Shifting logics: limitations on the journey from ‘state’ to ‘market’ logic in UK higher education

© Policy Press 2018. Our study of UK higher education institutions (HEIs) offers insights into the role of institutional logics in the adoption of organisational practices - specifically outsourcing. We identify two logics prevalent within HEIs: a public service 'state logic' and a 'market logic'. While adherence to the market logic supports commercial-based practices such as outsourcing, organisations enact competing logics in complex ways. Outsourcing is mainly limited to peripheral activities segmented from the core while a nascent cooperative solution is emerging as HEIs co-opt practices and discourse of outsourcing to justify hybrid relationships that marry competing logics in a process of selective coupling.


Introduction
Recent years have witnessed significant political and economic changes in UK higher education institutions (HEIs) with a 46 per cent drop in direct funding, other than research, between 2010 and 2014 (Bolton, 2014). As teaching funded from direct government grants continues to fall (UUK, 2016), tuition fees are expected to make up the shortfall and, appealing to a more discerning student 'market', HEIs increasingly face strategic decisions regarding identity, cost and quality of infrastructure and services. Against this backdrop, government policy has supported increasing use of outsourcing to seek efficiencies, however the implementation of this practice has been markedly limited (Elinder and Jordahl, 2013;Policy Exchange, 2010;UUK, 2011).
Drawing on institutional logics, representing deeply held norms, values and beliefs about what is and what is not appropriate (Friedland and Alford, 1991), we investigate how logics alter the propensity of HEIs to implement outsourcing practices. As such, we contribute to a more general theoretical question across the public sector of understanding the adoption of managerial practices depending on institutions' dominant logics. Our central contribution stems from delineating two ideal-type logics (Thornton et al, 2012) within HE. A state logic engenders practices consistent with collegial governance structures, communities of practice, public goods and organisational autonomy that privileges internal service provision. In contrast, the market logic supports more managerialist approaches associated with commercial objectives, efficiency, effectiveness and performance supporting outsourcing (Parker, 2012).
While our findings demonstrate HEIs adhering to the market logic are more inclined to develop commercial-based practices such as outsourcing, our contribution highlights complexity in how organisations enact competing logics. Outsourcing has been limited mainly to peripheral activities segmented from the core (Goodrick and Reay, 2011) and limits to the extent of outsourcing are evident. Further, a nascent cooperative solution is emerging as HEIs co-opt language and practices of the market to justify new hybrid relationships that marry competing logics in a process of selective coupling (Pache and Santos, 2013).
The significance of our study is not limited to the UK -there is growing adoption of neoliberal quasi-market mechanisms not only in HE sectors around the globe (Naidoo, 2016), but across public sectors, evidenced by the selective coupling and segmentation of commercial and non-commercial activities. Consequently, our study of competing logics feeds into public policy discourse on increased marketisation, apprising the need for governments and policy-makers to contemplate whether market-based mechanisms are always appropriate in bringing about improved performance in public services.
We begin by providing the context for the study, then examine the logics associated with market and state typologies. We examine inter-linkages between these and organisational behaviours to build our theoretical arguments of how institutional logics affect organisational practice. We then detail our empirical approach before presenting our findings and discussion.

The study context: UK higher education
Increasingly public service organisations have been affected by policy discourse and organisational practice associated with new public management (NPM) (Diefenbach, 2009;Pollitt, 2013). Deem et al (2007) argue NPM redefined the nature and legitimacy of certain forms of public service provision and organisation, justifying extensive use of market-based resource allocation mechanisms and control regimes, and creating an institutional shift to competition as the dominant imperative and global phenomenon in HE (Brown, 2011). Building on the work of Brown (2013), Johnston et al (2017) observe that, over the past 20 years, successive UK governments have been driven by market ideology to bring about policy changes in the HE sector shaped by 'policy-based evidence' as opposed to 'evidence-based policy': policy has been delivered and evidence then sought to support it.
NMP has challenged the extant logic of UK HE (Deem, 2004;Diefenbach, 2009;Parker, 2012;Parry, 2001) stemming from moves towards 'neoliberalism' within the public sector during Thatcher's conservative government in the 1980s (David, 2016). This saw more market-driven activities and the advent of managerial forms of governance; for example, the Jarratt Committee's inquiry into university management structures and systems and performance appraisal (Townley, 1997). Supplanting public good models of governance and social democratic values brought an ethos of accountability and monitoring, supported through incentives and performance targets and a focus on corporatisation, privatisation and reduced support from government (Bagley and Portnoi, 2014).
Major changes in 1992 imposed increased regulation through audit and quality mechanisms, and an expansion of the university system, as polytechnics achieved university status. A notable shift in the policy framework accompanied the Dearing Review in 1997. Alongside marketisation, the Blair government aimed to widen participation and instigated the introduction of top-up fees -stepping away from state funding for individual students (Savigny, 2013). The political agenda supported a policy of change within HEIs, rectifying perceived poor market orientation, expertise and lack of strategic vision (Lomer et al, 2016). Building on the Browne Report (2010), the coalition government introduced full tuition fees and HEIs were encouraged to adopt market-led behaviours and apply neoliberal quasi-market mechanisms and metrics. The Minister for Education, David Willetts, introduced changes leading to extensive marketisation, most significantly licensing private providers. Recent policy changes support performance metrics, including student satisfaction, drop-out and employment (BIS, 2016).
As part of this wider policy shift, the Diamond Report (UUK, 2011), identified activities (including outsourcing) through which HEIs could deliver efficiencies, improve quality and support core strategies. Outsourcing represents a highly strategic decision to reject in-house delivery depending on the activity's value to the organisation (Gilley and Rasheed, 2000). However, while university executives espoused the need for improvements and cost reductions (CVCP/HEFCE, 2000), adoption of outsourcing has been selective and limited (UUK, 2011) with the focus on operational support activities such as cleaning, catering, security, facilities management and accommodation, and professional services, including pension administration, aspects of HR, legal services and student IT support. As with public service organisations more generally, looking beyond outsourcing simple services towards more complex parts of their organisation, understanding determinants of outsourcing arrangements becomes critical. This is particularly in light of opportunistic behaviour, lower wages and quality of service from suppliers, potentially transitory savings (Jensen and Stonecash, 2005;Lonsdale et al, 2016;Lok and Baldry, 2015), and potential for greater private sector involvement in core activities.

Theoretical approach
Institutional logics, conceived as the 'rules of the game', are shaped by the presence of societal orders that provide the opportunity set for how organisations, and individuals within them, behave (Friedland and Alford, 1991). Following Thornton and Ocasio (1999, 804), we define institutional logics as 'the socially constructed, historical pattern of material practices, assumptions, values, beliefs, and rules by which individuals produce and reproduce their material subsistence, organize time and space, and provide meaning to their social reality…[they] are both material and symbolic'. Thornton et al (2012, 54) argue the bridge between institutional orders and behaviours can be understood by considering core principles as 'a governance system that provides a frame of reference that preconditions actors' sensemaking choices'.
The boundaries and conceptual domains of two institutional logics in HE are the traditional 'state logic' (Townley, 1997) and a market logic associated with NPM (Deem et al, 2007;Diefenbach, 2009;Parker, 2012). The root metaphor of the market logic suggests a mental mode experiencing and interpreting the world as transactional, predicated on individual interactions and value for money; the state logic interprets the world from a public welfare perspective. In the market logic, universities achieve legitimacy from performance against key metrics: research outcomes are systematically measured across the sector, guiding government research funding, while a national survey captures student perceptions of quality that contributes to league tables informing student choice. Both mechanisms can produce legitimacy-seeking behaviours and represent a far cry from the 'citizenship' ideal associated with the state logic. Changes in sources of authority reveal an increasing focus on scoring and holding to account organisational actions using these national measures, contrary to the ideal of universities as institutions held in public trust (Parry, 2001).
The root metaphors and sources of legitimacy and authority are associated with distinctive organisational behaviours and strategies as well as differing actor identities. The market logic associates with a more managerial identity, a focus on efficiency, effectiveness, and outcomes measurement, and strategies founded on commercial objectives (Parker, 2012). In contrast, the state logic associates with autonomy, collegiate governance, communities of practices, commitment to expanding and transmitting knowledge as a public good rather than a commodity, ensuring independence of intellectual thought and objectivity, serving the disadvantaged, and building societal and economic civil infrastructures (Lynch, 2006). Differences in logics characterised by more financial management and managerialist approaches to organisational behaviour, systems and strategies under a market logic, versus a more collegiate, decentralised and autonomous approach associated with the state logic, hold important implications for how HEIs respond to policy (for example, the Diamond Report) prescribing outsourcing (UUK, 2011). Representing an avowedly market logic, outsourcing represents a potentially contested area for university top management teams (CVCP/HEFCE, 2000).
A core principle of the institutional logics perspective is that organisational strategy, structure, and behaviours should align with the dominant logic, otherwise organisations may be subject to conflict and institutional weakening (Dunn and Jones, 2010). Consequently, decisions on internal resource allocation should support practices associated with the institutional logic. Hence, HEIs that have developed dominant market logics should be supported by associated sets of legitimate repertoires for action, upon which their members will draw, that support more market-based solutions, fostering a greater willingness to outsource. Conversely, HEIs with low orientations to the market logic are more likely to have goals and values that eschew market-based solutions, detracting from an HEI's willingness to engage in outsourcing: Hypothesis 1: HEIs are more likely to participate in outsourcing when they have a high market logic orientation.
The increasing marketisation of HE (Brown, 2011) is associated with the market logic's financial surplus/profit seeking behaviours and a root metaphor of the 'user pays' (Parker, 2012). In the UK's changed funding regime where income follows students, as Parker (2012, 259) notes, HEIs are 'thrust into the search for alternative revenue sources, while at the same time continually seeking cost efficiencies in [their] own internal operations'. The market logic has promoted a greater focus on generating income through tuition fees, reinforced by uncertainty regarding state funding and a heightened focus on teaching (Lomer et al, 2016). Combining these perspectives, we expect that HEIs that are more reliant on student fees for income in the domestic market, manifested in a higher teaching intensity, are more likely to adopt practices associated with a market logic.
As UK HEIs increasingly find home markets become contested, foreign markets represent a source of potential growth and revenue generation (Parker, 2012). Following the Browne Report (2010), governments have considered HE an export industry, generating income through tuition fees and partnerships with overseas institutions. Further, Lomer et al (2016) argue uncertainty regarding domestic funding raises the attractiveness of international students, paving the way for HEIs' current emphasis on funding from international sources and global competition in the sector. Graf (2009, 569) highlights the fact that the internationalisation of UK HEIs is 'strongly linked to the commodification and export of higher education services on a commercial basis', while Huang et al (2016) highlight competitiveness in international markets and increasing use by UK HEIs of agents to compete for a valuable share of the market. Hence international student recruitment may be driven by revenue generation, financial imperatives and strategic behaviours associated with the market logic (Parker, 2012).
Facing increasingly competitive markets and ever tightening margins, we argue for congruence between these manifestations of a market logic and efficiency-seeking behaviour to support outsourcing: Hypothesis 2a: HEIs are more likely to participate in outsourcing when they have a higher teaching intensity.
Hypothesis 2b: HEIs are more likely to participate in outsourcing when they have a higher international intensity.

Conflicting logics as limits to outsourcing
Research shows that logics are not cemented in stone; rather, being contingent on dominant social orders, they change over time, creating inconsistencies or conflicts (Lounsbury, 2007). A stimulus for change often derives from some exogenous destabilisation or trigger (Tolbert and Zucker, 1983), including regulatory, political or economic shifts that usher in a period in which existing practices are challenged and often unravelled (Greenwood et al, 2002). During such changes organisations confront institutional pluralism, contending with multiple and often conflicting, rules and associated norms and logics of behaviour; and in HE, universities will face multiple new identities (Parker, 2012). When society presents organisations with such multiple logics, organisations encounter alternative opportunities for action. However, while policy shifts can influence the direction of a field, organisations can refract such influences by drawing on higher-level logics (Greenwood et al, 2010). Further, the transition between logics is far from smooth and often associated with cognitive dissonance and 'stuckness or oscillation between logics' (Jay, 2013, 155).
Prior research suggests diversity of logics may result not only in contestation but also ceremonial accommodation, adoption or compromise of practices (Oliver, 1991) to manage or minimise legitimacy threats. For example, while publicly complying with a state mandate to adopt private sector business planning practices, museum service staff actively resisted these practices and values (Townley, 2002). Such resistance is not unexpected as organisational routines and practices become institutionalised and persist over time (Boeker, 1988), and professionals may resist logics that stand in contradiction to their own groups' interests (Stryker and Burke, 2000). Responses can be insufficient to embed norms, values and practices associated with a new logic and instead ceremonial accommodation reinforces commitment and salience of participants' goals. While new practices may be symbolically adopted, they may be decoupled from actual behaviour, and not internalised because they conflict with existing, normatively sanctioned practices (Tilcsik, 2010).
While many HEIs may engage with outsourcing, the extent of their activities may be limited, reflecting the underlying contradictions between logics within the organisation. Based on the preceding literature, we propose that even when HEIs engage in outsourcing, low adherence to a market logic will constrain development of further outsourcing activities: Hypothesis 3: Among those organisations that outsource, HEIs are less likely to engage in extensive outsourcing when they have a low market logic orientation.

Data and methods
Our mixed-methods approach involved surveying senior HEI professionals, conducting semi-structured interviews and holding a workshop with 24 HEI managers and suppliers, supplemented by secondary data from participants and the Higher Education Information Database for Institutions (HEIDI). Interviews focused on outsourcing decision-making and insights were reviewed against sector-specific practitioner literature to generate survey questions (Policy Exchange, 2010; UUK, 2011). To ensure data quality, findings were presented to a project steering group of experts and stakeholders at an HEI procurement Special Interest Group and a Leadership Foundation conference.
From the UK Leadership Foundation for Higher Education and UK mission group memberships we identified 131 HEIs and obtained complete responses from 56 (43%). Supported by the British Universities Finance Directors Group, the survey was administered online, targeting senior executives and professionals in finance, procurement and operations. Questions used five-point Likert scales to capture factors driving outsourcing behaviour, services outsourced, contractual arrangements, benefits and perceptions about challenges. Strategic benefits focused on improved productivity (33%), risk reduction (28%), and student satisfaction (25%). A higher concentration of outsourcing exists in front-line support services, mainly as single service providers in the private sector. Using operating data from HEIDI, we conducted a t-test to check for potential non-respondent bias, finding no statistically significant difference between the profile of respondent versus non-respondent organisations.

Measures
We created two dependent variables: engagement with outsourcing (a dummy variable) and total number of services outsourced (a continuous measure). We constructed two independent variables, outsourcing capability and market orientation. Relevant tests for factor analysis (Field, 2009) demonstrate an acceptable sampling adequacy, sufficiently large correlations for principal component analysis, and meet the criteria of 5 to 10 cases per variable. The scales had eigenvalues greater than 1 and accounted for 68.58 per cent of the variance. Factor scores were retained in the analysis.
The variable Outsourcing Capability controls for resources and capabilities available to an organisation that might influence its decision to outsource. Items include: 'sourcing is centralised and visible', 'we have the skills and tools', and sourcing is strategic and represented at the highest level. Higher scores represent a greater capability for outsourcing. The Market Orientation variable captures whether the organisation believes market solutions are appropriate. Items include: market-based services 'offer best value for money', there are 'benefits from outsourcing', it 'prioritises market sourcing', and holds the dominant belief that the 'market' is appropriate. Higher scores represent a stronger commitment to market solutions while lower scores reflect rejection.
We used HEIDI data to create the measure Teaching Intensity as the ratio of 'home and EU' domiciled student tuition fees and 'recurrent teaching grants' to total income. Higher scores indicate greater dependency on teaching income. The ratio of tuition fees from non-EU domicile students to total income measures International Intensity.
Higher scores indicate higher proportions of total income from student fees outside the EU. HEI administrative and governance structures may differ depending on whether HEIs existed before, or were formed after, the expansion of HE in 1992 (Deem, 2004) and this may have an effect on outsourcing decisions. Hence, we control for whether the HEI obtained university status after 1992. We also ran our models with a control for income but found high collinearity with Post 1992. Because income did not provide additional explanatory power we excluded it from our analysis. Other measures, such as endowment income as a proportion of total income and student staff ratios, were highly correlated, so they were excluded.
To test hypotheses 1 and 2, we utilised logistic regression in SPSS. Logistic regression is sensitive to sample size and the number of variables fitted as well as the effect of high correlation between variables. While our survey responses represent 43 per cent of the population, we took two actions to deal with sensitivity to small numbers: using variables that create the best fitting model and bootstrapping with 1,000 iterations to control for the effect of bias in standard errors (Field, 2009). To test hypothesis 3, we restricted our dataset to those undertaking outsourcing and utilised ordinary least squares regression.

Qualitative data
We conducted 31 semi-structured interviews (lasting up to two hours), analysed using NVivo. Participants were selected based on profession or expertise in strategic sourcing decisions (vice chancellors, finance directors, procurement and HR professionals). Questions discussed institutional sourcing decisions, activities suitable for outsourcing, core business activities, barriers to outsourcing, contractual relationships and contract management. We analysed these data utilising the organising principles of the two institutional logics.

Results
The correlations in Table 1 are modest and below the threshold of concern. We examined the Variance Inflation Factors and condition indices, finding no significant issues of multicolinearity, and the data indicate no violation of the regression assumptions (Field, 2009).
The results, modelling decisions to engage with outsourcing are shown in Table 2. The full model is significant and better fitting than the baseline (change in -2LL, 13.626, df(3), p<0.003). Measures of pseudo R 2 further demonstrate the full model provides greater explanatory power than the baseline (Field, 2009). Market Orientation is positive and significantly correlated with the decision to engage in outsourcing (1.27, p<0.01), providing empirical support for Hypothesis 1. HEIs with higher International Intensity (0.139, p<0.05), are more likely to outsource, supporting Hypothesis 2b while those with higher Teaching Intensity (0.045, p<0.1) are marginally more likely to outsource (providing limited support for Hypothesis 2a).
The results modelling the extent of outsourcing activities among those HEIs engaging in outsourcing are shown in Table 3. The change in R² between the baseline and full model (0.086, p<0.05) is significant, indicating the full model is  better fitting. Our main independent variable Market Orientation remains positive and significantly correlated with the extent of outsourcing activities (0.681, p<0.05). The result provides empirical support for Hypothesis 3 that predicted lower scores on this scale would lead HEIs to outsource fewer activities even when they have made the decision to participate in outsourcing. The full model reveals two interesting but not hypothesised effects that we discuss later: Teaching Intensity and International Intensity are negative and significant, indicating they are associated with fewer outsourcing activities than more, suggesting potential limits to the extent of outsourcing.

Qualitative results
Our qualitative findings elaborate upon the association between institutional logics and outsourcing practices exemplified in Table 4 and discussed below. Discourse surrounding outsourcing in the state logic reveals a consistent commitment to the core focus of HEIs as teaching and research, with a newer HEI acknowledging few activities are outsourced beyond cleaning services and that is 'really on the margins…It's nothing fundamental to the academic mission'. Indeed, adherence to the state logic reveals some HEIs provide services at a loss as part of an holistic view of their identity, as one large university highlights: 'there's some very clear sort of steers in the university about…the students paid £9,000 to come here to study, we're not going to sort of rip them off in terms of what they're paying for a cup of coffee. That would be sort of counterintuitive and against the [name] experience.' In contrast, key attributes of the market logic reflect the root metaphor of marketisation with an emphasis on value for money and consumerism, a large teaching university revealing: 'we actually will consider pretty much anything for outsourcing. So I know that's [about teaching] but we'll run the outsourcing rule over anything when we restructure.'  Basis of norms 'They looked at the typical saving that you could get from efficiency savings within the university applying commercial percentages, in terms of typical reductions and then looked at the further saving that you could get by the escalation volume. Then again, they're applying the typical stretch that you get within that volume.' 'You know the university committee is less sympathetic to that commercial mission than you would be perhaps in the high street or shopping centre for example.' 'That's a backdrop of us being a very paternalistic employer and thinking about the ramifications of outsourcing and the concerns that that creates and obviously very different terms and conditions of employment in the longer term.'

Basis of strategic imperatives
'Your acid test question is "Do we believe that we can make an adequate margin return on investment commercially?"' 'We need to focus on…why bringing competition into the process is important.' 'HEIs will make different decisions around what they deem to be core business and what strategies to pursue, for example how they interpret advancing public good or working with local communities.'

Nature of organisational actions and behaviours
'The culture here is towards outsourcing, and indeed, our Board of Governors constantly question, you know, why have you got so many support staff.' 'The VC is considering a radical departure from legacy structures, and removing schools and faculties -focusing on new ways of delivering academic services.' 'I think the VC and the sort of the senior team here, would be very concerned that if we sort of moved…let's just say all of our facilities over to [company] or somebody, we just wouldn't be able to respond quickly enough to the university's needs and I think that's the concern.' 'Each of the faculties have different views on what they'd like…and so we want to try and build that in a different way but I don't know what the market will make of this.' Despite a general consensus that academic provision is core, some contested views exist around the parameters. For example, a large teaching-oriented university highlighted the shift in thinking propelled by widening competition from private providers: 'I think certainly at the moment the academic provision is still regarded as core. So we have challenged that, considering how the private [institutions] provide. At the moment the economic is core.' Others contest this, perceiving the core to be protected with opportunities for outsourcing lying more firmly elsewhere: 'from a university perspective, clearly teaching and research are absolutely core and I wouldn't consider outsourcing, but in terms of our professional services I don't think anything's sacrosanct to be perfectly frank.' The marketisation thrust associated with student fees has led to a 're-imagining' and reevaluation of core activities with previous targets for outsourcing being incorporated into the discourse of the state logic. As one interviewee explained, 'accommodation and…the delivery of pastoral care and things like that are becoming increasingly important and increasingly core.' Several institutions highlighted contradictions between trying to provide a 'welfarebased' approach to educational services and seeking to use market solutions: 'it might be that it's not sensible to outsource catering because we know that we need to provide an out-of-hours catering service that's uneconomic. In which case, no commercial provider's going to take that on unless we guarantee their losses, so if it's going to run at a loss then we may be better off picking that up ourselves.' The state logic supports an inclination to provide services internally and, although some HEIs acknowledge the private sector might provide certain benefits, one felt that 'things like dyslexia services or academic skill support or student support services, these are things that we think we can do pretty well ourselves'. We found several HEIs use this language of pastoral duty of care to limit and appropriate the discourse surrounding outsourcing of student services.
While the market logic casts outsourcing in terms of responding to 'consumer' needs, the state logic interprets this differently. Rather than perceiving national student satisfaction rankings as a rationale to outsource more activities, they mobilise to limit the extent of outsourcing, citing loss of quality and service. As one representative explained 'the most important thing is student experience because we value it hugely here'. Associated with this lies a fear of loss of control and flexibility if outsourcing was used extensively as 'we just wouldn't be able to respond quickly enough to the university's needs and I think that's the concern'. Discussion of processes for entering into outsourcing contracts demonstrates the legacy of democratic participation in governance associated with the state logic as, 'you must also consider another layer of activity between central university functions and the 'faculties', which quite often have their own arrangements' and 'we made sure that the student union particularly were on board with it so…a lot of energy in communications, transparency and actually sort of...making sure that that happened the right way'. In other instances, appeals to external sources of authority are mobilised to support outsourcing actions, as a mid-sized HEI reveals, 'sometimes they do it because…a new requirement, a new piece of legislation that comes along on control or regulation'. The state logic demonstrates a fundamental resistance to 'profit' motives and external measurement, viewing universities as arbiters of the public trust.
On other attributes, actor identities and norms explain the adoption or otherwise of outsourcing practices. The market logic embraces the language of commerce with references to return on investment, competition and margins, and some HEIs recognise that people undertaking these functions are often recruited from other backgrounds. HEIs espousing these values also identified they needed 'to raise the profile of the procurement function' and get 'the right people in place' to professionalise and develop their capabilities recognising the: 'next part of the journey is how one manages the relationship…and that's the competency piece that the university needs to build up to have confidence in those actually managing the supplier in a mature, effective and sufficiently challenging fashion and is not being taken for a ride.' In contrast, the state logic suggests resistance to outsourcing by incumbent professionals steeped in traditional university ethos, seeking to exert power and restrain efforts to adopt commercial bases of operation. Such restraint is also couched in terms of supporting local communities and providing a public good, as outsourcing is sometimes usurped to justify a social mission directive, such as local sourcing or buying from social enterprises. For example, one university recounted its partnerships with a housing association and local council for student housing and sports provision. Another highlighted how adopting the living wage mandate created tension with contractors 'because it means the contractor has to operate [it for our contract] yet its other employees don't necessarily get the [living wage] and that gives them management issues in terms of flexibility and moving staff between their contracts'. Further, a presumption exists of a 'myth' that the private sector is more efficient at service provision, as one VC suggested, 'they just spend the money in different ways' and 'there seems to be public money going to them to do ridiculous things'.
Recognising the evident contested nature of outsourcing we also witness the emergence of shared services as an organisational form lying between traditional concepts of outsourcing and internal provision, as a mid-range HEI indicates: 'They [governing body] were very supportive of shared services -extremely supportive and actually saw it as a way of getting a step change.' One example highlighted the importance of adopting market-like management structures rather than academic committees 'with everyone getting around the table to make decisions'. Shared service provision aims to pool assets, avoid duplication and maximise economies of scale in operations, while creating a distinctly different organisational culture focused on professionalism of the service. A shared service provider recognised the state logic by demonstrating 'a charitable ethos and empathy which I think is really important in the university sector…we are close to our university partners…we understand them' while maintaining a strong market logic by rejecting traditional university single-pay structures to ensure agility to rapidly change service provision. Such shared services potentially enable HEIs to control provision through their ownership stake but deliver it in a way that, according to one large university, is 'less bureaucratic than you get at a university with a far sharper focus on service quality and KPIs'.

Discussion and contribution
This work aims to advance understanding of how institutional logics influence HEI responses to changes in the policy environment. Our interest has been on understanding how competing logics affect the uptake of an organisational practice (outsourcing) championed by governments pursuing neoliberal policies.
Our work demonstrates the presence of a dominant state logic translates into lower willingness to engage with outsourcing practices while, even for HEIs adopting these practices, stronger manifestations of the state logic engender resistance to expand outsourcing activities beyond initial limited experiences. In contrast, the market logic indicates willingness by organisations to engage with outsourcing and, moreover, to expand outsourcing activities, though some elements may still be subject to contestation. Material practices associated with market logics are not tightly coupled (Tilcsik, 2010) -many HEIs engage in limited outsourcing activities constrained to professional and support services. Loose coupling enables organisations to adhere to legitimacy by implementing government policy directives and practices without dealing with potentially difficult internal organisational conflicts.
Both teaching and international intensity are associated with adopting outsourcing practices related to market logics. First, institutions heavily reliant on student fees are likely to be more susceptible to changes in the funding regime and more attuned to seeking cost-reduction measures, such as a shift towards the outsourcing of some activities. However, because teaching intensity becomes negative in the models measuring the extent of outsourcing, we contend that these institutions may face contradictions. On the one hand they mobilise to engage in limited outsourcing, such as IT services or accommodation management, however the state logic is likely to kick in, motivating organisational responses inhibiting further expansion of outsourcing into core teaching activities. Likewise, some HEIs seeking new international markets adopt outsourcing, suggesting that such organisations enacted the market logic in response to changes in the funding regime. However, our findings demonstrate limitations as HEIs appear unwilling to engage in increasing levels of outsourcing. Such actions demonstrate complex responses to institutional pluralism, representing a market-based logic of greater commercial acuity constrained by the core educational mission.
Our study contributes to understanding how organisations respond to dual institutional logics. While most HEIs follow traditional outsourcing by contractingout a defined service to a third-party supplier, we find practices that reflect greater experimentation with new forms, including shared services and partnerships with social enterprises, local authorities and non-profits. This latter manifestation is consistent with the findings of Pache and Santos (2013) who highlight that hybrid responses selectively combine elements from competing logics. In so doing, organisations are better able to navigate across incompatibility and avoid challenges associated with compromise or costs and risks of ceremonial adoption of practices.
Interestingly, our study highlighted reticence among several HEIs about engaging further in outsourcing practices until they had evidence from others in the field of best practice that aligns to their goals. Hence, we anticipate that as alternative or hybrid models become better established, other HEIs may begin to replicate these.
Other research (Goodrick and Reay, 2011) has proposed the idea of 'constellations' of logics co-existing, sometimes in competitive while at other times in cooperative relationships. Such constellations were possible because key work practices were sufficiently discrete or segmented and aligned to one of the logics. In relation to our findings, HEI engagement in outsourcing can be segmented from core areas where the state logic is dominant in guiding organisational practices. Relatedly, conflicting logics may co-exist over time while a process of 'pragmatic collaboration' allows reconciliation as individuals within organisations retain their own self-identity while aiming to resolve differences (Reay and Hinings, 2009).
In considering implications of our research we should ask why it matters that HEIs appear to have limited ability to implement practices associated with the market logic. Recent literature suggests a multiplicity of potential outcomes associated with responses to institutional pluralism -some are potentially destructive (Dunn and Jones, 2010) while others range from accommodative (Goodrick and Reay, 2011;Reay and Hinings, 2009) to new organisational creation (Pache and Santos, 2013). Dunn and Jones (2010) identified when an institution comprises groups from distinctly different logics, for example logics of science-based knowledge versus patient-centred care in the medical education field, the potential for conflict is heightened. Such conflict may endure depending on the relative power of each group. In circumstances where core knowledge or skills are embodied in different groups essential to that institution, conflict can escalate to the point that the institution becomes weakened. Dunn and Jones (2010) suggest such weakening provides opportunities for 'invaders' to enter the field that might carry yet another competing logic. In this view, the presence of contested logics exerts negative effects on organisations that are untenable in the long term. In the context of UK HE, a recent expansion allowing private providers may bring an ethos more easily fitting the market logic. Indeed, our interviews revealed some HEIs mobilising language and behaviours associated with the market logic as a direct response to such perceived threats.
Our results provide additional insight to institutional theory by demonstrating that fields, such as HE, are not monolithic constructs in which organisations behave in homogenous ways. Rather our findings demonstrate that in a period of change, flux presents members of a field with complex, alternative practices emanating from differing societal orders each making claims on the direction in which the field could move. The diversity of response -from full contracting-out, through hybrid forms, to subversion for internal provision -highlights the need to understand why organisations respond in these alternative yet patterned ways within a field.
We identify the salience of the societal order of the state logic in HE that is both historically dominant and prevails across more recent expansions of HE. Our work provides evidence that many of the Post 1992 HEIs actively mobilise the state logic limiting the extent to which managerial practices such as outsourcing can be implemented, especially protecting core activities. Our work highlights that longevity of a field's dominant logic pervades even in the face of substantial changes. This suggests that theory needs to account for the perseverance of the dominant logic while examining how logics can evolve or mutate in responses to institutional pressures.
This study holds resonance for HEIs across country contexts as neoliberal quasimarket mechanisms are widespread: HE systems are experiencing deregulation, entry of for-profit organisations and increased quasi-market competition (Naidoo, 2016). Such reforms are not limited to HE as neoliberal policies have resulted in a range of quasi-market responses across many aspects of public service provision, including placing potentially conflicting responsibilities into separate institutions; separating commercial from non-commercial functions of the state; and separating the advisory, regulatory and delivery functions into different agencies (Olssen, 2016). These actions suggest selective coupling and segmentation to manage contradictions between market and state logics. Hence our study of competing logics is of broader significance in contributing to public policy debates surrounding increased marketisation.
Finally, our work suggests governments can learn from our findings that conflicting logics place limitations on the development of outsourcing practices. Our work challenges the policy assumption that public service organizations can be operated as if they are businesses (Cordella and Willcocks, 2012) and that practices will translate between logics. As such, policy makers should consider the appropriateness of mechanisms through which to improve the performance of public services, as highlighted in the IT outsourcing case of ASPIRE (Cordella and Willcocks, 2012). This is all the more important since the trend to marketisation of public services is only likely to increase over time (Elinder and Jordahl, 2013). In line with Johnston et al (2017) we suggest the right questions are asked and that any policy changes are fully evidence-based, using good quality objective data.

Further work and limitations
Further work could investigate micro-foundations of institutional logics that might account for limits to adopting new practices. Such studies might draw on social identity theory that suggests professional identities govern and provide consistency to behaviour (Stryker and Burke, 2000). Our study indicates that differences already exist in HEIs between the symbolic identification of professionals with the state logic and those identifying with the market logic. This difference could be explored more fully across occupations and levels within HEIs and more broadly across other public service organisations facing neoliberal policy shifts.
We provide some insight into how organisations make sense of and deal with the complexity of institutional pressures. Our work reveals how, by drawing selectively on the market logic, HEIs creatively integrate practices into existing structures and hi-jack the policy push towards outsourcing. For example, some HEIs have used student rankings to legitimise not the market-based outcome of outsourcing to improve service offerings, but instead to strengthen internal service provision by situating 'student experience' as central to core activities. Further work could examine other instances in which public service actors make strategic use of dual logics, thereby providing greater insight into the conditions and mechanisms that enable or inhibit policy implementation. While we focus on one distinct practice, outsourcing, opportunities exist to delve further into internal processes by which organisations adopt, adapt or refract such pressures in other aspects of their practices.
One of the main limitations of our statistical element in our mixed-methods approach has been a relatively small sample, despite achieving a good response rate from a small population. Despite this our results are robust and supported by our use of mixed methods that enabled us to provide qualitative insight to practices and explore the logics more fully. While the study is set in the UK, many countries face similar problems (Brown, 2011;Naidoo, 2016), hence our general findings should be transferable. Further, our work can be taken in the overall context of public sector studies of NPM and can inform that research more broadly.

Conclusion
Organisations that perceive themselves as largely following the state logic appear to prioritise actions consistent with this in their managerial practices. Conversely, those espousing a more market logic reveal limitations to which these outsourcing practices can extend. The ability to successfully blend or appropriate key elements of different logics provides organisations with an advantage in being more able to draw on wider repertoires of behaviours and actions when confronting pluralistic environments. We identify an arena of contested middle ground representing the interplay of these dual logics wherein HEIs engage in sense-making processes. In this space, our findings indicate the emergence of shared services and partnerships as new or hybrid organisational forms lying between market-based and state logics.