SEARCH WITHIN CONTENT
VOLUME 2017 , ISSUE 2 (June 2017) > List of articles
Citation Information : Evidence Base. Volume 2017, Issue 2, Pages 0-0, DOI: https://doi.org/10.21307/eb-2017-002
License : (CC BY-NC-ND 4.0)
Published Online: 03-August-2017
One-stop-shops (OSSs) are not a recent invention, but they are currently popular as governments attempt to address complaints about fragmentation in public service delivery. OSSs often struggle with multiple implementation problems. This review assesses ten years of research into government one-stop-shops, and seeks to determine drivers and barriers relevant to the adoption and implementation of OSSs in the public sector. OSS reforms are frequently promoted as win-win solutions, allegedly delivering seamless service at lower cost. Such promises are appealing to political decision makers, but typically understate the complexities, costs and risks of the OSS approach. When the time comes for implementation, managers often struggle with unrealistic political and citizen expectations, turf conflicts, inadequate resources, and dilemmas relating to the need for administrative specialisation. OSSs tend to be more successful where the goal of seamless service is pursued gradually, with sufficient resources, and where implementers address trade-offs between integration and specialisation.
One-stop-shops (OSSs) are a popular tool of government service delivery. An OSS is a place – physical, virtual, or both – where the public can obtain multiple products and services (Reid and Wettenhall 2015). OSSs are not a recent invention. The private sector department store model, which emerged in the mid-19th century, is one example, while government interest in integration of public service delivery dates back at least to the 1970s (Sharkansky 1979).
The OSS concept has gained new popularity with governments in recent years, for several reasons. There is growing concern that public service delivery is excessively fragmented, leading to duplication and therefore inefficiency, and to poor outcomes for vulnerable service users. There is also an increasing tendency to regard and treat government service users as ‘customers’ who should not be inconvenienced by having to deal with multiple dispersed and disconnected service providers (Dutil et al. 2008; Rosenthal and Peccei 2006). One-stop-shops have been implemented across the OECD, and examples can be found at all levels of government from local to national, and in service areas ranging from social assistance and health care to environmental regulation and business licensing.
Despite their growing use, one-stop-shops enjoy mixed support with governments. Furthermore, they encounter substantial implementation difficulties. This article presents a systematic review of a decade of research into the OSS model in government. It seeks to identify the factors that drive the adoption and successful implementation of OSSs. It therefore addresses two questions:
What explains government decisions to adopt the one-stop-shop model for service delivery?
What factors hamper or facilitate the successful implementation of the one-stop-shop model?
I define Successful implementation with reference to the three major benefits that are commonly claimed to flow from the OSS model. These are:
This article addresses peer-reviewed research into one-stop-shops published since 2005. I focus on work that explicitly addresses the drivers of and barriers to adoption and implementation of the OSS approach. In addition, and to provide the broader context in which OSS reforms have occurred, I cite general works on public administration and service delivery transformation where relevant.
The literature on OSSs identifies a large number of drivers and barriers. I summarise these findings, and in the process make two overarching theoretical observations. Firstly, decisions to adopt the OSS model tend not to be driven by ‘rational’ policy processes where the costs and benefits of the OSS option are rigorously weighed and compared to other service models. Rather, adoption decisions reflect the influence of administrative fashion (Bannister and Connolly 2012), combined with opportunistic efforts by policy entrepreneurs or ‘service champions’ to sell OSSs as ‘win-win’ solutions, allegedly enabling the delivery of better services at lower cost (cf. Halligan and Wills 2013).
The article’s second theoretical observation concerns the factors that drive successful implementation. I argue, following Christensen and Lægreid (2012), that OSSs tend to achieve service integration at the expense of ‘process specialisation’, leading to reduced effectiveness and efficiency in the delivery of some services. As a result, implementation is most successful and sustainable when the one-stop shop goal is carefully balanced with a traditional ‘siloed’ approach to service production and delivery.
Based on the currently available research evidence, the review finds that OSSs are not a win-win solution for the problems of government service delivery in a complex, fragmented and cost-constrained administrative environment. Instead, there are important trade-offs in their implementation. The review also points to several important limitations of existing research into the implementation of OSSs. These include a lack of systematic evidence concerning the role of administrative culture in service integration, inadequate quantitative data on the budgetary costs and benefits of OSSs, and a lack of attention to the impacts of OSS reforms on citizen satisfaction.
The review is structured as follows. The next section outlines the literature search procedure. I then investigate the reasons that governments decide to adopt the OSS model, and explore the factors that help and hamper implementation. The discussion addresses theoretical and practical implications of the review findings, and directions for future research.
I conducted a review of literature addressing the effectiveness of OSSs from 2005 to 2015. To qualify for inclusion, studies had to be empirically informed and subject to blind peer reviewing. The review was restricted to OSSs in OECD countries, to rule out the different considerations that come with administrative reforms in developing countries.
To locate appropriate articles for review, I searched the following databases for articles that discussed OSSs: Science Direct, EBSCHO Host, Annual Reviews, Web of Science, ProQuest, Informit, Taylor and Francis Online, and Google Scholar. The search criteria are outlined in Table 1.
The search for ‘one-stop-shop’ and cognate terms (Table 1) located a total of 4974 publications. I added the term ‘government’, and also reviewed abstracts and introductions to eliminate articles that used the keyword in contexts other than service delivery (for example, ‘one-stop-shop’ is often used metaphorically to describe a single location where information can be obtained, such as an online clearinghouse for statistical information). This reduced the number of publications to 138.
The 138 articles were then reviewed to remove those that did not discuss the drivers of OSS adoption or effectiveness. Some articles describe the implementation of the model without addressing what worked or did not work. Still others put forward the OSS model as an ideal to aspire to in a particular service delivery context. Twenty-six of the 138 articles discussed factors that might influence adoption and effectiveness.
Table 1 in the Appendix summarises the research approaches used in the 26 studies reviewed for this article. I have endeavoured to provide as much detail about methodology as was provided in the original publications.
The summary suggests several overarching themes concerning empirical research into one-stop-shops. Firstly, the literature is almost entirely based on case studies, where single examples of efforts to implement one-stop-shops are used to draw conclusions about the drivers and barriers to service integration. Case studies make sense in this area because the factors that drive success are often highly complex and require detailed understanding of the complexities of specific cases. While case studies have the potential limitation of being difficult to generalise, there are numerous examples of comparative case studies, where multiple jurisdictions’ efforts to implement OSSs are compared to find common and diverging drivers. The dominant methodology is in-depth qualitative interviewing, usually of senior policy makers and managers, but often also including middle managers and front line staff. Statistical research into OSSs appears to be rare. OSSs in the welfare administration field appear to have received significant attention in the literature, and general government service portals have also attracted considerable analysis. Studies rely heavily – sometimes exclusively – on official documents for information. Some studies had very small sample sizes, and some papers left key aspects of the research design, such as data sources and sample sizes, unspecified. Findings derived from studies with small samples and uncertain methods should be treated with caution.
A potential problem for the review approach is bias due to selection on the dependent variable. The literature search only included cases where an OSS was actually adopted and implemented. Since I did not study cases where OSSs were not adopted, I exercised due caution when drawing conclusions about the factors that drive the adoption and implementation of OSSs. However, the studies reviewed go some way to alleviating the dependent variable problem. Several studies explored cases where OSSs were formally adopted but then effectively abandoned (Dutil et al. 2010; Howard 2014). Furthermore, the literature is replete with discussions of implementation failures, where integration was attempted but not achieved or achieved only to a modest degree. There were also cases where OSS integration was realised but later reconfigured and partly reversed as implementation challenges emerged. Thus, while I only address cases of OSS adoption and implementation, these studies do provide an account of failures and limitations, thereby minimising dependent variable bias.
One-stop-shops (OSSs) are often seen as exemplifying a broad shift in governance and administration towards more citizen-centric principles and practices (Christensen et al. 2007; Reid and Wettenhall 2015). They ostensibly reflect increasing interest in bottom-up processes that take as their starting point the needs of users. Despite these post-bureaucratic, ‘grassroots’ pretentions, the OSS literature shows these initiatives almost invariably require top-down political endorsement to proceed (Anthopoulos et al. 2007; Askim et al. 2011; Blackburn 2014; Christensen et al. 2007; Flumian et al. 2007; Gagnon et al. 2010; Heenan and Birrell 2006; Reinwald and Kraemmergaard 2012). To explain the implementation of OSSs the first question is why do governments decide to adopt OSSs to deliver services? I suggest theories of administrative fashion and policy entrepreneurship best explain OSS adoption. As I shall point out later in this article, the opportunistic nature of OSS adoption decisions has significant implications for implementation.
OSSs appear to be a sensible solution to the challenges of contemporary service delivery. From this perspective OSS adoption decisions might be interpreted as “rational policy making” (Davis et al. 1988). Governments confront problems of service fragmentation and inefficiency, along with citizen dissatisfaction with public services leading to voter disapproval. According to the rational interpretation, governments consider various policy solutions such as contracting out, privatisation, decentralisation to regional and local government, and partnerships with non-government agencies. Each of these approaches has benefits and drawbacks, and governments choose OSSs where they regard integration and seamlessness as the best ways to realise their service delivery goals. In this view, OSSs are currently popular because of heightened citizen demands and expectations, alongside dissatisfaction with fragmentation, coinciding with the arrival of new technologies such as internet portals, which together make OSSs an attractive and technically feasible solution (Flumian et al. 2007; Gagnon et al. 2010; Kernaghan 2005).
The literature suggests this model simplifies and caricatures OSS adoption decisions. The rational policy making model has been widely discredited in the policy literature for its unrealistic assumptions that governments a) act as unified decision makers with one set of clearly defined goals and b) comprehensively consider the benefits and drawbacks of all alternative policy instruments (Allison 1971; Dror 1964; Cohen et al. 1972). The rational model has been extensively challenged by post-rationalist approaches as an inadequate description of how policy adoption occurs. Common to these post-rationalist approaches is the notion that governments do not clearly specify goals and problems in advance of choosing solutions, that it is impossible to engage in a comprehensive search for options, and that there are several decision makers rather than just one, each with their own perspectives and interests.
Post-rationalist explanations for policy adoption suggest policy decision makers make choices based on a high degree of uncertainty, pragmatism and opportunism. Cohen et al. (1972) propose the ‘garbage can’ as a metaphor, suggesting that policy adoption involves decisions by actors who are unsure of their own goals and of the consequences of choosing particular options. There are many solutions sitting in the metaphorical garbage can, having been developed in the past but dispensed with. OSSs are a prime example: the idea of a single storefront offering all services required by a particular client group is not new in government or the private sector (Askim et al. 2011; Gulick 1937; Reid and Wettenhall 2015; Sharkansky 1979). These solutions are occasionally retrieved from the garbage can by governments facing pressure to do something about a problem. Kingdon (1984) advances this approach by suggesting that problems, policies and politics exist as separate ‘streams’. Policy entrepreneurs, who may be politicians, bureaucrats or external actors, play a key role in creatively framing the problem so that it links up with an existing policy solution, and in securing political support for new problem framings and solutions.
Kingdon’s notion of streams helps explain the adoption of OSSs. The problem stream in this instance comprises a series of problematisations regarding ‘fragmentation’ of government service delivery. If one embraces NPM discourse and regards the service user as a ‘customer’, fragmentation leads to ‘inconvenience’ and ‘dissatisfaction’ (Clarke 2007; Dutil et al. 2008). If the service user is an entrepreneur seeking approval to start a business or develop natural resources, fragmentation leads in neoliberal discourse to ‘red tape’ and ‘economic burdens’, as multifaceted regulatory approval processes slow and discourage investment (Ongaro 2004). Welfarist service delivery discourse is concerned with the service user as an economically disadvantaged citizen potentially suffering a multitude of interconnected health and social problems, and thus fragmentation can mean vulnerable citizens are denied necessary supports (Hall et al. 2012). At the same time, in post-welfarist ‘activation’ discourse, fragmentation means citizens can get away with depending on government support without appropriate pressure to participate in government employment activation programs (Lindsay and McQuaid 2008). From a legitimacy point of view, fragmentation can be seen to undermine accountability of and trust in government (Heintzman and Marson 2005). Finally, fragmentation can be problematised as fostering inefficiency due to duplication and contradiction (Dutil et al. 2010; Heenan and Birrell 2006). Thus, the literature shows that the ‘fragmentation problem’ in service delivery is actually a series of (at best) loosely related problematisations.
The second stream is made up of policies, or solutions. In recent years there has been much interest in exploring ‘alternative service delivery’ models to solve the problems of existing service arrangements. These include models such as contracting out, internal markets, privatisation, regional and local devolution, self-service, and partnering and collaboration within government and between government and non-government providers (Ford and Zussman 1997). The notion of a one-stop-shop acts as a metaphor for a particular goal of service delivery – a single point where users can go to get the services they require (Reid and Wettenhall 2015). Other metaphorical terms used to describe this goal are ‘single window’ and ‘one window’ service.
These terms cover a spectrum of service arrangements. Kubicek and Hagen (2000) distinguish between three ‘levels’ of OSS. The first is a relatively superficial first stop shop (physical or virtual) where users enter and are then directed onward to service providers, who remain separate. The second level is the convenience store, where different agencies locate themselves together so users don’t need to move around. The third level is the true one-stop-shop, where users can obtain all services from the one organisation. One stops shops can be realised by creating new, dedicated service organisations, or as horizontal partnerships between separate organisations. They may involve non-government providers, and may integrate services from various levels of government. They have historically been offered through the physical ‘channel’, but the advent of e-government has opened the possibility of virtual OSSs. Thus, the OSS model is not a homogenous policy solution, but an umbrella term for a range of old and new strategies and technologies designed to achieve a degree of integration in service provision.
Our third stream is politics, that is, the political interests and agendas surrounding service delivery. The literature on OSSs suggests the political stream is dominated by several considerations. There is an interest in securing greater public satisfaction with government programs (Flumian et al. 2007; Hedestig and Söderström 2008). There is also a strong political desire to save money (Gagnon et al. 2010; Halligan and Wills 2013). Political actors tend to favour proposals that simplify administrative arrangements. Unsurprisingly, governments are especially receptive to reforms that can achieve all of these political objectives, i.e. increased service effectiveness, improved citizen satisfaction, simplification, and greater cost efficiency (Pollitt and Bouckaert 2004). It is worth noting, however, that service delivery reforms often fail to attract the interest of political actors – and where they do, this interest is hard to sustain, especially when reform turns to ‘unglamorous’ technical and organisation matters (Bannister and Connolly 2012; Dutil et al. 2010).
These three streams do not necessarily connect. The literature suggests connections are usually the product of a prominent problem recognition, coupled with a quite vague and ambiguous OSS proposal (Anthopoulos et al. 2007; Christensen et al. 2007; Rosenthal and Peccei 2006) and an appeal to key priorities in the political stream, namely increased citizen satisfaction, simplification, and reduced cost. The research finds numerous examples of policy entrepreneurs who engage in this linking exercise. In the service delivery literature these entrepreneurs are often called ‘service champions’ (Howard 2010; Ongaro 2004).
Champions of OSS reforms have been prominent in Australia and Canada, two early pioneers in the movement towards ‘citizen centred service’. These champions – mostly current and former public servants - sought new rationales for government involvement in service delivery (Bourgault and Gusella 2001). In Canada, champions succeeded in selling OSSs as solutions to problems of fragmentation, and as means to achieving political goals of user satisfaction and cost reduction (Howard 2010). In Australia, support for the Centrelink OSS emerged as the result of a fortuitous alignment of the interests and objectives of the government of the day, central agencies, and line departments in the welfare field (Halligan and Wills 2013).
Thus research suggests decisions to adopt the OSS model are not straightforward rational policy processes. OSSs have existed as policy solutions for many decades, but they have been recently resurrected by policy entrepreneurs seeking to defend and recast the role of government employees and organisations in the delivery of services. Their selective adoption in recent years reflects contingent alignments of problems, policies and politics, in which champions have used ‘symbolism’ and ‘figurative language’ (Askim et al. 2011; Christensen et al. 2007) to emphasise the rationality of OSS adoption, while downplaying complexities, contradictions and risks in the models. In such processes, decisions about the level, channel and organisational basis of the OSS may be ignored, or deliberately deferred to the implementation phase.
In the next section I explore barriers and drivers in the OSS implementation phase. As we shall see, the entrepreneurialism and opportunism of OSS adoption processes has important consequences for implementation dynamics.
This section addresses the barriers and drivers to successful implementation of one-stop-shops. The barriers and drivers identified in the literature can be classified into four groups: turf, culture, resources and capabilities.
In bureaucratic systems, turf – organisational responsibilities and resources – is at the forefront of most senior officials’ minds (Dunleavy 1985; Peters 2010; Wilson 1989). OSSs have potentially substantial implications for turf. Studies of OSS implementation find that concern with turf can produce bureaucratic resistance to integration. Askim et al. (2012, 1464) found that Scandinavian OSSs, involving collaboration between central and local government, were unstable because each level persistently sought to take control of the OSS. In the case of Service Tasmania, a first stop whole-of-government OSS, client agencies resisted cooperation with the OSS, delivering “slow processing times, inconsistent advice and poor support” (Blackburn 2014, 108). The effort to implement a similar gateway in Alberta, Canada, encountered resistance from social and health ministries who felt the generalist staff of the OSS would not be capable of delivering their programs at the street level (Howard 2015). Similarly, the OSS for health and social care in Northern Ireland has had to address professional jealousies, with some agencies and professional groups reluctant to collaborate because of fears the “hegemony of health” would monopolise funding (Heenan and Birrell 2006, 60). As Choudrie and Weerrakody (2007) note, “[f]or the success of a one-stop-shop environment each one of the parties must give up some power in order to truly reap the benefits of sharing across horizontal and vertical levels” (p. 36).
Turf issues are not restricted to bureaucratic rivalries. Involvement of non-government partners and contractors can exacerbate conflicts. Howard (2014) studied the attempt to implement OSSs in Alberta and found it was substantially impeded by the large number of established private service delivery contractors. In that case, while the government embraced the OSS idea in principle and agreed to implement it, political leaders were unwilling to bear the political cost of revoking contracts to private providers (cf. Aucoin 2002; Christensen and Lægreid 2011). Alberta ended up creating a compromise arrangement, with an electronic ‘first stop shop’ directing users to private contractors in their local areas.
What factors can mitigate turf issues? There appear to be two broad approaches: reorganisation and partnership. In the reorganisation model, there is an effort to overcome turf rivalries by combining functions into a single agency. Thus Heenan argues the OSS for health and social services in Northern Ireland has been relatively successful because “there is one agency, one employer, one vision, shared aims and objectives, [and] one source of funding” (Heenan and Birrell 2006, 55). Up until that point, “in the absence of a single health and social care budget, care managers were unable to commit resources from budgets that were outside their control” (Heenan and Birrell 2006, 55; see also Kernaghan 2005). Yet creating a new organisation has the potential simply to internalise turf conflicts. Part of the success of the OSS in Northern Ireland is attributable to managerial efforts to ensure different professional groups within the new OSS felt they had voice and could all aspire to management positions (Heenan and Birrell 2006). Turf conflicts can also be reduced where the client ministries themselves are merged, as was the case in Norwegian social and employment policy, meaning there was one national minister responsible for policy, programs and delivery (Fimreite and Laegreid 2009).
The possibility of merging ministries is forestalled where multiple levels of government are involved. Sometimes there is also too much political resistance to organisational integration. In these situations, informal partnerships and collaborations can be more effective. Partnerships have been used to achieve ‘convenience store’ OSSs (co-location) in areas of political turf sensitivity. This model worked successfully as a compromise in the Norwegian welfare administration for a period, but it tended to be quite unstable, with strong reassertion of distinct agency and jurisdictional roles over time (Fimreite and Laegreid 2009; see also Lindsay and McQuaid 2008). Fimreite and Laegreid (2009) observe that partnership works best where there is some hierarchical pressure on actors to cooperate, suggesting that ‘metagovernance’ can be a driver of effective service integration (cf Lindsay and McQuaid 2008).
Ongoing political support can be a powerful tool for addressing turf concerns. Continuous political support was viewed as critical to the success of initiatives in New Brunswick (Canada) and Tasmania (Australia) in overcoming the resistance of client agencies to the creation of a unified service interface (Blackburn 2014; Dutil et al. 2010). Political support can waver in the implementation phase, especially if the full complexity and costs of the OSS were downplayed during the adoption decision phase (Howard 2014). In this context, several studies point to officials who sought to make tangible short-term gains and to demonstrate these incremental milestones to politicians (Flumian et al. 2007; Howard 2014; Reinwald and Kraemmergaard 2012).
Culture is often cited as a challenge in one-stop-shop implementation. Observers suggest that cultural transformation is fundamental to the realisation of joined up service delivery (Flumian et al. 2007; Kernaghan 2005). Flumian et al. (2007: 566) argue that cultures of ‘vertical’ or ‘silo’ thinking are a major barrier to integration:
Not only do employees tend to look vertically rather than horizontally, they are also inclined to support departmental, rather than inter-departmental and inter-governmental, initiatives. Operational and managerial barriers also inhibit the development of shared values.
These necessary shared values include “a culture of service excellence among employees” (Flumian et al. 2007, 567). Heenan and Birrell (2006, 63) similarly suggest “culture must permeate all levels of service planning and provision in order to provide an integrated mindset.” Ongaro (2004) studied OSSs for business approvals in Italy and argues that a “user oriented culture” was essential to their successful implementation.
How can such cultures be fostered? One proposal is to institute various customer-service promotion mechanisms, such as service charters, satisfaction surveys and offices of customer satisfaction (Flumian et al. 2007). Performance standards and measures, along with quality audits of front line delivery, may change cultural attitudes by refocusing work effort towards client service goals (Gortmaker et al. 2005). Giving front line staff voice to express concerns about delivery systems, and allowing these staff input into the design of physical and virtual service interfaces, have also been proposed as positive drivers of cultural change (Blackburn 2014; Hall et al. 2012; Reinwald and Kraemmergaard 2012). Several case studies attributed these measures to the subsequent success of OSS initiatives, but they do not explore cases where these measures were not present, so we cannot draw firm conclusions about their importance.
Despite the prominence given to cultural drivers in the OSS literature, the empirical testing of the impact of such variables is limited. The review found no efforts to explicitly define culture as a variable, or to measure it systematically. There is little effort to distinguish the impact of culture from related factors, such as turf, resourcing and specialisation. Furthermore, the suggestion that replacing all ‘vertical mentalities’ with ‘customer cultures’ will lead to better outcomes is problematic. As Fimreite and Laegreid (2009: 295) argue,
it is, however, also important to pay attention to the fact that the ‘silo mentalities’ these structures and reform initiatives are supposed to attack often exist for very good reasons … Well-defined vertical and horizontal organizational boundaries are not only a symptom of obsolescent thinking … Very often they are justified by principles underlying our political structure, such as division of power, predictability, impartiality, rule-of-law, professional confidentiality and protection of privacy. Breaking these boundaries may pose new challenges to the political administrative system.
The claim that culture is a key driver of successful OSS implementation should therefore be treated with caution. The widespread attention to culture suggests it may play a significant role, but more research is necessary to isolate its importance for effective service integration.
As we have seen, OSSs are frequently sold as policy solutions that simultaneously enhance service quality and cut administrative costs (Halligan and Wills 2013). This review found no firm evidence that OSSs save money. Despite this, the assumption that OSSs reduce costs persists and has negative consequences for effective implementation because it leads to a variety of resource deficiencies and dependencies.
OSSs typically incur significant start-up costs (Howard 2014). Physical OSSs need to renovate storefronts or acquire new space; there are also costs for branding and marketing. Significant resources must be expended to reorganise management and staffing structures, and new staff may be required. Training in new processes needs to be designed and delivered. Substantial expenditures are usually required in the development of new ‘back end’ administrative processing systems. If the OSS includes or is primarily organised around a virtual portal, expenditures in web architecture can be very significant.
The literature suggests inadequate provision for start-up expenses is common in OSS implementation. For example, Service Alberta was required to meet these costs from its existing budget, on the assumptions that costs savings would be realised immediately and offset new expenditures (Howard 2014). A similar problem was in evidence at Service Canada, where there was a lack of funding dedicated to the integration process itself, meaning agencies had to take money from their own budgets – which they were reluctant to do (Flumian et al. 2007). When Australia’s Centrelink agency was created the government immediately imposed substantial budget cuts, on the logic that there would be significant savings from merging two agencies into one (Halligan and Wills 2013). Centrelink was forced to increase the size of front line caseloads (Howard 2012). This generally reduces staff and customer satisfaction, especially in social policy delivery where there are complex needs to address (Hall et al. 2012). In the case of virtual OSSs, lack of resources can be addressed to a degree by purchasing ‘off the shelf’ systems rather than having custom web architecture built, but these risk not being aligned with the specific constellation of services offered (Anthopoulos et al. 2007).
Resource deficiencies create other start-up problems. Even in OSS ‘success stories’, training is cited as a barrier to effective implementation (Blackburn 2014: 109):
training has been difficult due to time and financial restrictions … Interview transcripts indicate that many CSOs felt that training was inadequate and cite the hurried implementation as a cause. They felt that training was rushed, especially due to the multitude of services that they are expected to offer. This issue was a frequent occurrence in interviews.
Furthermore, many studies point to unrealistically ambitious timeframes for implementation: “experiences from other countries (among them Australia) that have established similar one-stop shop services show that attention must be paid to the fact that building a whole-of-government system is a long-term project that takes time to implement” (Christensen et al. 2007, 406). One study (Bannister and Connolly 2012) estimates that ‘desiloisation processes’ within public organisations, where systems are merged to make them ‘interoperable’, can take up to 15 years to complete. Time is particularly crucial where collaborations and partnerships are used to integrate services: “Forced collaborations that are not given adequate time to develop naturally are rarely effective” (Murray et al. 2014, 119).
Finally, a lack of resources is also frequently evident in the failure to invest in adjusting and redesigning administrative processes to cope with changed service delivery arrangements. Studies document numerous cases where little attention is given to ‘back end’ or ‘back office’ processes (Hedestig and Söderström 2008, 7).
research has shown that many one-stop-shop solutions in local governments have spent most of their effort on delivering fast and reliable service to citizens, i.e. resources have been spent on front-office. Spending most effort on front office activities has often resulted in less effort spent on the work of establishing an effective organisation that integrates front-office with back-office activities
Scholl and Klischewski (2007, 890) observe that when e-government projects are driven by an OSS goal they often downplay the “enormous obstacles of back end cooperation”. Hedestig and Söderström (2008, 1) conclude that successful OSS projects must:
(1) Pay equal attention to the front-office and the back-office perspective. Reengineered business processes and new work routines will not show up by themselves, and (2) When developing one-stop-shops as the CSU make sure to also develop models for the relationship between the one-stop-shop and the rest of the organization.
Thus while OSSs are often sold on their money-saving potential, there is little evidence such savings are realised in practice. However, the efficiency assumption powerfully shapes funding models for OSS implementation, leading to numerous start-up difficulties and ongoing problems, including inadequate back-end integration, ambitious implementation schedules, poor training, and insufficient staff-user contact time, resulting in dissatisfaction.
A final factor affecting OSS implementation is the capability of staff to deliver services that adhere to policy rules, meet clients’ needs, and are efficient. We have seen that these capabilities depend to a degree on the provision of adequate training and sufficient developmental time, both of which are often lacking in OSS initiatives due to budgetary restrictions. Yet there are additional structural obstacles to the development of staff capabilities necessary for integrated service provision within one-stop-shops.
The literature on specialisation and work division is relevant to this question (Christensen and Lægreid 2012; Gulick 1937). As Gulick (1937) emphasised, the complex work performed in modern organisations must be divided both vertically and horizontally. Vertical division is between those who oversee/coordinate tasks, and those who carry them out. Horizontal division is between different workers (or teams of workers) who carry out particular tasks. Division permits specialisation, meaning the increased competence and efficiency secured when workers concentrate their efforts on a subset of the full production process. Organisations face a series of choices about how work should be divided. The traditional ‘silo’ approach in public services divides work by ‘purpose’ and ‘process’, with staff arranged into organisations (departments) devoted to particular government purposes (e.g. health care, defence or environmental regulation), and then within those organisations, staff specialise in processes (surgery, nursing, hygiene, mechanical maintenance, accounting, etc.).
By contrast, OSSs divide work by ‘client’ and ‘place’; that is, offices carry out all functions for a group of service users in a particular location. The extent to which the OSS specialises in a specific sub-group of government client varies: some ‘whole of government’ OSSs like Service Canada or Service Tasmania have a broad ‘task portfolio’ (Askim et al. 2011) and see clients across the full range of government services, while others, including Australia’s Centrelink and the UK’s JobCentre Plus, focus on clients requiring social assistance.
A core problem that arises in the context of OSS implementation is the range of staff capabilities needed to carry out all the processes involved in serving the target group of users. While staff can in theory focus on providing ‘holistic’ service for the clients in question, their capacity to actually deliver all of those services depends on their ability to understand multiple complex government programs. A commonly implemented solution to this problem is to give up on the idea of getting ‘product closure’ at the OSS (Askim et al. 2011) and restrict the OSS to a first stop function, with OSS staff performing a ‘no wrong door’ or ‘triage’ function, and then redirecting users to specialist services (Askim et al. 2011). This still relies on OSS staff knowing the range of available programs, and performing effective initial needs assessments. Because of the breadth of service requests they receive, OSS staff may deal with some service processes very infrequently, and therefore be less efficient and effective at delivering such services (Blackburn 2014).
These specialisation dilemmas have in several cases led to a modification of the OSS arrangement, with a move away from clientele specialisation back towards process specialisation. For example, in the Norwegian welfare administration OSS, the original goal was to have a large number of the decisions made at the front line, but this has been modified, with key eligibility and processing decisions moved up to regional offices (Fimreite and Laegreid 2009). While this reduces some of the flexibility and holistic responsiveness at the front lines, it also frees time for staff to spend on clients.
In the Australian welfare administration, the emphasis on having ‘one main contact’, or a single caseworker responsible for all interactions and referrals, was dropped in 2001 and the OSS shifted back to internal process specialisation. In that case, the OSS remains but clients must now visit several desks within the agency if they need multiple services (Howard 2012; see also Christensen and Lægreid 2012). A dual management structure was introduced in the British JobCentres, with separate managers for employment and income administration functions, to cope with a growing portfolio of services (Askim et al. 2011). There is an ongoing effort in these cases to achieve a balance between the two conflicting imperatives of process and clientele specialisation, with recognition that there are practical limits to the seamlessness of service delivery.
This review sought to establish the factors that facilitate adoption and implementation of government one-stop-shops. There is a large literature that references one-stop-shops, but only a small portion of it addresses the drivers of adoption and implementation. This literature displays a series of methodological limitations. Some of the major aspects of OSS effectiveness are not tested. The claim that OSSs are more cost efficient has not been systematically addressed. Qualitative studies strongly suggest OSSs do not generate significant savings, but these rely on perceptions. Although improving citizen satisfaction is a major goal of the OSS model, this review did not find a significant body of research addressing citizen satisfaction with integrated arrangements. This is surprising given the general popularity of satisfaction surveys in the service delivery field.
This discussion makes several broader points about the drivers of adoption and implementation. I suggest opportunism and ambiguity in the design of OSS initiatives may serve a useful purpose. I further reflect on the notion that OSSs are ‘win-win solutions’ and suggest that a balance of integration and specialisation is required for effective implementation.
I have argued that OSS decisions are often not rational, in the sense that governments choose to adopt OSSs without full information on costs, risks and alternatives. Christensen and Lægreid (2012) and Fimreite and Laegreid (2009) criticise governments and senior officials for their limited awareness of trade-offs that arise in the implementaiton of OSSs. They decry the lack of ‘clarity of organistional thinking’, suggesting political and administrative leaders wrongly assume that fundamentally reorganising the divison of service delivery work will be straighforward and will only have positive implications for efficiency. While my review supports the notion that there are important structural trade-offs, and that leaders tend to be overly optimistic, I also suggest that this naive optimism may be a driver of OSS adoption. Proposals for OSSs have to overcome a series of political hurdles before they receive top-down endorsement, and it would not be surprising to scholars of the public policy process that OSS proposals are more likely to attract support when they are framed in ways that emphasise benefits and downplay costs and risks.
While optimism and oversimplification may facilitate OSS adoption, these elements of the decision process can create substantial headaches for implementers, who face unrealistic performance expectations. We saw that politicians can often lose interest in the OSS process at this point, leaving implementers with impractical expectations and little political authority to back them up in the difficult process of re-engineering the service interface. Yet it may be that this withdrawal of political interest is helpful, if it frees managers to modify and temper the initial OSS proposal so as to balance the structural trade-offs involved in integrating services. In this way, we might think of successful OSS implementation as having two distinct phases, one political, and the second managerial:
In the first phase of the process the politicians were the main reform agents, dominating both the experts and the central bureaucracy with respect to the organizational model chosen. In the second phase of the reform the actor constellation was the other way around … The politicians were now less active participants and the reorganization of the reform was mainly seen as an internal managerial process (Christensen and Laegreid 2012, 589).
While these authors identify serious deficiencies in the clarity of organisational thinking in both phases, the shift to a managerial emphasis allows a degree of repositioning and balancing, and better recognition of the importance of process-based specialisation and accountability.
OSSs appear to struggle with problems of process specialisation. In this respect, they confront the basic trade-offs of organisational design and work division long understood in the organisational literature (Gulick 1937). The promise of e-government to overcome or minimise these trade-offs (Askim et al. 2011) has not been realised. Such trade-offs are not restricted to complex services like health care and welfare case management. Indeed, I found evidence for the same kinds of specialisation dilemmas in ‘first stop shops’, where staff are largely referring users to other services. This is not surprising. Even though these staff do not require all the knowledge involved in achieving ‘product closure’, they have to deal with a much broader range of services (breadth instead of depth).
In the final analysis, one-stop-shops are not straightforward win-win solutions, even if they are sold to governments as ‘no-brainers’ (Howard 2014). OSSs face substantial trade-offs. In this context, a major driver of success appears to be a willingness to balance different specialisation imperatives. As Christensen and Lægreid (2012) argue, successful OSSs tend in the long run to “combine different organizational principles” (p. 593), resulting in hybrid arrangements.
Service integration is clearly an important public good, especially where vulnerable user populations are concerned. The choice to implement an OSS is often a valid one, but it requires some sacrifice of other service priorities, balancing of conflicting values, or both.
This review has highlighted the limitations of the existing evidence base concerning the implementation of OSSs. Further research is required to clarify the drivers of service integration. We need a systematic effort to define and measure the impact of culture on service integration. This could draw on the growing literature on measuring administrative culture and its impact on the performance of public organisations (Jung et al. 2009). There is also a need for quantitative analyses of the administrative costs of OSSs. Finally, future research should address the impact of rapid changes in information technology on the design and effectiveness of one-stop-shops.
The author wishes to thank Amber Chambers for assisting with the systematic literature search. An earlier draft of this paper was presented at the workshop ‘Collaborative Efficiency in Government: The Trend, The Implications’, ECPR Joint Sessions, Pisa, April 24-28. The author acknowledges the insightful comments from participants at this workshop. Finally, the author thanks the anonymous referees for their helpful suggestions.
Several terms were considered but not included. ‘Service transformation’, ‘service model’ and ‘public service’ were excluded because they would broaden the scope far beyond one-stop-shops. ‘Shared services’ is often associated with service integration, but it concerns the use of common service systems inside government, not public-facing OSSs. ‘One stop store’ was excluded because it is a private sector term, while ‘one stop source’ was excluded because it refers to a single point to obtain information.