Evidence to define the sustainability of a healthy diet

Defra

FO0430

Annex B – Social Sustainability

Contents

1.0Methodology

1.1Social sustainability

1.1.1Defining social sustainability attributes of foods

1.1.2Evidence of the socialsustainability attributes

1.1.3Evidence gathering and recording

1.2Expert Panel and workshop

2.0Results: Societal impacts

2.1Results: All foods

2.2Caveats

2.3Health and welfare dimension

2.4Ethical dimension

2.5Working conditions dimension

2.6Societaldynamics dimension

2.7Additional initiatives related to social sustainability

2.7.1The FAO e-fora on sustainability and the food chain

2.7.2The UNEP Report

2.7.3 Food Ethics Council: Food Justice Report

2.7.4 Marques and metrics

2.7.5The Rural Economy and Land Use Programme

2.7.6Future scenarios

2.7.7Other academic papers

3.0References

Appendices

B.1Evidence of societal impacts of food production and consumption

B.2References associated with Appendix B.1 on Social Sustainability Evidence

1.0Methodology

1.1Social sustainability

1.1.1 Defining social sustainability attributes of foods

What is true of sustainability in general (see Introduction) is particularly so of the concept of ‘social’ sustainability – one element or dimension of the broader concept. One definition often used comes from the Brundtland Commission (UN, 1987), which describes sustainable development as “development that meets the needs of the present without compromising the ability of future generations to meet their own needs", within which the three key factors are environmental protection, economic efficiency and solidarity in society. Although in this project we have adopted a similar three-pillar conceptualisation, it is important to recognise that other definitions make further distinctions, or disagree in terms of what aspects should be considered under which heading. For example, Van Calker et al (2005) considered four dimensions of sustainability (in the context of dairy farming): ‘ecological’, ‘economic’, ‘internal social’ and ‘external social’ sustainability. Within this, the key attribute related to internal ‘social sustainability’ was ‘working conditions’, and the key ones related to ‘external social sustainability’ were ‘food safety’, ‘animal welfare’, ‘animal health’, ‘landscape quality’ and ‘use of undisputed products’. Olesen, Groen and Gjerde (2000) however,defined sustainability in terms of resource efficiency, profitability, productivity, environmental soundness, biodiversity, social viability, and ethical aspects (the latter two perhaps being more ‘social’ than the former five). Others have discussed individual issues – for example, Broom (2010) has stated that ‘a system that results in poor (animal) welfare is unsustainable because it is unacceptable to many people’ (p.83) (a sentiment with which others, e.g. Keeling (2005), agrees).

One of the most internationally-known and widely tested approaches - the Response-Inducing Sustainability Evaluation (RISE) method (see Häni, Stämpfli and Keller,2003) - assesses the sustainability of farms based on twelve indicators for the economic, ecological and social situation: energy consumption, water consumption, situation of the soil, biodiversity, emission potential, plant protection, wastes and residues, cash flow, farm income, investments, local economy, andsocial situation of farmer family and employees. Interestingly, only the last of the 12 indicators would appear to speak to the social component. This is measured by considering: a) the relation between average Full Time Equivalent (FTE) compensation on the farm and Minimum Regional Income (MRI); b) Relation between lowest FTE compensation and farm income per FTE, and c) Assessment of the social situation of family workers and employees. While the first two aspects are economic aspects and reflect ‘driving forces’ of the indicator, the third concerns the ‘state’ of the indicator, and appears underspecified. Additionally, one might argue that this is a somewhat limited conceptualisation of social sustainability that misses a number of other potentially relevant issues identified by Van Calker et al and Olesen et al.

Furthermore, in the FAO (2011) document previously discussed, a kind of working definition of sustainability has been adopted for purposes of the consultation based on “an initial review, expert meetings, and subsequent review of corporate responsibility reports of dozens of food companies and retailers” (p.1). This identifiedfour pillars as having a role in creating the necessary framework conditions for ensuring sustainable development, namely: ‘environmental integrity’, ‘economic resilience’, ‘social well-being’ and ‘good governance’. (Although the latter is seen as somehow different: “Good Governance was considered by most experts to be an underlying, enabling concept rather than a pillar of sustainability (with) core issues identified therein (being) key components in the credibility of sustainability interventions.” (p.4)) Regarding social well-being, this is referred to as including: a) labour rights (the range of rights enshrined in the ILO Declaration on Fundamental Principles and Rights at Work); b) Non-discrimination and equity (equal access to opportunities and empowerment of women, reduction of discrimination and inequalities); c) education (access to, engagement in and attainment through education, knowledge sharing and awareness raising); d) health and safety (providing access to medical treatment, nutritional products and safe working conditions), and e) social commitment (acting actively to benefit society at large). This scheme specifies a total of 30 indicators to enable measurement of these ‘core issues’ (and note again, the diversity in language used to describe key concepts). What is particularly notable in this scheme is that issues such as animal welfare, landscape and ethics do not appear, and the emphasis is on issues such as child labour and education, that might not be apt for a UK-centric framework. Additionally: “A number of the issues… under Economic Resilience overlapped with those proposed under Social Well-Being and vice versa. In part this was due to the relatively greater difficulty of understanding what was intended by a number of the social core issues, and in part it was also due to the relative breadth of the concepts inherent to these issues. For example, what is understood as “Livelihoods”, and is it an economic or a social issue?” (p.4). ‘Transport’ and ‘employment’ are two core issues that appear under economic resilience as opposed to social well-being, which might be open to contention.

In short, the issue of social sustainability is a complex one, with little consensus as to what this phrase actually means. So, where does this inconsistency leave us? At the outset of the project we adopted a particular definition of social sustainability as the ability of an activity to endure in the social realm, which is to say, to be accepted by individuals and groups within society (as beneficial, or at least, non-harmful); not to be boycotted or banned or ignored.

Following this initial definition, and on the back of definitions in the literature, we selected four different attributes (though the term ‘dimensions’ is preferred here) that seemed to encapsulate the key aspects identified by others, namely:

  • Human health and welfare (perceived and real)
  • Animal health and welfare (perceived and real)
  • Ethics
  • Other social and cultural aspects (landscape, education, employment, cultural cohesion, etc.)

These four attributes incorporated aspects from Van Calker et al (2005) (e.g. animal health and welfare; food safety) as well as Olesen et al (2000) (e.g. social viability and ethics). They also seemed to closely fit with the understandings of the ‘Sustain’ network (UK), an alliance for better food and farming, which “advocates food and agriculture policies and practices that enhance the health and welfare of people and animals, improve the working and living environment, enrich society and culture and promote equity.” (The four dimensions are apparent here, with the ethical dimension related to the important concept of ‘equity’.)

Ideally, for this deconstruction of the concept to be meaningful, the four dimensions should be independent and comprehensive. One could argue, however, that animal health and welfare (certainly from an ‘anthropocentric’ sustainability position) might be incorporated within ‘ethics’, since the main reason to protect animals might be seen as to pander to ethical scruples rather than to protect animals for their own sake (e.g. as equally important as humans). One might also argue that the final dimension is rather a catch-all one, relating to more fuzzy concepts that still, at their base, are important from a ‘human health and welfare’ perspective. Nevertheless, this was the organization we initially chose to adopt, as its benefits seemed to at least equal its detriments.

Our initial data searches (described in the next sections) were based upon this system. However, following the expert workshop in December 2010, and comments from various experts about the focus on ‘acceptability’ being too narrow, with “social sustainability… not just about social acceptability, but also about preventing damage to society and key social groups” (personal communication) we amended our definition somewhat. (In fact, we did appreciate this element, but simply phrased the definition in an earlier draft in an inadequate manner.) Thus, we hereafter take social sustainability to involve developments that do not damage any aspect of society or any social group, particularly vulnerable groups (e.g. remote communities, those in the developed world, ethnic minorities, those with disabilities, the poorest members of society). By ‘damage’ we refer to both physical and mental damage – the latter aspect thereby introducing issues of acceptability and ethics.

The new attributes/dimensions were:

  • Health and welfare (perceived and real), with respect to a)food safety and b)nutrition
  • Ethics, with respect to a) inequity (amongst social groups) and b) animal health/welfare
  • Working conditions, with respect to employment conditions and worker safety
  • Societaldynamics, with respect to community cohesion, education, landscape

This re-framing seems to clarify links to further definitions, such as that of the FAO (2011), and takes into account comments received during and after the expert workshop. The specific changes are to elaborate that the health and welfare attribute is concerned largely with food safety and nutrition (not worker safety, which comes under working conditions), and to move the discussion on animal health and welfare into the ethics attribute (i.e. taking a more anthropocentric position). This change made sense from an analytical viewpoint, since we were originally faced with difficulties sometimes assigning research evidence to one of these attributes rather than the other. This definition also specifically notes that ‘inequity’ comes under the ethics heading (it did do so in the original analysis, though the lack of explicitly using the word ‘equity’ or ‘inequity’ caused disquiet in some of the expert workshop attendees). Finally, the ‘other cultural aspects’ has been better specified and split, with working conditions now a specific and separate attribute (which seems apt, given its status in other definitions of social sustainability), while ‘societaldynamics’is now used as a heading to capture the various other (often nebulous) issues relating to the protection of communities and society at large (their integrity and space). It is important to stress at this point that the reframing of the key attributes actually had little impact in terms of our charting of evidence of social sustainability (see the ‘results’ sections), and generally merely resulted in certain findings being moved from one cell or column of our matrix to another (that is, it didn’t lead us to any further sources of data).

1.1.2 Evidence of social sustainability attributes

The term ‘sustainability’ has a relatively precise meaning in terms of environmental aspects (though not without some degree of contention, as discussed in previous sections). Likewise, it might be argued that ‘economic sustainability’ is a concept that can be readily understood and measured (in terms of the costs and benefits of actions in a monetary sense). Unfortunately, ‘social sustainability’ is more qualitative in nature. That is, the clarity of the respective concepts is perhaps best indicated by the commensurate presence or absence of accepted ways of measuring those concepts. For example, environmental sustainability is frequently addressed through the application of ‘Life Cycle Assessments’ (LCAs) – a process that compares the environmental impact of some activity throughout a defined life cycle. Although there is no precisely named process for measuring economic sustainability, it would appear reasonable to apply measures of monetary cost to any activity, and so establish its relative sustainability versus other activities. However, social sustainability – being less well-defined, and occurring at a ‘higher’ level of analysis (i.e. at the level of human beings and societies, as opposed to the level of natural science components) – is rather elusive, with no universally accepted measures. This makes the comparison of pieces of evidence difficult – being in many cases based on anecdote or qualitative approaches.

More specifically, it can be seen from the definition of social sustainability adoptedhere that there are implied psychological and behavioural aspects in several of the dimensions, that is, an activity can be sustainable/non-sustainable because it has a ‘real’ impact (e.g. on human or animal health, on equity of access) or because it is perceived to have such an impact (the issue of genetic modification is one important example here, where sustainability is arguably compromised by perceptions rather than realities). As such, measures may include subjective, attitudinal ones as well as objective ones. While objective aspects can be established in terms of actual measured and recorded outcomes (e.g. worker accidents; proportion of animals that die in transportation; demographic characteristics of consumers of certain products), the more psychological measures are not so easy to establish. Data on attitudes can arise from qualitative, relatively unstructured research. Alternatively, such data can be attained through structured questionnaires – asking about attitudes and perceptions on scaled measures, or attempting to establish likely behaviour through measures of ‘intention’ to behave, or ‘willingness to pay’ for some aspect. Data on the sustainability of any activity might come in a variety of forms – with ensuing difficulties raised in terms of trading off data that might measure subtly different aspects of response measured through different research processes and ‘instruments’ about subtly different potentially sustainable activities (e.g. measures of impact on health versus qualitative and quantitative measures of perceptions/ attitudes). The issue of trading off evidence across the three sustainability pillars is discussed in the integrating sections of the report.

1.1.3 Evidence gathering and recording

In this project, we are concerned with different potentially sustainable activities carried out at different points in the food chain cycle (taking six specific stages, from ‘breeding’ to ‘consumption’), for a number of different food types or classes. One way to address this problem is to look at the activities related to one particular food at one particular stage and attempt to find evidence as to whether this might be socially sustainable or not. For example, one could look at ‘bananas’ at the ‘transport’ stage and look for some kind of related measures concerning health, ethics, working conditions, and societaldynamics. The problem with this approach is that much of the potentially relevant social research will not address problems in this particular way, and hence, we might find little of relevance. For example, it is unlikely that there is any specific evidence on perception and behaviour related to transporting bananas – it is not an issue likely to provoke research interest! As a case in point, a review for Defra on the sustainability issues around liquid milk, which focused on LCAs, but which did attempt to go further and consider social issues, concluded: “For milk retailing and utilisation, quantitative evidence for milk-specific impacts is sparse. As with transport impacts at the retail and consumer ends of the system, tying impacts to individual products presents significant challenges.”(Foster et al. 2007, p. X) As with milk, we have found, so with almost everything.Indeed, social sustainability indicators are linked mostly to how a food product is produced, processed, distributed and consumed, rather than the specific type of food. This is a very different approach to that used in LCAs and other methods for assessing environmental sustainability, and hence there may be a need to approach and conceptualise social sustainability in a different way.

A second approach is to thus consider generic issues first, and to try to slot any findings into the relevant cells of the matrix. For example, ‘food transportation’ generically might be a topic that has stimulated some research interest. It is this approach we take – finding generic research evidence, and then adding appropriate caveats to cells of the matrix to indicate whether there is or is not any information about a specific food type.

Aside from limitations due to lack of likely research on specific food type-food chain-social sustainability interactions, a second problem is that much research data on the broad issue is likely to be only tangentially relevant. For example, much of the research on human attitudes towards food concepts might have some implications for specific aspects of the food chain (for example, the extent to which consumers trust food risk managers might have implications for the sustainability of particular aspects of the food chain in which risk managers play a role), but teasing out full implications may require considerable conjecture and is not likely to provide clear-cut answers to the sorts of questions that Defra want answered. Rather than confusing matters and trying to interpret a vast amount of literature (not possible in the time frame of the project in any case), the focus of our analysis was on research and evidence that fairly directly addressed the central questions. In other words, our analysis has been pragmatic: we looked for major pieces of research, reviews, and established databases that appeared to address a clear aspect of social sustainability. What we havenot provided is references to every minor study with small and specific samples looking into a precise issue with uncertain generalisability.

And finally, our project does not have the capacity to purchase expensive market reports, and hence has focused on data that is in the public domain.

Table1 below shows the specific questions we attempted to answer. There are three questions in each cell of the matrix. The second question concerns the nature of evidence, and is shown within square brackets – essentially to indicate that this is not a question specifically required of us to answer. Associated with each food chain stage and societal dimension are a number of terms related to relevant concepts, which have been used as search terms in our analysis. Although the focus of this project is the UK, we did not necessarily discriminate against non-UK research (the bulk of that reported), although we did note when some issue seemed to imply national differences (and hence limited validity of data for the UK). With a UK emphasis, we were less interested in searching for details on – for example – educational benefits to people in the third world related to a particular food crop or system.