Criticisms of social capital theory And lessons for improving practice

Discussing criticisms allows us to focus our attention on the potential weaknesses of the concept & our application of it.

Social capital has been perceived as a miracle concept that is able to provide answers to a range of phenomena beyond an economic lens^[1]. Its rapid and wide application has made it one of the most popular concepts in the social sciences yet people using the term do not always mean the same thing. Social capital is ill-defined, with different authors attributing different meanings to the concept^[2]. Scratch beneath the surface of social capital and things get complicated rather quickly^[3]. Social capital is so complicated that entire books are devoted to exploring the conceptual and theoretical underpinnings. I have personally spent many hours, or even days, weeks, and months, analysing and synthesising the seemingly disparate literature on social capital in an attempt to gain a coherent understanding of the concept. This has been a gargantuan task over many years that has been regularly marked by moments of exasperation at the crass scholarship and cavalier application of the concept. Social capital is far from a unified theory; there is little agreement about definition, dimensions, measurement, or building. Social capital has been widely criticised, mostly for its ambiguity and variability. I believe that discussing the criticisms of the concept is very important since it allows us to focus our attention on the potential weaknesses of the concept and our application of it. This allows the opportunity to improve our practice and the validity or effectiveness of our projects involving social capital. Many of the criticisms are valid and failing to account for them could result in social capital projects being suboptimal, open to scrutiny, and even collapse under the weight of contradiction and assumption. Social capital is a metaphor that highlights the positive and productive aspects of sociability. However, it falls short of being a form of capital, and it is also widely accepted that social capital can have positive and negative consequences. This makes the term rather inappropriate and confusing. Economists can object to the inclusion of social with capital, and sociologists can take issue with the inclusion of capital with social. Yet despite these problems the concept of social capital has been widely applied to a seemingly ever-widening range of contexts.

Some authors have described the concept as “fundamentally flawed”

The main criticisms of social capital theory are that it is not social, not capital, and not a theory. This doesn’t leave the concept with much of substance, leading some authors to describe the concept as “fundamentally flawed”^[4]. In addition it has been claimed that it is impossible to measure, that problems of circularity make it a tautology, and that the possibility for positive or negative outcomes make it context dependent^[5]. Some aspects are objective, but others are subjective^[6][7][8]. Some are cognitive, but others are pre-cognitive^{[9] [7]}. Social capital can be rational, pre-rational, or even non-rational^[10]. This suggests social capital is more of an umbrella concept than a functioning theory^[11]. Yet these damning criticisms have not halted the use and application of the concept. Quite the opposite is true, with widespread and often cavalier use of the concept pervading much of the literature on social capital^[12].

Social capital seems almost immune to criticism.

Social capital seems almost immune to criticism. Its theoretical ambiguity and variability remains a problem for its rigorous application but also acts as an antidote to criticism. It renders most critiques both valid and invalid depending on theoretical perspective. For example, take the criticism that social capital is not social^[13][14][5]; while I can see that this could be true in the way some authors conceputalise social capital, it is most certainly not true from the way I and many other authors conceputalise it. This stifles intellectual debate because discussions of social capital often become conditional on theoretical perspective. There is potential for confusion with authors weighing in on discussions without appreciating the theoretical context. Someone may think they know what social capital is because they read Bowling Alone^[15] and so feel confident to comment or critique. But often comments about social capital apply only to a particular perspective on the concept^[16]. People who don’t know this can cause confusion and authors who do know will often not be confident enough of the context to contribute. And so, the popularity and widespread application of social capital continues, picking up speed as assumption and contradiction confounds future application of the concept.

Who is to say what is and is not rigorous research or good scholarship

This can even be seen in the peer-reviewed literature. There are numerous examples of publications that demonstrate poor connection to theory, or as Ben Fine^[14] has put it “vulgar scholarship”, that have been published in reputable journals. I can’t surmise exactly how this has happened, but one explanation is that they based their work on previous literature that was equally poor. Another possibility is that reviewers didn’t have the confidence to reject the work, either because they were not sure of the appropriateness of the theoretical perspective or because it was consistent with precedent. It would be great if there was a way to unpublish this work to prevent the continuation of poor social capital research. Since that is not possible another solution would be for this work to be discredited. But who is to say what is and is not rigorous research or good scholarship. This is especially difficult considering the variety of theoretical perspectives. The current situation is that researchers and practitioners approaching social capital for the first time must read widely to gain a thorough understanding of the concept from different perspectives. They need to evaluate the options and select an appropriate approach. In the past this has led to many people creating their own definition of social capital, and this has added to the current definitional difficulties. People who do not read widely run the risk of selecting a narrow perspective that is not sufficiently related to underlying theory. This leaves their work open to criticisms of circularity, tautology, exaggeration, assumption, and confusion.

The challenge can seem immense for people new to social capital

For someone approaching social capital theory for the first time the challenge can seem immense. There are numerous definitions, different levels of analysis, and different terminology that is often used interchangeably. Some authors refer to dimensions, but other authors use other terminology such as types, forms, or functions. These terms are even used interchangeably with the precise meaning remaining unclear. When reading literature on social capital one tends to find ambiguity, variability, inconsistencies, and contradiction. Unfortunately, the social capital literature is marred by ‘vulgar scholarship’^[14] where authors have used the concept without sufficient consideration of its theory and have failed to understand and account for its inherent complexity. The complexity partly stems from the broad nature of the concept since it relates to any aspect of sociability that has potential productive or perverse outcomes. This makes it virtually impossible for any project to take a comprehensive approach that includes all dimensions and levels. Therefore, researchers tend to narrow their scope to a particular area of interest, in the process they omit dimensions, aspects, or levels of analysis. This is not a criticism per se, but the way in which this is typically done has little bearing on our theoretical understanding of the concept. From another perspective the broad nature of social capital has made it applicable to almost any area of human endeavour. As such it has been described as a catch-all, for-all, cure-all term^[16][17] or an umbrella concept^[5]. Since it relates to human sociability it encompasses almost any concept related to social structure, social organisation, or social action. Social capital is a new term for an old concept. Issues related to the core themes of social capital have been investigated and discussed in the social sciences for hundreds of years. This means that there is an approximately similar term, or series of terms, to be found in every discipline of the social sciences. The table below includes a small sample of relevant terms, some are approximately equivalent to the concept of social capital while others are a subset or dimension of social capital.

embeddedness social support social cohesion social inclusion or exclusion social equality

group-based identification group dynamics social investment prosocial behaviour social solidarity

collective action social norms and sanctions reciprocity trust culture

Table. Terms related to social capital – this is not an exhaustive list The popularity of the term social capital has resulted in many authors using the term when it would be more appropriate to use a related term that has a more relevant meaning. Claude Fischer^[18] believed that the term social capital is unnecessary as other clearer and simpler terms, such as membership, sociability, and trust serve perfectly well^[11]. For example, an author may use the term social capital when their interest is social inclusion. They cherry-pick social capital theory for aspects that are relevant to social inclusion and discard the aspects that are not relevant. This causes several problems, not least is a failure to connect with existing literature on social inclusion and the resulting pollution of the social capital literature. Another example is an author interested in trust who instead uses the term social capital. They then equate social capital as trust, thereby ignoring the other dimensions of social capital. This is another example of cherry-picking social capital theory to suit an inappropriate application. In some cases, the popularity of the term social capital has led to authors using the term in the title of an article, but make little, if any, reference to the term in the body of the article. The term may be added without explanation or discussion in a way that can leave the reader unclear about what, how, or in what way the topic is relevant to social capital. It’s like reading an undergraduate student’s essay that employs a declarative conclusion such as “… and therefore social capital.” without discussion of how and why. While this may not cause further problems for the theory it demonstrates how the concept has become a catch-all term. The following sections deal with some of the common criticisms of social capital as a concept and theory. While some of these criticisms are applicable to social capital generally, many of them are applicable only when social capital is defined or used in a certain way. This means criticisms tend to relate to a theoretical perspective on social capital rather than social capital generally. It can be difficult to identify which comments relate to which perspective. That is, under what conceptual and theoretical approach does the critique hold true and when is the critique not relevant? This makes any critique complicated. Where possible I will identify when the criticisms are or are not relevant.

Social capital is not social

Many conceptualisations of social capital simplify, reduce, and abstract the concept to the extent that it almost ceases to be social. While many authors perceive social capital as a means to give social issues prominence in analysis and decision making it is often used for the opposite purpose. It gives economists the opportunity to ‘colonise’ sociological territory with fundamentally economic notions^[19][5]. This is not a criticism levelled at all economists, but is relevant to any researcher who  imports methodological individualism and reductionism that restrict social capital to economic rationality on the part of individuals^[20]. Where this happens, social capital could be accurately described as not social.

To measure social capital the ‘social’ is often reified and reduced to characteristics of something else

Social capital also tends to be abstracted because many aspects cannot be observed directly. Since they cannot be observed directly researchers tend to use indicators that can be measured and that are believed to have a causal relationship with the aspects of social capital being measured. This means that what is being measured ‘indicates’ the existence of social capital. Social capital is said to be measured by proxies because the proxies ‘stand in’ for the aspects of social capital we hope to measure. The quality of these proxies varies enormously depending on the theoretical and practical relationship to the aspect of social capital we are attempting to measure. Some proxies are merely correlational, such as the use of crime rates to measure trust, others are more widely accepted and theoretically robust, such as the use of group memberships to account for the size or nature of a person’s social network^[21]. Therefore, to measure social capital the ‘social’ is often reified and reduced to characteristics of something else. The reifying of the social may not be immediately evident but it is most problematic when there is a lack of connection between the theory and measurement practices. For example, measurement approaches often look at the behavioural manifestations of social capital, such as membership in associations as a proxy for trust. This is an oversimplification that contains assumptions and confusion over correlation and direction of causality between the aspect of social capital being measured and the proxy that is used. Trust is a good example of an aspect of social capital that cannot be observed directly. Many of the results of trust can be readily observed and measured, however this may or may not relate directly to the trust we are attempting to measure. The existence of trust could be inferred from the actions of individuals and often a lack of available data forces researchers to use an indicator of trust, such as crime rates as discussed above.

Many authors may fall into the trap of ‘explanatory reductionism’

By its nature the ‘social’ is highly complex and abstract. It relates to feelings, beliefs, and perceptions that are largely cognitive and pre-cognitive. So, to operationalise social capital we need to ‘flesh it out’ to make it more tangible or concrete. This is typically done by some degree of simplification, reduction, and assumption that distorts the essence of social capital^[22]. By trying to explain the operation of highly complex systems in terms of the properties of their constituent parts many authors may fall into the trap of ‘explanatory reductionism’^[23]. Indeed, social capital tends to be a reductionist concept, particularly where the focus is on the individual level. Ben Fine and Francis Green argued that the concept is “reductionist across a number of dimensions: to the individual, to utility maximisation and to universal categories”^[19]. This is not surprising considering reductionism is the dominant paradigm of many disciplines. It is very difficult to conceptualise particularly the relational and cognitive dimensions of social capital from a reductionist perspective. Methodological reductionism inevitably results in a focus on the individual, which diminishes the importance of the social setting in understanding human behaviour. The concern is that this makes the analysis of social capital less social.

Social capital is not capital

Social capital is an unfortunate analogy

The most obvious criticism of social capital is the term itself and the question of whether it is ‘capital’. This is a common debate that has pervaded the literature on social capital. It stems from the unfortunate analogy that the term social capital attempts to make. Not only does social capital fail to meet the traditional definition of capital, the fact that it can have both positive and negative consequences make it incongruent with the term capital. There are numerous critics of the inclusion of capital in social capital, including notable economists such as Kenneth J. Arrow, Robert M. Solow, and Samuel S. Bowles. For many economist’s social capital is an affront to their discipline due to the erosion of the concept of capital. Attaching the adjective ‘social’ to the economic term ‘capital’ is an attempt to modify ‘capital’ as ‘social’^[23]. To most economists this is both unnecessary and inappropriate. Robert M. Solow has been critical of the use of the term capital in social capital. He stated, “it is an attempt to gain conviction from a bad analogy”^[24]. He also stated that “I do not see how dressing this set of issues in the language and apparatus of capital theory helps much one way or the other”^[24]. Kenneth Arrow^[25] found that social capital failed to meet the three important characteristics of capital: (a) capital has a time dimension; (b) it requires deliberate sacrifice of the present for future benefit; and (c) it is “alienable”—that is, its ownership can be transferred from one person to another^[26]. Although he did find that (a) above may hold in part where building a reputation or a trusting relation^[25]. His overall conclusion was that social capital should not be added to the other forms of capital^[25]. The American economist Samuel Bowles discussed the issue of ownership of social capital:

“‘Capital’ refers to a thing possessed by individuals; even a social isolate like Robinson Crusoe had an axe and a fishing net. By contrast, the attributes said to make up social capital—such as trust, commitment to others, adhering to social norms and punishing those who violate them—describe relationships among people”.^[27]

This suggests that because social capital cannot be owned by an individual and therefore cannot be traded it doesn’t meet the traditional definition of capital. I remain dissatisfied with the term and I generally agree with many of the criticisms above. I find the inclusion of the word capital most problematic when discussing the negative outcomes of social capital. That said, the term now has a history, albeit a chequered one, that gives meaning and significance to discussions. To start again with a new term would likely lead to further complication and confusion. So, I believe we are stuck with the term and for better or worse it’s here to stay. It should be noted that not everyone disagrees with the use of capital in the term social capital. There is considerable controversy in the literature^{[28][29][30][31][32][33]}. Portes^[34] suggested the location of capital in relation to other forms of capital:

‘whereas economic capital is in people’s bank accounts and human capital is inside their heads, social capital inheres in the structure of their relationships’^[34].

This doesn’t add much conviction to the appropriateness of capital in social capital, but other authors have identified similarities to other forms of capital. Social capital is similar to other forms of capital in that it can be invested with the expectation of future returns^[35], is appropriable^[36], is convertible^[7], and requires maintenance^[37]. There are significant differences in that it resides in social relationships whereas other forms of capital can reside in the individual^[38]. Further, social capital cannot be traded by individuals on an open market like other forms of capital but is instead embedded within a group^[37][39]. It is clear from the literature that social capital has both similarities and dissimilarities with neocapital theories and is certainly quite dissimilar from a classical theory of capital. To discuss this in more detail it is necessary to further identify the characteristics of ‘capital’. Schmid^[32] identified that capital is not immediately used up in production but rather its’ services extend over time. The capital stock is subject to investment for future production and depreciation and decay from both use and non-use. Piazza-Georgi^[40] stated that capital produces income and encompasses the non-consumable, but depreciating, inputs into the production process. The author supported Schmid^[32] stating that capital is a productive resource that is the result of investment^[40]. Castle^[41] added that other characteristics of capital are usefulness and durability. If social capital is adherence to a norm and not affected by individual action as Fukuyama^[42] suggests, then it is not capital in the above sense. The main difference is that more than one person benefits from social capital^[43][32]. Smith and Kulynych^[44] believed that the word capital has a too broad, pervasive and honorific meaning and that the term blurs many distinctions which adversely affects the scholarly inquiry, whatever its implicit or explicit normative concerns. Inkeles^[30] suggested that the term capital is too limiting and would rather use the term social or communal resources.  The author argued this on the basis of:

‘capital being an element of production, in particular the production of goods, but also services. We want not only goods and serves but also social support, physical and social security, freedom of expression, opportunities to develop ourselves and a host of these outcomes not captured by the idea of goods and services’.

Hofferth, Boisjoly et al^[29] suggested that social capital is the result of altruism and therefore not capital as capital is a resource that is built up through investment and can be drawn upon when needed. Lin, Cook et al^[45] disagreed by identifying that social capital shares commonalities with other forms of capital, notably human capital. SCIG^[46] supported Lin, concluding that the consequences of social capital are capital in nature because capital suggests something that is durable or long lasting and suggests something that retains its identity even after repeated use, something that can be used up, destroyed, maintained, or improved. Many authors identify that both forms of social capital, structural and cognitive, qualify as capital because they both require some investment – of time and effort if not always of money^[47][48][49]. It can be concluded that social capital is unlike other forms of capital but also not sufficiently dissimilar to warrant a different term. Certainly, it is the use of the term capital that makes the concept attractive to such a wide range of people given the bringing together of sociology and economics^[50]. Perhaps a more appropriate term may be social solidarity as the notion connotes relations of trust, co-operation, and reciprocity just as much as social capital and might be used in place of it to overcome the problem identified above with using the term capital. It is interesting that the term capital should be used with social, considering capital is already a social relation. In the original sense of the word capital, an object is only capital under particular social conditions. In the same way the sources of social capital are only capital under particular social conditions. For example, a favor owed is only capital under certain, not necessarily favorable conditions. This idea brings in the notion of negative or perverse social capital.

Social capital theory is not a theory

The word ‘theory’ can mean different things in different fields of research, however it generally means an explanation or descriptive assertion related to specific events^[11]. Many approaches to social capital theory make significant generalisations to simplify the complex social environment with the aim of making measurement more practical and achievable. Unfortunately, by doing so they lose much of the explanatory power of the processes that make up social capital. Some authors even treat the concept as a coherent whole without explaining how meaning is derived from the various mechanisms that make up the concept. The opposite also causes problems. By attempting to distil social capital to simple discrete elements the essential meanings tend to be lost in the inherent assumptions that are typically involved. An example is the distinctions between bonding, bridging, and linking social capital. These typologies amalgamate a variety of contradictory aspects of both networks and norms into single categories, creating methodological blind spots that decrease the use-value of the concept^[51]. An example of this generalisation is how trust is typically treated in social capital measurement where trust is commonly equated with social capital. It tends to be measured by questions such as “Would you say most people can be trusted?”. This assumes that trust has no meaning independent of its ability to facilitate social relationships. A richer picture of trust suggests it is a psychological disposition independent of the relationship context^[21]. Other theoretical analyses suggest trust is a highly contextualised decision regarding the trustworthiness of potential partners, largely independent of the elements that are said to be part of social capital^[52][51]. A further problem with this approach is that trust tends to be pre-reflective so not fully known and understood by the person answering the question, but rather part of the background context of knowing and acting^[53]. If they have the time and motivation to reflect on this question their answer may be different to their immediate response. Responses are also influenced by a range of factors including personality, mood, and how they understand the question and its context. Experimental work suggested that social trust questionnaires may be conflating trust with caution^[54]. This discussion suggests that the way trust is typically included in social capital theory results in the loss of meaning.

Social capital is a renaming and collecting together of a wide variety of sociological concepts and processes

Social capital is a renaming and collecting together of a wide variety of sociological concepts and processes^[55]. Calling the encompassed processes ‘social capital’ is just a means of presenting them in a more appealing conceptual garb^[34], but by doing so it becomes a sack of analytical potatoes, merely a metaphor or a heuristic device rather than a robust and coherent theory^[14]. I can see how in this respect social capital has become a catch-all concept which allows each author to see what they want^[56]. Adler and Kwon^[35] made the point that social capital is “a wonderfully elastic term”^[57], a notion that means “many things to many people”^[58], and that it has taken on “a circus-tent quality”^[59].

The realm of what is “social” is extremely complex

The inclusion of the word “social” makes social capital incredibly general and widely applicable to virtually any area of human endeavour. The word “social” is one of the most widely and broadly used adjectives in the English language yet the realm of what is “social” is extremely complex^[60]. It is not clear what is assembled under the umbrella of social and what is the precise composition of the social domain. The use of the word social gives social capital enormous generality as well as ambiguity. When the concept is defined so broadly it seems to cover any social phenomena which could have potential effects on social and economic outcomes and therefore it lacks specificity and substance^[16]). When conceptualised as an undifferentiated mixture of multiple independent social dimensions the causal mechanisms of specific dimensions remain nebulous^[61]. A further result of this approach is the tendency for social capital theory to gloss over the multi-dimensional and conflictual nature of social stratification, especially where social capital ignores divisions of class, race, ethnicity, gender, age, etc.^[14]. The result is that social capital as a concept lies somewhere between grand systemic theory and mere description. It is often a heuristic since it is often not an optimal approach. Despite this, many consider it sufficient for their immediate goals. This adds validity to the claims that it is little more than a metaphor and heuristic device. As discussed previously, social capital encompasses a range of processes that have been subject to scholarly investigation in a range of disciplines. By combining these processes in a single concept the result is to reduce and homogenise the content of earlier diverse analyses^[14]. As such using the social capital concept is often not the ideal approach since in many circumstances a different concept would be more relevant and appropriate. However, the popularity and appeal of the concept has resulted in it being used in the place of other more suitable terms. When this happens often the author will cut and trim social capital to make it suit. This further reduces the explanatory power of the ‘theory’, especially compared to the established concept that should have been used instead of social capital.

Application tends to be flawed in view of model specification, multicollinearity, omitted variables, multiple equilibria, cross-section as representative of time series, and so on.

Other reasons why social capital has been criticised for not being a theory relate to the multitude of problems with conceptualisation and operationalisation. Social capital theory application tends to be flawed in view of model specification, multicollinearity, omitted variables, multiple equilibria, cross-section as representative of time series, and so on. The biggest problem tends to be confusion of the cause, function, and consequences of social capital^[62]. Following from Putnam^[63] and Coleman^[36][64] many authors do not really distinguish social capital from its product^[1]. Authors tend to confuse the existence of social capital with its functions and its causes with its effects^[65]. That is, they don’t differentiate between the identification of social capital and the resources obtained from or through it^[23]. This can result in researchers finding exactly what they set out to find since the dependent variables and independent variables measure the same thing. Mayer^[23] suggested that the cause, function, and consequences of social capital are often conflated in a circular argument. Often outcome variables and the underlying causal variables are not satisfactorily differentiated and possible alternative causal factors are not controlled for^[16]). The lack of distinction between source, form, and consequences of social capital has other problems. When social capital is defined in terms of the presence of desirable outcomes then researchers will logically find exactly what they are looking for. In this respect many authors treat social capital as an unalloyed virtue^[66]. Yet we know that this is not the case. The same social processes that result in benefits can produce undesirable outcomes, depending on perspective. Thus social capital research must deal with how socially desirable versus socially undesirable behaviours are selected^[66]. Ben Fine^[14] believed that the fact that social capital can also result in perverse outcomes depending on circumstances renders it unacceptable and subject to collapse under the weight of its own contradictions and inconsistencies. Further discussion on the problems of social capital measurement can be found here.

Lessons from criticisms of social capital theory

The criticisms of social capital may seem damning. After considering the problems with the concept one may be inclined to dismiss it entirely as fundamentally flawed and beyond salvage. But as an optimist I see the criticisms as an opportunity to learn and improve our practice in relation to social capital. Many of the problems with the use of social capital would be avoided if there wasn’t cavalier use of the concept. This can happen under several different circumstances. For example, where people use the term where a different term would be more appropriate, where people use the term without understanding or explaining how it is relevant, where people tailor it to suit their purpose by omitting or aggregating certain aspects, and where people use it with poor scholarship. It’s not clear exactly how to avoid making these mistakes given that every context is different and there are many different theoretical approaches to social capital. There are however several lessons that we can take from the above discussion:

• Don’t lose the social by allowing rationalism, individualism, and reductionism to reify the ‘social’
• Don’t treat social capital as capital, it’s not capital in the traditional sense, it’s more appropriately a glue, lubricant, or catalyst
• Don’t use social capital if a more appropriate term exists
• Don’t perpetuate the existing vulgar scholarship by oversimplifying or inappropriately aggregating aspects
• Don’t get caught in a circular argument – be very clear about what is source, form, consequence
• Don’t make assumptions about the indicators of social capital, particularly how the data is collected and what its significance is for the aspects of social capital being investigated
• Don’t ignore the context where social capital is being used or investigated and be clear about the scope of interest or relevance
• Don’t confuse or ignore the different levels at which social capital exists
• Don’t create your own definition or instrument where an existing one would suit

The application of social capital requires careful thought and consideration. In general, the best approaches tend to be those that maintain and embrace the specific context being investigated and those that allow for complexity to be maintained rather than simplification and aggregation that often involves assumption and confusion of causality. As such, qualitative approaches tend to achieve the best results, especially methodologies such as ethnography that allow understanding to emerge. Mixed methods also tend to be very popular due to the desire to collect data from a larger sample and this approach can have success.