Promoting solutions and co-constructing problems – management consultancy and instrument constituencies

Abstract The concept of ‘instrument constituencies’ has recently emerged in policy research to reflect actors and practices focused primarily on articulating and promoting policy solutions. A central component of this is that, with such an emphasis on promotion, policy-maker decisions are subject to supply push rather than being demand-led. In particular, the ‘solutions’ of instrument constituencies come before, rather than after, the problems they are held to solve – solutions chasing problems. Key actors here are external consultants. However, the extent to which their activities match the problem chasing of the instrument constituency concept is untested. As a start in addressing such neglect, this article draws on data from secondary sources from research, outside the policy field, in organisation studies. In particular, it compares promotional practices in consulting with the supply push element of the instrument constituency model, finding that they correspond, but only partially. Consultants do not always construct problems for their clients to fit their prepared solutions, nor do they simply conform to the conventional, ‘rational’ model. Rather, it is shown that they also simultaneously co-produce both problems and solutions with clients, combining supply-push and demand-led approaches. Thus, it is argued that the model need not be so strongly associated with problem chasing (and non-rational decisions). Rather, the case of consultancy draws attention to how decision-making through instrument constituencies can occur along a continuum of three positions, from problems pre-dating solutions, being (co-) constructed at the same time, to their production after the ‘solution’.


Introduction
Work by Voß and Simons (2014) and Beland and Howlett (2015) and others has begun to develop the notion of instrument constituencies (ICs). Although still an emerging concept in policy studies, ICs are seen as collective policy actors or 'heterogeneous practices' (Voß & Simons, 2014, p. 738) exclusively concerned with 'the articulation and promotion of particular kinds of solutions regardless of problem context' (Beland & Howlett, 2015, p. 2). Voß and Simons outline how ICs 'strategically market their solutions, for example, by engaging KEYWORDS Management consultancy; public policy; instrument constituency; policy solutions; promotion and sales practices; problem chasing; garbage can; decision-making The creation of need is … a general phenomenon of processes of choice. Despite the dictum that you cannot find the answer until you have formulated the question well, you often do not know what the question is in organizational problem solving until you know the answer.
However, the garbage can model has been hugely influential in other contexts including policy studies. In particular, it was integral to Kingdon's (1984) 'multiple streams' framework of policy-making which challenged the orthodoxy of rational decision-making (although rationality remains implicit in some accounts) (Cairney & Jones, 2015;Paton, 2014). Furthermore, despite the appeal, there is very little empirical work which sets out exactly how actors seek to find and/or construct problems for their solutions to fit and with what 1 it is recognised that for some, the meaning of policy instruments is quite distinct from that of policy techniques, tools or solutions although all are 'bearers of values ' (lascoumes & le Gales, 2007, p. 4). However, there are many different classifications possible (Hood, 2007). in this article, the concern is with how diverse prescriptions are connected to management and policy problems through action rather than the target, form or scale of those prescriptions.
effects (Beland & Howlett, 2015). This is a general limitation of current research on policy decisions, but is also especially acute if the concept of IC is to be developed further. One possible source of empirical insight would be to research the promotional practices of ICs, ideally in action. Different groups, such as administrators and technicians can act in this way, but given the market pressure they experience to promote ideas; their claims to political independence; and their role in policy implementation, consultants are seen as especially relevant (Beland & Howlett, 2015). Indeed, in many national and transnational contexts, external consultants of various kinds are recognised as playing a key and growing role in policy debate, formulation, implementation and evaluation (e.g. Beveridge, 2012;Gunter & Mills, 2017;Hilary, 2004;Saint-Martin, 2004). In health care alone for example, management consulting is said to be worth $6 billion per annum globally in terms of fees paid (Consultancy.UK, 2015), even if the extent of usage is notoriously difficult to establish (Howlett & Migone, 2013;Sturdy, 2011).
Consultants come in various guises ranging from those in the multinational accounting and professional services firms (e.g. PWC, KPMG and Deloitte) to sole practitioners, including academics, and span different domains such as policy, management and engineering (Lapsley & Oldfield, 2001;Prince, 2012) and different types of service and relationship from the standardised to the tailored and the transactional to the personal (Maister, 1993). Together, they are seen to offer not only knowledge, but also flexibility as an extra, temporary resource and, in particular, legitimacy for decision-makers (Martin, 1998). In the case of management consultants, this is especially the case given the valorisation of market-based approaches in neo-liberal regimes (Jupe & Funnell, 2015;Paton, 2014). However, debate continues over the extent to which consultants act simply in their clients' interests ('servants of power'); actively shape policy for their own ends, commercial or otherwise ('masters of the universe') (e.g. Hodge & Bowman, 2006); or play the role of policy partners (Saint- Martin, 2004).
With their strong focus on the articulation and promotion of solutions then, consultants appear to be an ideal focus to explore the concept of ICs, and yet, according to recent accounts, there is relatively little research in the policy field at the level of detailed practices (Howlett & Migone, 2013;Stone & Ladi, 2015). One exception is the work of Gunter and Mills (2017) in the UK education sector. They also see consultants as actively involved in the promotion of solutions and hint at problem chasing: In other words the (consulting) buyer in government or in a classroom may not know that they have certain needs or know what all their needs are, and so the vendor (consultant) can create retail opportunities' . (2017, p. 82) However, they note that detailed practices are hard to access and that this empirical neglect is widespread. Nevertheless, there is an emerging body of research on consultancy outside of policy studies, in organisation studies (Kipping & Clark, 2012). Some of this may shed light on the specific practices associated with ICs and policy innovation in general, and this forms the rationale of this article. In short, can selected empirical insights from secondary sources on consultants' promotional practices begin to develop our understanding of the concept of IC in policy studies and establish an agenda for further research?
The article is structured as follows. Firstly, by way of providing a context for the role of management consultants, we briefly introduce the field of management knowledge production and locate consultancy and its diverse forms, roles and practices within this field. We then selectively draw on data from the literature on management consulting practices which most closely relates to the articulation and promotion of solutions. Here, we identify three key practices or points on a continuum -the specific case of problem chasing that is explicitly associated with ICs; those which appear to follow the rational decision-making model of responding to pre-existing problems; and, more distinctively, the simultaneous co-production of problems and solutions between consultants and their clients or potential customers, which develops our understanding of ICs. We conclude with a brief discussion of the implications of the practices revealed for policy innovation and more generally.

The field of management knowledge production and consulting practices
In the context of management, there have been parallel debates to those in policy studies over the 'non-rational' nature of decision-making in adopting innovations (although little specific attention to problem chasing). Rationality was again the orthodoxy and still sometimes remains, even if it is recognised as being bounded (e.g. Abrahamson, 1996). Critics adopt various perspectives, pointing to political, cultural, psychological and institutional factors that appear to limit or undermine (formally) rational decision-making. Thus, over and above any technical merits from ideas addressing organisational problems, managers seek personal gain, psychological reassurance and, in particular, legitimacy from adoption, through emulation or isomorphism for example. Of course, rationality is also itself institutional, political, cultural and emotional in character (Sturdy, 2004).
This range of decision dynamics exists within a complex domain where various actors operate, performing different, overlapping roles. Suddaby and Greenwood (2001), for example, set out such an organisational field. This comprises: business schools; management gurus; consumers of management knowledge; large conglomerate professional services firms (PSFs) such as the large accounting firms; and other consulting firms. One might add the media and include the role of managers and civil servants as knowledge producers (Engwall, Kipping, & Usdiken, 2016). Nevertheless, different primary (not exclusive) functions are usefully ascribed to each actor. For example, consultants' primary role is seen as commodification -codification and abstraction of knowledge into portable tools and selling, translating or re-embedding them into new contexts (also O'Mahoney, Heusinkveld, & Wright, 2013). Commodification serves to intensify competition and to lay the ground for subsequent colonisation of commodified solutions to new professional jurisdictions (also Abbott, 1988) by large PSFs, such as accounting firms.
While intended as an overview and simplification, a number of things emerge from this management framing which are helpful in making sense of ICs in a policy context. Firstly, all of the actors, not just consultants, are involved in some way in the core function of ICs, albeit not exclusively so. For example, gurus popularise ideas and large PSFs extend the market scope of solutions. At the same time, superficially at least, the role of consultants seems to be closest to that of ICs in articulating, promoting and implementing solutions and thus the focus of this article appears justified on the basis of wider consulting research as well as from work on ICs specifically. However, this needs further scrutiny, not least because consultancy is not simply involved in commodification. Indeed, textbook accounts of consulting (e.g. O'Mahoney & Markham, 2013) point to a wide range of activities, including the following: • 'thought leadership' (e.g. authoring publications) and branding; • targeting potential clients and networking including via recruitment/secondments ('revolving doors'), external committees and attending social functions; • developing and internally promoting new products and services; • the consulting project cycle of proposal writing, problem 'definition' (see below) and planning; data collection and analysis; delivering and handover/withdrawal; review; • 'follow ups' with former clients to sell new products and services ('sell on'). Such a list is by no means exhaustive and also can vary significantly by project type and level of seniority (Kitay & Wright, 2007). In large firms, for example, the most senior consultants spend more time in business and client development (selling), while sole practitioners have to perform all the consulting tasks (also O'Mahoney & Sturdy, 2016). Furthermore, the list gives little indication of the dynamic nature of practices and variations in timescales. For example, consulting is often associated with short-term engagements through defined projects, but these can sometimes last years and both relationship and product development processes can also extend over very long time frames. Similarly, some consulting sales activity appears to occur at a distance with formal, Requests for Proposals published and submitted by consultants in an automated way. This would seem to preclude the possibility of problem chasing, but much public sector consulting at least, does not use this channel (e.g. NAO, 2016) and may sometimes occur after generic promotional practices such as thought leadership which may sew the seeds of a 'solution' and 'problem' .
Nevertheless, the list provides an initial overview against which to explore the practices of ICs and 'problem chasing' in particular. Here, certain domains of consulting activity are likely to more relevant than others such as new product development, marketing and sales, and client interactions more generally. For example, evidence of consultants developing new products explicitly in response to reported problems by clients would contradict the idea of problem chasing, while promoting solutions to clients regardless of (or in direct tension with) the problems they are experiencing would support it.
In the following sections, we attempt to explore these issues using secondary data. However, not all consulting research is helpful in this regard, focusing on issues such as working conditions, identities, careers and the technical outcomes of consulting (Barthélemy, 2017;Kipping & Clark, 2012). Furthermore, as noted earlier, consultancy is a difficult activity to research. Very often its subject matter is commercially or ethically sensitive although autobiographical or media exposes of consulting malpractice are sometimes helpful in this regard (e.g. O'Mahoney, 2011;Pinault, 2001). This is especially evident in the widespread use of consultants informally to legitimise or 'rubber stamp' client decisions (Martin, 1998;McKenna, 2006). Similarly, as we shall see, it is difficult to identify clearly and unequivocally, how consultants might seek to sell a service regardless of client needs. Thus, overall, given the paucity of research on detailed consulting practices, especially where there are sensitivities, we are obliged to be selective and can make no claims over generalisability or how such practices might vary according to different contexts. Nevertheless, some research data are available, which are drawn on selectively here and shed light on these and related practices.

Identifying and responding to client problems
In many respects, the problems experienced or anticipated by organisational actors, especially owners, managers, policy-makers, civil servants and other potential clients, are the lifeblood of consultants and consulting firms. The nature of these problems varies, but in general, a core service offered by consultants is knowledge to advise clients, facilitate changes (process consultancy) or directly implement them, in the form of flexibly deployed extra, expert labour. With management knowledge and practices largely unregulated, competition over the promise to solve problems is potentially fierce. Thus, and in keeping with the classic rational view of decision-making, identifying and predicting client problems and then devising solutions to them is integral to consulting. This is reflected in formal accounts of practices. For example, 'problem recognition' is held to be the first stage of consulting project cycle (O'Mahoney & Markham, 2013). It is also evident in research findings. Pettigrew for example, highlighted how, for commercial reasons, there is often great pressure on consultants to accept, in broad terms, the client's definition of the problem -for 'acceptability, legitimacy and impact' (Pettigrew, 1985, p. 478). This is reflected in consultants' own accounts of their practice.
Our approach is not very different from any of our competitors … The key thing is to listen and to find out what their (clients') problem is, where they are solving it and what their thinking is. (reported in Sturdy, 1997, p. 398) Furthermore, problem identification can be translated into a service in itself, as one practitioner observer (Scriven, 2012) of the leading consultancy, McKinsey & Co., noted: One of their tricks is brilliant: get the (client) board into a room and get them all to write down the 10 things that worry them most about the business, then take the answers and write a report around them, and send it back -with the bill. It's a bit like going to see a solicitor.
While this account is expressed cynically, the practice of identifying, organising and articulating client problems is not without significant potential value to clients. Such practices also feed into a longer term and wider research process for consultants of identifying problems across sectors which feeds into 'new product development' (NPD) through which firms and individuals develop 'solutions' . Here, in large firms, an internal selling process is also required to convince consulting colleagues of a new service's value before its full launch to clients, and proving early success in practice and likely market demand (i.e. existing problems) is considered vital here. As Heusinkveld and Benders observed, product champions 'seek to convey the impression that new concepts are able to address contemporary problems in many client-organisations' (2005, p. 301).
At a more general level too, consulting products can be seen to emerge as a response to various problems experienced by different client communities. In the recent growth of consulting around corporate social responsibility for example, 'CSR standards came about as a reaction to accidents, irregularities and negative publicity' (Jutterstrom, 2013, p. 69). And this is reported to have been the case historically across the consulting industry where, according to a leading historian of the sector, 'consultancy can be understood as a kind of reflection of prevailing managerial problems and definitions. Thus, when there was a major shift in the role of managers and in the focus of their attention, the kind of consultancy they used also changed' (Kipping, 2002, p. 29). In other words, at both the macro, sector level and in individual and firm practices, we can see variations of the rational model of decision-making that is served by consulting in identifying and responding to client problems (or demand) through service delivery and product development or developing solutions.

Problem co-production with clients -articulation, accentuation, adaptation and negotiation
The above account of consultants responding to clients' pre-existing problems is perhaps, unsurprising. It conforms to the rational model and, as we noted earlier, such 'prior problem definition' fits the IC model, even if emphasis is typically placed on problem chasing over demand-led policy decisions. However, it is partial, presenting the processes involved as overly rational and the client as wholly sovereign. This is not the case in consumer markets generally, but especially in the field of management and many other forms of consulting. While it is sometimes possible to identify the outcomes of technical consulting (e.g. Barthélemy, 2017), management and related policy knowledge is notoriously ambiguous (Alvesson, 2004) and uncertain (Czarniawska, 2013). Problems, solutions and their providers are difficult to identify and reliably assess (Glückler & Armbrüster, 2003). This gives some market power to knowledge providers such as consultants in their relationship with clients (Fincham, 1999). Thus, as the following examples show, rather than consultants simply responding to pre-defined client problems with tailored solutions (as above), they can be seen to co-produce problems simultaneously with their clients in a form which suits consultant interests (i.e. their solutions). In particular, they do this by articulating, accentuating and adapting problems, partly in competition with client groups.
As the earlier description of McKinsey collating client problems suggests, consultants seek to articulate or package as well as simply identify and respond to client problems. Abrahamson (1996), a key writer in the field of management innovations, for example, sees the first stage of management fashion setting as sensing consumers' 'incipient preferences' or those that have not yet been articulated. Indeed, given the length of time it takes to develop new management 'solutions' (Heusinkveld & Benders, 2005, p. 284), some speculative anticipation of future problems is inevitable. In some policy contexts too, consulting firms are able to identify in advance of clients what problems they will face through the practice of 'double-dealing' (Engwall & Kipping, 2006). Here, consultants co-construct problems with regulatory bodies such as EU Directorates and then, subsequently, advise the organisations affected by the regulation they have helped to devise (also O'Mahoney & Sturdy, 2016;Wedel, 2014).
Client uncertainty over the precise nature of their problems also gives consultants the opportunity to accentuate the significance and strength of problems in order to reinforce client need and anxiety -creating a 'burning platform' . This is evident in research across the sector where rhetorical devices deployed both before and after projects (see also below) appeal to different client wants and concerns, above and beyond those of technical effectiveness (also Christensen & Skaerbaek, 2010). Abrahamson (1996, p. 286) for example, suggests that fashion setters 'attempt to frighten managers by revealing threats of managerial demise if such a performance gap is ignored, and to entice them with sublime opportunities for managerial stardom if this gap is attended to' . Likewise, the following extracts from recent research on consultants' project proposals to clients and other presentational activities show how consultants seek to de-institutionalise client practices by identifying or reinforcing them as a problem to different client groups. They are worth quoting at length.
In the (consultants') project proposals, the (client's) existing practices … were positioned as the 'problems' by the consultants. Negative words such as 'problematic' , 'costly' , 'inappropriate' and 'unsafe' were used frequently … The negative impact of continuously adopting the existing practices was diffused to lower management layers and employees through training and the presentation of posters. (Hu, Williams, Mason, Found, & Esain, 2015, p. 5) In most instances, the consulting firms' presentations would move on to identify a particular problem or difficulty that their client faced. Often the client would have recognised this particular issue and called into the consulting firm in to help… However, the consultants would often be the first to collate and communicate aspects of their client's performance back to their client…… (As one consultant observed) 'There's no point informing them (clients) that mostly they are doing well. If that's the case then why do they need us? We are there to help them [the client] with a problem. You have to be clear on what that problem is and point out where they are going wrong' . These performance rankings did not simply communicate information, but often spurred action… The consultants' presentations appeared to present a problem to their clients, which often hinged on their relative performance to their competitors. (Gill, McGivern, Sturdy, Pereira, & Dopson, 2017) As the second example shows, a key way in which problems are accentuated is through reference to client competitors or other peer organisations. This appeals to a central feature in client demand for solutions -legitimacy within one's institutional context (isomorphism).
While both examples could be seen simply as consultants doing their job to stimulate and facilitate change in the client organisation, problems are also articulated or adapted in a way which is consistent with the 'solutions' that the consultants already have to offer. As Abrahamson (1996, p. 266), again, points out, 'fashion setters not only (a) sense and satiate incipient demand for new types of management fashions, but they also (b) shape and focus this demand by articulating for fashion followers the particular techniques that fit the types followers prefer' . As a consulting partner quoted in Sturdy (1997, p. 401) observed: What we're looking for in the first place is, what we call in the trade, 'hooks' or symptoms … and it is part of our skill to translate the symptoms into work (new consulting business) … and to make them aware that it is necessary … and we haven't done too badly! Similarly, in the context of contemporary Corporate Social Responsibility (CSR) policy, Bres and Gond (2014, p. 3) show how 'consultants act as social and environmental issues translators who are reconstructing and repackaging issues into "sellable" propositions' . As others have highlighted, this process is facilitated in part, by use of knowledge management systems which record materials such as proposals and data which can be accessed remotely and adapted to new clients (Boussebaa, Sturdy, & Morgan, 2014;Canato & Giangreco, 2011). However, as noted earlier, such commercially oriented adaptation needs to be obscured from clients in order to preserve the ritual or myth of rationality. As one consultant promoting IT strategy observed: We try and identify what the business issues are before we ever meet the (client) company and draw out some IT implications of that. So we don't go in and say 'we've got a lovely IT strategy for you!' We say 'is the fact that you are finding it hard to attract suitable labour an issue?' and 'yes it is' … (Sturdy, 1997, p. 399) This seemingly non-directive sales technique suggests that problem definition is under the control of the consultant. Indeed, much early consulting research in organisation studies presented consultants as dominant, even omnipotent, in their relations with clients. This view was corrected in part, by pointing to the market power and knowledge of clients in some contexts (Fincham, 1999). It has since developed further with the increasing role of specialist client procurement departments in commissioning consultants, especially in public sector contexts (Werr & Pemer, 2007). Here, a third party is introduced (also common in 'development' contexts) such that problem definition can be contested or negotiated and co-produced by at least three different parties, with consultants seeking to resist procurement managers, but not always succeeding. As one commentator reported in O' Mahoney et al. (2013, p. 225, 229) observed: ' At the heart of these [consultants' resistance tactics] is not only a question of protecting [consulting] margins … It's also about who has the right to define the project. The [client manager] thinks it's him [sic] because he knows his problem best. The consultant thinks it's him because he's got an MBA … the procurer thinks it's him because he can negotiate the best deal……' By establishing ownership of codified documents and by translating the words of both managers and consultants on the terms of procurement, procurers lay claim to being legitimate arbiters of the management problem and its solution.
In summary then, consultants by no means simply respond to pre-existing and client-defined problems with solutions in accordance with the rational model. Rather, the ambiguity of the knowledge at stake and market and other pressures on all parties allow for them to co-produce problems, through helping clients to identify and articulate them, but then accentuating their strength or significance and adapting their nature to fit better the 'solutions' they have already available. This may often be a contested process, as clients become more sophisticated and with the growing role of professional purchasers in particular. Nevertheless, what is noteworthy here conceptually in relation to ICs, is that such co-production effectively combines a demand-led approach with supply-push. It is therefore very close to (but not synonymous with) problem chasing or promoting solutions with little regard to problems experienced, to which we now turn.

Problem chasing (and solutions chasing solutions)
In this section, we explore examples of consultants acting in direct conformity with one of the central features claimed of ICs -creating and targeting problems and opportunistically seeking to fit available solutions to them. As already outlined, this is a distinct departure from rational or ethical norms of both clients and consultants and so is likely to be concealed in some way and is certainly difficult to access through research. Nevertheless, the conditions for this are present, with ambiguous knowledge; consultants under pressure to sell and; clients often keen to innovate and emulate through new 'solutions' . Furthermore, such supply-push practices are occasionally claimed in research on consulting. As Bloomfield and Danieli argued over 20 years ago, consultants 'do not so much target themselves at a particular niche as seek to create a niche and persuade clients that they are within it' (1995, p. 28). As we shall see however, this extends beyond manipulatively constructing problems with little, if any, resonance with clients, to the more widely reported practices of 'sell on' , regardless of establishing a 'need' , and to diversification. In the latter case, new solutions are extended to contexts which they were not designed for, and already established or dated solutions are adapted into new solutions -'old wine in new bottles' or what one might term 'solutions chasing solutions' .
While we have seen how consultants actively co-construct problems with their clients, it is rare to find empirical data of them cynically inventing problems which might otherwise have no traction with clients. Such practices are sometimes mentioned in confessional or popular accounts of management consultancy as a form of manipulation (Pinault, 2001). For example, the following was claimed in a recent observation of McKinsey (Scriven, 2012): They look at your business and say 'it's the opposite of what it should be' . If it's split by geography, they split it by customer type; if it's defined by customer type, they split by geography. There's no original thought whatsoever.
Similarly, and as we have already seen, consultants themselves sometimes claim powers of manipulation in selling, such as bringing a client to an appropriate choice or course of action as if it was their own idea -'you need to present the information to the … (clients) in a way that they can't fail to draw the message themselves' (consultant, quoted in Sturdy, 1997, p. 399). While it is important to retain some caution over such claims and to recognise practices as interactive and sometimes contested, there is certainly often huge pressure on consultants to sell and 'sell on' new business to existing clients. As noted earlier, this is especially acute at more senior consulting levels where promotion, to partner for example, and status are contingent upon sales, but it is also the lifeblood of sole practitioners. 'Sell on' is especially important in consulting because new prospects are more difficult to convert into sales given that much rests on trust between clients and the consultants they already know (Glückler & Armbrüster, 2003). Repeat business is a major source (often around two thirds) of income. It is achieved by identifying client conditions during projects which can subsequently be converted into problems (and future projects) to which the consultants already have solutions. This may often be in the form of co-constructing problems with clients, as we have already discussed, but it may also be more instrumental. Either way, 'sell on' is illustrated in the following consultant account (cited in Sturdy, 1997, p. 401) of translating a client problem of 'value for money' and needing a 'quick appraisal' into one of the 'need' for an IT strategy and, in the process of delivering this over two months, constructing the 'need' for a business strategy as well.
The (client) MD had an uneasy feeling that they weren't getting value for money from their IT and asked us to do a quick appraisal of essentially what was going on … Now it was quite clear when we got there that … it was a pretty messy situation … … (after two months) … so the question is what was going to happen after that? So, obviously, the ideal thing to do was develop an IT strategy, but there was no business strategy and it was quite clear that developing one in line with an IT strategy was going to take some time.
An additional and more conventional approach to extending sales beyond the domains for which solutions were initially identified is diversification at a market, rather than project level. This occurs in two main ways. Firstly, markets are extended by defining new fields of activity as being suited to solutions which had not been designed with that context in mind -solutions chasing problem domains. Secondly, existing solutions are re-packaged or re-branded to fit emergent client problems (also Anand, Gardner, & Morris, 2007, p. 412).
Both practices are evident in research on CSR consulting. For example, Bres and Gond (2014, p. 26) show how 'consultants expand their market by stretching the definitions and meanings of the concepts of sustainability and CSR as well as by infusing new entities (e.g. events, venues, purchases, standards) with these concepts' . Thus, conferences for example, including academic ones, have been reframed as sustainability issues and, as a result, consulting tools and international standards created. At the same time, Furusten, Werr, Ardenfors, and Walter (2013, p. 81, 91) show how a particular group of CSR consulting firms 'quite simply choose to present what they already do in terms of CSR … "a bit like the emperor's new clothes" … (Here,) well-established services in areas such as environmental impact assessment, change management and IT were labelled "CSR services" in order to ride the CSR concept's wave of popularity' . In short, this echoes Suddaby and Greenwood's (2001, p. 946) account, mentioned earlier, of colonisation and 'mass customisation' by large PSFs whereby competition drives firms 'to seek to extend the lifecycle of their knowledge products by extending their scope of application … (and) minor modifications, in new markets' .

Discussion and conclusion
Using selected data from secondary sources in organisation studies, we have explored some of the often hidden promotional and sales practices of management consultants. Our main purpose for doing this was to shed light on the concept of ICs in the distinct, but related field of policy studies, where consultants are highlighted as key actors, both in general and as important ICs, but where such data are relatively lacking. In particular, we sort to identify the extent to which management consulting practices match the IC concept and its central characteristic of a preoccupation with 'the articulation and promotion of particular kinds of solutions regardless of problem context' (Beland & Howlett, 2015, p. 2). Here, the idea of solutions 'chasing' problems or a supply-push approach, from the garbage can model of 'non-rational' decision-making, was a key reference point. We found that among the many practices of consultants, articulating and promoting solutions were indeed, very important, even if they were not exclusive concerns in the diverse set of consulting practices and roles. In this respect then, we confirm that consulting in management fits the idea of IC in policy contexts. As regards the relationship of consulting practices to problem chasing, we also found some correspondence in that it seems clear that consultants sometimes construct problems in a manner which fits pre-existing solutions, even regardless of client need, either in selling new business or 'selling on' business to existing clients. Indeed, consultants often claim this themselves. In addition, we found that it was achieved, not simply by creating (chasing) problems, but diversifying to new fields of activity by changing the meaning of, or repackaging, both new and old solutions to fit new problem-solution domains, ones they were not designed for -solutions chasing solutions. This is an otherwise unexplored aspect of the IC process.
While largely confirming the relevance of consulting to the concept of IC, our analysis has also revealed some differences and new practices by way of a contribution, but also raised some new questions for further research. In particular, although not contradicting the idea of IC or of garbage can decision-making, where various permutations are possible, it is important to highlight how consultants do sometimes follow the rational or demandled model and respond to defined problems with tailored or appropriate solutions. This is partly reflected in their new product development processes, but also claimed of their development as an industry overall, where practice has tended to follow client problems (Kipping, 2002). Furthermore, it is reflected in particular contexts where clients are able to resist or assert dominance in relations with consultants, such as through professional procurement and where Requests for Proposals are common and when long-term trust is threatened. However, and most importantly, establishing clearly which came first, the solution or the problem -the 'chicken or the egg' -is not always possible. Indeed, what was more evident, or easier to establish was the construction of problems by both consultants and clients together. In short, we found problems to be co-constructed through processes of articulation, accentuation and adaptation and that these could each be contested or negotiated by the parties concerned.
Co-construction is, once again, consistent with Cohen et al. 's (1972) model, but differs from discussions of ICs where the emphasis is on problems being defined to fit pre-existing solutions alongside a recognition of conventional demand-led decisions. Here, by contrast, while the rational model is still contradicted through having a solution already in existence, problems are defined through a combined supply-push, demand-led dynamic. Of course, this finding might simply be a consequence of the methodological difficulty of accessing the sensitive, non-rational and unethical practice of problem chasing or of sequencing the processes over time. So, for example, in-depth longitudinal studies may be better at revealing such dynamics. However, it may also be more of a conceptual issue with ICs in terms of downplaying or overlooking the interactive or co-produced nature of knowledge practices more generally. Indeed, Suddaby and Greenwood (2001, p. 940) make a similar point with regard to their model of 'a cycle of knowledge production and consumption that both stimulates and feeds consumer demand' . They argue that it 'is difficult and perhaps deceptive, to ascribe distinct and sequential stages to the process of management knowledge production. The events most likely occur contemporaneously …' At the very least then, it is important to explore problem and solution construction as ongoing, interactive and even simultaneous processes rather than assuming a blank slate upon which solutions or problems are first drawn.
If we accept some diversity of consulting practices, from demand-led to supply-push to combined processes, it is possible that they can be linked to specific contexts. Indeed, it is common to try and identify conditions which favour more or less rational decision-making. Abrahamson (1991) for example, following Simon (1960), argued that rationality was most likely in conditions of low uncertainty, which are comparatively rare in management (Salaman, 2002) as well as many policy contexts. Similarly, we have seen how client power may favour demand-led approaches and we can imagine how extreme pressures on consultants to sell, and on clients to buy, favour supply-push practices. In addition, there is some evidence of diversity corresponding to different actors within the field of consulting itself. For example, Canato and Giangreco (2011, p. 238) claim that practices can sometimes vary within a single firm, especially large ones, but mostly they distinguish between firms selling standardised solutions in diverse settings and 'focused service providers that do not sell pre-designed or pre-packaged solutions but rather develop ad hoc solutions for specific clients' . This is echoed elsewhere, where firms are sometimes classified as offering either personalised or commodified services (Maister, 1993). In CSR consulting, Furusten et al. (2013) identify three types of firm -idealists with an emphasis on education; specialists who are more commercial, but still serious about sustainability; and diversifiers such as the big auditing firms, who adapt or re-label their existing services (what Suddaby and Greenwood [2001] labelled colonizers). Although there is not the scope to pursue this further here, such classifications and internal diversity raise the question of whether consultancies should also be compared with other collective policy actors, including epistemic communities and advocacy coalitions as well as emerging ones such as individual 'issue professionals' in transnational policy spheres (Henriksen & Seabroke, 2015). Are such categories so easily distinguished in practice or better seen as ideal types? Indeed, actors' roles and organisational associations are often dynamic and ambiguous in policy contexts (Wedel, 2014).
Overall, the above analysis has added to our understanding of ICs in policy innovation through drawing on a related field of literature. It has done so primarily by confirming the connection between ICs and consultancy, but also extending the notion of problem chasing or sales-push approaches through the idea of the co-construction of problems (and solutions) between clients and consultants as well as supply-push through re-packaging solutions for new problem domains. Furthermore, we have drawn attention to the diversity of actors and contexts within the field of consultancy, with some matching ICs more than others. This prompts further questions and future research agendas, not least, to what extent are the management consulting practices identified here evident in policy consultancy? If there is a match, how, if at all, would co-construction or re-packaging solutions for example, be resisted or affect policy outcomes? Is co-construction preferable to problem chasing or is it simply inevitable that problems are produced jointly? This, in turn, prompts methodological questions, especially given the difficulty in establishing sequencing empirically and the sensitivity of non-rational practices. Would, for example, problem chasing be indicated by consultant indifference to the problems experienced by their clients that do not fit available solutions? How, if at all, can problems or solutions be classified as fully formed or complete so as to be followed or preceded by its other? What seems clear from the above analysis however, is that there is considerable scope and value in researching consultancy to reveal policy decision processes and that the IC concept helps generate different areas of focus than that of other policy actors.

Disclosure statement
No potential conflict of interest was reported by the author.

Funding
This work was supported by Economic and Social Research Council, followed by the grant number [RES 000 22 1980A].