Introduction
Benchmarks are weapons of symbolic violence.Footnote 1 They are a form of domination from an implied consensus on how to rank and order, to assess who and what is worthy, to inform ways of being. As symbolic weapons benchmarks are useful for both reformist and revolutionary agendas, for those who use benchmarks to adjust behaviour within conforming norms, as well as those who seek to provide a direct challenge to how we think about the world and how it should be. Benchmarks are available to a range of actors nowadays. Weapons of symbolic violence in the twentieth century were primarily controlled by the state through the organisation of ideational and material resources around official certification systems, as well as the establishment of jurisdictions of competence in cooperation with professional associations.Footnote 2 Struggles over symbolic violence include the official consensus from the state and those who support it, insults from individuals expressing their views, and expert critics who could claim authority to their viewpoints.Footnote 3 In the postwar period the rise of private capital and of international civil society led to benchmarks being increasing used as a challenge to state-led forms of certification and standard-setting. Firms began using benchmarks as technical tools in the 1950s, to identify weaknesses in their systems. These benchmarks transformed into forms of managerial strategy in the 1980s and then ‘calculative practices’ to focus our minds on competitiveness and performance in the contemporary period.Footnote 4 Now the use of benchmarks and ‘scorecards’ is common practice, with bodies such as the International Organization for Standardization producing benchmarks and standards to guide corporate, governmental, and civil society behaviour.Footnote 5 Non-governmental organisations (NGOs) and expert-activists also put benchmarks forward, seeking to influence behaviour by providing their own systems of evaluation and judgement. Be they used by firms, NGOs, or public agencies, benchmarks are primarily ‘reactive measures’ that seek to induce conformity towards a value system that is commonly represented as a measuring tool that reflects an otherwise unseen ordering of units within a system.Footnote 6 Occasionally benchmarks can also challenge us to change how we see the units.
In their introduction to this Special Issue, André Broome and Joel Quirk discuss four constitutive elements of benchmarks, which include: (1) the reification of normative standards into observable categories; (2) commensuration through homogenising entities into a common metric; (3) symbolic judgments that create focal points for action; and (4) reactions from those targeted by benchmarks.Footnote 7 We suggest that reification, commensuration, focal points, and reactions are all elements of what Pierre Bourdieu referred to as ‘symbolic violence’ and associated with gentle forms of control which define reality and include a system of domination in which those dominated are active or unwittingly complicit.Footnote 8 From this viewpoint, symbolic violence has risen as the ‘most economical mode of domination because it best corresponds to the economy of the system’.Footnote 9 We argue that this is indeed the case when it comes to benchmarks for global economic justice. Benchmarks seek to disenchant the observer and convey good science; they entail a process of accumulating economic capital through symbolic capital, trust in a benchmark, that has redistributive consequences for societies.Footnote 10 Ultimately benchmarking seeks to redistribute claims to material and symbolic forms of capital.
This article is interested in how NGOs and expert-activists coordinate and create benchmarks. We seek to highlight how benchmarks for global economic justice differ in how they are organised, how the benchmarks are articulated, and the process through which they are tested for three key elements: salience, will, and expertise. We place great emphasis on how salience, will, and expertise are important for benchmarking cycles that can change transnational norms. Salience is a key test of relevance for NGOs and expert-activists, both from stakeholders they seek to persuade as well as from a political audience that can pay attention and change practices through hard and soft forms of governance.Footnote 11 Will is the initial condition for NGOs or expert-activists to seek benchmark creation, and as a benchmark develops the original spur must be transformed into political will to move from evaluation to enactment. Expertise is a key ingredient because often NGOs select issues on which they are not fully equipped to analyse, and because expertise must be recognised by peers lest those creating the benchmarks be viewed as flawed and inadequate or simply barking at the moon. For benchmarking to be successful we suggest that the trinity of will, salience, and expertise must be present, since failing a salience of expertise test sends those propagating the benchmark back to square one, or to develop a benchmark that will attract little attention or policy traction.
Benchmarks align with dominant thinking about how things should be but can also challenge the normative order. As such we differentiate reformist and revolutionary benchmarks. Reformist benchmarks seek to create change within the dominant system. The rise of benchmarks to assess accountability and corruption are an obvious case of reformist benchmarking, where NGOs are adding new practices that flow in the same direction as well established normative trends.Footnote 12 Revolutionary benchmarks seek to challenge how the basic units of the system, and the logic of the system are perceived. These benchmarks are often driven by expert-activists and must persuade NGOs, stakeholders, and political authorities that they have a point. Revolutionary benchmarks, as we clarify below, are frequently challenged for their expertise and salience, especially on methodological grounds where frictions can easily emerge over good science rather than over politics. Claims to originate revolutionary benchmarks may be made from experience, from ‘bearing witness’ to injustices, but they also have to be supported by scientific expertise.Footnote 13 We suggest that reformist and revolutionary benchmarking cycles can be understood as ideal types, that is, as abstract heuristic devices that help us to understand a range of empirical cases and their differences.
Our specification of reformist and revolutionary benchmarks seeks to contribute to the extensive literature on how NGOs select issues to campaign on and how they muster ideational and material resources to push ahead. Our cases have been selected via inductive reasoning. In short, the mechanisms presented in the revolutionary case were presented to us over multiple interviews between 2011 and 2014 with tax justice campaigners and policy stakeholders. We then compared these dynamics with dominant themes in the International Relations literature and interviewed NGO professionals – those not tied to particular issues but generalists – about how they formulate benchmarks. These interviews were conducted in Brussels, London, and New York in 2012 to 2014. This piece focuses on the varied characteristics and mechanisms of benchmark entrepreneurship in the NGO world.Footnote 14 We seek to highlight how NGOs and expert-activists operate in a highly contested field of power where there are frictions between themselves and other types of organisations and groups, such as firms, states, and international organisations. Significant tensions exist within the organisations and groups as all seek a stake in the game.Footnote 15 Earlier work in this area highlighted how NGOs and transnational advocacy networks engaged in ‘information politics’ to get their messages across, and how they influenced states and international organisations by imposing reputation costs and shaming. To do so they selected issues linked to bodily harm and unequal access to rights, issues on which they could draw a large audience.Footnote 16 Should all this go well then it would be possible to establish a norm tipping point where transnational norms would be transformed.Footnote 17
More recent research has built on these earlier insights to show how NGOs engage in gatekeeping to only select issues they feel are salient,Footnote 18 with managers within NGOs carefully vetting issues for their attributes prior to committing resources to them to ensure that salience is present. The competitive rather than cooperative side of NGOs has also been explored at length, with excellent scholarship on how NGOs compete with each other for resources to the detriment of those they seek to help,Footnote 19 as well as how NGOs often act as coalitions and interest groups that display distinct national differences.Footnote 20 Extensive studies have also pointed out how NGOs have become professionalised, bureaucratised, and even corporatised in how they mount and operate campaigns.Footnote 21 From this perspective, NGOs and benchmark creation may be easy due to greater in-house capacity and expertise, or may also be much more difficult because many NGOs have become lumbering bureaucratic machines built to campaign rather than house expertise.
Rather the viewing the increasing professionalisation and corporatisation of NGOs as a rational process through which they seek to attain influence within the dominant normative order of our time, we argue that benchmarks are double-faced and volatile, depending on who is advocating them.Footnote 22 Reformist and revolutionary types of benchmarks contain different characteristics of salience, will, and expertise that change the stages in a benchmarking cycle and the conditions of possibility. Reformist benchmarking, we argue, commonly relies on NGOs outsourcing to experts who then seek to create stakeholder buy-in on the way to creating benchmarks, which are then conveyed to political authorities. Because these benchmarks tend to align with public and private sector norms reformist benchmarks have a much greater chance of passing the test of political engagement and leading to what Sidney Tarrow refers to as ‘certification’, when authorities validate claims and performances made by actors.Footnote 23 By contrast, revolutionary benchmarks do not come from larger NGOs, whose managers choose not to approach issues in a way that damages their membership and resource base. Rather, revolutionary benchmarks emerge from expert-activists that then seek NGO buy-in. These expert-activists often directly challenge how we see units in the system and the logic of the system. As such their benchmarks come up against ‘mature field’ opposition from NGOs and corporate interests that are well established and question the expertise behind, and salience of, the benchmark.Footnote 24 Given this conflict political authorities are less likely to certify these benchmarks. Still, expert-activists have the potential to use benchmarks as weapons in a fight over what modes of domination prevail.Footnote 25 Benchmarks are important weapons in global battles over the right to govern economic activity.Footnote 26
The article is structured as follows: first we provide the Reformist Benchmarking Cycle, establishing how it works and providing two examples of benchmarks through Transparency International’s Corruption Perceptions Index and the Gold Standard’s development of certification credits based on the benchmarking of renewable energy or energy efficiency projects. Both benchmarks conform to the normative order of our time and seek to reform the treatment of an issue by challenging core assumptions about corruption or environmental markets. Both benchmarks were also developed as offshoots from larger organisations. This leads us to the second section, where we provide the Revolutionary Benchmarking Cycle, establishing how it differs from the previous cycle. We provide an example of the Tax Justice Network’s Financial Secrecy Index (FSI), a benchmark that directly challenges assumptions about how core advanced industrialised states operate and the logic of how ‘tax havens’ operate in the international political economy. We highlight how expert-activists working through the Tax Justice Network have been key to the FSI and challenged by a ‘mature field’ in accountancy and politics, primarily on ‘non-political’ methodological grounds. Evidence for these cases has been gathered through extensive interviews with key stakeholders, as well as document analysis from public sources or private sources obtained from interviewees. In the final section we discuss the two benchmarking cycles and why it is important to differentiate reformist and revolutionary types.
Reformist benchmarking cycles
Reformist benchmarking cycles seek to push forward progressive agendas within the established organisational frameworks and normative order of the day. Reformist benchmarking is particularly useful to larger NGOs who concentrate on mounting and implementing campaigns on their selected issues. These NGOs, such as Oxfam, are highly bureaucratised and professionalised, with staff selected for their professionalism as much as for their passion in seeking to push forward a particular issue. As noted in the literature, such large NGOs are abundant with ‘management speak’ and a key concern for them is how to maintain lines of revenue and membership through the careful selection of issues.Footnote 27 Such agenda vetting is important for NGOs when choosing issues since those that are too complicated or carry mixed messages are unlikely to play well.Footnote 28 Issue selection is also guided by the industrial structure surrounding it,Footnote 29 with some issues easier to campaign on than others – compare, for example, ‘fair trade’ campaigns for coffee and chocolate and sex slavery in large hotel chains.Footnote 30 Since the fair trade mark was awarded in the Netherlands in 1988, NGOs supporting it have had a long history of cooperation with firms to provide a combination of salience, will, and expertise, especially in the identification of global value chains.Footnote 31 By contrast, initiatives such as the Global Business Coalition Against Human Trafficking have to make significant strides to combine their campaigns with the commercial interests and reputations of hotel chains in which sex slavery is an informal part of the supply chain.Footnote 32 The selection of issues will constrain how organisations can campaign and what institutional repertoires they have available to them.Footnote 33
Larger NGOs also have an interest in creating ‘scale free’ networks that have low costs for new members to join so that expansion and communication on issues is made more efficient.Footnote 34 Choosing issues with obvious political and moral salience is the key way to constructing these networks, as is involving key stakeholders in testing what counts as a salient issue. For these reasons, many NGOs cooperate closely with firms and non-civil society actors to harness greater support and pass salience tests.Footnote 35 Such ties provide access to those who may have the expertise that is missing from the NGO, and robust networks assist in the creation of benchmarks that can also gather political will. In short, there are many incentives for NGOs that are interested in creating benchmarks to conform to existing normative standards and then propose the strengthening of a particular element rather than advocating a revolutionary change in how the system is perceived.
Figure 1, below, presents our view of a reformist benchmarking cycle. To walk us through the various steps in the cycle, at the twelve o’clock position is the transnational norm which an organisation or activists seek to modify or change through the creation of a benchmark. This is an initial test of will that, if present, leads to the selection of an issue by the organisation at Step 1. As discussed, this process is carefully managed within NGOs and subject to a range of perceptions about what has political and moral salience and what does not. Following issue selection the organisation then moves to expert engagement that can take place within the organisation, outside the organisation (through external consulting), or through the establishment of an external expert group by the primary organisation. This is Stage 2. The next step is to engage stakeholders once the organisation has information from the experts. This is the first test of salience and stakeholders, be they other NGOs, firms, state agencies, or international organisations, can provide feedback on what will work and will not work, perhaps leading to an early death to this initiative. If the salience test at Step 3 fails then it is back to Step 1. However, if successful the next step is benchmark creation, where experts provide their expertise alongside the organisation’s normative reformist goals. This is Step 4 where the information from expert engagement is directly put to use. The next stage is political engagement where there is a second test of salience as well as a test of expertise. Political authorities may question the expertise of those involved at Step 2, especially when the issue is politically and economically sensitive, sending the benchmarking process back to the expert engagement stage.Footnote 36 Political authorities may also discount the benchmarking initiative as unimportant altogether, pushing it back to Step 1 where the organisation can vet and select the issue once more. However, if Step 5 is passed and there is a combination of salience, expertise, and also political will then the transnational norm can be transformed. We present this cycle as an ideal type that, as with all ideal types, should be broken down and reassembled once it fails to adequately depict reality. We also note that variation in benchmarking cycles can also be understood as emerging from the differences in the articulation of conforming and contesting norms for reformist and revolutionary issues at different steps in the cycle.
Reformist benchmarking cycle
Reformist benchmarking cycles are the most common form given that organisations can select issues that are likely to have salience, where expertise can be found due to the organisational and commercial benefits associated with established systems, and where will is more readily mustered as it already aligns with present norms. As such there are many examples to draw upon, but we highlight two prominent cases of reformist benchmarking that follows the cycle described above: Transparency International’s Corruptions Perceptions Index (CPI), and the World Wildlife Fund for Nature’s support for the development of the Gold Standard as certifiers of sustainability within carbon markets.
The Corruption Perceptions Index (CPI) was launched in the mid-1990s by Transparency International (TI). TI was established in 1993 by Peter Eigen who, frustrated with the impact of corruption on projects in Africa, retired from the World Bank to combat corruption. TI set out as an activist organisation led from the top.Footnote 37 Founding members included officials of international organisations, former ministers from developing countries, and elite businessmen who chose corruption as an issue to work on.Footnote 38 TI has more than 100 National Chapters who engage in domestic anti-corruption initiatives, while the international secretariat is located in Berlin. TI flagship projects include its ‘Sourcebook’, which was launched in 1996 to provide a step-by-step guide to combating corruption in different sectors. This works in conjunction with a best practice oriented ‘National Integrity System’, which provides a qualitative assessment of a country’s institutional capacity to provide good governance. By far the most visible TI product is the CPI launched in 1995.
The CPI was developed in an environment of increased attention on reforming polities in the post-Cold War period and the diffusion of democracy.Footnote 39 As suggested by Honghing Wang and James T. Rosenau, TI’s formation of a CPI would have ‘a better chance of success if it does not contradict the interest of powerful actors’.Footnote 40 In 1994, the OECD adopted a recommendation to criminalise foreign bribery and in 1997 its then 34 member states agreed to criminalise the paying of bribes to officials in other countries.Footnote 41 TI’s proliferation of anti-corruption norms was emboldened by ready support from multinational firms, powerful states, and international organisations who had a pre-existing interest in reducing corruption.Footnote 42
The CPI ranks countries and territories on a scale of 1 to 100 according to perceived levels of corruption among public officials and politicians. A score of 0 reflects very high levels of corruption. The CPI is composite index drawing from polls and surveys by political risk analysis firms, business schools, civil society organisations, and international organisations including the African Development Bank, the World Bank, and the Economist Intelligence Unit.Footnote 43 TI’s Berlin Secretariat does not make a strong claim to expertise in the fields of law, policing or justice, with professional skills in the promotion and dissemination of ideas. Rather, for TI ‘Expertise when needed was drawn in ad hoc.’Footnote 44 Index creation relies on the engagement of expertise embedded in existing analysis by a host of actors already promoting an anti-corruption norm. The index places emphasis on institutional quality, public officials’ proclivity to demand bribes, and the dependence of business performance on corrupt transactions. The CPI does not provide a full picture of corruption in a given jurisdiction, but focuses on political and administrative corruption.
In the last decade the CPI has been incorporated into governance measures by the World Bank, International Monetary Fund, OECD, and the Bank for International Settlements, which speaks to the ‘certification’ of the benchmark at the heart of global governance.Footnote 45 Improving CPI scores are also considered to be crucial in who receives overseas development assistance.Footnote 46 Yet, the index fails to capture the supply side of corruption and the culpability of private actors.Footnote 47 The 2013 CPI ranked 177 countries with Denmark heading the list with a score of 91 and Afghanistan, North Korea, and Somalia at the bottom with scores of 8. Countries deemed to be performing in the top twenty of corruption perceptions in 2013 included Switzerland, Netherlands, Singapore, Luxembourg, Hong Kong, and Barbados. These countries fare less well in the revolutionary benchmark we present below. Benchmarking issues of global economic justice for reformist purposes will often generate broad scope and traction, finding function in ready requirements.
A second benchmark that conforms to the reformist cycle is the 2003 Gold Standard initiated in the wake of the 1997 Kyoto Protocol by the World Wildlife Fund (WWF) to remedy the failure of Clean Development Mechanism (CDM) and Joint Implementation (JI) projects to deliver ‘additionality’ and genuine added environmental and social benefits. The WWF encouraged the initiative to become an independent body having realised the market ‘would be better served by an independent standard with really tough experts serving it’.Footnote 48 The Gold Standard exceeds the requirements of the UN’s CDM Executive Board and generates credits representing one tonne of reduced carbon dioxide equivalent and carrying a price premium in the voluntary market. The development of the standard relied on the work of a Technical Advisory Committee consisting of experts from government, civil society organisations, and firms. At the outset the advisory committee outnumbered Foundation employees by four to one. In an interview with the CEO it was noted that expert consultation helps reconcile potentially divergent market and non-market goals to augment legitimacy.Footnote 49 The benchmark emerged from a two-year consultation process, where stakeholder involvement was gathered from a large section of the NGO community. Without this support, and with the Standard now formally endorsed by over eighty NGOs, the value of certified carbon credit was at risk. According to the CEO:
We derive credibility and brand value from operating on behalf of CSOs [civil society organisations] and always held the view that the Gold Standard needed the support of many NGOs. There is real value associated with endorsements from a wide variety of NGOs. For a corporate investor this lowers risk, that NGOs will not turn around and beat you up about it, your CSR [corporate social responsibility].Footnote 50
A first round of consultations with members of environmental groups and representatives of other sectors were held before consultants with experience in project development and verification drafted a project assessment framework. There were also consultations with representatives of private sector firms active in the market for greenhouse gas allowances and government officials responsible for this area in South Asia, Philippines, Japan, South Africa, the United States, Western and Eastern Europe.Footnote 51 By 2014, the Gold Standard had certified more than 1,000 projects and taken 20 million tonnes of carbon out of the atmosphere. Now it is a recognised benchmark in the carbon commodity markets, with large firms supporting its development and specialist consultancies continuing to provide certification and technical expertise to improve further reformist cycles.Footnote 52 Revolutionary benchmarking cycles face greater challenges.
Revolutionary benchmarking cycles
Revolutionary benchmarking cycles seek to interpret how we view the system surrounding an issue, as well as the character of units within the system. They differ from reformist benchmarking cycles in that tests of salience, will, and expertise are tougher, confronting vested interests who have clear reasons not to overturn their understanding of how the system works. An issue for revolutionary benchmarking is that successful benchmarks often seek to remove much of the moral and political tension inherent in the benchmarking process. As noted above, the editors for this Special Issue note that benchmarks involve reification, commensuration, focal points, and reactions. Much reification and commensuration is about rendering technical, political, and moral complexity into easily read metrics and indices.Footnote 53 Reformist benchmarking cycles align interests between NGOs, experts, and those concerned in the public and private sectors, allowing reification and commensuration to be effective. This is quite different for revolutionary benchmarking cycles, where expertise is probed for flaws in methodological construction.
Figure 2, below, presents our view of a revolutionary benchmarking cycle. As with the reformist benchmarking cycle we have five steps, but the character of the steps in the revolutionary benchmarking cycle are quite different. At Step 1 expert-activists select an issue to campaign on. This selection is based more on expertise and transformation of moral intent than the typical gatekeeper function found in reformist benchmarking cycles, where key managers inside the NGO carefully vet what can be campaigned upon and what cannot.Footnote 54 Step 2 is benchmark creation where the expert-activists develop the methodology for benchmarking and may consult with other experts to verify its soundness. Rather than seeking wider stakeholder engagement the expert-activists seek to have other NGOs ‘buy-in’ to the benchmark and support its validity. This is a test of salience since the NGOs need to buy into not only the issue as politically and socially important but also have to agree with, or defer to, the expert-activists on questions of data gathering and significance, and especially methodology. That is Step 3. At Step 4 we note mature field opposition, where interests associated with the established institutions being challenged will confront the expert-activists on their expertise and fitness to judge and create a benchmark.Footnote 55 Given that mature field institutions have a lot of resources and established strategies to communicate with both the public and political authorities, opposition on the grounds of insufficient expertise will be treated seriously. The expert-activists will have to fight back to demonstrate that their benchmark and findings are legitimate and valid, leading to an interactive process. If there is enough momentum to carry on then, in Step 5, there is political engagement where political actors can also question the expertise within the benchmark. This is also a test of salience that is directed primarily at the NGOs who have bought into the benchmark rather than the expert-activists, since the NGOs are the established parties. If this test of salience fails then there can be a knock-on effect of NGOs reconsidering their own salience test, sending the expert-activists back to Step 1.
Revolutionary benchmarking cycle
A revolutionary benchmarking cycle can be found in the Financial Secrecy Index (FSI). The FSI contests both the CPI and alternative lists of ‘tax havens’. While the CPI focuses on public officials and politicians, the FSI highlights the issue of financial transparency as systemic. The expert-activists behind the FSI, the Tax Justice Network (TJN), initially sought cooperation with TI in 2008, but were unable to form an agreement on strategy. TJN representatives interpret the reluctance of TI to directly cooperate as due to American staff considering tax issues too difficult to communicate and not wishing to disturb the success of the CPI. A key member of TJN communicated to us that ‘TI had CPI running, they had this good selling point, they were all over the world with the index once a year … they depend on it for organizational survival.’Footnote 56 The OECD move strongly on tax issues with the blacklisting of 35 jurisdictions in 2000 for ‘harmful tax competition’.Footnote 57 The OECD furthered their efforts in 2002 with the promotion of Tax Information Exchange Agreements (TIEA) and their deployment in the wake of the global financial crisis has been viewed sceptically, with commentators questioning the mechanism’s substance.Footnote 58 TJN argue that the benchmarks provided by TI and the OECD were both ineffective.Footnote 59
For the world of NGOs, figuring out what to do about ‘tax havens’ is an ongoing thorny issue. While Oxfam identified ‘tax havens’ as a problem in the late 1970s there was no coherent movement on the ‘problem of offshore’ until the 2000s when activists organised through TJN. Established in 2003, TJN is a key actor on tax justice issues and strongly influential in defining campaign issues, targets, and policy content for the wider NGO community. TJN operates in an increasingly crowded policy space where many other NGOs and international organisations compete, including the OECD, the IMF, and private organisations such as the International Accounting Standards Board. This issue area is a mature field where public and private interests are well established. NGOs such as ActionAid have sought to mount a tax campaign on ‘Tax Power’, launched to increase focus on corporate tax abuse and encourage the increasing transparency of corporations and governments. Christian Aid has subsumed its tax efforts under the ‘IF campaign’ coalition targeting the G8. These well-resourced NGOs would like to tackle tax issues but have not developed revolutionary benchmarks. ActionAid have discussed the need for a benchmark on taxation, noting that ‘Benchmarking the responsibility of a tax policy’s content will inevitably be more controversial than benchmarking the management of that policy.’Footnote 60 As a consequence, ActionAid’s tax benchmarking is more of an investor guide to make them aware of risks associated with tax planning, a reformist agenda. In contrast to this, the FSI provides a revolutionary benchmark in seeking to reconfigure how the units in the system and the system as a whole are conceived.
The FSI places jurisdictions on an index based on the level of secrecy they provide and the scale of their activities in global financial markets.Footnote 61 It is built upon a qualitative component that evaluates a jurisdiction’s laws, regulations and tax treaties, and a quantitative component, the Global Scale Weight, which attaches a weighting to each jurisdiction on the basis of its share in global financial services exports. As a result the conventional imaginary of ‘tax havens’ is challenged. The 2013 FSI ranks the United States 6th, Germany 8th, and Japan 10th for financial secrecy, implicating some of the largest economies and most powerful nations in tax abuse and the facilitation of illicit financial flows. As a benchmark the FSI seeks to expose the hypocrisy of the G8 and place pressure on these core states to reform, rather than seeking to displace blame on the largely small island economies, those understood to populate the offshore world. The influential 1981 Gordon Report questioned the use of the term ‘tax haven’ early on, noting that the:
term ‘tax haven’ has been loosely defined to include any country having a low or zero rate of tax on all or certain categories of income, and offering a certain level of banking or commercial secrecy. Applied literally, however, this definition would sweep in many industrialized countries not generally considered tax havens, including the United States.Footnote 62
‘Tax haven’ has been a useful term precisely because it shifts blame away from the core of the OECD member states and prolongs a transnational normative position that Swiss private banking is acceptable while shell companies in the Caymans are not. Challenging this, the FSI is designed to draw in powerful states not usually associated with tax abuse, financial instability and illicit finance. Secrecy jurisdictions are defined as: ‘A jurisdiction which provides facilities that enable people or entities to escape or undermine the laws, rules and regulations of other jurisdictions elsewhere, using secrecy as a prime tool.’Footnote 63 The FSI points to hypocrisy in tax treatment within the OECD, something that experimental work involving the deception of tax avoidance and evasion service providers has also revealed.Footnote 64
For the FSI, the 15 ‘Key Financial Secrecy Indicators’ (KFSI) gauge a jurisdiction on the basis of the capacity to automatically exchange information and name the beneficial owners of companies, the extent to which country-by-country reports are required, whether provisions against money laundering are in place, whether a jurisdiction’s tax administration uses taxpayer identifiers for efficient administration, and if trusts with flee clauses are permitted. Equally weighted scores on the 15 KFSIs are combined to form a secrecy score, ranging for 0 to denote complete transparency and 100 to denote complete secrecy. In the 2013 FSI, Switzerland was ranked first with a secrecy score of 78 and a global scale weight of 4.916. Montserrat ranked last, at 82nd with a secrecy score of 74 and a global scale weight of 0.000. Notably, Nairu ranked 81st despite a higher secrecy score than Switzerland due to a global scale weight of 0. The dataset underlying the index is made available online and linked to data sources for verification.Footnote 65
The index measures secrecy rather than demarcating tax havens from virtuous jurisdictions or ‘offshore’ from ‘onshore’. TJN popularised the phrase ‘secrecy jurisdiction’ to more accurately capture the activities and institutions implicated in tax abuse and financial improbity. This linguistic move overcame the territorial implications of the term ‘offshore’. Any jurisdiction can be a tax haven for residents of another country.Footnote 66 Consequently, the FSI places jurisdictions on a secrecy spectrum, rather than nominating jurisdictions compliant or noncompliant. Dariusz Wójcik notes ‘tax havenry’ or ‘offshoreness’ is a matter of degree, not a binary variable.Footnote 67 In replacing the term ‘tax haven’, the FSI highlights that the activities of ‘secrecy jurisdictions’ are designed not so much as to attract mobile capital into a jurisdiction, but to provide a legal basis upon which to claim that an activity occurs nowhere or elsewhere for any interested fiscal authorities. Placing jurisdictions on a secrecy spectrum also broadens implicated practices beyond tax to include issues of financial stability and illicit finance.Footnote 68
Issue selection for expert-activist benchmarks is by the activists as opposed to gatekeepers within a NGO. Expert-activists must define an issue and then seek traction and uptake among governments, NGOs, regulators, and the media. This places considerable pressure on internal expertise and experience. The FSI was developed by a team within TJN with on the ground experience in accounting, corporate service provision, and law. While the reformist benchmarking cycle passes through stages of expert consultation and stakeholder engagement before benchmark creation, expert-activists select an issue and jump straight to benchmark creation. That expert engagement and stakeholder consultation is absent increases the benchmark’s vulnerability to scrutiny on methodological grounds. NGOs need to accept that the issue is salient and that the expertise behind the benchmark is sufficient before buying-in to it. Experts in mature fields will be unwilling to engage with benchmark creation where the benchmark challenges their position and revenue.
Only in the wake of the benchmark being launched does it come under concerted scrutiny. Mature field actors who are directly threatened by benchmark will question the expertise used in its creation. Geoff Cook, CEO of Jersey Finance Limited, responded to the 2011 FSI by contrasting it with appraisals by the IMF and OECD, stating that:
Jersey’s ranking in the Index is nonsensical. Jersey is one of the safest and best regulated international finance centres (IFC), as demonstrated by credible, independent assessments by internationally accepted organisations like the OECD and IMF.Footnote 69
The FSI is rejected by those in mature fields, who draw upon the weapons of stronger players in the system. With the rise of tax issues to the top of the international policy agenda, the 2013 FSI attracted considerable opposition. Caymans Finance commissioned an analysis by Aaron Smallwood, an economist from the University of Texas-Arlington, to dismantle the FSI. The critique targeted both the equal weighting of the KFSIs and the index weighting based on relative size of financial service exports. Smallwood argues that the 15 KFSIs do not contribute equally to levels of financial secrecy. For instance, while automatic information exchange is thought to contribute more to financial transparency than engagement in the tax treaty network the FSI weights both indicators equally. In calculating the index he notes that the:
United States, which has the sixth highest aggregate score for FSI, has the 56th highest score for S[ecrecy] … the jurisdiction with the worst secrecy score, Samoa, actually ranks 76th in terms of TJN’s total FSI, which is because of the small size of Samoa’s financial industry.Footnote 70
These interventions generate methodology battles that expert-activists are obliged to fight.Footnote 71 Opposition to revolutionary benchmarks also comes from international organisations that are heavily invested in seeing issues in a particular way.Footnote 72 On the FSI, the Head of Tax Policy and Administration at the OECD, Pascal Saint-Amans, stated in 2012:
I wish that their financial secrecy index would be a bit more robust. They have a secrecy index that built on a number of criteria that are disputable and based on information that is outdated … They came up with a financial secrecy level made up with a mathematical formula to get countries they want on the top of the list to be on top of the list. That's disingenuous, it’s not fair, it's not good and it sends the wrong message.Footnote 73
This attracted a direct rebuttal in the same journal from the Director of TJN, who argued that the OECD is unable to tackle financial secrecy issues.Footnote 74 Despite mature field opposition the FSI has gained political traction with private and public organisations.
The FSI has become a default tool for the broader NGO community now working on tax issues with ActionAid, Publish What You Pay, Christian Aid, and a host of NGOs all buying-in to the FSI in campaign materials. In 2013, the Centre for Global Development in the US began using the FSI as one component in the Commitment to Development Index, measuring the developmental value of the policies of 27 OECD members.Footnote 75 The Basel Anti-Money Laundering Index by the Basel Institute of Governance includes the FSI as one criterion in its assessment of money laundering risk, and has defended the FSI on its methodology.Footnote 76 The 2013 Bribery and Corruption Awareness Handbook for Tax Examiners and Tax Auditors mentions a high FSI score as an external environment indicator of a high risk country.
There has also been some interest from the private sector in applying the FSI. Two German sustainability rating agencies, imug and Oekom research, incorporated the FSI into their rating processes on bank bonds and country ratings.Footnote 77 The French bank Crédit Coopératif decided in 2012 not to maintain or establish affiliates, nor to invest or finance projects in non-European jurisdictions that received a secrecy score higher than 70, leading to the exclusion of 47 countries.Footnote 78 Interestingly, the FSI has also spurred an opposing benchmark, with the US-based Tax Foundation launching its International Tax Competitiveness Index in 2014. This benchmark ‘measures the degree to which the 34 OECD countries’ tax systems promote competitiveness through low tax burdens on business investment and neutrality through a well-structured tax code’.Footnote 79
Furthermore, the FSI has received some support from within the UN system. UNCTAD’s 2014 Trade and Development Report relies heavily on the work of TJN and explicitly recognised the FSI’s revolutionary content. In outlining the FSI methodology the UNCTAD report notes that with the FSI the ‘focus shifts, therefore, from governance issues within countries to the jurisdictions’ responsibility for offering financial secrecy at the global level’.Footnote 80 Despite mature field opposition and battles over expertise the FSI has passed necessary salience tests for NGO buy-in and there is will in the system to address these complicated tax issues, with some international and private sector organisations offering their support for this revolutionary benchmark.
Conclusion
Benchmarks can push forward reformist and revolutionary agendas, but must face tests of salience, will, and expertise if they are to create normative change. This article presents two ideal types of benchmarking cycles, reformist and revolutionary. The benchmarking cycles differ in steps actors must pass to continue, and especially on the drivers of benchmark creation and the opposing or supporting forces for tests of salience, will, and expertise.
Reformist benchmarking cycles seek adjustment within dominant transnational normative orders. They are pushed by NGOs that invest a great deal of time in carefully selecting issues they believe have salience and political and moral traction. As large and well-resourced NGOs are highly bureaucratised and professionalised organisations, they often lack the expertise to develop benchmarks themselves. Instead they outsource to and consult with experts, or they permit parts of their own organisation to become offshoots following a particular agenda. Once armed with expertise those pushing a reformist agenda also consider stakeholders before making their benchmarks, to fuse expertise with salience. These benchmarks are then taken to political authorities that, with an agenda that strengthens current normative dispositions, can use will to create transnational normative change. As discussed above, we provide two cases of reformist benchmarking cycles, through Transparency International’s Corruption Perceptions Index and the Gold Standard’s system for carbon credits. Both cases conform to our ideal type reformist benchmarking cycle and both seek to affirm current normative trends.
In contrast to reformist benchmarking cycles, expert-activists drive revolutionary benchmarking cycles. These expert-activists actively contest how we understand how the transnational normative order is composed. They do so by creating benchmarks that directly challenge how we understand how the system works as a whole, as well as how we should analyse the units in the system. Expertise is the key tool for these actors in creating a benchmark, and the obvious weak point for those who wish to oppose it. Expert-activists have to persuade other NGOs to buy-in to their framework to past the salience test. They also have to engage in methodological battles with actors in mature fields, who have strong normative, ideational, and material interests for defeating expert-activists in these contests and exposing them as frauds.Footnote 81 If expert-activists can get past mature field opposition then they can engage political authorities. This is also a difficult step given that these authorities are also strongly invested in the current normative order on the issue of concern. Opportunities for expert-activists to attempt revolutionary benchmarking are quite common in the NGO community, yet they commonly fail to pass the hurdles designated in our benchmarking cycle due to a lack of NGO buy-in, mature field opposition, and support from political authorities. Sometimes these attempts can garner support and have a political impact. We have provided Tax Justice Network’s Financial Secrecy Index as an example of a revolutionary benchmark. As discussed above, it fits with our ideal type, in benchmark creation, NGO buy-in, mature field opposition, and difficulties with political authorities.
The reformist and revolutionary benchmarking cycles include elements of reification, commensuration, focal points, and reactions, the criteria identified by the editors of this Special Issue as common to benchmarking in global governance. As argued, an approach is required that can distinguish what is at stake at points in different cycles, and how actors behave in different ways within the given social structures for benchmark creation and promotion. Benchmarks seek to actively reconfigure how we are governed and also how we seek to govern ourselves. In an era where physical violence between members of the OECD is low, it is through symbolic violence that advanced economies seek to dominate each other, as well as developing economies, as our discussion of ‘tax havens’ notes. We stress that benchmarks do not represent only the dominant normative order of the time but can contain conforming and contesting views.
Differentiating reformist and revolutionary benchmarks, and locating the tests for salience, will, and expertise, is important for the development of research on activist organisations. It is also important for the theoretical development of scholarship that seeks to understand how fields are composed and how conflict is generated and handled within them. Our work here contributes to Bourdieu-inspired scholarship on how actors are able to transform transnational economic governance,Footnote 82 especially where the ‘field’ is understood as an approach from which to ascertain position-taking within a social environment, rather than as a general theory.Footnote 83 Further research could usefully interrogate communication strategies, particularly media strategies to gain salience, which propel revolutionary benchmarks. While reformist benchmarks meet a ready audience, revolutionary benchmarks demand an audience to sit up and pay attention. Here we have provided heuristic tools to identify how different types of organisations must follow distinctive paths to create normative change, as well as how smaller groups can contest dominant forms of symbolic violence through their own expertise and activism.
