also been highlighted in a recent report by UN Habitat focussing on the value of sustainable urbanization (UN Habitat 2020).While this discussion is highly related to the one in this paper, we provide a more specific outlook of value from the perspective of infrastructure systems.
Although this review discusses the environmental, social, and economic dimensions of value in infrastructure systems, this does not imply they are commensurable or that these dimensions should have equal importance.This 'balance' obscures the nested nature between these dimensions of sustainable development: 'within the Earth's single planet limit, the environment nurtures our human society, which has invented the economy, to serve its needs' (Ainger and Fenner 2014).The discussion presented in the following sections seek to highlight the imbalances between different dimensions of sustainable development that are manifested through the different concepts of value.

Value and sustainability in infrastructure systems
The authors used Google Scholar (Google 2020), Scopus, and OneSearch (the University of Toronto's reference search platform) to search for references that contained different combinations of relevant 'value' and 'infrastructure' keywords, as presented in table 2. These queries resulted in an initial corpus of 853 publications.Then, we excluded duplicate references, publications that did not deal with large horizontal infrastructure such as vertical infrastructure (buildings) and 'soft' information infrastructure, and references that only referred to 'value' in the context of numerical amount or quantities.Additional references were included through crossreferencing, which allowed for the inclusion of relevant studies that refer to value implicitly and that may have been omitted otherwise.Some of the related keywords that were identified through the cross-referencing process and that were relevant to this review include 'preference', 'worth', 'benefits', and 'impacts.'After this initial review, the remaining 60 references were classified into the four concepts of value already highlighted.There are additional references cited throughout this review that complement the concepts and applications found in the existing literature.
Based on this selection process, the current section reviews the concept of value in the domain of infrastructure with an exploration of dimensions of sustainability.In particular, the environmental, social, and economic dimensions of sustainability are examined through Tadaki et al's framework of value described in section 2. Emphasis is drawn on the key conceptualizations rather than attempting to be exhaustive regarding methods, metrics, or case studies.Figure 1 summarizes the applied framework of infrastructure value and sustainability.There are clearly strong connections between these frameworks in the context of infrastructure: the 'value' of these systems drives its planning and operation, which in turn has a profound influence on the reality and evaluation of the dimensions of sustainability.

Environmental dimension of value
The environmental impacts of large infrastructure have long been recognized.).In general, these methodologies serve as supporting information for decisions related to infrastructure products, ranging from planning and design of new systems to retrofit and maintenance of existing ones.As such, they inform decision makers about key differences between alternatives in infrastructure decisions; they may also provide essential feedback and mitigation strategies in iterative design processes, although they are often part of linear assessment and development processes (Glasson andTherivel 2019 andVanclay et al 2015).There are several ways in which the environmental dimensions of infrastructure are assessed: integrating environmental impacts into monetary benefits/costs in cost-benefit analyses (CBA), measuring them through standardized metrics (e.g., CO 2 eq for GHG emissions, PM levels for air quality), and performing qualitative assessments (e.g., cause-effect descriptions of impacts in biodiversity).Such methods refer to different notions of value; however, the differences in meaning and implication are rarely acknowledged.This section reviews the ways in which environmental value has been discussed in the context of infrastructure, laying out examples of the four concepts of value described in section 2. Table 3 below provides a summary of the studies reviewed and their classification under different concepts of value and broad types of infrastructure (e.g., water, transportation, energy).
Some authors focus on quantifying the monetary value of natural resources, in an effort to integrate environmental improvements, benefits, and impacts in cost-benefit analyses often used for environmental decision-making (Costanza et al 1997, Daily et al 2000and O'Neill et al 2007).While these sources do not refer to infrastructure systems explicitly, they talk about the impact and the value of the built environment on the natural world and are a relevant example of how the environmental value has been assessed.In the context of infrastructure, there are numerous monetary measures.Some have studied how financial scenarios used for decision making are related to the environmental requirements for water infrastructure (Jenkins et al 2004).More recently, studies have explored monetary evaluations in the context of green/sustainable infrastructure solutions, highlighting the uncertainty that results from high parameter sensitivity in the evaluation of environmental benefits (Melo et al 2020), as well the irreducible nature of many uncertainties in the infrastructure provisioning process (Saxe et al 2020).Others suggest the use of causal chains that explain and validate users' willingness-to-pay to achieve more accurate results (Sunderland et al 2015).
Other studies focus on non-monetary metrics of direct preference such as utility or compound ratings instead of monetary measures.MCM, which express relative preference between alternatives through composite utility measures, have been widely used in the infrastructure literature (Kabir et al 2014).In general, these studies seek to capture the direct preference of decision makers for infrastructure alternatives based on their environmental performance.Further, these measures of value imply that impacts and benefits are commensurable and aggregable with other cash flows, or sources of utility; in terms of the value framework described in the previous section, this represents the notion of value as a magnitude of preference.Finally, some authors note that expressing value as a magnitude of preference often constrains the participation of larger groups of stakeholders.This is a consequence of measuring value in contexts with fixed sets of alternatives and predefined framings (O'Neill et al 2007).Additional examples related to environmental preferences may be found in (Beare et al 1998, Farber et al 2002, Foxon et al 2002, Foxon et al 2015, Freeman et al 2014, Funk et al 2019, Mell et al 2016and Williams et al 2017).
Other environmental assessments used in infrastructure development (e.g., impact assessments, life cycle analysis, carbon footprinting, among others) rely on dedicated metrics (e.g., energy efficiency, GHG emissions) instead of monetary or direct preference equivalents (Ainger and Fenner 2014).There are many examples in the infrastructure literature that describe both negative and positive 'environmental value' of large infrastructure through specific, expert-designed metrics.Some studies have focussed on resource utilization (e.g., energy, water, land use) and production of waste and pollutants (e.g., GHG emissions, wastewater, solid waste) (Sahely et al 2005).Other authors have used externally-defined goals as a framework of assessment, e.g., mapping the role and influence of infrastructure systems on the United Nations' SDGs and using dedicated performance metrics that enable the tracking of SDG targets linked to infrastructure provision (Adshead et al 2019 andThacker et al 2019).These studies show that infrastructure is an essential enabler of the SDGs, being directly or indirectly included in 72% of the targets (Thacker et al 2019).While the SDGs draw important broad connections between infrastructure development and environmental, social, and economic impacts, their materialization is open to interpretation and often presents context-based challenges.Consequently, some authors have drawn more specific connections between the SDGs and environmental assessments for infrastructure projects (Castor et al 2020, Kørnøv et al 2020and Mansell and Philbin 2020).
These studies are used to find quantitative 'intrinsic' values according to a specific goal that is independent from the utility of an individual or a group of stakeholders.Consequently, the metrics used in the methodologies presented above are the result of expert recommendation and are not tied directly to the perspective of decision makers.According to the value typology discussed in section 2, this follows the definition of value as contribution to a goal.In contrast to the notion of value as a magnitude of preference (which sources value in the decision maker), the concept of contribution to a goal implies value is a property of the object or system.Additional examples that showcase environmental value as contribution to a goal in infrastructure may be found in (Allende et  The notion of values as priorities discussed in section 2.3 may also be found in studies of the environmental dimension of infrastructure.This concept refers to the hierarchy and structure of priorities for individuals and societies, instead of pointing to the preference over a particular system or object.In the context of environmental assessment of infrastructure, this is often generically labelled as 'environmental values'.Some studies describe how environmental priorities have shifted over time regarding specific infrastructure systems.For instance, transportation infrastructure planners in the Netherlands shifted from expansion-oriented demand management to demand-reducing strategies, acknowledging and prioritizing the negative environmental impacts of new construction (Geels 2007).In terms of planning, studies show that integrated and sectoral planning affect the representation of collective environmental value systems.While the former gives a clearer view of the interdependencies and environmental consequences of infrastructure projects, it often complicates project materialization (Busscher et al 2015).Others have highlighted the disconnection between the value system and management practices expressed by companies in the construction sector, which hinders the tracking and realization of those values and objectives (Hamilton-Foster 2014).
Infrastructure plays a critical role in moving into a low-carbon society, since improved environmental performance of large systems may result in more fundamental and widespread changes than changes to individual value systems (Webb 2013).This is supported by studies that show that individuals do not strongly relate their individual well-being with the environmental performance of infrastructure systems (Hienuki et al 2019).Additional examples of discussions of value systems and priorities in the domain of infrastructure may be found in (Chan et al 2012, Foerster et al 2015, Ives and Kendal 2014, McAndrews et al 2018, Morison and Brown 2011, Wolsink 2010, Wolters et al 2020and Zajchowski and Brownlee 2018).
Environmental impacts of infrastructure are often complex and go beyond the scope of quantitative assessments (O'Neill et al 2007).Some authors have studied infrastructure projects from a qualitative perspective, relating specific historical contexts with the built environment.This provides a complementary exploration of the environmental impacts of infrastructure development, often focussing on the local regulatory context and the implications for specific communities; see, for instance, (Fearnside 2002).Others highlight how infrastructure development often overlooks local value systems and historical connections to traditional infrastructure and ways of life (Grubert 2018 andKennedy Dalseg et al 2018).The recognition that 'valuable solutions' in infrastructure are not universal-highlighting the importance of historical relationships between users and their surrounding environment-is closely connected to the notion of value-as-relations.However, infrastructure examples that explicitly recognize the notion of relational values are rare; this could be due to the more abstract and qualitative nature of the concept, which raises challenges regarding its incorporation into quantitative assessments such as EIAs, LCA, and CBA that have lead the assessment of sustainability in infrastructure systems.

Social dimension of value
Large infrastructure almost inevitably has profound social impacts-including those on health, liveability, material well-being, culture, family and community structures, institutions, politics, equity, gender relations, among others (Vanclay 2002).Through the provision of essential services, infrastructure enables individuals to engage in activities that align with their priorities (Sen 1999).Specifically, these systems provide physical factors of urban social sustainability (e.g., decent housing, local environmental quality, accessibility); and nonphysical factors (e.g., health, equity, employment, cultural traditions) (Dempsey et al 2011).The assessment of these impacts is usually a part of planning processes and regulations for infrastructure, and is usually done in parallel with EIAs through Social Impact Assessments (SIA) (Government of Canada 2019, Tilt et al 2009, Vanclay et al 2015and Vanclay 2002).In this context, the notion of social value appears as a framework to understand the social impacts, benefits, and assessment of preference related to infrastructure (Raiden et al 2018).Similarly, social value in the domain of infrastructure is operationalized through a range of different metrics, from monetary equivalents and quantitative metrics of utility to qualitative descriptions of historical processes.
Social impacts usually have an abstract and qualitative nature, which makes it challenging to provide quantitative measures to track them.This also leaves wide space for interpretation and alternatives to measure these impacts: monetary equivalents, quality of life indices, qualitative descriptions, among others.Some aspects relevant to social value, such as the material well-being of individuals and communities, are directly related to economic prosperity, employment, income, and debt, among others (Badasyan and Alfen 2017, Samli 2011and Vanclay 2002).In this sense, economic and social value are intertwined, and thus, monetary proxies have often been used to measure social outcomes.However, many social outcomes related to infrastructure development are not intrinsically monetary, and as such, they can-and should-be explored separately from economic considerations (UN Habitat 2020).The economic objectives of infrastructure value will be explored in section 3.3.This section explores and classifies the social dimension of value in infrastructure in existing studies.Table 4 summarizes the reviewed literature, classifying publications according to the concept of value discussed and its broad infrastructure application.
Some authors approach social value by measuring the magnitude of direct preference of a decision maker over an alternative.Some early works in this topic estimate the monetary value of social impacts (e.g., walkability for transportation infrastructure) through direct assessments of preference such as market surveys, valuation of marginal changes in supply and demand, or spill over effects on property values (Litman 2003).More recently, studies have explored how technological applications such as building information modelling can be used to measure equity impacts based on monetary equivalents and willingness-to-pay functions (Parker 2014).Others have measured social value through adaptations of financial metrics, e.g., a Social Return on Investment (SROI) to quantify the delivery of social goals (Purwohedi and Gurd 2019).These studies assume the value of social outcomes related to infrastructure development is captured by individual preference.Further, they use monetary estimates to calculate aggregate measures of well-being that are normally used in the context of CBA (e.g., cost-benefit ratio, discounting, net present values).
Other studies focus on quantifying social impacts with respect to external goals (e.g., equity, culture, liveability) using dedicated metrics that do not reflect direct preference.Specifically, risk management, equity, and safety have been identified as central goals behind the development of sustainable urban forms (Eizenberg and Jabareen 2017).For instance, some studies use surveys in which users can express varying levels of wellbeing related to infrastructure alternatives through expert-defined needs (e.g., safety, comfort, convenience) (Kalyviotis et al 2018).Other authors use social network analysis to quantify the flow of information and communication of perceptions between infrastructure stakeholders (Doloi 2018).Assessment matrices that integrate different goals and objectives of social sustainability are also used to integrate different measures of social value into an overall estimate (Reddy et al 2014 andSiew et al 2013).These studies explore social value as contribution to a goal, acknowledging that the complexity of the social impacts of infrastructure may go beyond what is captured by individual preference.Consequently, they propose or adapt metrics to measure the contribution of infrastructure alternatives to specific goals (e.g., meeting predefined needs, positive public perception).Additional examples may be found in (Cui and Sun 2019, Edum-Fotwe and Price 2009, Hertogh and Bakker 2017and Zeng et al 2015).
Another perspective of social value are individual and collective priorities and how they relate to public preference over infrastructure projects.The engagement with stakeholder priorities and needs during earlyphase planning has been found to influence the perceived usability and the delivery of value during later stages of infrastructure projects (Grum and Kobal Grum 2020).Some study the relationship between collective value systems and the concept of critical infrastructures, highlighting the importance of the cultural and social perceptions of threat and risk (Burgess 2007).Others have described how mitigation efforts for negative impacts are dependent on the value systems of the infrastructure developers, which often do not align with the priorities of local communities (Tilt et al 2009).Consequently, some identify the importance of mapping and understanding the value systems of stakeholders to successfully navigate their expectations and align project performance goals with them (Vallance et al 2011).
Regarding relational values, some authors discuss the social impacts of infrastructure from a contextual and historical perspective.However, there is little or no literature on large infrastructure that explicitly refers to 'relational value'.This is most likely due to the recent nature of the concept, formally developed in the context of ecosystem services within the last five years (Chan et al 2016 andTadaki et al 2017).Rawluk et al (2019) describe the concept of value-as-relations as a tension between realism and relativism, acknowledging that '. . .a comprehensive understanding of how people value nature requires multiple ways of knowing (epistemologies)' (Rawluk et al 2019).Still, some discussions on the social value of infrastructure implicitly explore this concept including contextual or historical analyses, even if they do not formally mention it.Some studies in the domain of infrastructure acknowledge that a challenge in SIA is dealing with the changing cultural and political conditions that shape the perception of impacts that result from infrastructure development (Tilt et al 2009).Others recommend highly localized assessments of social impacts that shift away from 'universal' metrics of social well-being, highlighting the importance of scale selection in the valuation of infrastructure products (Mulholland et al 2020).In broader sustainability discussions, the preservation of historical practices within social sustainability is described as 'maintenance sustainability' (Vallance et al 2011).Additional examples of studies that touch on the concept of value as relations in infrastructure may be found in (Mulholland et al 2019 andRaiden et al 2018).

Economic dimension of value
Infrastructure is closely related to economic development-it enables most economic activities by providing necessary transportation options, energy, and communication platforms.Further, infrastructure itself represents a significant portion of the global economy: estimates show that, in 2013, existing infrastructure accounted for around 70% of GDP in most countries, projecting a global infrastructure investment need between $57 and $67 trillion US dollars through 2030 (Dobbs et al 2013).Consequently, economies throughout the world depend on the adequate selection, development, operation, and maintenance of infrastructure.
When approaching the economic aspects of infrastructure, it is important to make a distinction between economic impacts and the monetary evaluation of those impacts.The former refers to the positive or negative economic effects that follow the development and operation of infrastructure, while the latter refers to the action of assigning a monetary value on a given impact by a decision maker.While many economic impacts are expressed in monetary terms, these can also relate to non-monetary objectives such as wealth distribution, growth, and economic risk.In this subsection, we review the study of economic value in the domain of infrastructure-both from the perspective of monetary evaluation, and through other notions of value presented in section 2. Table 5 summarizes the reviewed literature for the economic dimension, classified again into different concepts of value and by domain.
Many studies have focussed on assessing the economic value of infrastructure as measured by measures of direct preference (Badasyan 2018, Badasyan and Alfen 2018, Conrad and Seitz 1994, Funk et al 2019, Jones et al 2014, Mousakhani et al 2017, Rezvani et al 2015, Uddin 2013, Williams et al 2017and Zhang 2006).Some have made a clear distinction in the literature between financial performance (which relates to the investors in an infrastructure asset) and economic performance, i.e., whether or not an infrastructure asset has added socioeconomic value (e.g., time savings, increased productivity) (Badasyan and Alfen 2017).Others measure the economic value of infrastructure through changes in property and land market values, studying revenue mechanisms that capture land value increases that result from infrastructure development (Bujanda and Fullerton 2017, Mathur and Smith 2013and Rybeck 2004).Moreover, some have explored the effect of infrastructure performance in indirect costs such as medical expenses, which are related to air pollution levels that result from different modes of transport (Sarkar et al 2007).
The integration of several economic factors such as job and social equity, economic growth, and method of financial procurement has been explored through MCM that quantify the relative preference of decision makers between infrastructure alternatives (Martin-Utrillas et al 2015).These studies seek to aggregate different potential economic benefits such as employment generation and healthcare expense reduction through measures that are directly comparable to construction and operation costs.This provides important insight on how economic goals relate to individual preference, and the willingness of stakeholders to engage in the development of infrastructure products that align with these goals.
Several authors have studied the economic value of infrastructure in terms of its contribution to larger economic goals such as wealth distribution, employment, and equity (Munnell 1990 andStraub 2011).Some have shown that higher government expenditure on public infrastructure results in a significant reduction in unemployment (Leigh and Neill 2011).However, others highlight that improvements on these goals is rarely distributed equally among the public (Bivens 2014).Thus, the distinction between economic growth and economic prosperity becomes key when examining the effects of infrastructure investment.While economic growth is an important motivation for infrastructure investment, ensuring equitable and sustainable productivity in the long-term is also critical to oppose inequalities in income distribution (Samli 2011).
The concept of value as a framework of priorities has not been extensively studied in the economic dimension of infrastructure.In general, existing studies and collective objectives such as the UN's SDGs call for infrastructure investment itself as a priority (United Nations 2015).Some draw attention to the tensions between economic and environmental priorities related to infrastructure development (Baklanov et al 2018 andOECD 2010).Others study the relationship between culture and value systems with respect to economic policy in infrastructure; see, for instance, (Wilson 2012).However, the existing academic literature lacks an exploration of the economic priorities of stakeholders related to infrastructure development beyond growth itself (e.g., employment quality, wealth distribution, equity).These aspects may hold key insights to improve conflictive dynamics between different groups of stakeholders involved in the infrastructure domain.
Finally, to the best of the authors' knowledge, the concept of value as relations has not yet been studied through the economic dimension of infrastructure.This is likely due to the origin of this concept of value, which explores preferences that are not explained by rational measures or motivations such as economic performance.While some communities have specific economic relationships to the development of infrastructure (Wilson 2012), the concept of relational value highlights the importance of traditions and preferences that stand apart from what are normally labelled as 'rational' motivations (Grubert 2018 andTadaki et al 2017).

Towards an integrated notion of value in infrastructure
Separating the relevant dimensions of infrastructure performance into different value concepts highlights how these dimensions shape discussions and evaluations of infrastructure systems.Firstly, there is a noticeable tendency to quantify the value of infrastructure through measures of direct preference such as monetary and utility measures.These studies are useful and contribute to very well-studied and widely used methodologies such as cost-benefit analysis and multi-criteria assessments.However, these methods also come with shortcomings, including the lack of consistency in accounting for impacts across projects, aggregation of impacts that are not commensurable, biases towards project development, and implications on external participation and communication (Atkins et al 2017, Flyvbjerg 2014and Tadaki et al 2017).
Since infrastructure systems impacts a wide range of stakeholders in ways that go beyond market quantification, it is often challenging to express value as an accurate monetary or utility magnitude, making comparisons across projects difficult and less reliable (Atkins et al 2017).Moreover, it may overlook the dynamic nature of impacts and the hierarchical nature of some of the potential benefits of infrastructure-an important feature in related fields such as ecosystem services (Fisher et al 2009).Even sustainability's most well known framing, the TBL framework, has been questioned in recent years as it has been used to justify financial assessments, rather than to perform holistic multi-dimensional analyses of the ways in which infrastructure and businesses create or destroy value (Elkington 2018).In addition, the evaluation of economic value as the contribution to external goals provides valuable insight, showing that even when infrastructure development has been tied to economic growth, this does not necessarily translate to the delivery of economic prosperity.
In terms of impact aggregation and comparison, measures of direct preference may give the illusion of commensurability, distorting the overall view of infrastructure projects for decision makers.Clearly, this limitation biases assessments against impacts that are not readily monetizable, as is the case of many environmental and social impacts (Tadaki et al 2017).For instance, some of the multi-criteria decision-making methods directly aggregate financial performance (e.g., net present value) with other goals that are not oriented towards direct preference (e.g., impacts on biodiversity).While it may be useful to have an overall view of the preferences of decision makers in some contexts, these methods may also distort both the interpretation and comparison of preferences.Further, they imply that different dimensions impacted by infrastructure products (i.e., social, environmental, economic) can (or should) be balanced through aggregation.In contrast, a broader understanding of the value of infrastructure systems can provide a useful framework that focuses on the trade-offs and alignments across value and sustainability dimensions instead of balancing performance across them.This vision of value can be explored, for instance, through the application of non-compensatory decision-making methodologies (e.g., ELECTRE, PROMETHEE) for the design and management of infrastructure systems (Hassan et al 2015).
Cost-benefit analyses-usually performed from the perspective of the parties in charge of the development and operation of the infrastructure-often happen in a context where there are already strong incentives to go forward with the implementation of projects, further distorting the value of infrastructure projects for stakeholders (Atkins et al 2017 andFlyvbjerg 2014).This is exacerbated by the limited and regulatory scope of the assessment of other impacts (e.g., social, environmental), which is often perceived as a hurdle by project developers (Ward and Skayannis 2019).Recognizing the importance of all the notions of value in the context of infrastructure decision-making could result in key improvements to future developments.In order to design and operate successfully sustainable infrastructure, we need to have a clear view of the aspects that are best accounted for through each of the different forms of value (e.g., as magnitude of preference, as contribution to a goal, or as relational values).
Studies that understand value as contribution to goals (particularly for the environmental and social dimensions of infrastructure) have often paid little attention to the ways these goals reflect the priorities of different stakeholders.This may separate their view of high-level collective goal from what is measured as valuable in existing studies.While these goals seek to capture complex preferences that may go beyond individual preference, it is important that they address the priorities of the involved stakeholders (Hienuki et al 2019).This plays a central role in meeting the service needs at the end user/community level; it also aids in setting appropriate performance thresholds for the infrastructure beyond standard regulations that may not reflect contextual objectives.Thus, it is necessary to study the similarities and tensions in stakeholders' views of socioeconomic and environmental performance.In other words, collective goals speak to the varying priorities of communities, and acknowledging these differences is central to a successful management of the trade-offs and the impacts of infrastructure.
Furthermore, it is critical to consider the dynamic nature of value.Infrastructure systems are unique and long-lasting endeavours, and thus the priorities that drive performance goals and the goals themselves will likely evolve throughout its life cycle.Further, infrastructure products face significant deterioration and ageing throughout their life cycle, which often impacts their ability to perform reliably and efficiently (Sánchez-Silva and Klutke 2016).However, infrastructure is many times designed and analysed from a static framework of objectives and performance goals (Ward and Skayannis 2019).This contributes to the obsolescence and underwhelming performance of projects in the long term, where many of the social and environmental impacts of infrastructure become significant.For instance, the motivations behind the development of transportation infrastructure focussed on private vehicles have evolved in many societies, where there is now a greater concern for the environmental and social consequences of these systems.The perception of value of the involved parties-as well as the ways that this value is delivered-transformed over time in these infrastructure projects.For example, Greiman and Sclar (2019) describe how shifts in users' mode choices as well as local regulations for environmental assessments led the 'Big Dig' project in Boston, Massachussetts to be outdated from the start of its operation (Greiman and Sclar 2019).Further, the impacts of infrastructure are also evolving constantly and materialize at different points in time.Rigid definitions of priorities, objectives, and system boundaries may be detrimental to the understanding of the impacts of infrastructure development.For instance, social and environmental impacts often result from long-term processes with fuzzy boundaries that are tied to deep uncertainties and triggered by unexpected circumstances such as extreme natural events that affect a wide range of systems and stakeholders.Consequently, some researchers have called for the development of flexible and adaptable infrastructure systems that embrace change and uncertainty over time, avoiding potential negative technical lock-in; see, for example, (Chester andAllenby 2018 andSánchez-Silva 2018).However, existing systems still play a key role in the delivery of value, and improving its management and operation is key to achieving better economic, environmental, and social outcomes (Saxe and MacAskill 2019).Moreover, endof-life and legacy considerations are also relevant in the delivery of value given the very significant amounts of materials, logistic effort, and capital embodied in existing infrastructure.
The selection of an appropriate scale for an infrastructure product also plays a central role in the definition of the stakeholders (and their preference), collective objectives, and priorities.Large-scale infrastructure systems, which are the focus of this literature review, usually rely on centralization and economies of scale that increase efficiency across large populations.However, this can hinder the ability to adjust for individual preferences and priorities of specific communities, impacting the value perceived at the individual and/or community level.In turn, smaller-scale solutions that rely on redundancy and wide distribution may reduce efficiency at large population scales, but in turn, can provide increased flexibility for individuals and communities and reduce vulnerability to failure events compared to centralized systems (Alanne and Saari 2006, Farmani and Butler 2014and Makropoulos and Butler 2010).
Finally, it is critical to recognize the interdependency between the delivery of different dimensions of value in infrastructure.The existing literature clearly reveals a close connection between the social and environmental dimensions of value; for example, improvements in environmental objectives are often associated with better health outcomes and increased well-being (e.g., improvements in air quality or noise pollution) (Coutts and Hahn 2015).However, little attention has yet been drawn to the modelling of interdependencies nor to the conditional materialization between different benefits and impacts.Such an awareness could offer greater transparency for decisions related to infrastructure development, better revealing trade-offs and interactions that are often obscured when separate impacts are simply aggregated.

Conclusions and recommendations
Infrastructure provides key services and goods that enable a vast array of socio-economic activities.Yet, the development of infrastructure requires significant capital and has long-lasting impacts on societies and their surrounding environments.Thus, not surprisingly, there is a large body of literature that explores the notion of value for different aspects of infrastructure including social, environmental, and economic considerations.However, the way 'value' is understood across the infrastructure literature is shown to be diverse, a reality that results in disparate conversations regarding the impacts and performance of infrastructure projects.The current review of value uses a conceptual framework that separates the different concepts and assumptions of value for infrastructure systems and explores how the concept of value is represented in environmental, social, and economic dimensions, all notably relevant to conversations of sustainability.Ultimately, the notion of value contains-either implicitly or explicitly-key expectations and objectives of all stakeholders relating to infrastructure development, operation, and evolution.Value as direct preference is more prevalent in planning and operation spheres in which infrastructure alternatives are evaluated; value as contribution to goals emerge both in infrastructure regulation and for long-term societal planning; value that articulates priorities reveals that stakeholders have different hierarchies of priorities; value as embedded in relations focuses on the historical bonds between communities and their built environment.
This review illustrates that the notion of value as a magnitude of direct preference is prevalent throughout the different dimensions of sustainability.However, such notion of value is shown to have problematic implications regarding the aggregation and direct comparison of benefits and impacts that are not readily interchangeable.Further, such an approach tends to exclude relevant impacts that are not readily quantifiable.Other studies measure the value of infrastructure projects from the perspective of contribution to collective or external goals, recognizing that some aspects of preference go beyond the preference of individual stakeholders, and result from more complex relationships between collectives.Two additional notions of value-namely, as an evaluation of priorities and contextual relations between communities and their environment-are as yet rarely present in existing studies of large infrastructure systems.This has implications on the ways collective preference for infrastructure projects is assessed; namely, it tends to ignore historical relationships between communities and their traditions, and rarely examines the priorities of different groups of stakeholders.
Additional research is needed to integrate the different notions of value of infrastructure studied here.These efforts are critical to successfully address present challenges such as improving the sustainability of infrastructure products.Some of the main research opportunities on the topic of value in infrastructure systems include: • Addressing the interdependencies between different values in the context of system performance.This distinction should lead to a better representation of how impacts materialize throughout an infrastructure system's lifetime.In other words, the aggregation of independent objectives is too coarse, since social, environmental, and economic benefits are often closely coupled and strongly interdependent.• Exploring the dynamic nature of preference towards environmental, social, and economic objectives, so that systems can be transparently and accurately evaluated at different stages of their life cycle.Dynamic decision-making frameworks are critical to successfully achieve long-term objectives, since both the objectives and the metrics used to track infrastructure achievement often transform over time.Further, some of the recommendations for existing practice of infrastructure decision-making include: • It is critical that existing frameworks of value integrate alternative notions of preference-such as contribution to objectives, or historical relations to the built environment-into the decision-making processes of infrastructure systems, which mostly focus on direct-preference measures of value (e.g., monetary performance, utility metrics).• Identify and quantify the interdependencies in the delivery of value between infrastructures systems.This could improve the integration between decisions made for interdependent systems, leading to improved performance and less disruptions in their operation.The ultimate goal is to develop infrastructure that is better equipped to meet broad expectations across different dimensions and at multiple levels, ranging from the direct preference of stakeholders to high-level collective goals and historical relations of communities.Yet, the evaluation process must also seek to avoid being mired by complexity, as infrastructure systems continue to operate and provide essential services.Such a balance is a difficult to achieve.

Table 1 .
(Tadaki et al 2017) concepts of value discussed in this article; adapted from(Tadaki et al 2017)with infrastructure examples.

Table 2 .
Query terms used during the literature review process.
'Value' + 'direct preference' 'Value' + 'relations' Figure 1.Conceptual frameworks of value and sustainability used in this review.

Table 3 .
Reviewed literature for the environmental dimension of infrastructure, classified by concept of value and application.

Table 4 .
Reviewed literature for the social dimension of infrastructure, classified by concept of value and application.

Table 5 .
Reviewed literature for the economic dimension of infrastructure, classified by concept of value and application.