How Institutional Economics May Support the Analysis of Individual and Collective Capabilities

ABSTRACT

The analysis in this article starts from the recognition that institutions make an important part of the Capability Approach, as conversion factors from resources to capabilities and as affecting agency. The tradition of Old Institutional Economics is critical of the neoclassical view of the individual and agency and emphasises the importance of social context. Hence, there is some common ground between the Capability Approach (CA) and Old Institutional Economics (OIE). The purpose of this article is to explore how insights from OIE might enrich the CA both conceptually and empirically. This may be done for individual capabilities as well as for collective capabilities. A better understanding of collective resource-institutions and collective agency-institutions will also contribute to the analysis of capability expansion in the community economy through collective productive capabilities in commons, cooperatives and mutuals. The conclusion is that an understanding of institutions in the tradition of OIE may help to clarify the relationships between social norms and agency; complement the notion of constrained choice with enabling institutions; and point at how institutional transformation may support individual and collective capability expansion.

KEYWORDS:

• Capability Approach
• Old Institutional Economics
• institutions
• Thorstein Veblen‌
• community economy
• collective capabilities

Introduction

Institutions play an important role in the Capability Approach (CA). First, and most explicitly, in the conversion from resources to capabilities. Here, institutions are generally portrayed as conversion factors, which determine to what extent resources can be turned into capabilities for people with different characteristics and in different circumstances. In general, CA research tends to recognise that these resource-transforming institutions can: constrain or stimulate behaviour and through that, enable or disable capabilities; they can be universal (having the same effect on those involved) or targeted (working out differently for different groups); and they may be formal or informal. But these roles of institutions remain largely implicit in the CA, as Nambiar (2021) has also noted recently. Secondly, institutions also play a role at the stage where agency combines with opportunities in the choice of valued functionings. There, they often tend to be seen in the CA as constraints on choice, while OIE also recognises institutions which support agency and enhance the wellbeing of oneself or others. A third way in which the CA might benefit from insights from OIE, is that both the CA and OIE rely not only on individuals but also on social structures in their explanations, as Hodgson’s (2003) analysis of Thorstein Veblen – the founder of OIE – shows. In OIE “explanations always involve both individuals and social structures (Hodgson 2012: 287)” [emphasis in original].

In the next section, I will summarise the role of institutions in the capability literature and I will explore key insights from Old Institutional Economics in the following section. This will be followed by a section on what we might learn from OIE for the CA. In the section thereafter, I will discuss how institutions as conceived in OIE might also support the analysis of collective capabilities (which I will define in that section). The paper will end with a conclusion.

The Role of Institutions in the Capability Approach

The concept of institutions in the CA is not used regularly and consistently. For example, in the recent Cambridge Handbook of the Capability Approach (Chiappero Martinetti, Osmani, and Qizilbash 2021), the index does not have an entry for institutions. Also, there is very limited literature in the CA focusing on institutions and an important type of institution such as social norms. In this section, I will briefly review the role of institutions as they appear in the CA.

If we start from a schematic overview of the CA, as is done for example by Ingrid Robeyns (2017 – see f.e. the flow chart on p. 83), we can distinguish various stages. Institutions feature in two of these stages. The capability process starts with (access to and control over) resources. The transformation of resources into capabilities is influenced by institutions, which are sometimes labelled as conversion factors. These differ between persons due to different characteristics and circumstances. “All conversion factors influence how a person can be or is free to convert the characteristics of the resources into a functioning” (Robeyns, 2017, 46). Following Robeyns with a small adaptation, the CA distinguishes three types of conversion factors: personal (f.e. constraints as a consequence of a disability), socio-economic (f.e. discrimination in labour markets), and environmental (f.e. missing roads in remote rural areas). These conversion factors are also referred to as “structural constraints”, including “social institutions”, “social and legal norms” and “other people’s behaviour and characteristics” (idem, 83). In turn, structural constraints are defined as “the institutions, policies, laws, social norms and so forth, that people in different social positions face” (idem, 65). Conversion factors (institutions) generally tend to be regarded as constraining capabilities. At the same time, various CA researchers use the notion of entitlements to refer to supportive structural social arrangements (which are generally also institutions) which are supposed to remove or compensate capability reductions (see Sen 1984,¹ and for a recent discussion, Reddy and Daoud 2021). In the next section, I will argue that all of the terms referring above to conversion factors and entitlements are generally considered as institutions in OIE.

The next stage in the process of capability expansion is the centre stage: the capability set which provides an individual with the freedom to achieve certain valued functionings. This involves a choice-process referred to as “constrained choice” as phrased by Robeyns (2017, 83). These constraints seem to refer to similar factors as conversion factors, and are framed, again, exclusively as negative. The difference is not always clear in research in the CA. The first set of institutions as they are acknowledged in the CA is resource-related (as conversion factors and entitlements to address these) whereas the second set is agency-related (as constraints on choice). But which type of institutions should be regarded as affecting the transformation of resources into capabilities and which institutions should be seen as affecting the choice of capabilities into functionings? Let me give an example of why the lack of a clear distinction of the two types and roles of institutions in the CA may be unhelpful. Gender discrimination in hiring in the labour market could be regarded as a resource-institution, affecting women’s access to paid employment, while the agency of women in promotions may be constrained due to the glass ceiling, possibly having to work twice as hard as men to obtain a promotion. In the first case, their lack of access to secure employment or a living wage seems to be the constraining institution, whereas in the second case, their agency is constrained, which may also affect their sense of agency in trying (or giving up on) moving up the company job ladder. A distinction between these two roles of institutions matters for the analysis of capability expansion.

Let me now discuss these two types of institutions in the CA a bit further, to see how we might disentangle them. The CA refers to various forms of entitlements that give people access to and control over resources that, through production and consumption, would provide them with a valued capability set. Sen (1992; 1996) explicitly refers to the social character of entitlements, as social arrangements, and CA scholars often link these to the state, through the constitution, laws, regulations and governance (Gelber 2021). These all refer to formal institutions, backed up with fiscal resources and enforcement by the state. Other CA scholars have suggested that such entitlements may not be sufficient to realise nontangible capabilities such as dignity (Alkire 2005).² Drydyk (2021) notes that the freedom of advantaged groups can reduce the options (capability set) of members of disadvantaged groups through their abuse of power. This may happen through discriminatory informal institutions, through habits and routines leading to the treatment of certain groups with contempt, lack of dignity or other forms of disrespect, as various CA scholars have discussed (Burchardt 2021; Reddy and Daoud 2021). Which informal institutions could be drawn upon to address such capability failures? In other words, which non-state entitlements might play a role here? It remains unclear how dignity and respect, and suitable informal entitlements to support them for everyone, feature in the capability framework. Are they conversion factors? Or constraints on agency? In order to be able to shed some light on these and related questions, I will now discuss the concept of institutions as developed over more than a century in the tradition of OIE.

What are Institutions in Old Institutional Economics?

Institutions and Agents

Institutional economics started with the original, or old, institutional economics, by Thorstein Veblen at the end of the nineteenth century and flourished in the first half of the twentieth century when it made an impact on economics in the US (Ayres 1932; Commons 1924; 1931; 1934; Hamilton 1919; Veblen [1899] 1934; 1919). It starts from the recognition that some social structures affect economic behaviour and outcomes. “The original institutional economists, in the tradition of Thorstein Veblen and John R. Commons, understood institutions as a special type of social structure with the potential to change agents, including changes to their purposes or preferences” (Hodgson 2006, 2). More precisely, they perceived “institutions as systems of established and prevalent social rules that structure social interaction” (idem, 2).

Today, institutional economics includes a variety of perspectives beyond OIE. In particular, there is the New Institutional Economics (NIE) (see, for example, Coase 1937; North 1990), which rejects some and adopts other elements of neoclassical economics (see, f.e. Hodgson 2004). In this article, I will rely on OIE because it seems closer to the CA than NIE in two ways. First, in terms of the critique of mainstream economics (such as the rationality concept – see f.e. Veblen’s (1919) criticism of mainstream rationality as “a lightning calculator”).³ Second, because of the foundation of both OIE and the CA in other ethical approaches than utilitarianism (in particular deontological ethics, see, f.e. Searle 2005, and virtue ethics, see, f.e. Nussbaum 1998). This, of course, does not imply that there are no useful links between NIE and the CA. But here I have decided to focus on what OIE might bring to the CA.

I will start from a broad definition of institutions as “durable systems of established and embedded social rules that structure social interactions” (Hodgson 2004, 424). This happens through the normative dimension of institutions: in circumstances X do Y, as pointed out by Hodgson (2006, 3). Interestingly, Elinor Ostrom (2005, 3), a political economist working on commons and their institutions, has not declared that she is part of OIE but has provided a similar definition of institutions⁴: “the prescriptions that humans use to organize all forms of repetitive and structured interactions.” In the sections on collective capabilities, I will return to Ostrom, because her application of institutions to common-pool resources is relevant for analysing collective capabilities.

In OIE, institutions can reflect power and status, and therefore can benefit a dominant group rather than society as a whole, as Veblen ([1899] 1934) noted sharply in his Theory of the Leisure Class (published in 1899). They tend to be internalised through socialisation, peer pressure and habituation, which may occur through group membership – a process in which the individual and the social level are intertwined, as also has been recognised in social economics (see, f.e. Davis and Dolfsma 2015; Etzioni 1990; Granovetter 1985). This is why the terms ‘norm’ and ‘social norm’ are often used to point at an important component of institutions. This immediately makes clear that institutions are normative, in the sense that “norms delimit a ‘how’ and a ‘should’” (Morgan and Olsen 2011, 429). This is not to say, however, that all institutions can be best described as social norms – think, for example, about laws and organisations. Norms generally tend to refer to informal institutions, such as gender norms (see, for example, Van Staveren and Odebode 2007) or the behavioural norms belonging to an ideology, such as neoliberalism (Wrenn 2021), or the consumerist norms that Veblen ([1899] 1934) already criticised (well-known from the “Veblen effect”) and which Starr (2008) argues help to explain under-saving by many Americans.

OIE’s concept of rationality tends to be partially different from that in neoclassical economics and includes ideas which were at the time or later phrased in terms of bounded rationality, uncertainty, endogenous preferences and status, in relation to rules and norms. According to Hodgson, working largely in the tradition of OIE, this implies that rationality involves meaning, which in turn relies on institutions. “The attribution of meaning (…) requires the use of acquired concepts, symbols, rules and signs” (Hodgson 2004, 422). This influences our behaviour. “We rely upon routines and habits, which are both forming and formed by social institutions” (Hodgson 1988). Hence, the relationship between agency and institutions is pictured as two-directional, in which individual behaviour and institutional context mutually influence each other (Hodgson 2004). According to Veblen ([1899] 1934), life is a process of selective adaptation to circumstances – not so much by individuals, each on their own, but through selection of institutions. This is understood in OIE as a continuous two-way process in which individuals show purposeful behaviour and institutions emerge, or are created deliberately, to enable the achievement of their purpose, while changing circumstances make some institutions obsolete and urge others to be changed or created, while the intentions and purposes of individuals may then change as well (Hodgson 2003). This implies that institutions are not merely constraints on given preferences and choices, but also affect “the very wants and preferences of individuals themselves”, as Hodgson (2003, 17) has argued.

Six Broad Categorizations of Institutions

The tradition of OIE seems to imply at least six broad categorizations of institutions, which seem relevant for the CA. Each refers to a contrast or opposition. A first broad categorisation may be seen between universal (affecting everyone involved in the same way, such as driving on the left side of the road in the UK for drivers of vehicles) and targeted institutions (affecting different social groups differently, such as official menstruation leave in Spain for pre-menopausal employed women). Targeted institutions often reflect power and have been elaborated particularly by feminist economists in relation to gender (Van Staveren and Odebode 2007) and in stratification economics in relation to class and race (Darity 2022). Such institutions serve to achieve, preserve or strengthen the power of some groups over others and may therefore also be labelled as asymmetric, as in a zero-sum game.

A second broad categorisation of institutions recognises that they might stimulate or constrain certain behaviours (Veblen [1899] 1934). This does not imply, however, that constraining institutions are naturally bad for wellbeing (a religion which does not allow the use of alcohol may have followers who are happy to feel protected in this way from the seductions of drinking). To quote Hodgson (2006, 2): “Regulation is not always the antithesis of freedom; it can be its ally.” What matters in the CA is that institutions are expected to contribute to capability expansion, and hence, that they should be enabling for this purpose. Sometimes, this requires agents to be aware of the extent to which institutions are disabling, as Morgan and Olsen (2011) argue. In particular, they point at “the critical awareness of the agent that follows, breaks, bends and transforms rules” (Morgan and Olsen 2011, 428). Hence, the distinction between stimulating and constraining institutions does not fit one-to-one the distinction between institutions that enable or disable capabilities, as is illustrated in Table 1.

Table 1. Four options of how institutions may enable or disable capabilities.

	Stimulating institutions	Constraining institutions
Enabling capabilities	A. Freedom to expand capabilities	B. Constraints leading to expansion of capabilities
Disabling capabilities	C. Freedom to limit capabilities	D. Constraints leading to limitation of capabilities

The four options in Table 1 concern both one’s own capabilities and those of others. Cell A refers to supportive institutions providing the freedom to enable capability expansion, for example, equal rights legislation for LGBTQIA persons. Cell B refers to constraining institutions with the effect of expanding capabilities, for example, restrictive gun-laws. Cell C refers to constraining institutions with the effect of disabling capabilities, for example, discrimination or criminalisation of same-sex couples. Finally, cell D refers to supportive institutions for people’s freedom to reduce (likely other people’s) capabilities, for example, permissive gun-laws or the absence of legal protection of same-sex couples.

A third way of broadly distinguishing between types of institutions is between formal and informal institutions (see for an example of a description, which is not necessarily generally adopted, Kaufmann, Hooghiemstra, and Feeney 2018, 387). Although the terminology of formal and informal was not used by the original institutional economists, today it is employed regularly in institutional economics, including in OIE. For example, formal and informal gender institutions of entrepreneurship (Warnecke 2013); formal and informal finance in pre-communist China (Dong 2023); a comparison between formal and informal institutions influencing FDI (Seyoum 2011); or a clash between informal and formal institutions in nation-state building (Tomass 2012). To simplify, formal institutions may be seen as concerning explicit or codified rules and structures, such as organisations, laws, programmes and regulations, whereas informal institutions may be seen as reflected in social norms, routines and shared beliefs.⁵ Examples of widespread informal institutions include the family, social movements and systemic discrimination by employers.

A fourth broad differentiation of institutions recognises that, although all institutions are normative (when X do Y), not all institutions are moral in the sense of Y being about good versus bad or right versus wrong (Samuels 1977). For example, there is nothing inherently good or bad, or right or wrong, by a traffic rule requiring that traffic should drive on the right side of the road. In other words, it would not be bad or wrong if other countries would choose the rule to drive on the left side of the road. What matters is that the rule is either left or right – not that one of these would be morally superior to the other. But in the case of moral institutions, do Y reflects not only a normative directive but also a morally good or right option as compared to an alternative. In the case of such a moral institution, alternative options Z are considered to be bad or wrong in a moral sense, as is the case, for example, for laws against homicide or norms about the respectful treatment of all groups of people in public spaces. The distinction between normative and moral is important, also for the understanding of the role of institutions in the CA. Of course, driving on the left side when the rule is to drive on the right side would be considered wrong in a moral sense (f.e. irresponsible, reckless and/or criminal), but that refers to not following the rule. The same counts for driving right when the rule is left. There is nothing inherently morally commendable about driving on the left side in countries where that is the rule, as compared to driving on the right side in countries where that is agreed to be the rule. This distinction between normative and moral was recently revived by moral philosopher Dannenberg (2023) who explains that all moral statements are normative but not all normative statements are moral. He refers, for example, to the philosopher Philippa Foot, who clarified already decades ago: “Moral judgements are normative, but so are judgments of manners, statements of club rules, and many others” (Foot 1972, 310). Hence, some institutions are simply useful or meaningful rules, whereas others carry, in addition to that, moral meaning.

A fifth way of generally distinguishing between types of institutions is the extent to which they are created deliberately or emerge spontaneously (Veblen 1919). Of course, formal institutions, such as organisations and regulations, are created deliberately. But this may be the final step in the consolidation of rules which have initially emerged spontaneously, such as basic traffic rules about priority (which eventually led to codified rules with exams, driver’s licenses and fines). Informal institutions are generally the result of spontaneous order as, “emergent social structures” (Hodgson 2004, 425). For example, we observe less kissing on cheeks when greeting since the covid-19 pandemic.

Finally, a sixth broad categorisation of institutions is that of symbolic versus material institutions. Veblen ([1899] 1934) criticised the wasteful status institutions of the wealthy as “conspicuous consumption”. Here we see that wasteful material consumption acts as a symbol for status so that some symbolic institutions may have material implications. Other symbolic institutions do not involve material resources, for example, cultural expressions such as singing the national anthem at the start of the school day. Symbolic institutions do not necessarily require resources, or at least, they do not depend on them. They include, for example, social cohesion through levels of trust and cooperation in a society, or ceremonial procedures in academia around graduation. Material institutions involve access to and/or control over material resources, such as social housing, the pension system or the minimum wage.

Obviously, the six broad categorizations of institutions are often found in combination. For example, the government bureaucracy of India might be characterised as a formal, (seemingly) universal, deliberately created material institution with budgets, civil servants and offices, whereas the caste system might be portrayed as a targeted, or asymmetric, informal, (im)moral, social and symbolic institution.

Table 2 summarises the six broad categorizations of institutions that were discussed in this section.

Table 2. Six broad categorizations of institutions.

Universal	Targeted and/or Asymmetric*
Stimulating**	Constraining**
Formal	Informal
Moral	Neutral
Deliberate	Spontaneous
Material	Symbolic

*Whereas some institutions are targeted to a specific group, others may benefit one group at cost of another, reflecting a zero-sum game – so that those institutions may be labelled more appropriately as asymmetric.

**As was explained in Table 1, both stimulating and constraining institutions can enable or disable capabilities; here the focus is on whether an institution stimulates one to do something or constrains one from doing something.

The Logic of Institutions: Functionality, Efficiency and Power

One of the explanations of why we have institutions, is functional. “Generally, institutions enable ordered thought, expectation and action, by imposing form and consistency on human activities” (Hodgson 2004, 425). Some functional explanations of institutions rely on an efficiency argument, defined as reducing costs in relation to outputs or results.⁶ For example, institutions can be an efficient response to bounded rationality. As Herbert Simon (1983) has argued, our capacity to make rational choices is limited by the amount of information we can access and process and the time it takes to weigh all costs and benefits, leading to what he termed “bounded rationality”. Critical consumer platforms or shopping Apps can be seen as institutions which respond to such bounded rationality. Another efficiency gain of institutions lies in the reduction of transaction costs (Coase 1937). Think, for example, about community-based farms with shareholder-members which ensures secured sales. Other institutions may be solutions to coordination problems, which generate high individual risks and costs when not solved collectively. Some coordination problems simply require a low-cost convention, such as driving on the left or the right. Other coordination problems, however, are typical social dilemma’s, involving costs for individuals to support them and therefore risking free riders. They involve constraining institutions which, paradoxically as I have mentioned earlier, can expand freedoms, as Ostrom (2005, 170) has explained: “So, individual freedom of action is also expanded by enforceable rules since rules enhance the predictability of the actions taken by others in a decision situation.”

But functional explanations often tend to be insufficient or even reductionistic as Hodgson’s (2004, 34) critique of Gidden’s understanding of routines only in terms of their functions explains. Some institutions are not, or no longer, functional, due to path-dependence of outdated institutions, which were functional in the past but no longer in changed circumstances. Or, alternatively, they are beneficial for some groups at cost of others. An example of a path-dependent institutions is the reliance on import-tariffs in many African countries. This was an important source of tax revenue before the era of globalisation. But since import tariffs were slashed by the WTO, this source of tax revenue has been reduced significantly. But the power of elites prevents the tax systems in Africa to shift to progressive income taxation. Another way in which institutions may not be functionalistic is when institutions affect people’s values and preferences (f.e. conspicuous consumption or the gender difference in the choice of science in schools or in technical professions). In the OIE tradition, existing institutions are therefore not regarded “as necessarily functioning to promote the social benefit – in fact, rather the opposite” (Rutherford 2001, 174). A case in point are gendered institutions, which reflect a patriarchal culture, and which tend to benefit men but are inefficient because they reduce the full use of human resources (this may even show in lower GDP growth rates: Kan and Klasen 2021). This points at the role of power behind some institutions. This may concern market power by big brands in the case of consumerism (which may trigger the Veblen effect), the power of elites (who benefit from a status quo tax system), or patriarchal power. The cases of elite power and patriarchy are exemplar for nonfunctionalistic institutions, which tend to persist because they sustain the power of one group over another.

The logic of institutions as sometimes functionalist and other times not, or beneficial only for some groups and not others, is important for understanding their role in the CA because it points at different entry points for institutional change. Institutional change may be geared towards improving the functionality of institutions in general (f.e. dealing with externalities through fines or taxes) or for certain groups in order to eliminate their disadvantage (f.e. anti-discrimination laws or mandatory transparency rules about gender wage differences in firms). Hence, sometimes institutional change may be a positive-sum game, whereas other times it seems to involve a zero-sum game.

Institutions for Capability Expansion

Since the CA is a normative approach to economics – considering the good to be in the expansion of capabilities for all – it is relevant to focus on enabling institutions for capability expansion.⁷ Going back to the distinction between resource-institutions and agency-institutions, we can now further detail the types of institutions that may support capability expansion.

Enabling Resource-institutions for the CA

At first sight, resource-institutions that are universal, f.e. through general laws, might be thought of providing equal entitlements to resources for all. But given the differences in people’s personal, socio-economic and environmental circumstances, universal resource-institutions not automatically result in equal capabilities. So, universal institutions, such as equal rights, need to be complemented by a set of targeted institutions which compensate for personal differences and unequal circumstances. For example, poor women without access to credit in the banking system due to lack of collateral could be compensated by a government programme or NGO project of microcredit for women, which helps them to expand their capabilities (Kabeer 1999; Nambiar 2021).

Resource-institutions can be formal or informal. Formal resource-institutions include laws and regulations, organisations and programmes, as well as structures such as fiscal systems. Within these, there could be a distinction between universal formal resource-institutions (f.e. gender equality in inheritance rules) and targeted, or specifically asymmetric, formal resource-institutions (f.e. unequal inheritance rights for women in nineteenth century Europe). Formal resource-institutions which support capability expansion are often characterised as entitlements in the CA, as “the means of establishing command over the commodities and resources needed to achieve valuable capabilities” (Reddy and Daoud 2021, 684). In contrast, informal resource-institutions assign access to resources based on traditions and conventions. These may include asymmetric institutions benefitting more powerful groups, as in traditional land inheritance systems in the Global South which exclude daughters from inheriting land or assign them a smaller part as compared to their brothers.

Resource institutions may be created deliberately or could emerge gradually. Both deliberate and spontaneous institutions are relevant as conversion factors for resources. For example, asylum seekers have a right to food and shelter and may indeed be supplied with both by the state or a humanitarian aid organisation, but they may be treated with contempt or aggression by members of local communities, who may believe that there are too many asylum seekers.

Many resource-institutions will have a moral dimension because access to resources is imperative for the expansion of capabilities. This counts for land, material, financial and social resources. For example, the rights of asylum seekers to protection, housing and food represent a moral institution, which needs to be upheld even if they come from countries which are generally regarded as safe, until their application process is finalised.

Finally, most resource-institutions will be material – governing access to and control over income, credit, land, and natural resources. But there are also symbolic resource-institutions, such as social capital, which may appear crucial for capability expansion for groups that are marginalised (see, f.e. Migheli 2011). However, as Veblen emphasised, symbolic institutions may also reflect status and even generate emulation among disadvantaged groups at the cost of their own material wellbeing (for example, when peer-pressure in disadvantaged urban neighbourhoods leads youth to buy big-brand sneakers even though they cannot afford purchasing healthy food).

In sum, resource-institutions in the CA can be understood in terms of all the six categorizations of institutions of Table 2, in order to enable capability expansion for all, despite different and unequal circumstances. Although they often need to be functionalistic, they do not necessarily have to be efficient. Moreover, the capability expansion of disadvantaged groups may require a transformation of nonfunctionalistic institutions which merely reflect oppressing traditions or discriminating social norms. Or, to state it differently, resource-institutions should be effective as entitlements for capability expansion for people, irrespective of their circumstances.

Enabling Agency-institutions for the CA

Before going into agency-institutions, I would like to clarify the concept of agency as it has been understood in feminist economics in relation to empowerment – a concept that is closely related to the agency aspect of capabilities. Naila Kabeer (1999, 438) has defined agency as “the ability to define one’s goals and to act upon them” which “also encompasses the meaning, motivation and purpose which individuals bring to their activity, their sense of agency, or ‘the power within’.” This reflects OIE’s recognition that institutions embody generally accepted ways of thinking and doing. “Thus, institutions worked to mould the preferences and values of individuals brought up under their sway”, according to Rutherford’s (2001, 174) overview of the key characteristics of OIE. Kabeer subsequently explains empowerment as the combination of agency and access to and control over resources and defines it as “the process by which those who have been denied the ability to make choices acquire such an ability” (Kabeer 1999, 437). In a new contribution, she made an explicit link to the CA when stating that women’s capabilities are constrained by hegemonic gender ideologies (asymmetric informal institutions) affecting their agency by “undermining their sense of self and social worth” (Kabeer 2021, 100). Hence, agency is more than choosing among given options: it is also concerned with the imagination of the possibility to choose among all options, as well as the self-confidence of expanding these options (individually or collectively).

Just like enabling resource-institutions, enabling agency-institutions support capability expansion, even when they imply constraints on certain behaviour. They are often informal, as they shape, through a variety of social norms, what individuals value and choose to do and be. Some agency-institutions have a moral character, as they are based on moral commitments (non-preferential goals), as Sen (1973; 1977; 1985) has argued. Or they may be immoral in the sense that they may be harmful and thereby wrong, such as the adaptive preferences of some poor women in India who have internalised the patriarchal norm that they should eat less and less nutritious food than men (Sen 1990).

There is an important role for symbolic agency-institutions, both at the individual level and the interpersonal level. As Kabeer already indicated for the individual level, a sense of agency reflects meaning, motivation and purpose. This connects to the extent of reflection on one’s options, the courage to expand them and the sense of self and social worth that may result of it, which in turn strengthens one’s agency. At the interpersonal level, enabling symbolic agency-institutions refer to attitudes of others which affect people’s sense of agency and, subsequently, their self-worth. It is here where respect and dignity may be located in the CA. Being treated with respect by others and being able to appear in public with dignity support one’s sense of agency. Hence, such attitudes in public life can be included in the CA as enabling agency-institutions. They enable one to make bold choices even when dominant social norms would disapprove them for members of disadvantaged groups. Think, for example, of women with small children seeking a paid job outside the household in a culture in which mothers are expected to stay at home. They may succeed even if this is frowned upon by many, but when they are in addition treated with disrespect and scorn, they may give up or seek paid employment by working at home (f.e. making handicrafts which are sold by a brother or through a webshop). In other words, symbolic agency-institutions directly matter for capabilities involving meaning (Alkire 2005).

To sum up, enabling agency-institutions in the CA are mostly informal, they may be moral as well, and can be symbolic in terms of providing meaning and purpose. They include efficient routines that deal with bounded rationality but also include non-functionalistic agency-institutions with symbolic meaning. These may, however, also disable capability expansion, for example through disempowering social norms and widely shared harmful beliefs.⁸

Institutions for Collective Productive Capabilities

The CA recognises both individuals and collectives, and both agency and structures, as the basic units of analysis, but the emphasis tends to be at the individual level (Lessman 2022). This focus has received criticism from various CA scholars (see, f.e. Gasper and van Staveren 2003; Stewart 2013). Stewart (2013, 82) rightly argues that “the primacy of individualism in the capability approach is at odds with the flourishing of social beings”. The recognition of the relevance of collectives is important for the CA because it is the community (as a social group or organisation), which provides the immediate opportunities, or lack thereof, that are “central to our life because the local society we live in constitutes the most proximate socio-institutional setting that most directly shapes agents’ daily life and wellbeing” (Biggeri, Ferrannini, and Arciprete 2018, 128). It is in communities that people work, invest, collaborate or dominate, where institutions operate, and where capabilities are disabled or enabled. Communities may also affect the capabilities of outsiders – those belonging to other communities. Finally, communities may enable collective action, which is recognised by some CA-scholars (see, f.e. Ibrahim 2017) as crucial to expand collective capabilities and/or to enforce institutional change. This may be for the sake of enabling individual capability expansion of those who are part of a collective (f.e. better labour conditions for domestic workers)⁹ or for others at whom the collective action is targeted (f.e. in the case of humanitarian aid). So, a crucial difference between a focus on individual capabilities and one on collective capabilities is that institutional change away from disabling capabilities towards enabling capabilities often comes about by collectives – either in an ad hoc way or through pre-existing collectivities.

To analyse the impact of collective capabilities on collective wellbeing, Andreoni, Chang, and Estevez (2021, 180) propose to think of collective productive capabilities as “productive capabilities that cannot be held by individuals and can only be held by groups (e.g. productive organizations, cooperatives, social groups).” These may also involve non-members of the community, for example through hired labour or external customers. Productive groups include, of course, for-profit firms. These are organised hierarchically (Coase 1937) and in a capitalist economy generally driven by shareholder value, and hence, profit-maximization. As a consequence, part of a firm’s profit leaks out to external individuals or investment funds. In this article, I will limit myself to collective productive capabilities that are rooted in the community economy – where the benefits largely remain within the community because the collective is self-owned, self-managed, and decisions are made by members in a relatively democratic way (Gibson-Graham 2006). This is possible because it keeps market exchange at arm’s length for investment and/or production and/or consumption. In the community economy, various (though not necessarily all) production factors, such as land, labour and capital, have no supply and demand sides meeting in a market exchange, but instead are combined through economic role combination. For example, in worker coops, workers own the capital stock rather than capital hiring labour, and in mutual insurance organisations, the policyholders are at the same time the owners of the capital invested (Van Staveren 2020a). Other community economy organisations produce for their members, so that also a sales market for commodities is absent. An example is the collective use of solar panels (in particular when stand-alone, off-grid, so that there is even no power supplied to and taken from an external electricity network). This commonality of collective productive capabilities – economic role combination across the supply and demand side of the market – does not imply the absence of power in the community economy. Also collectives may create, support and use institutions which marginalise or exploit subgroups, such as migrants or a religious minority, through oppression and disempowerment by a majority or elites.

Hence, the community economy is distinct from the market and a large share of the private sector operating exclusively in markets, but it is also distinct from the state and its regulation and redistribution, with its economic role combination through cooperation (Gibson-Graham 2006; Polanyi [1944] 2001). It is a third economic domain, already distinguished by Adam Smith ([1759] 1984), with different values from those of the market and the state, based on cooperation, caring, and a purpose to which those involved are committed, supported by collectively agreed rules (Van Staveren 2001). The community economy includes commons, mutuals, worker cooperatives, non-profit (or social) enterprises, community currency systems, time banks, and community-supported agriculture (CSA) (some of these are also referred to as the Social and Solidarity Economy, c.f. ILO 2022, but these include government-subsidized organisations, which therefore are dependent upon the state). There are also hybrid forms of community economy and the market with for-profit firms, such as, for example, family enterprises, where the capital is owned by a family and the firm is often managed by family members but employees are hired from outside and commodities sold in markets.

The community economy is the economic domain in which collective productive capabilities are required (f.e. collective labour) but also where they are developed (f.e. food security for the community), in a mutually reinforcing process through relevant institutions, as for example, in commons (Ostrom 1990). Collective capabilities have been defined by Ibrahim (2006, 398) as “the newly generated functioning bundles a person obtains by virtue of his/her engagement in a collectivity that help her/him to achieve the life he/she has reason to value.” This can happen entirely internally, in community economy activities which contribute directly to the wellbeing of their members. For example, ROSCAs (Rotating Savings and Credit Associations), which are bottom-up, self-managed, relying on collective decision-making, and thereby a community-economy alternative to market-based or state-supported microcredit programmes. Other collective capabilities may be externally oriented, for example various forms of collective action such as group protests to demand social policies from the state or consumer boycotts against bad labour conditions.

Collective capabilities have been understood in relation to individual capabilities as the sum-total or average of the individual capabilities in a group, taking into account interaction effects (Comim 2008; Stewart 2005), or as distinct from individual capabilities precisely because of the interaction effects, such as social capital (Ibrahim 2006; 2017). The interaction effects in the development of collective capabilities occurs at the meso-level, precisely the level where institutions emerge and operate (Hodgson 2000; Van Staveren 2020b; Von Jacobi 2018).

Collective capabilities have both instrumental value and intrinsic value (Evans 2002; Ibrahim 2017; Murphey 2014). The instrumental value is important because collective capabilities tend to be directed toward social change, that is, capability expansion for a group or society as a whole (Agarwal 2000; Alkire 2008; Ibrahim 2006). But they may also be limited to particular groups only. For example, schooling for boys only in poor villages, which will provide men in the village with economic and social advantages that women will not have, due to their collective capability to read and write. This inequality is not caused by the capability of reading and writing as such, but by a patriarchal institution – in this case a stereotype gender norm – which prevents girls from learning to read and write as well.

A similar issue is at stake for the intrinsic value of collective capabilities. Some scholars have criticised the intrinsic value of collective capabilities by pointing at the inequalities, oppression, social exclusion, and preference-distortion which groups may exhibit (Alkire 2008; Robeyns 2005; Sen 2005).¹⁰ But this critique confuses capabilities with institutions affecting them. The problem is not collective capabilities as such, but the social norms (and sometimes even legal rules), which enable some groups to exercise institutional power over other groups, constraining their access to resources, exploiting their labour, stigmatising their identity, or otherwise disadvantaging them. As various CA-authors have shown, certain institutions only benefit the powerful, while other institutions enhance the agency of the disempowered (Thorp, Stewart, and Heyer 2005). In other words, the problem of inequality is with the particular institutions embedded and defended in certain communities or by powerful groups in those communities – institutions, which can be disabling and immoral, preventing capability expansion of some to the benefit of others. And that reflects precisely the role that institutions have in the CA: as mediating how resources translate into capabilities and as influencing agency, as I have indicated at the beginning.

As Nambiar (2021) has reminded us, institutions can enable or disable capability expansion. Hence, we need to focus our attention on institutional change in order to contribute to the expansion of capabilities and how collective action and its collective capabilities can contribute to such institutional change. Because “groups and communities possess rules that may restrict capability improvement (…) since they mould preferences and values, and any attempt to improve capabilities must start from the informal rules that hold the groups together” (Nambiar 2021, 1982). Peter Evans was probably the first scholar pointing at the need to address institutions and institutional change for achieving collective capabilities. “Institutional strategies for facilitating collective capabilities are as important to the expansion of freedom as sustaining formal electoral institutions.” (Evans 2002, 59). By identifying and challenging disabling, asymmetric and immoral institutions, they can become subject to institutional change with the help of collective action so that they may be transformed into enabling, universal (and targeted when needed to address unequal circumstances), and morally right institutions for capability expansion.

A theory of collective action that focuses on institution building and institutional change is Ostrom’s institutional framework for the management of common-pool resources. Even though Ostrom does not put herself in the OIE tradition, there are some commonalities, for example in the understanding of collective action by one of the founders of OIE, John Commons, in which he explicitly included unorganised and customary groups. “Collective action ranges all the way from unorganised custom to the many organised going concerns, such as the family, the corporation, the trade association, the trade union, the reserve system, the state” (Commons 1931, 649). Ostrom’s theory of collective action seems particularly helpful for the understanding of the role of institutions for achieving collective capabilities, especially for reducing the probability of free riding and power asymmetries (Ostrom 2000).

Ostrom (1990; 2005) has studied commons across the world and has uncovered eight design principles – institutions for collective action – which make them successful for members over time. As in OIE, they arise at the meso-level (Ostrom 2000, 154, explicitly uses the word “evolve”), they are enabling for the collective (also by putting constraints on individual behaviour such as free-riding), they are often informal (but some may be formalised over time, such as conflict resolution mechanisms) and combine the material (resource management) with the symbolic (finding meaning in collective self-management and a common purpose). She characterises the challenge of sustainable common-pool resource management as involving two dilemmas: first of sustainable resource-use and second of following agreed rules. This distinction is quite similar to the distinction between resource-institutions and agency-institutions that I have made for a better understanding of the role of institutions in the CA.

The first three institutions for collective action that Ostrom has uncovered concern the collective resource (Ostrom 1990, 90):

1. what the resource and its benefits are and who are allowed to use them

2. rules for contributing collective labour and distribution of the benefits

3. nested commons in case of very large or complex common pool resources.

The other five institutions refer to collective agency:

(4)	collective choice arrangements;
(5)	monitoring procedures;
(6)	graduated sanctions;
(7)	conflict-resolution mechanisms; and
(8)	minimal recognition by external authorities of the right to organise.

I will discuss these two types of institutions (resource-institutions and agency-institutions) below in relation to collective capabilities.

Collective Resource-institutions and Collective Capabilities

Collective resources include formal collective property rights (as in worker coops) as well as informal, traditional rights of access, use and control without the right to sell common property (as in common pastures). A collective may agree to pool resources for collective investment and/or production, as in mutuals, while members may keep an individual claim to their resources. In addition, collective resources may not (only) be owned or used collectively, as in the case of land or aquifers below individually owned land, but they may (also) be produced collectively, as in hay from joint meadows or electricity from community solar panels.

All these forms of collective resources are backed-up by collective resource-institutions, as Ostrom has pointed out, which assign rights and responsibilities by workers/investors/users/owners over the collective resources. They involve rules about the distribution of well-defined tasks (f.e. maintenance), when these tasks need to be performed (f.e. a certain number of days per month) and by whom (f.e. by one able-bodied adult per household), as well as rules about harvesting or other benefits, often according to a rule of equality (f.e. the same amount per head or per household) (Ostrom 1990), reflecting a universal type of institution.

For collective capabilities to develop and benefit all participants, the collective resource-institutions should be functional. This implies that they need, for example, reduce uncertainty about the resource’s availability, prevent free riding through smart incentives for monitoring and transparency, crowd-in commitment by ensuring fair distribution and reduce transaction costs through cooperation instead of hiring external supplies or labour. This requires a process of learning and trial-and-error, with the institutions generally not being created in one go but rather emerging relatively spontaneously through a process of adjustment, (Hodgson 2000; Ostrom 2000). Human beings are “complex, fallible learners who seek to do as well as they can, given the constraints that they face and who are able to learn heuristics, norms, rules, and how to craft rules to improve achieved outcomes.” (Ostrom 1998, 9).

An example of successful collective capability expansion using collective resource-institutions is given by Vicari (2014) about a small rural cooperative in Brazil. This coop has engaged in collective action vis-à-vis the municipal government in order to enforce a legal change which has shifted a disabling institution into an enabling one for the female members who collect and sell babassu nuts: after collectively presenting their demands to local authorities, they were allowed to collect the nuts from any tree in their area, even when the land is owned by a private landowner. This has expanded their collective resource-base. Also, they used collective action in the sales market to circumvent middlemen by entering fairtrade value chains to sell their nuts at a higher price in the UK. This has increased their collective livelihood security as well as their individual earnings, which could only work with a collective agreement with the foreign fairtrade buyers.

This example also demonstrates how closely collective resource-institutions are related to collective agency-institutions, to which I will now turn.

Collective Agency-institutions and Collective Capabilities

Collective agency is often linked to collective action for social change, involving institutional change. This is also how it is often used in the CA: as protests, demonstration or strikes. But collective agency is also necessary for decision-making in a collective, monitoring of collective activities, sanctioning of free riders and obtaining recognition by authorities, as Ostrom has revealed in her study of commons worldwide. That is why Ibrahim (2017, 203) defines collective agency broadly “as the capacity of a group to define common goals and the freedom to act to reach the chosen goals”. Alternatively, collective agency has also been framed as collective intentionality, which emphasises the group’s common goals, or purpose, as irreducible to individual goals (Lessman 2022).

Collective agency requires, as Ostrom has shown, inclusive decision-making bodies, which tends to involve democratically elected representation. The equality, or symmetry, involved, is crucial for members’ commitment: one member one vote, which is a moral dimension of the decision-making rules. Collective agency also involves monitoring to prevent appropriation by non-members or free riding by members, often followed by graduated sanctions – which represents another moral dimension to these institutions, derived from a purpose and protecting the common good. Hence, collective agency-institutions should be functional for the whole, which implies they should in general be universal (for all the group members), while they tend to be moral, as expressed by a purpose, and may have symbolic meaning as well, as the example below will illustrate. The equality and morality and/or symbolic meaning of collective agency-institutions are likely to involve the sharing of information (transparency) and rotating monitoring among members (equal role sharing), and self-reinforcing sanctions (f.e. by allowing rotating monitoring teams to keep the fines for trespassing rules), which can result in surprisingly low levels of free riding (Ostrom 2005). Finally, as Ostrom has pointed out, collective agency-institutions should include conflict-resolution institutions that are strictly universal so that they do not benefit some over others.

Several studies employing the CA have illustrated how collective agency institutions work in practice. For example, Schlosberg and Carruthers (2010) present two case studies of indigenous communities in the US and Chile fighting for justice for their communities and their land. They argue that the members of the two indigenous communities challenge exploitative institutions which affect their collective resources of native land with traditional burial places, spiritual value, and food and herbal medicine resources. The authors point out that the collective agency is led by a common goal, defending their collective resource by challenging disabling formal institutions forced upon them by the state, and that they do this based on a strong awareness of what is at stake (purpose) and through effective collaboration in their protests, demands, and actions. “By fighting for autonomy, land, respect, or political voice, indigenous activists are fighting for the capabilities necessary for their communities to function fully” (Schlosberg and Carruthers 2010, 18). Interestingly, the authors explicitly refer to the lack of recognition by the state as experienced by the indigenous communities through a downgrading of the community’s economic, social, cultural and religious values by authorities. This example shows that the struggles for institutional change go deeper than protecting collective resources but also include the legitimacy of their shared values and practices in the eyes of external agencies – an institution which Ostrom included as the final institution in her list (see p. 19 and 20) that tends to make commons successful forms of collective action. Hence, respect, recognition and legitimacy are all collective agency-institutions which are inherently linked to achieving collective capabilities.

Conclusion

I have explained the concept of institutions as it is used in Old Institutional Economics with six categorizations. I have tried to argue how this understanding of institutions could be integrated in the Capability Approach by enriching the concepts of conversion factors and entitlements as resource-institutions, and by elaborating the concept of agency with the help of agency-institutions. Such integration could help us to think of institutions as always having two sides: enabling and disabling capabilities, and that policies aimed at eliminating disabling institutions will only be successful when they replace them with, or transform them into, enabling institutions. These may be universal, or targeted in case that certain individuals or groups need to be compensated for personal, socio-economic or environmental characteristics or inequalities.

The concept of institutions as it is used in OIE also helps to tidy up some loose ends in the capability framework, in particular notions related to agency, such as self-worth, dignity, respect and legitimacy. In particular, the concept of agency-institutions seems to be helpful in giving a place to public values such as respect and dignity, by characterising them as enabling, moral and symbolic agency-institutions.

Finally, I have tried to show how OIE combined with Ostrom’s theory of institutions for collective action may support the analysis of collective capabilities. First, by distinguishing between collective resource-institutions and collective-agency institutions as sets of formal and informal rules which may enable or disable the expansion of collective capabilities. Second, by recognising that collective productive capabilities in groups may serve the group members and/or an external purpose. And third, by arguing that inequality, stigmatisation and social exclusion are not merely a consequence of the capabilities of powerful groups but rather the result of their power as expressed in discriminating social norms, unfair rules, oppressive traditions and unequal laws, which are all institutions supporting existing power relations and benefitting their capabilities over those of less powerful groups.

Of course, this discussion is just an exploration of the possibilities of taking institutions more seriously in the Capability Approach. Not every type of institution that was distinguished here may be useful in empirical studies. But that is not the point. What matters, is that researchers employing the Capability Approach may be better informed and might feel supported in making better use of the concept of institutions, relying on more than a hundred years of research in the tradition of Old Institutional Economics.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Irene van Staveren

Irene van Staveren is Professor of Pluralist Development Economics at the International Institute of Social Studies in The Hague, The Netherlands (part of Erasmus University Rotterdam). As a pluralist economist, she works at the intersection of a variety of economic approaches, including the capability approach, institutional economics, social economics and feminist economics. Her research focuses on the entanglement of economics and ethics, through concepts such as gender norms, social cohesion and efficiency. She is the author of a pluralist textbook, with a freely accessible online course on Coursera, Economics after the Crisis – an Introduction to Economics from a Pluralist and Global Perspective (Routledge, 2015). She recently published Alternative Ideas from Ten (almost) Forgotten Economists (Palgrave, 2021). She is also the director of the freely available online cross-country database Indices of Social Development.

Notes

1 Sen (1984, 497) defines entitlements as “the set of alternative commodity bundles that a person can command in a society using the totality of rights and opportunities that he or she faces.”

2 A recent example from the Netherlands is about a single mother with functional impairments, living in the city of Tilburg. She has been dependent on welfare for years, and when her vacuum cleaner broke down, the rules entitled her to receive money to buy a replacement. But the municipality did not trust that the appliance had really broken down. Hence, the civil servant of the municipality asked the woman to come to the municipality offices with the vacuum cleaner to demonstrate that it indeed was not working. Lacking a car and saving money on a bus fare, she struggled carrying the heavy load on her bicycle to the city centre. There, the civil servant required her to plug it in a wall socket in the main hall of the municipality building, under the eyes of citizens visiting the building. She emotionally described this experience in a TV show on inequality as very embarrassing. Although she was granted the money to buy a new vacuum cleaner, she felt deeply humiliated by how the civil servant had treated her.

3 Hodgson (2004, 422) emphasises that in OIE, “the understanding of the (rational) individual presupposes an understanding of specific social structures, just as the understanding of social structures presupposes an understanding of individuals.”

4 Ostrom (2004) did, however, acknowledge that she was influenced by Commons (1924).

5 Note that beliefs are also institutions and not simply limited to the minds of individuals. “Beliefs are part of social reality, because human actions can not only depend upon beliefs but also have to take into account the supposed beliefs of others” (Hodgson 2004, 424). An example is a strategy of modesty by women during high-end job interviews when they expect the interviewer to have stereotype gender beliefs about women negotiating for a higher salary in the same way as men do, labelling them in their minds as “bitches” (Leibbrandt and List 2004).

6 Efficiency is here used in a general sense and not as in neoclassical welfare economics where it refers to Pareto Optimality.

7 As indicated in Table 1, institutions that enable capability expansion may constrain certain behaviours.

8 An example of such a belief is that in southern Africa albinos are sometimes considered as witches, which may result aggressive attacks on albino children or expulsion from the community.

9 Collective action by various groups of domestic workers worldwide has eventually resulted in the adoption of the ILO (2011) Convention on Domestic Workers.

10 See f.e. Comim (2008) for a good discussion on this point of Sen (2005).