Business as Usual in Financial Markets? The creation of incommensurables as institutional maintenance work

This paper aims to contribute to the literature on ‘institutional maintenance work’. Focusing on the institutional disruption resulting from a regulatory project of market rationalization, it enriches the description and analysis of the specific institutional maintenance work performed by powerful actors who engage in resistance against what they perceive as a threat to their discretion. Built on an in-depth qualitative study, our case concerns an attempt to change the form of over-the-counter markets as part of a recent financial reform. The paper contributes to the expanding literature on the maintenance of institutions by suggesting, in particular, that the creation of incommensurables should be added to the list of strategies available to powerful incumbents seeking to resist institutional change. Bridging the gap between the literatures on institutional maintenance and commensuration, it also demonstrates that specific institutional changes can usefully be understood as changes in commensuration systems. This innovatively suggests the existence of degrees of commensuration and calls for a finer-grained understanding of the institutional work required to maintain institutions in a context where the degree of commensuration experienced by a field or a market threatens to increase under coercive pressure.


Introduction
Recently, the concept of institutional work proposed by Lawrence and Suddaby (2006) has been broadly embraced by organizational scholars pursuing an interest in institutional approaches. Extending the theoretical and empirical agenda beyond the creation of institutions to the issues of persistence and deinstitutionalization (Hwang & Colyvas, 2011), it proposes to re-examine the taken-for-granted reproduction of institutions envisioned by early institutional scholars. It suggests that efforts (Lawrence, Suddaby, & Leca, 2009) may be required to engage in the disruption of institutions, or conversely, to ensure their stability. Responding to the call that institutional maintenance should be considered as a fundamental question for institutional research (Lawrence, Leca, & Zilber, 2013), scholars have begun to provide insightful accounts of the purposeful work institutional actors engage in to protect or restore the integrity of institutions in various situations (e.g. Currie, Lockett, Finn, Martin, & Waring, 2012;Dacin, Munir, & Tracey, 2010;Micelotta & Washington, 2013;Zilber, 2009).
We propose to contribute to this as yet emerging stream of literature, by looking at the way powerful actors face changes aimed at coercively transforming institutional arrangements in the name of rationalization. Such situations are not rare and can be found, for example, in reforms of accounting standards (Carpenter & Feroz, 2001), public sector management reforms (Ezzamel, Robson, & Stapleton, 2012;Oakes, Townley, & Cooper, 1998), changes in the administration of healthcare services (Dent, 1995;Kitchener, 2002), changes in transfer pricing systems or in the definition of relevant indicators for the not-for-profit sector (O'Dwyer & Unerman, 2008). Institutional theory has long recognized the role of the state or delegates of the state as central agents of institutional change (Edelman, 1992;Lawrence & Suddaby, 2006;Russo, 2001) and carriers of structural devices of organizing (Hasselbladh & Kallinikos, 2000). Meeting the rational orientation of the contemporary social world, state reforms in the name of rationalization and modernization moreover combine normative elements -containing evaluative and obligatory dimensions -with coercive power through rule-setting and sanctioning (Scott, 2008), making the state a uniquely influential agent of institutional transformation. This might explain why a good deal of the literature related to state attempts to create or transform institutions has focused on either ex-post compliance or acquiescence with coercive isomorphism (e.g. Townley, 2002) or adaptation to the new rules (e.g. Currie et al., 2012). Less attention has been paid to situations where powerful incumbents put up fierce resistance to maintain institutions threatened by proposed changes to which they are not willing to adapt (for an exception see Micelotta & Washington, 2013).
Yet, resisting institutional change the representative of which is a highly legitimate (public) authority, acting in the name of the widely shared assumption that rationalization is a desirable goal, issues specific challenges. As will be shown, rather than the construction of a number of distinct strategies, it requires a capacity of elaborating a complex system of mutually reinforcing arguments, to resist, despite the double legitimacy of the initiator of change and of the cognitive framework s/he relies on. Previous institutional theory has not addressed situations where resistance to institutional change happens, notwithstanding the fact it can be directed against neither the proponent of the change nor the underlying assumptions. Understanding the work such resistance requires might therefore provide new theoretical insights into the cultural and social power processes needed to maintain less rational patterns in societies, repetition valuing rationalization as a highly desirable objective. Moreover, in a practical perspective, disregarding the potentially frontal resistance of powerful actors (Lounsbury & Rao, 2004) may prevent states or regulators from efficiently addressing the issues for which they are accountable to constituencies that might differ from those to which market or field participants are accountable. Our research question is therefore as follows: What are the means used by powerful defenders of the status quo who are willing neither to adapt to nor to compromise with coercive rationalization-driven institutional changes?
Empirically, we focus on the proposed changes to over-the-counter (OTC) financial markets 1 resulting from the regulatory reform currently in progress in Europe. The European Union (EU)'s intention is to transform these financial markets in the aftermath of the financial crisis into organized, transparent and 'more effective' markets. 2 Because those who defend the status quo have much to lose in the reform, they are putting up fierce opposition, providing us with interesting data through their response to the consultation organized by reformers. This offers us an opportunity to analyse the institutional disruption which ensued and how powerful incumbents become involved in institutional work of maintenance at an early stage of the upcoming change.
We find that, at that stage, one strategy for resisting actors is to construct and defend the idiosyncrasy of the initial institutional arrangement and to leverage on the inertia of existing practices on the markets before the reform by contending that these are in clients' best interests.
In developing our analysis, we make two main contributions. First, we contribute to the burgeoning literature on work devoted to the maintenance of institutions (Currie et al., 2012;Dacin et al., 2010;Lok & de Rond, 2013;Micelotta & Washington, 2013;Quinn Trank & Washington, 2009;Riaz, Buchanan, & Bapuji, 2011;Sminia, 2011;Zilber, 2009) by underlining the significance of 'resisting' as a type of institutional work whereby actors are unwilling to compromise with or adapt to institutional change. We suggest that the creation of incommensurables is one of the strategies available to those involved in this type of institutional work and describe the means by which incommensurability can be constructed in such situations. Second, we demonstrate that specific institutional changes can usefully be understood as changes in commensuration systems. This innovatively suggests the existence of degrees of commensuration in fields or markets and calls for a finer-grained understanding of the institutional work required to maintain institutions in a context where the degree of commensuration experienced by a field or a market threatens to increase under coercive pressure. Our paper suggests that some features of such a context entail specific maintenance endeavours for resisting actors, who find themselves having to support some widely accepted principles while rejecting the application of these very principles in the specific situation concerned. We interpret the creation of incommensurables as a way to resolve this tension.
The rest of the paper is divided into five parts. In the first, we focus on the theoretical foundation of our research, mainly the literature on institutional maintenance work and state rationalization projects. The second section describes the research context and the third our research method, a qualitative study over the period 2010-11, based mainly on analysis of secondary data. We then study the case of proposed changes to OTC financial markets and the rejection of reforms by elite financial institutions. In the fifth and final section, we discuss the main conclusions and contribution of our research.

Institutional Maintenance Work: Struggling Against Coercive Changes Driven By Rationalization Objectives
Institutions are usually defined as taken-for-granted and enduring sets of practices (e.g. Greenwood, Oliver, Sahlin-Andersson, & Suddaby, 2008). In this view, many scholars consider processes of institutional stability to happen through repetitive social behaviour and processes of self-reproduction (Jepperson, 1991). Since not all institutions are self-reproducing (Hwang & Colyvas, 2011), the introduction of the notion of institutional work has however served to highlight the 'purposive action of individuals and organizations' so as to analyse the processes of 'supporting, repairing and recreating which ensure compliance' (Lawrence & Suddaby, 2006, p. 230), which are central to actively maintaining existing institutions. Lawrence and Suddaby (2006, pp. 230-234) highlight different types of work that are considered as central in understanding the acts aimed at preserving existing institutional arrangements. The creation of various rules which support an institution (enabling work), the enforcement, and monitoring in order to ensure that participants comply with existing rules (policing), or the creation of coercive barriers to participation (deterring), constitute the first types of social processes related to the coercive pillar of institutions. The second type, more cognitively oriented, includes tactics such as valorizing and demonizing existing beliefs, mythologizing by sustaining myths regarding the history of an institution, or embedding and routinizing, i.e. the process of infusing the foundations of an institution into the day-to-day routines and organizational practices of its participants.
While the issue of exactly how institutions are maintained has long been understudied in the empirical literature (Lawrence et al., 2009;Lawrence et al., 2013), some recent research has emphasized the idea that even deeply institutionalized practices and social relations require purposeful work to keep them running (Currie et al., 2012;Dacin et al., 2010;Lok & de Rond, 2013;Micelotta & Washington, 2013;Quinn Trank & Washington, 2009;Riaz et al., 2011;Sminia, 2011;Zilber, 2009). In relatively stable contexts and in the absence of any disruption, this process of maintenance takes the form of preserving and reinforcing existing institutional arrangements that favour incumbents, through everyday routines and practices (e.g. Dacin et al., 2010;Zilber, 2009).
However, in contexts involving upheavals, such as those related to new regulation and similar coercive disruptions driven by rationalization objectives, the efforts of actors in maintaining institutions run up against the state (or delegate of the state) as a powerful agent of institutional change. Such state reforms are justified by the willingness to introduce a more reasonable and rational social order (Townley, Cooper, & Oakes, 2003). They generally include standardized rules of conduct, documentation for structuring tasks and the definition of standards in impersonal and decontextualized terms (Hasselbladh & Kallinikos, 2000;Porter, 1996;Townley, 2002). They often increase transparency and surveillance (Samiolo, 2012). Through their unique capacity to establish rules, construct rewards and sanctions, confer property rights or impose standardized definitions (Lawrence & Suddaby, 2006) and thanks to their coercive right, the state and its delegate enjoy consequential power in creating or disrupting the institutions that enable and constrain organizational behaviour (e.g. Edelman, 1992;Holm, 1995;Leblebici, Salancik, Copay, & King, 1991;Russo, 2001). This is probably why much attention has been paid to the efforts of actors to comply or acquiesce with coercive change. Townley et al. (2003) highlight, for example, how the introduction of performance measures in the Province of Alberta was promoted by the government as an attempt to develop greater transparency, a process that was accepted or presupposed as valid by most participants. Along the same lines, Kitchener (2002) observes uncritical adoption of managerial innovations based on the rationalizing logic of mergers in academic health centres.
Beyond compliance or acquiescence with coercive isomorphism (Townley, 2002), and in spite of state intervention to promote changes, the difficulty in introducing global reforms has more recently attracted the attention of scholars (e.g. Taupin, 2013). In some circumstances, the state, as the actor of the institutional change efforts, faces opposition from the promoters of the status quo, who become involved in a complex and labour-intensive social process of confrontation. This process can encompass more or less compromise and adaptation (Currie et al., 2012). It can also involve fierce resistance aimed at preventing the change from happening or even reversing change and re-establishing previous institutional arrangements (Micelotta & Washington, 2013). Situations where powerful defenders of existing institutions fiercely resist any attempt to implement reforms driven by rationalization objectives have so far received insufficient attention. We believe this is detrimental, as the analysis of the resistance undertaken by incumbents against a coercive change which they perceive as a threat to their interests might provide interesting insights. It could enhance our understanding of the kind of institutional work which is required to maintain institutional arrangements favouring local knowledge, discretion or personal relationships. It could help us understand the kind of social and cultural power processes that enable less rational patterns to survive in societies which see rationalization as a highly desirable aim. Studying how institutions can be preserved, despite a coercive disruption driven by rationalization objectives, may highlight new forms of institutional maintenance work beyond compromise, negotiation or repair work. We therefore investigate the type of work involved in fiercely resisting a reform project that threatens to produce a disruptive coercive change, so that no change can even occur and the existing stability remains. To address this issue, we study the EU's Markets in Financial Instruments Directive (MiFID), an effort to reform OTC financial markets in Europe and more specifically its attempt to rationalize and introduce more transparency. We focus on resistance efforts by the most powerful actors in these markets and we highlight the creation of incommensurables as one strategy of resistance aimed at defending the specificity of the initial institutional arrangement.

The proposed regulatory change
The setting for our study is OTC financial markets, examined in the light of proposed regulatory change. OTC markets cover all transactions in financial instruments that occur off-exchange and involve direct agreement between two parties -generally banks -that do not usually make the price of the transaction public. OTC trading allows for considerable freedom in the design of contracts and is of special interest to parties looking for customized instruments to hedge or invest. OTC markets are highly concentrated and are dominated by large investment banks. 3 The growing dependence of the whole financial sphere on OTC markets went unnoticed for almost 30 years until the subprime crash suddenly turned the spotlight on them, revealing the quiet dominance they had obtained in modern finance (Huault & Rainelli, 2013). Under the pressure of negative public opinion in the aftermath of the financial crisis, politicians, regulators and public authorities had to take determined steps towards regulating these markets. In a collective move, regulators on both sides of the Atlantic have adopted an approach favouring the displacement of as many OTC transactions as possible to electronic organized markets. The September 2009 G20 summit explicitly stated: 'All standardised OTC derivative contracts should be traded on exchanges or electronic trading platforms, where appropriate, [our italics] and cleared through central counterparties by end-2012 at the latest.' Seen as a means 'to tackle less regulated or more opaque parts of the financial system and to improve the organisation, transparency and oversight of all market segments', 4 the philosophy of the new regulation will be implemented through the Dodd Frank Act in the US. In Europe, the project has been developed mostly through MiFID 2. 5 The aim is to ensure that 'where appropriate, trading in standardised OTC derivatives moves to exchanges or electronic platforms', which, 'at a minimum, would imply that trading on exchanges and electronic platforms becomes the norm when the market in a given derivative is suitably developed'. 6 This reform is seen as a way to construct 'more robust and efficient market structures'. 7

Financial reform as a change in market form
The transformation proposed by the regulators is revolutionary. It amounts to converting bilateral, concentrated and non-transparent financial markets into a modern electronic form of Walrasian market where transactions are organized on a multilateral and transparent basis -numerous potential sellers facing numerous potential buyers in an auction where prices are public information. Before reform and over almost three decades, regulators had no real-time information whatsoever on transaction prices, and a buyer had to ask a series of brokers about potential prices, unaware of the context in which his/her request would be treated. This market form, according to regulators promoting the reform, is suspected of potentially having favoured 'competitive distortions and reduce(d) the overall efficiency of the price discovery process'. 8 In post-reform markets, prices will be common knowledge, readily available to regulators as well as potential buyers, allowing free competition between buyers and sellers to play its part fully in price-setting. The intention, explicitly formulated as being to promote 'effective' price formation processes, builds implicitly on the financial theory of market efficiency, also known as the efficient market hypothesis (EMH). Since Bachelier's early work in 1900, market efficiency has provided the paradigmatic framework for theories about financial models and innovation. EMH has been taught universally for the last 50 years and provides the theoretical basis for most financial industry practices. Transparency enjoys the positive status of a virtue in this paradigm, as it is the prerequisite for market efficiency. Transparent markets provide free and equally accessible information, including information about prices, to all investors, allowing them to trade in fair conditions. This is how markets attain informational efficiency, which transforms into general efficiency. Regulators' analysis of the post-crisis situation adopts this perspective as they envision OTC markets as having drifted away from the well-functioning, efficient market model and needing to be transformed into more effective markets. The scale of the proposed change however is huge due to the size of OTC markets. The total notional amounts outstanding of OTC derivatives stood at $632 trillion at the end of December 2012. 9 Experts from the TABB Group 10 suggest that 'as much as $40 billion of annual revenues (excluding credit derivatives) [are] at stake in global OTC derivatives for the 20 largest broker dealers', mostly large investment banks. They point out that the proposed reform is likely to entail compressed margins and reduced profitability for major players. Unsurprisingly, the industry's reaction has been to resist these regulatory projects.
In sum, in proposing a new kind of price mechanism inspired by the EMH, in order to construct more efficient European financial markets, the regulators intend to make a radical change in the form of OTC financial markets. As the resistance to these reforms indicates, changing the price discovery mechanism involves far more than merely altering the technical functioning of the market. Previously, legitimate prices on OTC markets resulted from private bilateral negotiations involving limited comparison. Prices were proposed by a handful of major dealers who were known for their competencies in designing complex financial deals. The new price mechanism now proposed will constrain the discretion of large brokers in pricing deals and promote increased impersonality in transactions. It will allow more surveillance by regulators, who seem willing to assume a disciplinary role in the aftermath of the recent financial crisis. This regulatory initiative inspired by a rationalizing logic based on the EMH can thus be interpreted as an attempt to change the core of the institutional arrangement governing OTC markets. 11

Methods
Our research is based on a qualitative case study, which is appropriate for many reasons (Greenwood & Suddaby, 2006, p. 31). First, as the theory of institutional maintenance is at an early stage in its development (Lawrence et al., 2013;Micelotta & Washington, 2013, p. 1144, when it is difficult to isolate key variables (Yin, 2008), a case study is useful so as to analyse a complex social setting in which causal dynamics are not immediately apparent. Second, the time period covered includes a regulatory jolt driven by motives of rationalization, in other words, an attempt to change the way prices are established on a financial market. We focus on the early stage of the institutional disruption, which provides an opportunity to gain insight into the way powerful defenders of the status quo engage explicitly in institutional work in order to fiercely resist institutional change and to maintain the previous institutional arrangement. Third, the study of OTC financial markets is particularly appropriate because, following the financial crisis and various subsequent public initiatives, banks, experts, regulators and the media have had little hesitation in expressing their views. There is therefore a large amount of discursive material containing their opinions and arguments about the advantages and limitations of likely future developments.

Data sources
We used several discursive and textual sources of empirical evidence, which can be divided into three main categories: archive material, press articles and MiFID consultations.
The first step consisted in collecting the available documentary material from a wide array of sources such as official documents -from the Bank of International Settlements or the International Organization of Securities Commissions -and industry publications -from the International Swaps and Derivatives Association or Tabbgroup. We also drew on the Factiva database, which provides business news collected from 14,000 sources. In total, we analysed 150 articles related to the reform for the period 2008-2011, beginning with research on the keywords 'OTC markets' and 'regulation'. This first step was to improve our understanding of the context of the OTC market, retrace key events and provide initial textual accounts of debates and discussions.
The second step was to analyse contributions to the public consultation organized by the European Commission (8 December 2010 to 2 February 2011) (see the summary in Table 1 and Appendices 2 and 4). The purpose of this consultation was 'to consult market participants, regulators and other stakeholders on possible changes to the regulatory framework established by MiFID in the field of investment services and activities as well as markets in financial instruments'. It invited responses to 'provide important guidance for preparing a formal Commission proposal'.
Two interviews with an International Swap and Derivatives Association (ISDA) member and an MEP's assistant also helped us understand their respective views of progress in the MiFID legislative process. Table 2 presents details on each of the data sources and their use in our data analysis.

Data analysis
Our analysis was conducted in several stages. We analysed the data inductively, following the procedures established for naturalistic inquiry and grounded-theory building (Glaser & Strauss, 1967;Golden-Biddle & Locke, 2001). Starting with the press articles, we first arranged the data in order to identify the main actors and events in OTC markets between 2008 and 2011. We then set out to capture the discourses of the various actors (Greenwood & Suddaby, 2006;Huault & Rainelli, 2009, 2011 involved in OTC markets and their evolution. This first step was the basis for delineating themes through the analysis of key events (Isabella, 1990) (see Appendix 3). We then focused on an initial set of texts from the MiFID consultation, subsequent ISDA comments on MiFID and articles published on TabbFORUM. We coded each text independently for words, phrases, terms or descriptions offered by the different texts. These elements constituted first-order codes. We reread each text several times, each time marking phrases and passages that revealed similarities and differences among sources (Nag, Corley, & Gioia, 2007, p. 828). After working through multiple texts, we collapsed these codes into first-order categories. For example, comments on the 'impossibility of benchmarking, differences in price for identical products, establishment of a price for a solution…' could be grouped into a category labelled 'A price is not a Contextualize the analysis in terms of the evolution of the financial industry price'. Examples of first-order categories include 'EMH does not apply', 'A market for specialists' and 'Incentives to trade'. The next step of our analysis involved looking for links between first-order categories so that we could collapse these into theoretically distinct clusters, or second-order themes (Dacin et al., 2010(Dacin et al., , p. 1401. For example, categories where sources alluded to 'EMH does not apply' or 'Incentives to trade' were collapsed into a theme labelled 'Contesting the universality of the EMH theory'. Second-order themes also included 'Describing the specificity of the market structure', 'Invoking the Darwinian argument' and 'Disputing the feasibility of the practices'. The final step involved organizing the second-order themes into the overarching dimensions that underpinned our theory. As in Micelotta andWashington (2013, p. 1148), our analysis was driven by the following questions: What type of work is being done? What is the purpose of the work being done? Two dimensions emerged. The first was 'Invoking market nature' and included the second-order themes 'Market structure specificity' and 'Contesting the universality of the EMH theory'; the second focused on 'Levering on the inertia of practices' and included the following second-order themes: 'Invoking a Darwinian argument' and 'Disputing the feasibility of practices'.
To ensure that we could be confident in assigning codes to categories, we each assessed the other's coding and reached agreement on categories. We took a number of steps to ensure the trustworthiness of our interpretations, including distinguishing explicitly between first-and secondorder data in frameworks and reports, and conducting 'member checks' with key informants to be sure that our interpretive scheme made sense to the people involved (Nag et al., 2007, p. 829). We also conducted two additional interviews with an MEP's assistant and a member of ISDA. Table 3 illustrates our final data structure, showing the categories from which we developed our findings and the relationships between them.

Findings
When we first began analysing the contributions to the December 2010 MiFID consultation, focusing on market structure and pre-and post-trade transparency issues, we were struck by the relative convergence of the answers provided by various members of the industry. There was no apparent dissent among the actors; banks, bank associations, hedge funds, financial services providers or corporates all seemed to subscribe to general criticism of the proposed regulation, describing it as useless, ill-suited, difficult to enforce and potentially detrimental to the functioning of financial markets. 15 The more developed and articulate answers, however, came from banks and bankers' associations, which is no surprise, given their concentration in OTC markets and the vested interest at stake. 16 We concentrated our analysis on the responses of these powerful actors and observed that they anticipated the likely outcomes of the proposed reform to be lack of innovation, drying-up of liquidity in a period of market stress, increased costs for market participants, and risk mismatches or increased exposure of retail investors to market risks. Further analysis soon revealed one major feature of interest: the issue of price mechanism seemed to play an important role in the debate.
In the next section, we describe how we used the resistance of incumbent banks to change in the price mechanism to reveal in greater detail the many dimensions they mobilized. Overall, we find that they contested the regulation attempt and rationalization project along two main lines: invoking market nature and leveraging on the inertia of practices, two strongly interconnected strategies.

Invoking market 'nature'
Our analysis of the answers to the MiFID consultation suggests that market actors often refer to the specificities of OTC markets, i.e. market 'nature', in terms of the market structure specificity and how it should work.
Describing the specificity of market structure. Incumbent banks describe OTC markets as a specific species of market, the nature of which is insufficiently understood and not taken into account by the regulators. They focus on price, which lies at the core of the implementation of pre-trade price transparency, which in turn is obviously pivotal to reform. One of the strongest attacks on the spirit of the reform is the ISDA's defence that a price on OTC trading markets is not the same as a price on organized markets. Contesting the term 'quote' used in the proposed legislation, the ISDA suggests that the price of OTC derivatives should be seen as a price for a 'solution' that can 'take weeks, if not months to finalise' and is therefore impossible to make public immediately. There will be 'by definition no benchmark price' for highly bespoke products, and apparently identical products will differ in price according to the 'perceived creditworthiness of the counterparty'. The ISDA and major banks contend that a price on OTC markets 'will always reflect the situation at the time it is made and therefore will not necessarily be comparable'. This amounts to rejecting the very premise of the proposed price mechanism. They insist on retaining a personal and idiosyncratic dimension in the transaction. This runs contrary to the impersonality of Walrasian price mechanisms, which picture atomistic sellers and buyers acting as price takers. Maintaining that prices are not comparable represents clear resistance to the spirit of the reform, which presupposes that idiosyncratic values can be expressed in standardized ways: The Commission should take into account the fact that the non-equity markets are qualitatively and structurally different markets -the equity markets should not be used as the yardstick against which other markets should be judged. It is not appropriate in our view to 'transpose' exactly the equity model to OTC derivatives markets. (HSBC-MiFID consultation) Another dimension of actors' representation is liquidity. The products exchanged are depicted as 'quite illiquid', because of the nature of the product or the transaction: OTC markets are less liquid than the market for shares. (Citi-MiFID consultation) The key characteristic of the OTC derivatives markets is that a low number of large trades are executed between market professionals. (Goldman Sachs-MiFID consultation) Opponents argue as follows: because OTC markets are not liquid markets, making price transparency mandatory will be a disincentive to liquidity providers and ultimately impair the functioning of the market. This argument deserves further analysis, but first we should note that the notion that some markets are illiquid by nature does not fit well with the EMH, which implicitly posits that when there is sufficient transparency on financial markets, liquidity should ensue 'naturally'. The idea that there are 'naturally' illiquid financial markets is a breach of the EMH that paves the way for the unorthodox view that, on such markets, transparency will harm the already fragile liquidity and impair market efficiency.
Finally, the notion that OTC markets have a specific structure is partly developed in the consultation under the label 'markets for specialists'. The main counter-argument here is not that the reform could be detrimental but that it is unnecessary, because market players are professionals or 'specialists' rather than retail investors. This line of justification implies that, while the need to protect retail investors is critical to pre-trade transparency requirements on equity markets for example, it is superfluous for 'professional' or 'wholesale' markets. According to some bankers' associations, making OTC markets more transparent might even make it more difficult to limit their access to 'all sorts of investors', who might (unwisely) 'feel as if it would be good for them to invest in derivatives instruments because they are traded on exchanges'. OTC markets should be kept as 'specialists' markets', and no regulatory requirement about transparency will be necessary because 'market participants', who are 'principally institutional and professional in nature', can access pre-trade transparency through multiple routes. They already have all the information and data they need and do not need the kind of protection given to the layman or retail investor. We interpret this argument as an attempt to protect elite financial institutions from the attempt to replace their discretion with the mechanical objectivity of a Walrasian pricing system: A clear distinction should be made between primary and secondary markets, retail and wholesale markets and the different client categories as to applicability and, where applicable, to the type and degree of regime that should apply. (HSBC-MiFID consultation) However, the suggestion that the very structure of OTC markets might be an issue is more revolutionary than it might appear. At no point does financial theory envisage the possibility of several forms of financial markets existing simultaneously. The notion of professional or wholesale, as opposed to retail, financial markets does not feature in the classical corpus of financial theory, although professionals, of course, experience the multiplicity of market forms in their practice. So defending what large brokers experience as a practical reality against the objective set by regulators implies developing arguments that maintain that EMH does not apply to OTC markets because of their specific nature.
Contesting the universality of the EMH theory. While the systematic resistance of major investment banks to any change in the status quo is not surprising, a close analysis of responses to the MiFID 2 consultation reveals an unexpected argument additional to those already mentioned. In contradiction of the traditional EMH, in which financial markets ideally and unequivocally belong to the typical Walrasian auction model, opponents of the new price mechanism develop a particular theory of the market and how it should function. First, they deny the virtues of transparency by highlighting how it plays against efficiency. Second, they query the motives and incentives for trading on specific OTC markets. One commentator, while acknowledging the traditional EMH vision of the benefits of transparency, challenges the universality of the theory to suggest that there are products and markets where it does not apply: In general terms, transparency in markets can help to build confidence, by ensuring that participants have access to information. However, there are products and markets which are so illiquid that revealing trade information could actually be detrimental to buyers and sellers. We have to balance the benefits of transparency versus the potential downsides. (Citi-MiFID consultation) Many respondents to the consultation develop other unorthodox theories, building on the same idea. Mandatory transparency imposed on OTC markets is likely to entail a reduction of liquidity, which in turn will impair the 'sought-after price transparency'. This runs contrary to the EMH notion that transparency, especially pre-trade price transparency, is a condition for market liquidity, which in turn conditions market efficiency.
In this way, opponents of regulation drift further and further away from the traditional theory of market efficiency and end up maintaining that the absence of pre-trade transparency, not to say opacity in OTC markets, actually provides an incentive to trade on them. Market actors repeatedly express concern that mandatory transparency could have a detrimental impact on liquidity and the availability of the OTC products to investors, as 'banks would be reticent to trade them'. There could be 'a disincentive for liquidity providers to publish prices'. This amounts to saying that a certain degree of opacity is necessary on OTC markets if some participants, such as 'liquidity providers', are to have incentives to trade. The clients' interests are also invoked to emphasize the specific nature of OTC markets: Mandating trading of OTC products on regulated markets would cause a decrease in liquidity as end-users prefer to trade OTC. Too much transparency will create cost, noise, and, as discussed, could have a negative impact on market liquidity, in particular for large OTC trades. (Deutsche Bank-MiFID consultation) Transparency requirements can result in decreases in order/transaction size and increased trade frequency. These can be signs of an inefficient market, as they can be the result of the unwillingness of market participants to perform effective risk transfer functions. (BBA-MiFID consultation) On such markets, transparency could prevent clients from keeping their market position secret and possibly 'breach client confidentiality'. Only a certain degree of opacity seems to provide participants with incentives to trade, thus producing some liquidity (although apparently not much, according to the argument above) and ensuring a kind of market efficiency.
In sum, our first finding is that in their attempts to resist the market form proposed by regulatory bodies, markets actors contend that OTC markets must be understood differently from more traditional financial markets. They have their own irreducible 'nature', with a specific market structure: an individual price discovery mechanism, liquidity that is distinct from liquidity on standard markets, and a wholesale organization that naturally fits their nature. Overall, the strategy of large international banks can be interpreted as the defence of the idiosyncrasy of OTC markets and their incommensurability with other markets.

Levering on the inertia of practices
In this section, we elaborate on actors' strategy of levering on the inertia of practices in their struggle against the evolution of the market form. First, they promote the defence of the status quo in the name of an evolutionary argument. Second, they discuss at length the feasibility of the practices involved in the reform.
Invoking a Darwinian argument. The issue of the market form is developed in responses to the consultation that resort to a kind of Darwinian argument, according to which the current market form has emerged naturally and survived because it best fits the interest of 'clients' or 'customers': There have not been, to the best of our knowledge, any serious academic studies demonstrating that the transposition of the equity model to OTC derivative products will be economically more efficient and adapted to clients' needs than the current trading model. (BNP-Paribas-MiFID consultation) Regulation is rejected as detrimental to the harmonious development of means of meeting investors' demands: transforming OTC markets into organized markets will 'reduce consumer choice'.
According to this line of argument, clients badly want the confidentiality provided by bilateral trading or customizability. The reform is criticized for not respecting the 'natural' emergence of the kind of market that fits these demands: OTC markets provide for bespoke risk management. Firms want to manage their risks correctly, with minimal slippage. As long as regulators are provided with the transparency they need … and removal of systemic risk (through clearing between the major players in the markets) then there would appear to be no reason to change the OTC nature of the product, or the trading venues where deals are executed. (Deutsche Bank-MiFID consultation) Disputing the feasibility of the practices involved in the reform. This involves challenging the criteria defining which financial instruments will be eligible for moving from trading OTC to electronic venues. The principle of the MiFID 2 proposal is that, once it is adopted, all 'eligible' OTC derivatives should move to organized platforms and be cleared through a clearing-house. This should take place, 'where appropriate', 'when a market in a market derivative is suitably developed' and it should concern all 'standardised OTC derivatives'. As all these expressions require further definition, the major industry players devote significant amounts of space in their consultation responses to contesting the definitions and categories proposed by the regulators. Most of their effort goes into presenting the practices needed to implement the reform as not operational, impossible to apply and unachievable. For example, the ISDA discusses the difficulty of defining which OTC products would be 'suitable' or 'eligible' for specific items of regulation. It insists that suitability for trading on an electronic venue should be defined independently from the suitability for central clearing. This supports the theme that it will be very difficult, if not impossible, to implement the new practice of technically defining which financial products will remain traded on pure OTC markets and which will move to a platform. The regulator proposes that standardization is a suitable criterion for deciding whether a product should move from OTC to electronic trading. Opponents of the reform insist that standardization is 'not always appropriate' and in practice very difficult to define. Liquidity is not considered any better a criterion, because it 'would be difficult to administer in practice'. The ISDA repeatedly warns regulators that including a liquidity threshold in the assessment of whether a product should be traded on a venue will be fraught with difficulties. It insists the threshold should be: 'set at a realistic level that differentiates between products, capable of being calculated and predicted, subject to periodic review and able to accommodate temporary changes in the market'. Other actors support the ISDA's position: 'We envisage that defining what is 'sufficiently liquid' could be problematic' (Deutsche Bank-MiFID consultation). It appears that the consensus among opponents to the reform is that in practice, the computation of thresholds will be extremely intricate, if not unfeasible.
The argument that complexity is an equally irrelevant criterion is emphasized several times in responses from the ISDA, BBA and other investment banks: 'A false link is sometimes made between product complexity and product risk, which leads to the illusion that complex instruments are automatically high-risk instruments.' The same example, described in many banks' responses, is intended to demonstrate how product complexity in OTC derivatives markets often results from products being designed to reduce risks for the customer. This means that it is not possible to identify in advance products that warrant regulatory treatment because of their supposed complexity.
Overall, these arguments set out to undermine the feasibility of the practices required to implement the proposed legislation. Opponents attempt to convey that the market form proposed by regulators does not fit OTC markets, depicting the proposed reform items as irrelevant, nonoperational or, more generally, impossible to implement on OTC markets, and by insisting on differentiating between types of market and product. If standardization, liquidity or complexity cannot be adequately measured on OTC markets, if they do not exist in OTC markets in ways that are comparable to organized markets, then the transformation intended by the regulator is not only potentially detrimental -it also becomes unachievable. The struggle over the change in the pricing mechanism that we observe in responses to the consultation thus goes well beyond the technical debate it appears to be. When the ISDA suggests redrafting article 7 of the MiFID, 1 it tellingly proposes 'removing the references to "depth of trading interests" and "on a continuous basis" … and inserting the phrase "where appropriate".' This amounts to attempting to undermine the categories and thus the structure of the reform itself. Asserting that 'this would enable an OTF operator to determine the appropriate pre-trade information to be made available to clients, based on the needs of those clients', the ISDA reveals the political struggle at the heart of the debate over the new price system. At stake is nothing less than the preservation of the discretion of elite financial institutions and by extension the entire possibility of reform and regulation.

Discussion
In this paper, we studied the change in the market form and pricing mechanism involved in a regulation project concerning financial OTC markets in Europe. We focused on the resistance that incumbent market actors engage in and observed that it takes the form of resistance to a change that will reduce the discretion of elite financial institutions in structuring and pricing financial deals. This specific setting provided us with an opportunity to document the type of early stage institutional work required from powerful defenders of the status quo facing coercive and rationality-driven institutional disruption, which they choose to resist fiercely rather than seek a compromise.
Our case study allows us to identify two channels of resistance. First, large international banks try to oppose the reform by constructing and defending the idiosyncrasy of the initial institutional arrangement. To convince the regulator, institutions opposing the regulation define what they perceive to be the market nature and contest the universality of the usual financial theory of reference as not valid on the markets under attack. Second, market actors lever on the inertia of existing practices on the markets before the reform by contending that these practices result from a demand from end-users. In addition, their resistance to the changes proposed in the reform mostly involves demonstrating that the proposed practices are not feasible on OTC markets, cannot be implemented and are non-operational.
Overall, these levers of action are strongly interconnected and seem to build on each other. Describing the specificity of the market structure is the foundation for contesting the universality of the EMH theory. Invoking a Darwinian argument supports the claim of a specific type of market, because it provides reasons for its specific nature. Disputing the feasibility of practices is finally a direct consequence of the inapplicability of the EMH. By and large, the actors we observe can thus be envisioned as constructing a complex and elaborate system of arguments that mutually reinforce and support each other. 18 While previous institutional theory has documented various types of maintenance work (including valorizing and demonizing, mythologizing or embedding and routinizing), which bear some links with the efforts we document, we believe it provides no clear understanding of the systematic character of the strategy we highlight. To understand the efforts of actors resisting a change, which proponent and underlying assumptions benefit from high legitimacy, we therefore propose to interpret the core of the institutional work of maintenance they perform, by borrowing a notion from the literature on commensuration (Espeland & Stevens, 1998), that of the creation of incommensurables. Incommensurables in essence are things that are defined as so unique that they cannot be expressed in terms of a standardized measure. Espeland (1998) provides as an archetypal illustration the refusal of the Native American Yavapai people to sell their land at any price to government authorities who wanted to build a dam; as the Yavapai perceived it, they were being asked to sell something priceless, namely their own identity. We argue that the market actors' claim -that OTC products, without being priceless, cannot be valorized through the mechanical Walrasian encounter of supply and demand and require a specific market form instead -is not so different. We thus propose that suggesting the irrelevance of existing theories, attacking the merits of taken-for-granted institutions like the EMH or the definition of a market price, or linking unorthodox views to broadly shared theories can be interpreted as various means by which powerful actors involved in resisting institutional change attempt to construct incommensurables. The tentative 'identity building' (our markets versus other market forms, i.e. us versus them) we observe is therefore not unlike contingent valuation targets resisting being framed as 'quasi-consumers' and instead demanding to be treated as 'citizens' (Lohmann, 2009), or Venice rejecting a scheme for flood protection involving impersonal, 'hypothetical' modelling in the name of its perceived identity (Samiolo, 2012). Such an approach is mostly about defending the idiosyncrasy of a threatened institutional arrangement. As such, we believe it should be of specific interest to scholars studying institutional work.
Interestingly, building incommensurables along the lines we have documented and in the type of context we have described must be achieved in a situation where the moral dimension is almost entirely absent. 19 This demonstrates that it differs from the valorizing/ demonizing work depicted by Lawrence and Suddaby (2006) (see also Angus, 1993) as work in which actors maintain institutionalized beliefs by purposefully assessing the moral status of participants in the field.
Moreover, although the type of attempt to create incommensurables by large investment banks involves many of the forms of institutional work listed by Lawrence and Suddaby (2006), including defining, identity-building or theorizing, it bears specificities that shed light on less familiar aspects of the institutional work required to maintain institutions. First, it is both favoured and constrained by a pre-existing situation, as large banks draw on the networks they have already built up over the years in the field of OTC markets (Huault & Rainelli, 2009;Morgan, 2008). While networks are obviously useful for almost all forms of institutional work (e.g. Lawrence, Hardy, & Phillips, 2002;Orssatto, den Hond, & Clegg, 2002), banks succeed in actively mobilizing existing ties in the financial industry to resist the proposed regulation, to the detriment of the promoters of the reform.
Second, and more importantly, it appears as the answer to an original and particularly intricate question regarding institutional transformation. Meeting the rational orientation of the contemporary social world, reforms by the state (or delegate of the state) in the name of rationalization and modernization combine normative with coercive power, making the maintenance of the threatened institutions especially challenging even for powerful defenders of the status quo. The creation of incommensurables might therefore be one of the few strategies available to them.
Originally, the creation of incommensurables is a way to resist attempts at commensuration, usually defined as a specific form of standardization whereby different qualities are measured using a single quantitative common metric that transforms differences into quantity (Espeland & Stevens, 1998). However, prices as a commensuration system on financial markets, including OTC markets, are hardly a novelty. What metric are investment banks resisting when they contend that a price on OTC markets is not the same as the price on another financial market? Our interpretation is that what they are actually trying to protect from the reformers is an original commensuration system that relied on the high degree of discretion enjoyed by elite financial institutions, which allowed idiosyncrasies, and did not achieve the degree of impersonality we take for granted, especially in price systems governing financial markets. We know from the literature that commensuration is most often seen as a way to reform institutional systems based on personal relations and local knowledge (Samiolo, 2012) and to introduce rationality or what Porter (1996) calls 'mechanical objectivity'. It replaces trust in individuals with trust in numbers (Porter, 1996), often allows increased surveillance and is sometimes perceived as threatening a cherished identity (Espeland, 1998;Lohmann, 2009). We also know from previous studies that for those who benefit from an existing system of authority depending on expert judgement, character or informal knowledge (Espeland & Stevens, 1998), as well as for those who value ethical systems that depend on personal relationships (Desrosières, 1998), there is no shortage of reasons for resisting commensuration. As our findings show, although not akin to a mere commensuration attempt where a single metric would be imposed all at once to quantify qualities in a given field, the institutional change we study can be interpreted as threatening to increase the degree of commensuration in the market, which defenders of the status quo interpret as endangering their discretion.
Adopting this perspective, our work on the reform of financial markets provides insights into the institutional work required to maintain institutions disrupted by a coercive rationalization project aimed at changing or increasing the degree of commensuration in a given field or market. As rationality and objectivity benefit from de facto normative legitimacy in modern societies, the advocacy by promoters of change is particularly difficult to counter. In the case we study, price commensuration is a normatively legitimate part of the initial institutional arrangement, so that it becomes awkward to struggle against an increase in price objectivity and impersonality -which amounts to struggling against the degree of commensuration of the pricing system to maintain the status quo -both in terms of normative associations and theorizing. This difficulty is all the more acute as promoters of institutional change use mimicry to align the functioning of organized markets to that of OTC markets. The creation of incommensurables along the path we describe thus appears as a potentially fruitful way to defend the status quo by rejecting the proposed institutional arrangement as irrelevant in the specific situation under debate without having to reject the normative grounding on which this status quo partly relies.
We believe our paper makes two contributions. First, we contribute to the burgeoning literature on institutional work devoted to the maintenance of institutions (Currie et al., 2012;Dacin et al., 2010;Lawrence & Suddaby, 2006;Lok & de Rond, 2013;Micelotta & Washington, 2013;Riaz et al., 2011;Zilber, 2009) by proposing that the creation of incommensurables be added to the list of strategies available to actors attempting to maintain institutions and by documenting the institutional work required to implement this strategy in a context involving a coercive and rationalitydriven institutional disruption. We suggest that instances of the process masterfully depicted by Espeland (1998) in her account of the Yavapais' refusal to sell their land might be more universal than previously understood and apply to situations where moral dimensions are almost entirely absent. Our work shows that the core of incommensurables creation relies on constructing insularity, on erecting a fence around a territory of exception where the institutional change proposed is demonstrated to be irrelevant. We show that the specific institutional work this requires involves suggesting the irrelevance of existing theories, attacking the merits of taken-for-granted institutions, and linking unorthodox views to broadly shared theories. It also involves purposefully betting on the difficulty in changing embedded routines. We know from the literature on commensuration how crucial the adoption of practices is in explaining the success or failure of commensuration attempts. Examples include commensuration of housework with paid labour, which failed to be incorporated into practices despite repeated attempts (Espeland & Stevens, 1998), unsuccessful efforts to build reporting mechanisms for carbon disclosure (Kolk, Levy, & Pinkse, 2008) or the issuance in the US in 1984 of guidelines for standardizing sentencing practices to limit judicial discretion, which judges, unsatisfied with the results produced, never adopted (Espeland & Stevens, 2008). Conversely, the ranking of law schools was so rapidly adopted that sceptic deans soon had to take the practice seriously (Espeland & Sauder, 2007).
Documenting the opposition efforts of the large investment banks trying to avoid reform -legitimizing and naturalizing existing practices while attempting to make the proposed practices look impractical and discrediting alternatives -we show that building on the inertia of practices might be purposefully used by skilful and powerful actors, to claim exception and struggle against the coercive disruption that threatens the previous institutional arrangement organizing the market.
Our second contribution is in bridging the gap between the literatures on institutional maintenance and commensuration, and demonstrating that specific institutional changes can usefully be understood as changes in the degree of commensuration experienced in a field. As commensuration is often perceived as a dichotomy, envisaging a continuum of degrees of commensuration opens the question of the specificity of the institutional work required to resist an institutional change aimed at increasing the degree of commensuration. Our paper suggests that some features (like a pre-existing network of actors) might render resistance easier, while others require the institutional work of maintenance to support some widely accepted principles while rejecting the application of these same principles in the specific situation concerned. The creation of incommensurables can be interpreted as a solution to this issue.

Conclusion
The main limitation of our research regards the time period of our study, which allows us to document the process of resistance to institutional change at an early stage of the institutional disruption but not the outcome and further developments. We are aware that the fact that we focus on 'early' institutional work, when actors are given an opportunity to discuss the proposed rules, might influence the weight devoted to resisting work as opposed to the work required for maintaining institutions through adaptation, compromise and repair work as documented, for example, by Currie et al. (2012) and Micelotta and Washington (2013). However, while the length of the regulation process of financial markets in Europe prevents us from reaching a conclusion about the success or failure of the actors studied herein, we believe our data provides useful insights into the type of institutional work in which powerful defenders of the status quo facing coercive and rationalitydriven institutional disruption engage.
As with any case study, inevitable idiosyncrasies need to be accounted for when it comes to generalizing our results. However, we have discussed how we think our findings are theoretically relevant and can be applied more generally to situations of institutional disruption that can be seen as producing a change in the degree of commensuration experienced in a field.
We believe one important direction for future research suggested by our study is in better understanding the relationships between the kind of institutional work documented herein and that documented in settings involving de novo institution creation, or the maintenance of existing beliefs in the absence of coercive threats to them. More research is definitely warranted in order to generalize the differences in the type of identity construction (Lawrence & Suddaby, 2006) or theorizing (Greenwood, Suddaby, & Hinings, 2002) required in different contexts.
To conclude, our study shows that the transformation of market forms is necessarily a complex and labour-intensive social process, bound to meet with resistance as powerful actors engage in efforts to maintain previous institutions. We believe this finding is important, as disregarding these difficulties may prevent states or regulators from efficiently addressing the issues for which they are accountable to constituencies that might differ from those to which market participants are accountable. The assumption that 'mechanical objectivity' is a given of any market risks ignoring where the discretion lies in organizing the market and eventually setting market prices. It also risks overlooking the likelihood of resistance to reforms in market forms, infrastructures or pricing systems and the potential channels to which actors might resort in organizing resistance. On a more theoretical level, we argue that the analysis of resistance to coercive institutional change proposed in the name of rationalization provides interesting insights into the cultural and social power processes that favour the maintenance of patterns in modern societies that keep on relying more on local knowledge and personal relationships than on mechanical objectivity. 14. TabbFORUM is a discussion forum for the financial industry, where major financial actors publish their ideas about what is happening in the marketplace and their opinions about the impact of new events, such as regulation, on the industry. It is a relevant source of data for this study, since it reflects the main discourses and justifications of major financial actors. 15. Most of these actors are represented in their relationship with the regulator, including as regards lobbying activities by the ISDA, which is in charge of defending common interests. As regards the regulation we discuss, no internal conflict is apparent. 16. For the role played by large incumbent banks in the development of OTC markets, see Huault and Rainelli (2009). 17. ISDA paper, MiFID/MiFIR and Transparency for OTC Derivatives, 15 February 2012. 18. We really thank one of the anonymous reviewers for having attracted our attention to this important point. 19. Although moral issues were raised in the context of the crisis as banks were exposed to severe criticism from public opinion, moral arguments are utterly absent from the specific debate documented herein. Idiosyncrasies are presented as technical and never moral.
(11) Which market features could additionally be taken into account in order to achieve benefits in terms of better transparency, competition, market oversight, and price formation? Please be specific whether this could consider for instance, a high rate of concentration of dealers in a specific financial instruments, a clear need from buy-side institutions for further transparency, or on demonstrable obstacles to effective oversight in a derivative trading OTC, etc. (12) Are there existing OTC derivatives that could be required to be traded on regulated markets, MTFs or organised trading facilities? If yes, please justify. Are there some OTC derivatives for which mandatory trading on a regulated market, MTF, or organised trading facility would be seriously damaging to investors or market participants? Please explain the reasons for your views. (37) What is your opinion on the suggested modification to the MiFID framework directive in terms of scope of instruments and content of overarching transparency requirements? Please explain the reasons for your views.

Pre-and post-trade transparency
(38) What is your opinion about the precise pre-trade information that regulated markets, MTFs and organised trading facilities as per section 2.2.3 above would have to publish on non-equity instruments traded on their system? Please be specific in terms of asset-class and nature of the trading system (e.g. order or quote driven).
(39) What is your opinion about applying requirements to investment firms executing trades OTC to ensure that their quotes are accessible to a large number of investors, reflect a price which is not too far from market value for comparable or identical instrument traded on organised venues, and are binding below a certain transaction size? Please indicate what transaction size would be appropriate for the various asset classes.
(40) In view of calibrating the exact post-trade transparency obligations for each asset class and type, what is your opinion of the suggested parameters, namely that the regime be transaction-based, and predicated on a set of thresholds by transaction size? Please explain the reasons for your views.
(41) What is your opinion about factoring in another measure besides transaction size to account for liquidity? (42) What is your opinion about whether a specific additional factor (e.g. issuance size, frequency of trading) could be considered for determining when the regime or a threshold applies? Please justify. Could further identification and flagging of OTC trades be useful? Please explain the reasons.