Introduction

To support sustainable development implying ‘economic and ecological system sustainability’ (Faucheux and O’Connor 1998, p. 4), the economic valuation of ecosystem services (ES) was developed to incorporate natural capital into decision-making and economic development (Munda 2000) and to illustrate dependence of human well-being on ES (TEEB 2010). Environmental economic valuation is predominantly conducted within the theory of environmental economics whose analytical foundations are based on neoclassical economics. Neoclassical economics expresses value to society as social welfare which is defined as aggregated preferences of self-interested rational individuals. In this context, sustainability is characterised by the maximisation of social welfare over time. It has been argued that the valuation of ES has reached the limits of mainstream welfare economics, circumscribed by the utilitarian framework (Bowles and Gintis 2000; Gowdy 2004; Parks and Gowdy 2013) (see Table 1 for commonly criticised assumptions and their link to the social value discourse, and Table 2 for a glossary of terms relating to economic value and economic valuation). Furthermore, it is acknowledged that economic valuation fails to account for the total value of ES to society (Irvine et al. 2016; Kenter et al. 2016b). This underestimation has been a substantial factor regarding the continuous degradation of ecosystems (Millennium Ecosystem Assessment 2005, b).

Table 1 Conceptual core and specific assumptions of environmental economic valuation and links to the current discussion about social values
Full size table
Table 2 Glossary of terms
Full size table

Therefore, identifying social values of ES is perceived as an important challenge in environmental valuation (Parks and Gowdy 2013). However, this requires a solid theoretical foundation. The existing literature does not provide a coherent conceptual framework for the identification of social values (Bunse et al. 2015; Kenter et al. 2015; Parks and Gowdy 2013). Recapitulating, many questions about the ontology, elicitation and aggregation of social values remain (Kenter et al. 2016a, 2019).

The discussion about social values in the context of the ES concept and/or sustainability appears to be a relatively newly discovered topic within environmental and ecological economics. However, there are roots in economic theory that explicitly address social values which can be traced back to more than a century. The current critique against economics regarding (neglect of) social values is usually limited to the neoclassical economic conception of value and human behaviour. Likewise, current frameworks fail to account for economic theories which remained outside of the neoclassical realm (Hansjürgens et al. 2017). Consequently, traditions of research on social values in economics have been hardly given appropriate attention in the present literature discussing social values of ES.

The aim of the paper is to review economic traditions outside the scope of mainstream (neoclassical) economics with reference to social values. These economic traditions may advance understanding the notion of social values. They may also contribute to the establishment of a theoretical foundation for the assessment of social values. Furthermore, the review aims to identify recurrent questions and attributes associated with social values in these economic theories.

The remainder of this paper is organised as follows: the next section introduces the historical development of economic value theory especially concerning the role of nature. The historical context is important for two reasons: first, it illustrates the diversity of schools of economic thought and illustrates the path taken within the theory of value towards what some critics refer to as value monism as opposed to accounting for a plurality of values (see, e.g. Gowdy and Erickson 2005). Second, the general discourse in economics on the origin of value affected also the discussion on social values. The following section describes early discussions of social values in economics which mainly centred on value-in-use and value-in-exchange. The subsequent section briefly depicts the theories of externality and social goods which contributed largely to the development of environmental economics while remaining within the realm of neoclassical economics followed by which economic theories outside the realm of mainstream (neoclassical) economics that may enhance the theoretical foundations of social values are presented. Implications for the theory of social values and sustainability are discussed before the concluding section. Finally, conclusions derived from the review.

Historical context: development of economic value concept and the role of nature

The scientific discourse about ES did not emerge before the late 1970s (Gomez-Baggethun et al. 2010). Still, the awareness about services provided by ecosystems or rather negative impacts due to degradation of ecosystems dates back to the ancient world when, for example, Plato wrote about negative impacts of deforestation of the hills of Attica (Mooney and Ehrlich 1997). The current discussion related to social values of ES is inseparable from the historical development of economic value theory and the role of nature. Therefore, this relationship will be illustrated shortly.

Historically Physiocracy is considered the first scientific school of economics (i.a. Bell 1953, p. 121; Ware 1931). The Physiocrats recognised nature as a unique form of wealth while the cultivator was regarded as cooperator that was needed to produce wealth (Bell 1953, p. 131; Quesnay 1962/2003, p. 232). Shortly after Physiocracy and with the dawn of the industrialization began the development of the classical economic theory (Hubacek and van den Bergh 2006). Although ES did not literally appear in classical economic literature, some of its scholars referred to them as natural agents (Gomez-Baggethun et al. 2010; Say 1821/2008, p. 74). Natural agents were considered to be free of charge and, therefore, only their value-in-use was appreciated (Ricardo 1821/2001, p. 207f.). In his work, Capital, Marx (1867/1967) considered labour as the only source of exchange value which is in line with Ricardo’s value theory of labour and implies that the value of an object can be measured by the hours needed to produce it (Hubacek and van den Bergh 2006). He likewise assumed natural agents as costless and ‘spontaneously provided by nature’ (Marx 1867/1967, p. 178). Generally, the rapid industrial and technological development and capital accumulation during the nineteenth century caused a change in the economic rationale resulting in a limited appreciation of nature in economic analysis (Gomez-Baggethun et al. 2010).

In line with this trend and likely shaped by the long-term persistence of the industrial revolution arose the marginalist revolution (Hubacek and van den Bergh 2006). It initiated an essential alteration of the economic methodology by incorporating mathematical maximisation which originated in physics. While production dynamics were at the core of the classical economic analysis, neoclassical economic theory emphasises the static analysis of exchange (value) (Christensen 1989; Mirowski 1991, p. 195ff.). This affected the theory of (economic) value in two ways: first, economic value and commodity prices were derived from subjective preferences of rational individuals. Preferences are assumed to be exogenously given implying that the formation of preferences and process of choice is irrelevant to economic analysis (Bruni and Sugden 2007; Christensen 1989). Hence, psychology was banished from economic analysis by deducing rationality from (internal) consistency of choices (Bruni and Sugden 2007; Samuelson 1938) Second, regarding the role of nature, the marginal revolution restricted economic analysis to exchange value. Thus, the focus shifted towards marketed goods (Parks and Gowdy 2013).

Early economic discussion on social values: value-in-use, value-in-exchange, and social components of value

The general discussion about the economic theory of value intensified again with the rise of the marginal utility theory around the late 1880s (Kurz 1995/2003, p. 71). The debate was not limited to the source of value; instead also the role of society as value provider and possible methodological implications were examined.

The discussion was partly influenced by socialist scholars’ reaction to marginal utility theory such as Rodbertus. He adopted Ricardo’s (labour) theory of value (see Rodbertus 1842) and primarily investigated the role of production cost and amount of labour regarding a good’s value. Rodbertus argued that value-in-use is the only value type, suggesting that value-in-exchange (or exchange value) is social value-in-use (see Wagner 1878, p. 223ff.). According to Rodbertus, value-in-use can be either individual or social. Individual value-in-use relates to individual wants while social value-in-use considers the value-in-use of a social organism (‘sociale Organismus’). Although this social organism is composed of many individuals it has own wants which are beyond the aggregation of individual wants. According to Rodbertus, a social organism’s wants (e.g. a nation’s wants) dominate individual wants and not vice versa (see Wagner 1878, p. 222f.).

Yet, also non-socialist economists contributed to the discussion by debating society’s influence or rather role regarding values, e.g. Gärtner (1887) disagreed with Rodbertus’ definition of social value-in-use. Gärtner (1887) kept the general distinction between value-in-use and value-in-exchange and argued that the latter is not independent of value-in-use. Value-in-use is seen as importance a good is given regarding its ability to satisfy wants, whereas value-in-exchange represents the importance of a good regarding its value-in-use, the good’s availability and need for the good (‘Bedarf’) (Gärtner 1887, p. 422). He also disagreed with Rodbertus about value-in-use being purely social. Instead, Gärtner distinguished between individual and social exchange value and individual and social value-in-use. He highlighted that individual value-in-use may also arise if the individual can use a good to satisfy exchange wants (‘Tauschbedürfnisse’), whereas social value is importance ascribed to a good regarding the satisfaction of societal wants or rather of societies’ average person (‘Durchschnittspersönlichkeit’) and can be either social value-in-use or exchange value (Gärtner 1887, p. 423f.).

Some scholars argued that economic activities such as production, exchange and distribution are social processes (Schumpeter 1909) and that values are socially constructed (see, e.g. Anderson 1911; Clark 1886/1894; Seligman 1901). Thereby, prices and values can be considered to be social phenomena (Clark 1899/1914, p. 40ff.). Clark (1886/1894, p. 83) supported the idea that society and not the individual determines value. Additionally, he was in favour of the assumption that value is an absolute magnitude which is quantifiable by measuring marginal utility (Clark 1886/1894, p. 74, 1899/1914, p. 237) and thereby, emphasised the importance of exchange value. He argued that ‘final utilities to society’ determine prices (Clark 1899/1914, p. 243) and that society would estimate the utility ‘which constitutes a social or market valuation’ (Clark 1886/1894, p. 83). This implies that market value is the utility of society as organic being. So, goods with equivalent market value imply that society estimates identical utilities for these goods. In contrast, the individual measurement of utility represents only value-in-use. Hence, a good which society considers cheap may be beyond price for an individual (Clark 1886/1894, p. 81f.).

Seligman (1901) disagreed with earlier scholars such as Rodbertus and partly with Clark about the general definition of value, in particular about the definition of (social) value-in-use. Seligman (1901) denied that the distinction between exchange and use value is correct because he argued that only (social) marginal utility expresses value. Furthermore, he claimed that his concept of social marginal utility would make the exchange-use-value distinction redundant. He argued that the ‘foundation of value is independent of exchange’ because also an individual that cannot interact with others—like a castaway isolated from society—values goods due to satisfaction of individual wants and consequentially the weighing-off of desires (Seligman 1901, p. 327). Hence, in an individual setting, only two goods are required for the existence of value. Yet, he considered this setting as artificial and ‘actual life’ is about living in a society in which goods are exchanged and humans are ‘social beings’ (Seligman 1901, p. 323ff.). Therefore, he argued that society and not the individual sets value on goods. He reasoned that the aggregate wants of all society members determine the value of the good in ‘actual life’. The subjective wants of the individual (marginal consumer) can only (marginally) affect the aggregate wants of society (Seligman 1901, p. 323). Therefore, according to Seligman value is not individual but social and value in society is expressed by social marginal utility.

Further Seligman (1901, p. 323) argued that individuals consider not only individual wants but also other society members’ wants. Thus, even a good that has no direct utility to the owner may still have value if the good has a ‘social purpose’ (Seligman 1901, p. 324). A good that is useful for another society member has indirect marginal utility for the owner as it has direct marginal utility to society, implying that individual marginal utility is ‘a reflection of social marginal utility’ and that value is the result ‘of a socialization of wants’ (Seligman 1901, p. 325). Additionally, Seligman reasoned that in a social setting only exchange value is relevant because ‘value is a social conception’ which compares different commodities by trading them between individuals (Seligman 1901, p. 326). He concluded that economics as a social science should account for the social conception of value instead of relying on ‘individualistic’ theories (Seligman 1901, p. 347).

In his analysis ‘On the Concept of Social Value’ also Joseph A. Schumpeter (1909) discussed the question if social values are a value category that (partly) substitutes individual values. Schumpeter (1909, p. 213ff.) emphasised that the concept of marginal utility applies only to individuals who value things based on the quantity they have and not based on the quantity that is available for the whole society. Therefore, he supported the principle of methodological individualism because marginal utility can only apply to individuals and not to society as a whole. Schumpeter agreed with Seligman that exchange value can only arise due to interaction between multiple individuals and that for an isolated individual the marginal utilities depend only on the individual herself. Yet, if the individual is part of a society the possibility of trade arises and, therefore, value assigned to a good by the individual is also affected by the wants of other society’s members. This signifies that the individual utility curves are directly affected by social influence. Thus, social influence may form individual demand curves and affect marginal utilities. Still, Schumpeter stressed that the interaction between the individuals is driven by self-interest. Further, Schumpeter opposed Rodbertus’ and Seligman’s idea of society determining values. If value is considered as exchange value, Schumpeter argued that the conception of social value only describes social interaction and influences of ‘mutual interaction and interdependence’. According to him, this social influence on the individual does not oppose individualistic methods (Schumpeter 1909, p. 217f.). He argued that individual valuations determine value and prices because society is not a conscious being ‘having no brain or nerves in a physical sense, cannot feel wants and has not, therefore, utility curves’ (Schumpeter 1909, p. 215).

In contrast to other scholars, Anderson (1911, p. 9ff.) rejected the idea of social (marginal) utility to investigate the nature of social value and to quantitatively measure it. He discussed that individual motives alone cannot explain economic value because motivation relates to ‘something superindividual’—what he refers to as social values—so ‘ends, aims, purposes, desires’ are all affected by the interaction of society’s members (Anderson 1911, p. 199). Also Clark (1886/1894, p. 36) questioned the assumption that human behaviour is only motivated by self-interest and highlighted moral principles and unselfishness as further motives. According to Schumpeter, who was more concerned with methodological issues, the only wants which are ‘strictly social’ are expressed by a community which consists of individuals that act collectively and consciously as such (Schumpeter 1909, p. 216). Although recognising the existence of altruistic or social wants Schumpeter (1909) reasoned that they can only be accounted for by individuals. Consequently, he argued that the individuals’ motivation for demand is irrelevant for the analysis. Therefore, he concluded that the value theory of marginal utility should be based on an individualistic methodology.

Externalities and social goods: beyond market prices

The focus of the above-illustrated economic value theory was primarily on exchange value, value-in-use and market prices. Thereafter, Pigou (1920) initiated a discussion about uncovered (social) costs which are not reflected by market prices especially regarding environmental goods. The central argument of Pigou’s analysis of social costs is that for an economic activity, the ‘marginal private net product’ and the ‘marginal social net product’ may diverge, e.g. due to uncompensated costs of people not directly involved in the economic activity (Pigou 1920, p. 114ff.). In other words, a person’s activity, e.g. consumption or production has a negative or positive impact on a third person that is not directly involved in the activity. The person taking the action does not take into account the costs or benefits imposed on the other person. This uncompensated damage (or “spill-over” benefit) is usually referred to as externalities. Pigou (1920, p. 168) suggested that these externalities can be internalised by state intervention, e.g. in form of taxes or subsidies.

Howard Bowen (1943, p. 27) contributed to this discussion by distinguishing between private goods and ‘social goods’, commonly referred to as public goods. He defined public goods as indivisible and non-excludable as they are part of the environment individuals live in (Bowen 1948, p. 173). He stated, without further explanation, that the marginal utility theory of individual goods is transferable to social goods. Hence, the aggregation of the individual marginal rates of substitutions reflects societies’ ‘curve of total marginal substitution’ which closely corresponds to the total demand curve (Bowen 1943, p. 30). In other words, individual preferences for the provision of a public good can be illustrated in a curve showing how much money the individual is willing to give up for the good’s provision. The aggregation of all society members’ preferences then expresses how much society as a whole is willing to spend. Bowen emphasised the difficulty in finding a reasonable unit to quantify social goods that are often complex. He considered voting as the best procedure to reveal and aggregate individual preferences which are not observable by consumer choice to optimally allocate public goods. Yet, he stressed that voting on marginal changes in a good’s quality does not necessarily determine the optimum output.

These theories developed by Pigou, Bowen and others extended existing neoclassical theories by important aspects and contributed significantly to the development of environmental economics. Yet, they remained within the realm of neoclassical economics. In contrast, the following two sections will present selected economic theories that challenged the neoclassical paradigm. These theories intended to (1) link individual behaviour, preferences and values with embeddedness in institutional social and cultural contexts, (2) account for complex human behaviour, morality, social influence and culture, and (3) thus go beyond the neoclassic rational choice theory and concepts of utility.

Social values in (old) institutional economics: Clark’s and Kapp’s holistic approach

Social values have also been discussed in the realm of institutional economicsFootnote 1 challenging the narrow assumptions of neoclassical economics. One of the pioneers was Veblen (1898) criticising the assumptions of homo economicus and utility maximisation. Generally, institutional economists were especially concerned with developing a more holistic theory of human behaviour and social value compared to the above-presented concepts based on marginalism.

In contrast to neoclassical value theory, Clark (1936, p. 54) called for a concept of (social) value which does not depend on market valuations. He considered social value to be ‘value to society’. The assessment of the latter requires a ‘truly organic social valuation’ which is not just the aggregation of individual utilities (Clark 1936, p. 49). He argued that market value does not measure social value and that market value is only social because the context is ‘an organic social situation’ (Clark 1936, p. 50). He reasoned that net products may diverge from market value due to interdependent utilities (Clark 1936, pp. 45, 56) and due to ‘intangible utilities’ (Clark 1936, p. 45). Hence, social value is seen as ‘anti-marginal’ because exchange cannot be considered to be independent of complex social states and relationships within a society. Therefore, marginal utility cannot determine the price because the price relates to the ‘value of the whole’ (Clark 1936, p. 59). Yet, Clark noted that social value will never be comparable to the quantifiable approach of exchange value in form of market prices. Nevertheless, he argued, a theory of social value may still deliver important insights (Clark 1936, pp. 44, 60). Especially, when value is linked to motives and desires which may be achieved by incorporating insights obtained by psychology, ethics and sociology (Clark 1936, p. 60f.).

Another milestone in the theory of social values was developed by Karl William Kapp. Kapp’s (1950/1975) theory of social values or rather social cost broke with the work of early neoclassicists such as Clarks, Anderson and Seligman which argued—as illustrated above—that society determines the value of a commodity. Kapp disagreed with these authors who identified market prices as social value indicating the value of a good to (all individuals of a) society (Kapp 1950/1975, p. 256). Analysing the neoclassical theory of externalities, especially referring to Pigou, Kapp (1950/1975, pp. 37ff, 256f) criticised that subjective value theory recognises only individual preferences and that a social valuation must incorporate social benefits (exchange value on the market) and social cost (negative impact on society which is not accounted for by the market exchange) which jointly represent the social value to society.

Although Pigou’s definition of externalities and Kapp’s definition of social cost seem similar, their concepts differ. Kapp’s concept extended the scope of social costs beyond the market sphere. Kapp (1970, p. 841) argued that the analysis of externalities abstracts economic activity into an ‘autonomous “economic” sphere’. Thereby, the (neoclassical) economic analysis of externalities neglects power relations which cause non-autonomous behaviour, and it does not account for the severe environmental and societal effects caused by production and distribution (Kapp 1970, p. 841f.). Kapp (1969) questioned markets as efficient institutions for coordinating behaviour because costs can be shifted towards other individuals and the environment due to asymmetric power relations. Additionally, profit-maximising behaviour would induce cost-shifting as rational behaviour (Kapp 1970, 1950/1975, p. 13).

Concerning the translation of individual preferences scales into collective preference scales Kapp (1950/1975, p. 257) questioned earlier theoretical analysis of social evaluation. He acknowledged Bowen’s (1943) analysis that formally the individual marginal rates of substitutions can be combined to a collective marginal rate of substitution in order to express the collective WTP for a marginal social benefit. However, he argued that formal marginal analysis and analysis of externalities might fail to account for social preferences due to the impossibility to measure or calculate marginal costs and rates of substitutions. Therefore, he argued, also social cost cannot be measured in practice (Kapp 1950/1975, p. 259f.).

According to Kapp (1978, p. 288ff.), social values must be based on collective decision-making instead of individual rationality, and democratic and participative processes are indispensable because of conflicting objectives and interests (Kapp 1978, p. 317). Furthermore, social well-being cannot only be based on maximisation of aggregated individual utilities because in democratic societies these individual utilities are socially evaluated to judge social well-being. These judgments may also consider desired social ends (Kapp 1950/1975, p. 260).

Kapp supported the idea that economic activity is tightly linked to the natural and social system (Kapp 1970, 1985, p. 150ff.). He emphasised that humans can be seen either as isolated individuals or as social beings embedded in society (‘Gesellschaftswesen’) (Kapp 1936, p. 43). Dependent on the perspective the satisfaction of individual needs or societal needs will be relevant. He further argued that this distinction is not about constructing two opposing views but to recognise the needs of society as needs of all individuals. Yet, if a human considers herself to be isolated, the subjective preferences will be individual. This behaviour is evoked by the market which sets the focus on exchange value and thereby, reveals preferences of isolated individuals while neglecting societal interests (Kapp 1936, pp. 42–44). Therefore, Kapp (1950/1975, p. 260f.) called for the development of methods to discover individual preferences regarding social ends. So, Kapp (1950/1975, p. 260) held the view that for theoretical and practical reasons a justification of social preferences has to extend the utilitarian concept which defines total welfare as the aggregate of individual utilities. Kapp (1977, p. 538) reckoned that because of environmental problems it is necessary to consider the utilitarian premise of pleasure maximisation as secondary objective while the primary objective is the ‘social and moral imperative of minimizing human suffering’. The principle to minimise suffering is referred to as negative utilitarianism (Popper 1962, p. 284f.; Smart 1958). In other words, the ‘individualistic moral principle’ is subordinated to a social one that is in line with social and ecological sustainability. To achieve this, Kapp (1977, p. 538) considered appropriately designing institutions and policies. Yet, he notes the difficulty to agree on certain minimum standards and recommends political process to achieve consensus.

As illustrated above, (old) institutional economists emphasised human embeddedness in natural and social context implying that individuals may have interests that transcend individual values, e.g. towards social ends. Yet, the presented theories remain vague how these aspects could be included in the economic analysis. Therefore, the following section presents theories that elaborated more detailed contributions.

Complex human behaviour, multiple preference orderings and interdependent preferences: beyond self-interest and individual values

The criticism of the concept of homo economicus has a long tradition in economic theory (see, e.g. Veblen 1898). One of the arguments brought forward by critics is preferences’ interdependency (see, e.g. Veblen 1918). It implies that preferences may be affected by social environment and culture. Yet, these arguments did not find their way into mainstream economics. According to Duesenberry (1949, p. 17ff.), marginal utility theory ignores the nature of preferences. If preferences are interdependent, it is essential to understand their nature and how they change. Therefore, he called for the analysis of motivation and incorporation of psychological assumptions to account for the influence of social factors and culture on preferences (Duesenberry 1949, p. 20). The theories by Harsanyi, Sen, Buchanan and Musgrave presented hereafter transcend individual values and self-interest, and/or do not assume that individual choice and individual welfare are necessarily linked.

John Harsanyi’s utilitarianism

The early work of Harsanyi (1955, p. 315) distinguished between an individual’s ‘subjective preferences’ and ‘ethical preferences’. The former reflect what the individual actually prefers and affect only the personal utility function. In contrast, ethical preferences regard what is preferable from the societal perspective. These preferences are only expressed when the individual ‘forces a special impartial and impersonal attitude upon himself’ (Harsanyi 1955, p. 315). The individual adopts social considerations by interpersonal comparison. In other words, the individual puts herself in the position of all other society members and assumes to have an equal chance to be in any of their positions. This concept is similar to Rawls’ (1971/2009) ‘veil of ignorance’ regarding the concept of fairness. However, the outcome of Harsanyi’s preference adoption and equal probability assumption differs from the outcome obtained by the uncertainties of the ‘original position’ defined by Rawls. The latter hinders the individual to know her position in society and, therefore, the individual does not know her individual utility function. In contrast, the idea behind Harsanyi’s utilitarianism is that the (expected) social welfare is the weighted aggregation of the (expected) individual functions. Hence, it has to be highlighted that in Harsanyi’s theory the individual still acts as a rational utility-maximizer, however, he/she acts just under uncertainty.

Armatyra Sen’s theories of meta-rankings, commitment and sympathy

Armatyra Sen (1977) opposed the idea that utility maximisation or rather the idea that an extended utility function can fully explain human behaviour. Instead, Sen (1977) distinguishes between different types of altruistic behaviour: commitment and sympathy. The notion of sympathy is similar to neoclassical conceptions of altruism (see, e.g. Becker 1974) where the increase of another person’s welfare increases the individual’s own well-being directly. In contrast, actions based on commitment are motivated by a sense of duty, are non-egoistic and may even affect personal welfare negatively. Hence, behaviour arising due to commitment is not in line with the assumptions of self-interest and individual utility maximisation. Commitment is not relevant for private goods characterised by perfect excludability and rivalry but is relevant for public goods (Sen 1977).

If the individual has “multiple selves”, the question is again, as in the case of Harsanyi’s subjective and ethical preferences, which preferences the individual considers. While Harsanyi assumed that social preferences must be enforced by the individual, Sen (1977, 1982) held the view that preferences are hierarchical: he referred to the concept of meta-rankings. Meta-rankings rank preference rankings and imply that individuals do not only maximise their own utility but also incorporate moral judgments. Therefore, not only preferences under certain constraints can be ranked in order to maximise utility, but the individual can also reason what to maximise and may include, for example, non-utilitarian aspects. Sen (1977) considered Harsanyi’s dual structure of preferences unsatisfactory. While sympathy may be captured by subjective preferences, it remains unclear to him how commitment relates to them. In contrast, meta-rankings allow transcending the context and constraints existent for the actual choice and provide insights into the individual’s morality.

James Buchanan’s constitutional economics

Inspired by the work of the Swedish economist Wicksell (1896), James Buchanan largely contributed to the theory of constitutional economics. In contrast to neoclassical economics which investigates human ‘choice within constraints’ constitutional economics focuses on ‘the choice of constraints’. In other words, it aims to explore choices which are made in alternative hypothetical social frameworks. The latter are characterised by constitutional, legal and institutional rules (Buchanan 2008, p. 1f.).

Buchanan’s concept of social choice remained on the basis of individual rationality and he emphasised the divergence between the philosophical foundations of individualism and the organic concept of society. Both are considered to be useful for certain problems but social rationality can only be discussed referring to a social organism which itself has values or ends—and not in the case of individuals obtaining value orderings (Buchanan 1954b).

Beginning with the focus on individual decisions, Buchanan (1954a) argued that individuals behave differently in different contexts such as the market and the political arena because distinct preference scales influence behaviour. In the market context, individuals act as atomistic beings which do not incorporate interdependencies. In comparison, in a political context, the individual is aware of the decision-making-process’ social character and her participation in this process as well as her vote’s influence. Hence, individuals may consider a ‘more inclusive value scale’ and may consciously choose for the group, e.g. by accounting for interdependencies—internalising their actions’ externalities (Buchanan 1962, p. 24). Yet Buchanan argued that a dichotomy of behaviour is unrealistic and supports methodological consistency regarding human behaviour. In his opinion, behaviour may be based on ‘moral or ethical principles’ which leads to other-regarding behaviour and may ‘inhibit individual utility maximizing behaviour’ (Buchanan 1961, p. 340). Yet Buchanan opposed the idea that individuals act socially or based on self-interest due to a duality of selves; instead, he argued that behaviour is dependent on the context which defines guiding principles (Buchanan 1954a, 1962).

Buchanan argued that these decisions constrained by rules are made on the post-constitutional level, i.e. the level of daily decision-making (see, e.g. Buchanan 1959). Yet, collective decision-making has a second layer, the constitutional level, on which the “rules of the game” are chosen. According to Buchanan, the normative criterion for selection of “good” rules is not based on the efficiency criterion; instead, he introduced the ‘unanimity rule’ implying that collective decision-making cannot be justified if an individual is worse off. Unanimous consent may be achieved throughout the process of decision-making if everyone expects to benefit (Buchanan 1954b; Buchanan and Tullock 1999, p. 85ff.). To define “good” or fair rules of collective decision-making, Buchanan and Tullock (1999, p. 78ff.)—similar to Rawls (1971/2009)—argued in favour of a veil of uncertainty meaning that the individual does not know her position/role in society when agreeing on rules. Therefore, the individual cannot have any particular interests besides collective or social ones. Yet also the role of deliberation as a method to reveal preferences is highlighted. First, directly by stressing that preferences may not be ex ante given and may be formed through discussion and social interaction (Buchanan 1954b); second, indirectly as the unanimity criterion is grounded on similar arguments to those of deliberation within theories of consensus (Hansjürgens et al. 2017).

Richard Musgrave’s theory of merit goods

As described above, economic theory distinguishes usually between private goods and public goods, often with a particular focus on externalities. Musgrave (1957, 1959) introduced with his concept of merit goods an additional category. While a clear definition of this concept does not exist (Andel 1984; Musgrave 2008), merit goods often refer to a good’s evaluation which involves norms different from consumer sovereignty (Musgrave 2008). Common examples are healthcare and education that could be provided by the market but would be under-consumed or in the case of demerit goods the restriction of drugs.

Musgrave (1959, p. 8f.), in his concept of merit wants, differentiated between ‘social wants’ and ‘merit wants’. Social wants refer to public goods which are characterised—following the definition of Samuelson (1954)—by non-excludability and non-rivalry. As people will not voluntarily pay for these goods, the market cannot satisfy these wants. Any intervention of the state regarding social wants aims to supply the optimum amount of a good and to satisfy consumer preferences (Musgrave 1959, p. 10f.). Thus, the public good intervention has consumer sovereignty as underlying norm.

In contrast, merit wants refer to situations in which interventions aim to correct consumer preferences. Resources are allocated to satisfy wants which could be provided by the market but individuals choose differently (Musgrave 1959, pp. 8–14). Hence, the concept of merit goods implies that for certain goods the market demand does not correspond to the optimum demand and that such goods involve value judgments different than consumer sovereignty (Ver Eecke 1998). Musgrave (2008) identified five settings in which either consumer sovereignty is difficult to implement but still preferred or the evaluation of the good is based on different norms and, therefore, violate the norm of consumer sovereignty. First, ‘pathological cases’ in which the individual does not choose what is best for her, e.g. due to time discounting. Second, situations in which individual preferences are conditioned by society. Musgrave referred to this as ‘rule of fashion’. These two cases do not dismiss consumer sovereignty as preferred but rather aim to correct individual preferences. A third case is “community preferences”. Private preferences and community preferences may diverge if the individual considers herself as member of the community. An example would be payments for preservation of historical monuments (Musgrave 2008; Musgrave and Musgrave 1989, p. 57f.). Referring to Colm (1965); Musgrave (2008) argued that the formation of common preferences can be explained without the existence of a social organism. In a society, ‘common concerns’ are developed due to social bonds and culture which may lead to the development of ‘common wants’ and, therefore, consumption of private goods or support of public goods may diverge from individual preferences (Musgrave 2008; Musgrave and Musgrave 1989, pp. 55–58). A fourth argument is “paternalism in distribution”: society maybe concerned with redistribution of income to cover society members’ basic needs but do so by providing the actual goods instead of the monetary equivalent. Fifth, as discussed above, value judgments may also refer to some “higher values” or multiple preference orderings as discussed in the theories of Sen or Harsanyi. In this context, merit goods are chosen due to ethical preferences or rather commitment, and the assumption of consumer sovereignty remains.

Musgrave (2008) considered the case of diverging community values and individual preferences as most relevant application of the concept. While Musgrave’s or rather Samuelson’s concept of social wants is individualistic. Merit goods may transcend the assumptions of an individualistic conception. Hence, Musgrave broadened the perspective towards a societal focus by emphasising that humans are social beings suggesting that preferences and actions cannot be separated from the social environment. He assumed that individuals are able to evaluate private and social wants. He supported this view by noting that otherwise democratic processes such as voting could not function (Musgrave 1959, p. 10f.).

Discussion

As mentioned above, many questions about the ontology, elicitation and aggregation of social values remain (Kenter et al. 2016a, 2019) and a consensus about the understanding of shared and social values does not exist (see, e.g. Irvine et al. 2016). The current debate contributed to the understanding of preference construction and to the development of new techniques to elicit shared and social values (see, e.g. Kenter et al. 2016b; Orchard-Webb et al. 2016; Raymond et al. 2014). Nevertheless, as Irvine et al. (2016, p. 1) emphasised the ‘fundamental questions’ are equally important. One of these questions is in how far current economic approaches are able to deal with complex issues such as social values for sustainability. According to Dasgupta (1985), economic theory can be circumscribed by epochs which emerge due to changing circumstances. Each epoch poses new questions regarding different problems and hence, old theories cannot be dismissed but are relevant dependent on the context (Dasgupta 1985, p. 143). While Dasgupta’s analysis may well hold regarding the most influential schools of thought in Britain, it does not account for the heterogeneity of economic theory within these epochs as illustrated above for the case of social values. Further, recent contributions to the economic valuation have illustrated the potential of combining different theories. With reference to the critique against conventional environmental economic valuation, especially stated-preference methods, Deliberative Monetary Valuation (DMV) and Deliberative Democratic Monetary Valuation (DDMV) were developed. These approaches combine economic with deliberative processes (Orchard-Webb et al. 2016; Spash 2007) to inform preferences and to account for value plurality (Lo and Spash 2013). Deliberation is a promising approach to form and express shared values (Kenter et al. 2016a). However, the theoretical basis of DMV remains vague (Bartkowski and Lienhoop 2018; Bunse et al. 2015) and the combination of two different theories entails conflicts, e.g. mix of communicative and individual rationality (Vatn 2009). Bartkowski and Lienhoop (2018) illustrated how the integration of Sen’s relevant theories can strengthen the theoretical foundation of DMV and may overcome some problems of conventional economic valuation.

Although most of the theories presented in the last two sections did not discuss social values explicitly apart from the old institutional economist, they touch upon topics which often serve as starting point for the current discourse about social values of ES. Incorporating these theories into the discussion of social values provides opportunities for developing a more solid theoretical foundation, may contribute to the understanding of the ontology of shared and social values and may further strengthen the theoretical basis of DMV. In the following, some links to the current debate will be highlighted (see also Table 3).

Table 3 Contributions of the identified theories to the theoretical foundation of social values in response to common criticism against the neoclassical economic framework
Full size table

In the context of environmental goods, people are likely to have “impersonal preferences”, preferences which are considered to be independent of the personal state, besides individual preferences (Dworkin 1981). Thereby, the “isolated individual approach” does not correspond with the complexity of ES and social values (Kenter et al. 2014). In the identified economic theories, humans are not necessarily seen as atomistic individuals. A more holistic view is adopted emphasising human embeddedness in society and nature. This embedded conception is characterised by interdependencies, institutions and cultural factors. By conceptualising the individual in society the theories transcend individualism in form of individual contexts and values. In the current debate, values that transcend specific contexts or situations are referred to as transcendental values (Kenter et al. 2015). The theories identify values which are not only originating due to individual utility maximisation and/or individual preferences but allow for the recognition of plural values, e.g. values and wants of the community and/or preferences beyond self-interest. A more complex human behaviour is outlined by incorporating preferences or rather types of values based on, e.g. deontological ethics, considerations of justice, and norms. Human behaviour may not only be instrumental and rational (utility-maximising homo economicus), but may transcend the common assumption of means-ends instrumentalism. This implies that values may not only be based on individual preference satisfaction but may also be affected by morality, value hierarchies (different rankings of values may exist leading to conflicting values), multiple preference orderings (imply that each individual has “multiple selves” which cause context dependent preferences) and/or interdependencies. Hence, the identified theories challenge the assumption of narrow individualism and regard humans as social beings.

As illustrated above, a long discourse deals with the question if society or only conscious beings can hold values (see Schumpeter 1909) and how social values may be elicited, through individuals or through a supra-individual entity (see Kenter et al. 2015; Musgrave 1959)? The identified theories suggest that individuals are able to hold or rather express social preferences. The arguments brought forward to support this assumption were diverse such as preferences behind a veil of ignorance, commitment, meta-rankings and common wants. Hence, individuals may not only consider their personal wants but also what is desirable from a societal point of view. Further, especially the theories of Kapp, Buchanan and Musgrave reason about social ends and/or social value judgments with regard to social well-being and individual utility. In this respect, these economic theories share many aspects that overlap with the broad definition of sustainability such as long-term focus due to social ends, justice, fair distribution and allocative efficiency (Norton et al. 1998).

The identified theories illustrate that values beyond the individual may exist. Yet, the question remains what the implications for policy-making are. Some scholars argue to limit individual choice sets due to incompatibility of consumer sovereignty and sustainability (Menzel and Green 2013) or rather promote solutions based on societal preferences (O’Hara and Stagl 2002) to enhance sustainability. Others question whether individual decisions should define public choices due to diverging individual and social time preferences (Marglin 1963). As the identified theories illustrate that society may hold values different from individual values, for example, due to longer existence and/or merit wants, state intervention to correct individual preferences which are against sustainability may be justified if the socially accepted norms transcend consumer sovereignty. In the evaluation of certain goods different norms are already considered (Musgrave 2008). Yet, it is not clear how individual preferences should be corrected if consumer sovereignty is dismissed; thus, the challenge remains to identify desired social ends. Based on the assumption of preference endogeneity, participation and deliberation could play an important role for the development of criteria for correction of individual preferences and determination of socially preferred end-states. Ravenscroft (2019) suggested a new normative approach inherently linked to sustainability which emphasises the formation and articulation of shared social values for society to express how resources ‘ought to be’ allocated before their actual allocation.

Further, Renner (1999) argues that linking theories of constitutional economics to sustainability policy will make normative assumptions explicit so alternative policy options can be deliberated upon. This may lead to more democratic decision-making, and criteria in line with sustainability may be guiding policy-making. Theoretically, institutions could be designed in a “good” way to incorporate sustainability as a factor.

In general, policy implications derived from these theories are far from clear. Recommendations based on the neoclassical paradigm have the advantage that the only normative aspect is the efficiency criterion and policy-makers can fairly easily include obtained insights into decision-making. In contrast, the above-presented theories would introduce different criteria and, therefore, involve also more complex normative considerations. Thus, the ability to provide policy recommendations regarding the current societal framework are limited (Buchanan 2008).

Concluding remarks

In this article, contributions of selected economic theories to the concept of social values were reviewed. It was found that the discussion is far beyond the scope of neoclassical economic theory and that economic theory is not as narrow as some critics claim. Still, the current literature on social values does not account for insights that can be derived from the past while the literature should take account of the long discourse about social values in economics and reflect on the progress already achieved. Furthermore, the review illustrated that identical or similar topics reoccur, resulting in the identification of recurrent attributes associated with social values: (1) complex human behaviour and multiple preferences; (2) relevance of human embeddedness in nature, social relations and culture; (3) value pluralism and hierarchies; (4) public participation and social learning; (5) preference aggregation; (6) interdependence of preferences and utility; (7) issues of distribution, power and justice.

Several strands in economic theory that touch upon social values were identified (among others): Kapp’s theory of social cost; Harsanyi’s utilitarianism; Sen’s theories of meta-preferences, commitment and sympathy; Buchanan’s constitutional economics; and Musgrave’s theory of merit goods. The review illustrated potential contributions of these theories to the theoretical foundation of social values and that they may strengthen the theory of DMV to address commonly voiced concerns regarding economic valuation of the environment. Since the intention of this paper was primarily to identify relevant theories and illustrate their potential contributions, there is still a need to integrate these theoretical insights into a conceptual framework.