1. INTRODUCTION

Regional economic development is an evolving policy field as it responds to new social, economic and environmental opportunities and challenges (Pugalis and Gray, 2016; Stimson et al, 2006). As a highly politicised policy domain it takes different shapes and forms over time; moulded by the interests, ideologies and power dynamics between different constellations of actors competing and interacting in, through and across particular places. Hence, analyses of regional economic change and development require territorial sensitivity, which supports the use of case studies.

In spite of a chequered history in Australia, where "successive governments at the federal level have shied away from committing significant resources to regional development" (Pape et al., 2016: 913), regional economic development, in its various guises, is deployed throughout the majority of the country's diverse landscape--notably from a national perspective, although certainly not exclusively, through the Regional Development Australia (RDA) program. According to the Commonwealth Government, the RDA program "is a national initiative which seeks to grow and strengthen the regions of Australia. Comprising 55 Committees, the RDA network covers metropolitan, rural and remote regions across the country" (Commonwealth of Australia, 2015a: 3).

Notwithstanding, entities such as RDA Committees attempt to promote an ethos of working in partnerships, yet a shared understanding of the nature and purpose of regional development is lacking (Regional Australia Standing Council, 2013). Even so, extant literature, such as the definition provided below by the Institute of Economic Development, indicates that regional economic development is generally considered to refer to:

"A set of policies and actions designed to improve the performance of a spatially defined economy for the benefit of all residents".

Such definitions and interpretations of regional economic development reveal the dual objectives of economic growth and social equity, which are not necessarily complimentary, and also demonstrate the interactions between attaining policy order in the face of societal complexity through seeking to implement a 'set of policies and actions' and 'define' spatial economies.

Whilst it remains a contested activity, the conventional wisdom of orthodox regional development practice--often pursued by local, state and national governments--extols the need to embrace complexity, recognise ambiguity and account for multiplicity, whilst simultaneously managing uncertainty and risks through imbuing policy order and control. To embrace and work with complexity, uncertainty and fragility (see, for example, Sorensen, 2016), which according to Taleb (2012) are key features of contemporary society, those leading and practicing through the manifold types of regional economic development organisations (RDOs) are expected to navigate intricate spaces and networks at the interface of complex governmental and regional socio-economic systems. For example, the RDO faces evolving government systems, which include new modes of service delivery, fiscal tools, accountability frameworks and networked-based forms of governance (see, for example, Ziafati Bafarasat, 2016). Nevertheless, many governmental functions and policies, which are instrumental to the development of regions, such as the 'backpacker tax', are 'space-blind' and remain well beyond the purview of RDOs (Pugalis and Gray, 2016).

Changes to regional socio-economic systems are also altering the functions and performance of 'regions', including changing methods of industrial production, the ongoing decline of manufacturing as a share of the regional Australian economy, demographic shifts and evolving patterns of migration, and entrenched forms of uneven development (Commonwealth of Australia, 2015a). In terms of the latter, this same report suggests that the further an Australian lives from metropolitan areas the less income they will have (except in mining dominated remote areas), and they will have on average worse health outcomes and suffer more from mental health issues.

The ambit of this conceptual paper is to investigate what we refer to as the regional economic development paradox, which attempts to instil policy order in the face of societal complexity. The research is intended to inform contemporary debates concerning the efficacy of one notable type of RDO; the RDA Committees, which at the time of writing the program is in the midst of an independent review (See: https://rda.gov.au/review/). New South Wales (NSW), which has a long history of regional economic development policy, which can be traced back to at least the 1940s, provides the setting for empirical analysis. The New England North West Region of New South Wales Australia, located inland between the state capital cities of Sydney and Brisbane, provides a case study to investigate the interface between government and the regional socio-economic systems as we examine the role of the RDA Northern Inland Committee (RDANI). This case study provides an entry point to consider the suitability of the present configuration of RDA Committees.

The research approach passed through four key phases. Firstly, a twin-focussed review of scholarly literature about RDOs and complexity theory was undertaken to help situate the study in a broader theoretical context. Secondly, RDANI was selected as a case study as the literature and soundings with practitioners indicated that the 14 RDA Committees in NSW share some commonalities in terms of organisational structure and priorities, and to a lesser extent exhibit some similar traits with the 54 other RDA Committees in Australia. Given the aim of this article, one RDA Committee was considered sufficient to highlight the influence of the single phenomenon (Gerring, 2004) of complexity and its potential influence on managing uncertainty. The case study approach was used to provide a boundary for analysis and to increase our understanding of social problems (Gomm et al., 2009), which in this instance was an indication of present and future challenges facing RDA Committees (however findings from a single case are not intended to be representative of all RDA Committees). In addition, the researchers' familiarity with RDANI and the region was also a factor in the selection process. Thirdly, empirical material from RDANI was analysed according to the conceptual perspectives identified in phase one. Fourthly, the conceptual framework was applied to RDANI as an organisation to identify key new insights of wider relevance. The remainder of this paper loosely mirrors the key stages of research.

2. CONCEPTUALISING STATE-SPONSORED REGIONAL DEVELOPMENT ORGANISATIONS IN AN AGE OF COMPLEXITY

The practice of regional economic development in Australia has been and remains institutionally and operationally complex, and involves a multitude of actors from different civic, economic and social spheres (Collits, 2008). Similar to trends in other States and Territories of Australia, and consistent with some international experiences (e.g. English regional development, see Pugalis and Townsend, 2012), regional development policy in NSW has ebbed and flowed as it has been found to be in (political) favour and out of (political) favour at various junctures since the twentieth century. In view of this experience, and informed by a view of regional development as a state-sponsored activity, Beer (2007: 131) acknowledges institutional complexity by arguing "that regional development in Australia is hampered by the outcomes of our system of federalism, by political ideologies grounded in neo-liberalism that are wary of direct interventions in regional economies, and by an emphasis on short-term political responses, rather than longer term strategic interventions".

The diversity of RDOs can be bewildering and includes consultative committees, advisory boards, statutory bodies, development agencies, aboriginal corporations, voluntary bodies and partnerships, branches or agencies of government (e.g. local or state), business chambers, and private companies. Mountford (2009) distinguishes between five types:

1. Development and Revitalisation Agencies (lead urban redevelopment processes).

2. Productivity and Economic Growth Agencies (lead employment, entrepreneurship, and innovation processes).

3. Integrated Economic Agencies (integrate a wide range of interventions).

4. Internationalisation Agencies (focus on international positioning or promotion).

5. Visioning and Partnership Agencies (focus on long term strategy making).

Focussing on state-sponsored RDOs (i.e. those initiated, promoted and/or predominantly resourced by a branch of government), this particular type of RDO can be conceptualised as institutional entities that are established, mobilised and deployed to oversee, lead and/or implement regional development ambitions, variously codified in a plethora of documents, such as, strategies (NSW Department of Industry, 2015) and regional growth plans (NSW Government, 2016). These institutional arrangements are continually being reconstituted to greater or lesser degrees; shaped, for example, by competing, complimentary and coevolving actor interests, and changing spatial, economic and political conditions. State-sponsored RDOs are, therefore, constellations of diverse policy networks: "more or less stable patterns of social relations between mutually dependent actors which form themselves around policy problems or clusters of resources and which are formed, maintained and changed by a series of games" (Klijn et al., 1995: 439). Whilst these are often comprised of the 'usual suspects', including political and economic elites, more marginal actors (i.e. those outside of the political and economic elites) can also be influential (Pape et al., 2016). Hence, not all state-sponsored RDOs are top-down bodies.

The nature of institutional arrangements performs a powerful role in shaping understandings of economic development (Douglas et al., 2009; Pugalis and Tan, 2017). Thus, whilst the organisational forms of some RDOs can appear bounded, in institutional terms they are products of relationships, multiplicity, and heterogeneity, and, thus, always under construction (cf. Massey, 2005: 9).

It has been observed that state and federal governments have preferred to deploy 'independent' RDOs (at arms-length from local politics), although these independent entities have often struggled to engender local influence and commitment (Beer et al., 2005). However, not all RDOs are state endorsed or sponsored, and their variegated institutional forms, include informal networks, soft spaces of governance and fleet-of-foot partnerships as well as 'official' government entities and statutory bodies. Based on the results of previous research, which identified over 1 000 RDOs operating throughout Australia (Beer and Maude, 2002), this quantum suggests that RDOs are intended to be the primary interface between complex governmental and regional socio-economic systems (See, for example, the mission statements and goals of RDOs). Nevertheless, given that multiple RDOs typically operate across a single geography, it is not always apparent which RDO takes primacy.

The multiplicity of state-sponsored RDOs is an outcome of complex intergovernmental relations. The roles and interactions between each tier of government in regional development policy remains ambiguous and, thus, open to considerable contestation (Pape et al., 2016). Historically, the relationship between the NSW State Government and local government could be characterised as being adversarial, although this fluctuates over time, space and policy area. Pape et al. note that:

"The current regional development policy process is characterized by state and federal government commitment to establishing regional agencies, who are given the task of developing--though not implementing--strategies of social and economic change. This enables governments to demonstrate a commitment to social and economic capacity building at the regional level, even though empowered institutions of regional governance are absent" (Pape et al., 2016: 917).

The 'power', autonomy and capacity to act of regional entities mirrors earlier critiques of some RDOs (see, for example, Wilkinson, 2003). In part, this can be explained by the complexity of institutional environments within which state-sponsored RDOs operate and the complexity of the fundamental task at hand. In the case of RDA Committees, for example, Pugalis (2016) has recently referred to them as part of a 'patchwork regional institutional architecture'. According to some, such complexity negates against state-sponsored RDOs developing programs and interacting with stakeholders in a predictable manner (Patton, 2011). Instead, state-sponsored RDOs must contend with many variables, competing objectives and timescales, ephemeral political priorities, changing policies, interrelationships and evolving socioeconomic landscapes (i.e. complex systems). This further complicates regional development activities and, may, limit the capacity of state-sponsored RDOs to navigate their role through the complex systems of regional economic development.

Operational complexity, on the other hand, can be observed from the role of the economic development practitioner. The twenty-first century economic development practitioner is often expected to exhibit a wide range of skills to help navigate multifaceted processes (e.g. institutional, political, financial etc.), and to broker and facilitate sustainable regional economic development. Consequently, practitioners are expected to be dexterous, nimble and resourceful; influencers, networkers, brokers and facilitators; marketeers, program managers and project promoters; business savvy and versed in government processes and community needs; ethical, transparent and accountable; knowledgeable (in a codified and place-based sense); boundary-spanners, collaborators and builders of trust; strategic and politically sensitive; action-orientated and outcome-focussed; and understand what works, why and where, as well as what does not (Colitts, 2008). The onus is, therefore, on economic development practitioners through institutions, such as RDOs, to be agile, flexible and responsive to multifarious changes; expressed through the notion 'fleet-of-foot' (Pugalis and Townsend, 2013). Making sense of and 'managing' such complexities is a major public service challenge, but is particularly acute in the field of regional development which is multidisciplinary and transcends any single policy domain (such as land use planning, tourism or education).

In the following analysis, a basic complex systems framework has been applied to the case study. Deploying this framework, we examine the function and role of a specific RDO in a new way (Mitleton-Kelly, 2003), rather than repeat or reinforce past critiques of these organisations. Articulations of complex systems vary, although some key features include: non-linearity, feedback, spontaneous order, robustness and lack of central control, emergence, hierarchical organisations, and numerosity (Ladyman et al., 2013). Firstly, regional systems emerge and evolve in a non-linear manner as people move, factories close, houses are built, and land uses change, for example. Secondly, there is feedback within complex systems, such as the relationship between water availability and crop production. Thirdly, spontaneous order emerges within regional systems as change is absorbed by the many actors who have adjusted their behaviour. Fourthly, regional systems are deemed to be robust when they withstand change (e.g. floods or industry closure) with limited central control. Fifthly, emergence of new states of equilibrium and disorder engender further system changes. Sixthly, each entity is organised hierarchically, with its own level of interaction with the system and within the entity itself. Lastly, numerosity refers multiple parts interacting in manifold ways.

Of interest to this study is the notion of complex adaptive systems, which is a particular type of complex system. The complex adaptive system is has been defined as "a collection of individual agents with freedom to act in ways that are not always totally predictable, and whose actions are interconnected so that one agent's actions changes the context for other agents" (Plsek and Greenhalgh, 2001: 625). Consequently, "the system is constantly adapting to the conditions around it. Over time, the system evolves through ceaseless adaptation" (Lewin and Regine, 2003: 168). Similarly, Holland (1992) describes complex adaptive systems in terms of an 'ensemble of many elements', which is the prerequisite condition of interaction. Firstly, for systems to interact or communicate with one another requires an exchange (i.e. energy, matter, information etc.). Secondly, elements of the system require the means to interact for a 'common cause'. The mediating mechanism can include forces, collision or communication. Without interaction a system merely forms a 'soup' of particles, which necessarily are independent and have no means of forming patterns or establishing order. Thirdly, disorder is a necessary condition for complexity as complex systems are those whose order emerges from disorder. Fourthly, robust order can be observed as whilst the above three conditions are all necessary for a complex system to emerge, they are not sufficient. Hence, a system consisting of many similar elements, which are interacting in a disordered way has the potential of forming new orders, patterns or structures that can persist. Lastly, a complex system displays memory. Holland (1992) notes that it is "the persistence of internal structure", which engenders memory.

In summary, these features of complex adaptive systems highlight the paradox between policy order and complex practice. The existence of complex systems poses a challenge to state-sponsored RDOs to develop an advanced understanding and appreciation of their role vis-a-vis governmental and socio-economic regional systems, which we illustrate through the case of RDANI.

3. THE CASE OF RDANI

In NSW, 14 Regional Development Australia (RDA) Committees were established by the Commonwealth and NSW Government, and they are intended to be:

"strong advocates for their region and drivers of change and, as such, have a pivotal role in fostering regional economic development. The RDA network strengthens partnership across all three tiers of government, regional business and the wider community to boost the economic capability and performance of regions. RDA Committees are active in promoting Australian Government and NSW Government programmes and shepherding communities towards appropriate funding sources that will assist projects that work towards economic development" (Commonwealth of Australia, 2015b).

RDANI is a not-for-profit organisation, said to be 'community based', which receives joint funding from the Australian and NSW governments. It "is contracted by the Australian Government ... and the New South Wales Government through the Department of Trade, Industry, Regional Infrastructure and Services to act as a primary link between government, business and the community in the Northern region" (RDANI, 2016: 8). RDANI's 'core responsibilities', laid out in a charter, include:

* Creating regional business growth plans and strategies which will help support economic development, the creation of new jobs, skills development and business investment.

* Identifying environmental solutions which will support ongoing sustainability and the management of climate change.

* Identifying social inclusion strategies which will bring together and support all members of the community.

Box 1. Regional Development Australia Committees
Committees are responsible for achieving five key Outcomes:

Outcome 1: Regional Plan

A current three to five-year plan that focusses on economic
development of their region. The plan must take into account
any relevant Commonwealth, state, territory and local government
plans.

Outcome 2: Critical issues

Advice to all levels of government on critical issues affecting
their region.

Outcome 3: Priority activities

Advice to governments on priority activities that will drive
regional economic development, on future economic opportunities
and on their region's comparative advantages, based on consultation
and engagement with community leaders.

Outcome 4: Project proposals

Assistance to local community stakeholders in order for them to
develop project proposals; and referral of stakeholders to
appropriate public and/or private funding sources--including the
$1 billion National Stronger Regions Fund.

Outcome 5: Promote Australian Government Programs

Increased awareness of Australian Government programs in their region.

Source: adapted from Commonwealth of Australia (2015b).

In this respect, RDANI emulates many state-sponsored RDOs in both Australia and internationally that focus on business, environmental and social activities (i.e. sustainable economic development). The regional plan has the potential to perform an important function in this respect as it is intended to "help focus governments and communities on strengths, issues, and a work plan to achieve those priority regional development goals" (Stewart in RDANI, 2016: 2). Moreover, RDANI utilise "a regional economic development framework to promote long-term regional economic growth and provide a practical focus for strengthening our region" (RDANI, n.d.a). Complementing or perhaps competing with RDANI's strategy is the draft New England North West Regional Plan, which was recently issued by the NSW Government for community consultation (NSW Department of Planning and Environment, 2016). This plan also constructs a vision and identifies goals and actions for the region to achieve sustainable economic development leading up to the year 2036. Furthermore, most local governments in this region have economic development departments and staff to support development initiatives, and have produced their own economic development strategies (see, for example, Narrabri Shire Council n.d., and Tamworth Regional Council, n.d.), not to mention regional organisations of councils and a plethora of other RDOs. The dense networks of multiple and intersecting policies, strategies and bodies serve to demonstrate the complexity of regional development policymaking, which, arguably, mask under-resourcing and implementation inertia.

RDANI functions in an environment with many elements, with varied institutional and individual participants within a geographically diverse region of NSW. The Northern Inland region of New South Wales is located inland between the state capital cities of Sydney and Brisbane (see Figure 1). This region has an estimated population of over 180 000 people, a gross regional product exceeding $AU10 billion, 66 233 local jobs and key sectoral strengths in manufacturing and agriculture (RDANI, n.d.b) (see Figure 2). The Northern Inland region is defined by three distinct geographic areas, the New England Tablelands and gorge country, the North West Slopes and the North West Plains, which includes the following 12 local government areas: the new merged Armidale Regional Council between Armidale and Guyra Councils (Armidale Council, n.d.), Tamworth Regional, Liverpool Plains. Tenterfield, Glen Innes Severn, Inverell, Moree Plains, Uralla, Gunnedah, Narrabri, Walcha and Gwydir.

RDANI has a mission "to ensure long-term sustainable growth and development of the region by driving and contributing to initiatives that will support economic development and jobs" (Stewart in RDANI, 2016: 2). It aims "to be the driver of business growth, new jobs, skills development and business investment within the Northern Inland region" and "to improve the economic, physical and social conditions of communities across the region" (RDANI, n.d.a). According to RDANI's website (ibid.), recent projects include: Moree Plains Business Workshops, Northern Inland Regional Investment Profile, Namoi Investment Prospectus, Northern Lights Project, 2016 Northern Inland Innovation Awards, and Northern Inland Business Energy Assessment. The Committee includes a board of 'local people' supported by 'a small team of dedicated staff. In terms of skills sets, both members and officers are claimed to possess 'extensive networks' transecting different sectors:

"Committee members are volunteers drawn from the region and who bring leadership, skills and experiences from the community, business, local government and not-for-profit organisations" (RDANI, 2016: 8).

RDANI's Complex Operating Environment

According to Holland's (1992) conceptual framework, RDANI operates within an ensemble of elements directly interacting with each other and capable of communicating and exchanging matter, and creating disorder and order (patterns or structures) as a result. To illustrate this point, Figure 3 reproduces the Northern Inland innovation network, which seeks to identify key stakeholders (AusIndustry, 2016). A total of 75 organisations and programs, excluding individual actors, are identified as primary participants. This network is indicative of the extent of this ensemble of elements, although one suspects that the actually existing Northern Inland innovation network is much more dense, diverse and dynamic.

Applying Holland's (1992) theoretical lens to the Northern Inland innovation network, the basic elements of ensemble, interaction, disorder and robust order, and memory can be observed. Each of the elements has the capacity to directly interact and exchange with other elements (although the connections between the actors has not been identified), and each interaction is mediated through communication channels (e.g. information technology) or physical connections (e.g. events or infrastructure). For example, the University of New England may connect to the NSW Department of Industry to create an incubator program, which then creates new connections to other elements of the ensemble. Once established, the incubator program will create disorder, and then order will remerge as the elements self-organise to adapt to the influence in this system. The persistence of the incubator program within this system will then create a memory, which will inform elements in the future.

4. POLICY ORDER AND COMPLEX PRACTICE: ANALYSING THE ROLE OF RDANI

Through this section we analyse the tensions that RDANI is grappling with as it seeks to impose some semblance of order on the complex practice of regional economic development. In doing so, we draw upon this analysis and empirical insights to inform a broader discussion concerning the efficacy of RDA Committees.

Informed by the conceptualisation of RDOs presented in section two, our analysis of RDANI focussed on six policy-relevant perspectives utilised to help navigate uncertainty, which compares and contrasts methods that seek to engender policy order and those that seek to work with complexity (see Table 1). We apply Holland's complexity framework as new way of thinking about RDOs- exchange, mediating mechanism, disorder, robust order and memory--to help explain complex practice for RDANI.

The first perspective which pervades regional economic development discourse is values. The principle of producing policy order tends to favour narrowly conceived value-for-money-based assessments. An implication is that those projects with readily identifiable economic or financial outcomes are favoured, whereas projects that might exhibit broader values, including the potential for wider disruption leading to new periods of adjusted equilibrium, are often disregarded.

The first perspective which pervades regional economic development discourse is values. The principle of producing policy order tends to favour narrowly conceived value-for-money-based assessments. An implication is that those projects with readily identifiable economic or financial outcomes are favoured, whereas projects that might exhibit broader values, including the potential for wider disruption leading to new periods of adjusted equilibrium, are often disregarded.

Given RDANI's emphasis on being 'community based', this might imply that social considerations will shape their institutional values. However, we find in practice the rich layers of public values are ordered and managed through formal processes, such as formal consultation exercises, with economic or financial objectives and measurements prioritised, although not always in an explicit manner. This is consistent with neoliberal policy perspectives, where a narrower articulation of regional economic development, preoccupied with quantitative economic growth outputs, often comes to the fore. For example, RDANI review their performance against a narrow set of key performance indicators, or KPIs, which are a key instrument that emerged from New Public Management thinking. In Holland's (1992) terminology, exchange is focussed between economic actors, whereby the primary mediating mechanism becomes economic growth. Yet, this narrow approach to values misses the necessary features of creating disorder, which after self-organisation leads to a new robust order. In other words, a complex public values-based approach would embrace disorder by departing from prior practice as part of the process of searching for new states of robust order.

While value-for-money rationales are a common feature of regional economic development practice and neoliberal policymaking, a sign of encouragement is RDANI's attempt to achieve 'net benefit'. For example, their Regional Freight Study "identified the most worthy road upgrade projects ... utilising a benefit-cost framework" where "[p]rojects were ranked in order of net benefit, and the public versus private benefits of upgrades were identified" (RDANI, 2016: 8). The work has been used to advance road infrastructure projects in the region (e.g. the Mt Lindsey Highway, Thunderbolts Way). This points to the possibility for RDA Committees to develop their own values-based methods, which might depart from orthodox economic approaches. However, reconciling such bottom-up values with a top-down operational framework is likely to be a considerable challenge.

The second perspective--organisational form, governance and accountabilities--highlights the tension between organisational parameters (e.g. legal, policy, budgetary etc.) and the porosity of RDA Committees, which are themselves ensembles of elements within a broader complex system. RDANI seeks structure and order through, for example, structured review processes and planning documentation, although these were necessary requirements to satisfy government (RDANI, 2016: 7), in terms of providing an appearance of managing risks and uncertainty. RDANI embodies the principle of policy order in that it is an Incorporated Association in operation under the Incorporated Association ACT 2009 (NSW), where the organisation structure is predefined and where Committee members are appointed by the relevant Minister in the Commonwealth and State government, who also define the priorities and roles of RDANI (Commonwealth of Australia, n.d). This bequeathed structure imposes some limitations on RDANI, although achieving compliance may not necessarily be at the expense of working with complexity.

RDOs are nodes in partnership networks, yet RDANI does not have a directly elected democratic mandate, like Commonwealth, State, and Local government RDOs. Whilst some Committees include local elected members, many do not. A board of members is selected to 'represent' the views of diverse and often competing actors. Yet, it is not clear what mediating mechanisms beyond the pursuit of economic growth are deployed to attempt to reconcile the views of a multitude of actors from different civic, economic and social spheres. Indeed, formal consultation processes are ordered (and circumscribed) by institutional norms, processes and procedures. An alternative perspective, nevertheless, could challenge the relevance and efficacy of mediating mechanisms. In practice, the co-option of local policy elites (whether from public, private or voluntary sectors) is deployed as a favoured mediating mechanism. However, there is scope for RDOs to search out and interact with more diverse networks and actors. In such a circumstance, new actors would complement and enrich the complex system, which might in turn engender disorder and feedback, and subsequently new order.

The role and importance of geography in economic development is crucial to framings of regional economic development. As the third perspective, the policy order approach to defining operational territories features prominently in the work of RDANI. For example, the Regional Plan (intended in part to 'describe the region'), states that "RDANI is confident of improving the economic wellbeing of the more than 40 towns and surrounding districts in our region" (Stewart cited in RDANI, 2016a: 2). However, the geography of many RDA Committees bears little resemblance to functional regions, evidenced in some areas by competing regional structures and networks (Pugalis, 2016). However, this in itself is not a necessary precondition of working with complexity. In seeking to embrace and acknowledge disorder, RDANI could look to radically reconstitute its bounded geographical template. In a practical sense, this might include identifying and engaging with how other places interact, complement and compete with one another. Such networks, interactions and relationships might include finance, education, trade and charities. This approach would necessitate mediating mechanisms to facilitate disorder, feedback and new states of order.

Perspective four is concerned with actors and institutional relations. RDANI is positioned for policy order as the identified primary institutional space for mobilising different partners, capabilities and resources; performing "a pivotal role in fostering regional economic development RDANI will take a leadership role in bringing together organisations to take advantage of government programs, policies and initiatives, and will be an effective conduit between governments and regional communities" (RDANI, 2016: 8, emphasis added). In Holland's (1992) terms, RDANI seeks to impose, at least in a rhetorical sense, robust order on a complex system that has already developed new states of robust order. The issue for RDANI is that has been created from previous processes and interactions of disorder, feedback, adjustment and order. As such, this creates a constant tension between trying to impose an orderly and sequential project and trying to adapt to new states of robust order while maintaining the memory of RDANI's role in the past or previous programs. In other words, as the broader socioeconomic landscape evolves, there is a need for RDOs to adapt accordingly and not simply regurgitate prior practice. However, the extent to which RDA Committees are embedded within broader systems of government and governance is a moot point. For example, whilst the "Regional Plan is a key document for Regional Development Australia Northern Inland (RDANI)" (RDANI, 2016: 9), it is less clear if the plan is a key tool for the region. If this plan does not create exchange between the local ensemble of actors, then disorder and new states of order are unlikely to emerge. In this sense, the function of RDA Committees is often akin to a delivery agent for the state (acting as a post office of sorts or 'administrative post offices' in the words of a reviewer), whereby attempts to mobilise distinctive regional cultures and capabilities tend to be more circumscribed. The implication being the need for RDOs to recognise complex systems and manage complex relationships in new ways. If RDOs are operating as 'administrative post offices', then this has formed as memory in the system.

In an attempt to engage a broader range of actors, during 2010/11 RDANI held 21 meetings with 'community' members as part of 'exploratory conversations' intended to identify key strengths, assets, successes, challenges and opportunities. RDANI is also "looking at ways of obtaining structured feedback on both interaction and project activity during the course of each year" (RDANI, 2016: 7). Yet, given the points noted above about its organisational form and governance, which is similar to other RDA Committees, we contend that RDANI is often a marginal voice in policy decisions impacting regional economic development. In other words, it is a recipient of policy dictates, which reacts to funding calls and other 'opportunities'. Yet, this does not need to be the case. For example, RDA Committees might seek to utilise their memory to mediate between elements of the system and foster change as order and disorder within the system, particularly in terms of transforming intergovernmental relationships.

The fifth perspective relates to planning, prioritisation and projects. Interventions such as those with the aim of "attracting] at least five new enterprises providing tangible employment outcomes" (RDANI, 2016: 7), are somewhat divorced from the broader, complex challenges that the region faces. Such a goal is emblematic of trying to impose policy order rather than engage with societal complexity. A key issue for state-sponsored RDOs is that they become too focused on the needs of the provider of funds (Johnson et al., 2005), which have their own priorities and award funds with strings attached. Too often, 'chasing the funding' has become an end in itself (Pugalis, 2016). The result can be a situation where spending funds becomes a priority. It could therefore be argued that much practice is ordered by the funding envelope, which is an issue of much broader relevance beyond the case of RDANI and, even, Australia. The above analysis suggests that the operation of RDA Committees is driven by transactional relationships predominantly with government, which is indicative of top-down regional policy and which tends to ignore the day-to-day contemporary experiences of complex systems. Indeed, funding alone is unlikely to engender new states of disorder.

The final perspective relates to temporal matters. The Regional Plan claims to be 'comprehensive' as well as being a 'living document' that "presents a snapshot of Northern Inland NSW" (RDANI, 2016: 9). Yet, there is little evidence to suggest that it is a dynamic 'living' tool. As a published plan, it is intended to be finalised (albeit recognising the need for review). Indeed, the Regional Plan 2016-19 is a refreshed version of Regional Plan 2013-16 and Regional Plan 2010-13. Hence, there is sufficient memory (e.g. priorities, processes, implementation and lessons), which at present appears to be underutilised, and it is not clear how this regional plan embodies the diverse values, resources and capabilities of RDANI's ensemble of competing actors, and it is even less clear how these tensions might be reconciled or engaged in order to enhance outcomes. The principle of a living strategy document, nevertheless, offers substantial potential, and could for example result in new forms of engagement with diverse actors in the system, which in turn might create enough influence over the local complex system to initiate the emergence of new memory in the system. Admittedly, given that the current funding models available to RDANI often reflect the short-term nature of the policy cycle and the international move towards performance-based contracts, projects are unlikely to embrace all elements in the system, disrupt and engender new equilibriums.

5. CONCLUSION

Both government systems and regional socio-economic systems have been perpetually transforming and, these changes appear to have been accelerating over recent times. Observing a regional economic development paradox, where state-sponsored RDOs, such as RDA Committees, attempt to instil policy order in the face of societal complexity and complex practice, this paper has introduced complex systems as a frame to analyse the working methods of RDOs, by drawing on the case of RDANI in NSW. In seeking to inform contemporary debates concerning the efficacy of RDA Committees, which are in the midst of an independent review at the time of writing, we conclude, with some broader points derived from our findings, that regional economic strategies are preoccupied with providing the appearance of policy order and reflect a bias towards structured processes and transactional relationships, which eschew practical complexities. We have put forward an alternative way of understanding the operation of state-sponsored RDOs; considering their design, objectives, activities and potential for change.

At least four reasons can be put forward to help explain why there is a bias towards formalisation and simplicity as part of an attempt to impose or, at least, give the appearance of policy order in the face of complexity. The first reason is what Rubin (1988) has referred to as the "search for administrative certainty and task closure". This approach appears to override the capacity of the organisation to embrace complex practice, whereby interactions with other actors involve exchange and order and disorder, which may occur differently across time and space.

The second reason is that practitioners tend to favour systems and identifiable outputs to overcome a perceived lack of management and stakeholder understanding of their roles (Rubin, 1988). In the case of RDANI, this is apparent through the focus on tightly framed KPIs and quantifiable outputs that are divorced from addressing some of the most fundamental regional economic development challenges, such as industrial diversification and resource efficiency (RDANI, 2016), which transcend narrowly conceived, time-limited programs. This defensive, albeit pragmatic, approach limits an organisation's engagement with complex systems, and perpetuates traditional working methods and orthodox practice as it hinders the creation of new memory in the system.

The third reason is that RDOs avoid undertaking activities that seek to grapple with societal complexity, such as radical and/or high risk experiments for fear of failure (Ackoff, 2006). Stepping beyond the relative safety provided by the appearance of policy order and into complex and unknown realms heightens the chance of bureaucratic failure (i.e. targets not met).

The fourth reason relates to a lack of knowledge and understanding of complexity and, thus, limits its application (Ackoff, 2006). For example, complex systems are multifaceted, complicated and convoluted. This can be bewildering for actors to comprehend, and also makes for complicated communication with the general population. Consequently, there is a preference to stick with the status quo, whereby orthodox practice prevails. Taken together, these reasons constitute tactics deployed by state-sponsored RDOs that seek to help to minimise mistakes arising from their operations, which is an 'act of commission'. Embracing complexity would require a more radical shift in practice, which could result in an 'act of omission'. Therefore the default position is to maintain current operations and seek to minimise 'risks' (particularly reputational risks).

Some RDOs, similar to RDA Committees, are at risk of being stuck-in-the-middle of two evolving complex systems. On the one hand, governmental systems are being recast, which is altering the role of government, including the scope and delivery of services. On the other hand, the complexity of regional socio-economic systems continues to swell. It could be argued that the impact and performance of many RDA Committees is restricted by their design--they are often creatures of the state that lack democratic legitimacy, statutory powers, special purpose policy tools and goal independence (Pugalis, 2016). Compliance is rewarded over creativity. Crude output frameworks, measures of 'success' and KPIs insufficiently reflect transformational changes and encourage regional actors to game the system--adopting and striving for the appearance of policy order. Indeed, RDA Committees lack sufficient autonomy and credibility, as they have struggled to engender robust order. The implication is that many state-sponsored RDOs face the challenge of becoming more 'user-centred' rather than 'government choosing', whereby the 'benefactor' becomes an 'enabler'.

Notwithstanding the variegated nature and performance of specific RDA Committees, it is argued that the overall RDA program is tangential to local, regional and national development activities and ambitions. The national network of RDA Committees, for example, is largely divorced from other Australian Government policies (and public sector policies/services more generally) that are central to stimulating growth and development. Moreover, the influence of RDA Committees in shaping, stimulating and developing local economies is incidental, at best. As a result, RDA Committees are ill-equipped to provide a 'mediating mechanism' as they jostle for position alongside other bodies in "a complex and fragmented set of institutional arrangements for regional governance, intervention and decision-making processes" (Pape et al., 2016: 913). Indeed, RDA Committees often compete with rather than cooperate with other regional structures, such as, Regional Organisations of Councils and/or other state-sponsored RDOs as is the case in parts of Victoria. This can diminish leadership capacity, and can result in the duplication of activities. There is a major risk that some RDA Committees are peripheral actors in the 'ensemble of many elements' contributing to regional economic development. Alternatively, the RDA program could be repurposed as a key vehicle to help improve intergovernmental relations, encourage joint-working, and support local initiative and creativity (Pugalis, 2016). Yet, for this to be realised creativity would need to be on a par with compliance.

One perspective suggests that if RDA Committees are unable to help regions navigate the complexity of governmental and regional socioeconomic systems then they are likely to face an existential crisis. An alternative perspective suggests that the RDA Committee--including its antecedents--persists because of 'memory' in the complex system; changes have been absorbed and the elements have created a new equilibrium, which is what we observe of RDA Committees in late 2016. The robustness of the regional development committee despite the disorder created by periods of popularity and unpopularity, existence and non-existence, suggests a memory or pattern has been created that maintains this organisational form.

RDA Committees and similar state-sponsored RDOs can therefore choose to tread one of at least two paths: adapt and survive, or continue with business-as-usual and risk fatal demise. A greater appreciation of complex systems may be a pragmatic step as part of the evolution of state-sponsored regional economic development. After acknowledging the importance of complex systems, regional organisations can then begin to consider their role in such a system. Subsequently, interactions can be prioritised that engender adaptive responses, thereby disrupting order and orthodox practice. Whilst such a cultural shift (emphasising freedom to act) may be challenging, it might be a fruitful way of effecting regional performance and development in a context of financial restraints. One of the key learnings from complex adaptive systems theory is that actions, whilst not necessarily predictable, are interconnected in ways that "changes the context for other agents" (Plsek and Greenhalgh, 2001: 625) through a perpetual state of adaptation.

ACKNOWLEDGEMENTS: This paper is a revised version of a paper presented at the ANZRSAI Conference 2016 and subsequently published in the conference proceedings under the title--The regional economic development paradox: policy order and complex practice. We acknowledge the insightful comments and suggestions provided by Tony Sorensen, although the usual disclaimers apply and the views expressed in this paper are those of the individual authors and do not necessarily reflect the views of their respective institutions.

REFERENCES

Ackoff, R. (2006). Why Few Organizations Adopt Systems Thinking. Systems Research and Behavioral Science, 23, pp. 705-708.

Armidale Council (n.d.). Armidale Dumaresq and Guyra will Merge, Online version accessed 30 May 2017, http://www.armidale.nsw.gov.au/news/armidale-dumaresq-and-guyra-will-merge.

AusIndustry (2016). Innovation Stakeholder Map--Northern Inland. Internal Working Document, NSW, Australia: Australian Commonwealth Government.

Beer, A. (2007). Regionalism and Economic Development; Achieving an Efficient Framework. In A. J. Brown and J. A. Bellamy (Eds.) Federalism and regionalism in Australia: New approaches, New Institutions (pp. 119-134). ACT Australia: The Australian National University Press.

Beer, A., Clower, T., Haughton, G. and Maude, A. (2005). Neoliberalism and the Institutions for Regional Development in Australia. Geographical Research, 43(1), pp. 49-58. doi: 10.1111/j.1745-5871.2005.00292.x

Beer, A. and Maude, A. (2002). Local and Regional Economic Development Agencies in Australia. Adelaide, Local Government Association of South Australia.

Collits, P. (2008). The Learning Practitioner. Paper presented at the 12th annual SEGRA (Sustainable Economic Growth for Regional Australia) Conference, 18-10 August 2008, Albury, NSW.

Commonwealth of Australia (2015 a). Progress in the Australian Regions: State of Regional Australia. Department of Infrastructure and Regional Development, ISBN 978-1-925216-90-5

Commonwealth of Australia (2015b). Regional Development Australia Better Practice Guide. ACT Australia: Department of Infrastructure and Regional Development.

Commonwealth of Australia (n.d). My RDA. Regional Development Australia. Online version accessed 30 May 2017, rda.gov.au/my-rda/nsw.aspx.

Douglas, J. Carson, E. and Kerr, L. (2009). Local Government and Local Economic Development in Australia: An Iterative Process. Just Policy: A Journal of Australian Social Policy, 50(Apr 2009), pp. 54-61.

Gerring, J. (2004). What Is a Case Study and What Is It Good for? American Political Science Review, 98(2), pp. 341-354.

Gomm, R., Hammersley, M. and Foster, P. (2009). Case Study Method. Sage Publications Ltd. doi: 10.4135/9780857024367

Holland, J. H. (1992). Complex adaptive systems. Daedalus, 121(1), pp. 17-30.

Johnson, G., Scholes, K. and Whittington, R. (2005). Exploring Corporate Strategy; Text and Cases (Seventh ed.). FT Prentice Hall.

Klijn, E.-H., Koppenjan, J. and Termeer, K. (1995). Managing Networks in the Public Sector: A Theoretical Study of Management Strategies in Policy Networks. Public Administration, 73(3), pp. 437-454. doi:10.1111/j.1467-9299.1995.tb00837.x

Ladyman, J., Lambert, J. and Wiesner, K. (2013). What is a Complex System? European Journal for Philosophy of Science, 3(1), pp. 33-67. doi:10.1007/s13194-012-0056-8

Lewin, R., and Regine, B. (2003). The Core of Adaptive Organisations, Complex Systems and Evolutionary Perspectives on Organisations: The Application of Complexity Theory to Organisations. Chapter 8, Pergamon.

Massey, D. (2005). For Space. London: Sage.

Mitleton-Kelly, E, (2003). Ten principles of Complexity and Enabling Infrastructures. In E. Mitleton-Kelly (Ed) Complex Systems and Evolutionary Perspectives on Organisations; The Application of Complexity Theory to Organisations. Elsevier, pp. 23-50. ISBN 9780080439570

Mountford, D. (2009). Organising for Local Development: the Role of Local Development Agencies. Summary Report. Working Document. CFE/LEED, Paris, OECD.

Narrabri Shire Council (n.d.). Narrabri Shire Economic Plans and Strategies. Online version accessed 30 May 2017, http://narrabri.com.au/economic-development-1062.html.

NSW Department of Industry (2015). Economic Development Strategy for Regional NSW. Online version accessed 30 May 2017, http://www.industry.nsw.gov.au/invest-in-nsw/regional-opportuniti es/economic-development-strategy-for-regional-nsw.

NSW Department of Planning and Environment (2016). Draft New England North West Regional Plan 2036, NSW Government, Online version accessed 30 May 2017, http://www.planning.nsw.gov.au/~/media/Files/DPE/Plans-and-policies/new-england-north-west-draft-regional-plan.ashx.

NSW Government (2016). Hunter Regional Plan. Online version accessed 30 May 2017, http://www.planning.nsw.gov.au/~/media/Files/DPE/Plans-and-policies/hunter-regional-plan-2036-2016-10-18.ashx.

Pape, M., Fairbrother, P. and Snell, D. (2016). Beyond the State: Shaping Governance and Development Policy in an Australian Region. Regional Studies, 50(5), pp. 909-921. doi:10.1080/00343404.2015.1055461

Patton, M. Q. (2011). Development Evaluation: Applying Complexity Concepts to Enhance Innovation and Use (Vol. 1). New York London: The Guildford Press.

Plsek, P. and Greenhalgh, T. (2001). The challenge of complexity in health care. British Medical Journal, 323, pp. 625-628.

Pugalis, L. (2016). IPPG Submission to RDA Independent Review. Sydney, NSW: Institute for Public Policy and Governance.

Pugalis, L. and Gray, N. (2016). New Regional Development Paradigms: an Exposition of Place-Based Modalities. Australasian Journal of Regional Studies, 22(1), pp. 181-203.

Pugalis, L., and Tan, S. F. (2017). Metropolitan and Regional Economic Development: Competing and Contested Local Government Roles in Australia in the 21st Century. Proceedings of the ANZRSAI Conference 2016--Towards the Future: Emerging Priorities in Regional Policy and Practice, RMIT, Melbourne, 5 -7 December, pp. 121-136.

Pugalis, L. and Townsend, A. R. (2012). Rebalancing England: Sub-National Development (Once Again) at the Crossroads. Urban Research and Practice, 5(1), pp. 159-176.

Pugalis, L. and Townsend, A. R. (2013). Trends in Place-Based Economic Strategies: England's Fixation with 'Fleet-of-Foot' Partnerships. Local Economy, 28(7-8), pp. 693-714.

Regional Development Australia Northern Inland (RDANI) (2016). Regional Plan 2016-2019. Online Version accessed 18 January 2017, http://www.rdani.org.au/files/pages/our-region/regional-plan/RDANI_Regional_Plan_V4_Web.pdf

Regional Development Australia Northern Inland (RDANI) (n.d.a). About us. Online Version accessed 18 January 2017, http: //www. rdani .org. au/about-us.php

Regional Development Australia Northern Inland (RDANI) (n.d.b). Northern Inland Information and Data Resources. Online Version 18 January 2017, http://www.rdani.org.au/our-region/regionaldata.php.

Regional Australia Standing Council (2013). Framework for Regional Economic Development. Canberra: Department of Regional Australia, Local Government, Arts and Sport.

Rubin, H. J. (1988). Shoot Anything that Flies; Claim Anything that Falls: Conversations with Economic Development Practitioners. Economic Development Quarterly, 2(3), pp. 236-251. doi:10.1177/089124248800200304

Sorensen, A. (2016). Community Development in an Age of Mounting Uncertainty: Armidale, Australia. In G. Halseth (Ed) Transformation of Resource Towns and Peripheries: Political Economy Perspectives, Oxon, Routledge.

Stimson, R. J., Stough, R. R. and Roberts, B. H. (2006). Regional Economic Development: Analysis and Planning Strategy. Berlin, Heidelberg: Springer Berlin Heidelberg.

Taleb, N. (2012). Antifragile: Things that gain from disorder. London: Penguin.

Tamworth Regional Council (n.d.) Economic Development. Website 30 May 2017, http://www.destinationtamworth.com.au/Economic-Development-/ tamworth-regional-council-economic-development-strategy.

Wilkinson, J. (2003). Regional Development Outside Sydney, NSW Parliamentary Library Service, NSW Government. Online version 18 January 2017, https://www.parliament.nsw.gov.au/researchpapers/Documents/reg ional-development-outside-sydney/13-03.pdf.

Ziafati Bafarasat, A. (2016). Meta-Governance and Soft Projects: A Hypothetical Model for Regional Policy Integration. Land Use Policy, 59, pp. 251-259. doi: http://dx.doi.org/10.1016/Uandusepol.2016.09.004.

Lee Pugalis

Professor of Urban Studies, Institute for Public Policy and Governance, University of Technology Sydney, Ultimo, NSW,

2007, Australia.

Email: lee.pugalis@uts.edu.au

Darren Keegan

Independent practitioner-researcher

Caption: Figure 1. Location of RDANI and Local Government Composition. Source: RDANI, 2016.

Caption: Figure 2. Structure of the Northern Inland Economy. Source: care 2011.

Caption: Figure 3. Northern Inland Innovation Network. Source: Ausindustry, 2016.

Table 1. Policy Order and Complexity: Summary of Analysis.

                   The appearance of
Perspectives       policy order              Working with complexity

Values             Narrow value-for-money    Broad, evolving and
                   assessments typically     dynamic public values
                   confined to specific      reflecting qualitative
                   programs and projects     as well as
                   cloaked in broader        quantitative metrics
                   public value discourse

Organisational     A formal and planning-    A dynamic and
form, governance   based not-for-profit      unbounded
and                organisation operating    organisational
accountabilities   within certain legal      network, multi-actor
                   parameters, a             involvement and
                   governance board and      governance, and
                   accountabilities in       multiple
                   terms of performance      accountabilities to
                   management                different constituents

Geography          A defined territory       Multiple, overlapping
                   based on composite        and dynamic
                   local government          functional'
                   administrative areas      geographies working
                                             with geographical
                                             complexity

Actors and         Hierarchical              An ensemble of often
institutional      reporting relationship    competing actors with
relations          with formal tiers of      diverse objectives
                   government and formal     that mutually self-
                   consultation with         adjust
                   'other' stakeholders

Planning,          Formalised, scheduled     Entrepreneurial,
prioritisation     and transactional:        flexible and
and projects       given shape 'from         relational interactions
                   above' by national        from multiple
                   and state policies        directions and
                   and 'from below' by       dimensions
                   local government
                   policies

Temporal           Funding parameters        Porous and emergent
                   conditions delivery       visions and strategies
                   timescales, which         to deliver a mix of
                   tend to be short-term     long-term and short-
                   --reflecting political    term projects
                   budgetary cycles

Source: Authors' analysis