Empowerment via delegation? The administrative capacity-building potential of Cohesion Policy urban development strategies

ABSTRACT

The study examines the implementation of urban development strategies in Scotland, UK, and the Veneto Region of Italy and fills an important gap in the knowledge about the capacity-building potential of such strategies. Following a principal-agent approach, the investigation shows that narrow delegation models incentivize compliance, are less conducive to capacity-building and reinforce an administrative capacity paradox. By contributing an original framework for the study of delegation models, a fine-grained understanding of administrative capacity that acknowledges the importance of agency for the success of capacity-building initiatives and policy recommendations for the period 2021–27, the research will be of interest to scholars and practitioners alike.

KEYWORDS:

• administrative capacity
• government quality
• regions and cities
• European Union Cohesion Policy
• sustainable urban development

JEL:

• R58
• H83

INTRODUCTION

Cohesion Policy is one of the most studied European policies. Research abounds and has focused on the policy’s implementation and impact, which have been examined from multiple theoretical and disciplinary angles (Piattoni & Polverari, 2019). Within this broad field of study, the topic of administrative capacity has received increasing attention in recent years (Milio, 2007; Terracciano & Graziano, 2016; Mendez & Bachtler, 2017; Surubaru, 2017), for at least two main reasons. The first is a growing awareness of the important role that institutional factors play in shaping the policy’s performance. For example, a recent study by Rodríguez-Pose and Ketterer (2020) demonstrates that improvements in regional levels of institutional quality can be a powerful predictor of economic growth for lagging regions and that improving institutional quality should be part of development strategies in these contexts (see also Rodríguez-Pose, 2013; Rodríguez-Pose & Garcilazo, 2015). A second reason is that Cohesion Policy presents several characteristics that make administrative capacity a particularly central, and sometimes problematic, issue for policymakers. Cohesion Policy is implemented under ‘shared management’ and a so-called ‘subsidiarity principle’, whereby European Union (EU) and domestic institutions, across different territorial scales, share responsibility for the implementation of national and regional operational programmes, and are required to engage with partners and stakeholders. These factors mean that public actors should possess vertical and horizontal coordination abilities, brokerage and communication skills, and the capacity to integrate different visions of regional development (Bachtler et al., 2014; Bailey & De Propris, 2004; Dąbrowski, 2014; Potluka & Liddle, 2014). Cohesion Policy is also inherently cross-sectoral, which means that policymakers need to overcome ‘silo-mentalities’ within and across government structures (Catalano et al., 2015; Bachtler et al., 2016; Polverari et al., 2017). Moreover, Cohesion Policy has introduced over time themes that were novel for many domestic administrations (e.g., gender equality, climate change, the digital agenda, Smart Specialisation). All this has required public administrations at multiple levels of government to learn new ways of doing things, and to adapt domestic approaches to evolving policy paradigms and rules (Dąbrowski & Graziano, 2016; Polverari et al., 2017). These features have made it essential to consider whether public administrations possess the capacities that are necessary to implement Cohesion Policy successfully and how these capacities can be enhanced (European Commission, 2017; NEI Regional and Urban Development, 2002; Polverari et al., 2020).

While there is a solid rationale for the increasing focus on administrative capacity, existing research presents important shortcomings. First, a variety of indices and indicators have been produced to measure different aspects of institutional performance – such as the European Quality of Government Index (EQI) (Charron et al., 2019, 2021), Bertelsmann Stiftung’s Sustainable Governance Indicators (SGI); Hammerschmid et al.’s (2014) Management Capacity Index; and the World Bank’s Worldwide Governance Indicators (WGI).¹ However, these as well as other studies (e.g., European Commission, 2017; Lodge & Wegrich, 2014; Rodríguez-Pose, 2013) focus predominantly on concepts such as state capacity, good governance, government quality or institutional quality, and lack clarity about which aspects can be directly linked to the organization and functioning of public administrations. Even when administrative capacity is the focus of analysis, this term is ‘often used either too narrowly to be understood as the simple training of employees, or too broadly to include the entire functioning of government activities’ (El-Taliawi & Van der Wal, 2019, p. 9).

Further, while there has been increasing recognition that in Cohesion Policy, as in other contexts, public administrations are much more than simple ‘implementers’ and that they are active agents in the policymaking process (Peters, 2015), analyses of administrative capacity in relation to the management and implementation of Cohesion Policy, such as the studies quoted above, have tended to focus on the measurement of administrative capacity. This has often involved resorting to proxies (absorption rates), citizens’ perceptions or the quantification of the input required to fulfil administrative tasks (time, human resources). This focus has sometimes led to a neglect of the role of agency and the impact that motivational factors can have on civil servants’ willingness to engage in capacity-building activities.

When considering specifically the implementation of urban development strategies, another weakness of extant research relates to its territorial scale, insofar as most of the data refer to the national level. The creation, since 2010, of the aforementioned EQI has undoubtedly represented a ‘breakthrough’ (Rodríguez-Pose & Ketterer, 2020, p. 975). Nevertheless, notwithstanding country-specific exceptions (e.g., Nifo & Vecchione, 2015; Sacchi et al., 2019), there is still a lack of comparable empirical data on the levels of administrative capacity within member states, especially at the subregional scale. Besides, the regional EQI itself presents limitations. It ‘aims at capturing average citizens’ perceptions and experiences with corruption, and the extent to which they rate their public services as impartial and of good quality in their region of residence’ (Charron et al., 2019, p. 1928; original emphasis). Thus, while it is a useful general measure of perceived government quality at the meso-scale, it provides only a partial and context-dependent understanding of this.

The present study addresses the above-discussed weaknesses, namely: (1) the lack of conceptual clarity about administrative capacity as opposed to related, but non-equivalent, concepts; (2) the limited available evidence about administrative capacity at a local scale; (3) the neglect of agency; and (4) the failure to appreciate the consequences of these knowledge gaps on the design of capacity-building initiatives.

Following a ‘most similar’ comparative research design (Przeworski & Teune, 1970), the paper examines the implementation of 2014–20 urban development strategies in Scotland (UK) and Veneto (Italy), with an in-depth comparative case study methodology. The focus on urban strategies is particularly salient, not only because of the above discussed gaps in the comparative knowledge about administrative capacity at a local scale, but also because cities have attracted increasing attention in EU policymaking and have become significant recipients of EU Cohesion Policy, starting with the launch of a dedicated ‘Urban Community Initiative’ in 1994 (Ramsden & Colini, 2013). However, despite the increasing share of EU funds devoted to urban areas, scholars have emphasized that there is not yet a ‘clear European strategy for sustainable urban development’, and the challenging adoption of an ‘Urban Agenda’ has been interpreted as proof of the reluctance by some member states and regions to devolve more powers to cities (Atkinson & Zimmermann, 2016, p. 414).

In this light, the 2014–20 Cohesion Policy regulations represented a watershed, at least in principle. To respond to criticism on the mainstreaming of urban development support during the 2007–13 period, the 2013 regulations introduced important innovations, increased the financial resources to be allocated to sustainable urban development, and assigned new and more significant tasks to cities, as well as more autonomy on resource allocation (Atkinson & Zimmermann, 2016). A new ‘European code of conduct on partnership’ was also established to enhance the involvement of partners, among which cities, in the drafting of national partnership agreements, and in programme design and implementation.

However, the regulations entered into force when many countries were close to finalizing their partnership agreements, and the European Regional Development Fund (ERDF) and Common Provisions Regulations provided domestic authorities with considerable room for manoeuvre with regards to implementation. Domestic authorities were free to choose: the urban areas; the thematic objectives; whether to go above the mandatory 5% threshold for resources to be allocated to sustainable urban development; and whether to use the options provided by Art. 7(2) of ERDF Regulation 1301/2013. Moreover, the managing authorities had significant leeway in deciding the tasks, above the selection of operations (projects), to be delegated to the cities.

While scholarly research on the application of these new provisions is still limited, a few studies show that there has been considerable variation in the policy responses of member states and regions (Mendez et al., 2021; Van der Zwet et al., 2017). However, there is hardly any evidence yet about the administrative capacity-building outcomes of these strategies. To fill this gap, we examine the administrative capacity shortcomings and capacity-building measures of the 2014–20 urban strategies of Scotland and Veneto to address the following research questions: first, whether the new emphasis placed on urban authorities has actually improved administrative capacity at this level; second, whether capacity-building outcomes might have varied depending on the roles and responsibilities assigned to the cities; and third, whether such outcomes might paradoxically strengthen those cities that are already better endowed with administrative capacity.

To address these research questions, we adopt the definition of administrative capacity provided by NEI Regional and Urban Development (2002), which was developed specifically for the examination of administrative capacity within the context of EU Cohesion Policy. According to this:

administrative capacity can be defined as the ability and skill of central and local authorities to prepare suitable plans, programmes and projects in due time, to decide on programmes and projects, to arrange the co-ordination among principal partners, to cope with the administrative and reporting requirements, and to finance and supervise implementation properly, avoiding irregularities as far as possible.

(p. 2)

Through the adoption of this definition, we provide: a fine-grained understanding of administrative capacity for Cohesion Policy implementation at the urban scale; new insights on the kinds of administrative capacity gaps that can hamper cities’ ability to implement urban strategies; an assessment of whether implementing these strategies actually improves administrative capacity at this level; and whether different delegation choices have an impact on this.

The remainder of this paper is organized as follows. The next section presents the analytical framework and methodology. This is followed by empirical sections illustrating in detail the delegation models adopted for the implementation of urban development strategies in Scotland and Veneto, and their impact on the administrative capacities of cities. Comparative findings and conclusions on the administrative capacity-building potential of different delegation models are provided at the end.

THE ANALYTICAL FRAMEWORK

Research design and hypotheses

This investigation stems from the acknowledgement that the 2014–20 Cohesion Policy regulations place an unprecedented emphasis on urban development and that they provide considerable room for manoeuvre to the domestic authorities on what and how much to delegate to cities. For this reason, we opted to frame our study within rational choice institutionalism and the related principal-agent perspective. This analytical approach has been mostly used for the study of regulatory policies (Bevir & Rhodes, 2001) and has only rarely been employed for the analysis of redistributive policies such as Cohesion Policy (notwithstanding notable exceptions, e.g., Bachtler & Ferry, 2015; Blom-Hansen, 2005). However, as underlined in the introduction, one criticism of the literature on administrative capacity in the context of Cohesion Policy is that it underplays the role of agency. Rational choice institutionalism, instead, places agency at the centre of the analysis. It ‘presupposes that actors choose a particular action or course of actions because they believe it to be the most efficient way of realizing a given end’ (Bevir & Rhodes, 2001, p. 4).

In public policy, delegation can be an ‘attractive option’ for policymakers because it can reduce the ‘transaction costs’ of policymaking, that is the ‘costs related to searching and processing information, bargaining and negotiation, and monitoring and enforcement (see Thomson and Torenvlied 2011, pp. 142–44)’ (Tosun et al., 2019, p. 401). When the cost of delegation is perceived to be lower than the benefits expected, then delegation occurs through a formalized relationship between one or more agents who will act on behalf of a principal. In these contexts, principals would ‘prefer agents who possess the appropriate talents and skills to accomplish the task to be delegated’ (Blom-Hansen, 2005, p. 629; see also Pollack, 2002).

From these theoretical premises, we frame our research around three core assumptions:

• Meso-level authorities base their decisions on how many and which functions to delegate to cities on an assessment of whether the latter possess adequate capacities (Blom-Hansen, 2005).² Where the meso-level authorities do not consider cities sufficiently well-equipped, they will not delegate extensive tasks because the transaction costs of doing so would be too high.

• Different delegation models entail different approaches to administrative capacity-building. Linked to transaction costs (Tosun et al., 2019; Pollack, 2002), we posit that where meso-level authorities deem cities weakly equipped to deal with the functions delegated to them, they will: grant the cities only limited discretion; enact stringent controls; and foresee capacity-building measures that have been defined in a top-down manner and which are aimed primarily at compliance. On the contrary, where meso-authorities perceive cities to be well-resourced, not only will the delegation model be more extensive, but also cities will be credited with more autonomy.

• Urban policymakers, acting as agents, will be more motivated to engage in the implementation of urban strategies which they had an active role in shaping than in contexts where their role is one of simple policy recipients (Atkinson & Zimmermann, 2016).

From these assumptions, we draw and test two research hypotheses:

• The first relates to the cities’ response to delegation and it is that narrower delegation models tend to incentivize a compliance answer by the cities, while more extensive delegation models, characterized by the attribution of wider ranging responsibilities and more autonomy, determine greater engagement, resulting in a more extensive, widespread and lasting improvement of administrative capacity. (RH1)

• The second descends from the previous: if RH1 holds true, then the implementation of urban development strategies determines higher levels of administrative capacity-building in contexts where urban authorities already possess relatively higher levels of administrative capacity. (RH2)

Independent and dependent variables

Our independent variable is the delegation model adopted for the transfer of functions from the meso-level authorities (the principals) to the cities (the agents). The dependent variable is the level of administrative capacity within the cities after the implementation of the urban development strategies. We appraise whether and to what extent the level of administrative capacity of cities has changed from T1 to T2 as a result of the implementation of the strategies.

We define our independent variable as follows: delegation models can range along a continuum from narrow to extensive, based on the scope, depth and style of delegation.

• The scope of delegation relates to the programme management functions that are delegated to the cities. It can range from narrow (only selection of operations) to extensive (all programme management functions).

• The depth of delegation relates to the autonomy that cities enjoy in fulfilling their delegated functions, that is, the level of discretion that they have in accomplishing their activities. Cities can exercise full discretion or share their responsibilities with the meso-level authority. The depth of delegation can range from thin, when cities share their delegated competences with the delegating authority and the latter oversees them on an ongoing basis, to thick, when cities have full autonomy, and controls are mainly ex post and focused on results.

• The style of delegation relates to the approach taken to the accompanying capacity-building measures provided to the cities to enable them to exercise the delegated functions. The delegating authority can choose to support the cities based on its own assessment of their needs (hands-on style of delegation) or allow the cities to deploy additional financial, human or organizational resources in a more independent way (hands-off style of delegation).

Narrow delegation applies when: cities are responsible solely for the selection of operations (narrow scope); responsibility for delegated tasks is shared with the meso-level authority; the meso-level authority puts in place stringent controls on an ongoing basis (thin depth) and provides targeted support based on its own assessment of cities’ weaknesses (hands-on style). Extensive delegation applies when: cities are assigned a wide range of functions, above and beyond selecting operations (extensive scope); responsibility for delegated tasks is exercised by the cities in full autonomy, with the meso-authority intervening only ex post to check whether results have been achieved (thick depth); and, supporting measures are established in dialogue with the cities who are afforded significant autonomy in order to tailor capacity-building efforts to their needs (hands-off style). We consider these two delegation models as ideal types. We opt for this dichotomization, following a similar approach as in Peters (2015), given the impossibility of precisely measuring these dimensions.

The dependent variable is the level of administrative capacity within the cities’ administrations. We appraise whether there were gaps in the cities’ ability to implement their delegated functions (e.g., selection of operations, public procurement, financial management) and whether there were improvements as a result of implementing the urban development strategies. In our appraisal of administrative capacity, we focus on human resources, organizational structures, and systems and tools (NEI Regional and Urban Development, 2002). In this framework:

• Human resources include the personnel employed for policy management and delivery. This includes civil servants and external consultants hired, for example, through the use of Technical Assistance (i.e., funds that are allocated to programme authorities to support implementation). Human resources must be adequate both in number and quality (skills and competences).

• Organizational structures relate to the organizational assets of the bodies charged with policy implementation functions, namely their internal organization and whether they foresee any vertical and horizontal coordination arrangements (e.g., working groups, committees, networks).

• Systems and tools are the operational instruments that the bodies responsible for policy management and implementation use to support the fulfilment of all tasks, from strategy design to project delivery. They include software, hardware, information technology (IT) infrastructures, codified procedures, guidelines and so forth.

Case selection and methods

Scotland and Veneto present several commonalities: both enjoy considerable autonomy over a comprehensive range of economic development matters. Both are facing similar challenges of economic modernization, and both are run by governments that are strongly committed to pursue further autonomy (Scottish National Party and Lega). Crucially, both have a lengthy experience with the implementation of Cohesion Policy and, within this, of urban development programmes. Further, in both cases, the meso-level governments devoted significant financial resources to urban development in their 2014–20 ERDF programmes, above the minimum 5% threshold (7.3% of total programme funding in Scotland and 12.8% in Veneto). Yet, to do so they adopted different approaches from amongst the spectrum of possibilities allowed by the regulations, which makes comparing the administrative capacity-building performance of these alternative models particularly insightful.

As will be discussed in more detail in the empirical analysis section, the Veneto Region asked selected cities to develop sustainable urban development strategies, according to Art. 7(2) of Regulation 1301/2013, within the scope of predetermined thematic objectives and provided them with concrete support towards their design and implementation (but not with financial resources that they could spend to hire staff or acquire specialist support). Additionally, the Veneto Region is especially interesting by virtue of the operation in the city of Venice of two separate programmes: the Veneto ERDF Regional Operational Programme (ROP) and the National Operational Programme for Metropolitan Cities (NOP Metro), managed centrally by the national Agency for Territorial Cohesion. Contrary to the ROP, this latter devolved financial resources to the city of Venice to strengthen its programme management structure through dedicated Technical Assistance funding. The Scottish Government, on the other hand, opted not to engage with Art. 7, and to implement urban development projects via a dedicated ‘Strategic intervention’ and the devolution of management and coordination functions to the City of Glasgow (which in turn coordinated and supported the other cities). These different choices make the comparison between Veneto and Scotland particularly salient to appreciate the capacity-building effect of different urban development delegation models.

The research was operationalized through a comparative case study design (Della Porta, 2008; Yin, 2003, 2011). It entailed desk-research, comprising the examination of literature and policy documents (programmes, urban strategies, reports), as well as fieldwork, and the triangulation of secondary and primary data. Fieldwork included 14 face-to-face semi-structured senior-level interviews, conducted between March and June 2019, following established research standards to ensure confidentiality, anonymity and reliability (Lancaster, 2017). Interviews involved policymakers from the meso-level governments of Scotland and Veneto, a selection of Scottish cities, the Scottish Convention of Local Authorities (COSLA), all Veneto urban authorities, and the managing authority of the NOP Metro (for the full list, see Table A1 in Appendix A in the supplemental data online). Additionally, in the preliminary stages of research, a scoping interview with the Veneto ERDF ROP managing authority and a focus group with all Veneto urban authorities were used to finetune the scope and aims of the study, and inform interview checklists.

DELEGATION MODELS COMPARED

This section analyses the delegation models adopted for the design and implementation of urban development strategies in the two meso-level authorities examined. For a summary overview of the two models, see Table A2 in Appendix A in the supplemental data online.

Scotland’s delegation model

The Scottish Government was the managing authority of the two Cohesion Policy programmes that operated across Scotland, one funded by the ERDF and the other by the European Social Fund (ESF). Implementation was undertaken via 14 ‘Strategic interventions’, namely ‘groups of projects of significant financial scale led by organisations with the capability and scale to manage the EU audit obligations and deliver significant outcomes’ (Scottish Government, 2019, p. 135). Their delivery was entrusted to ‘lead partners’, ‘typically the existing organisations and legal vehicles which already manage domestic funding in the same policy area’ (p. 8). Lead partners, in turn, coordinated ‘delivery agents’, who were tasked with running operations. Reliance on a limited number of strategic interventions and lead partners was a strategic choice introduced to streamline implementation, given the reduction of EU funding and the alignment of EU programmes with overarching domestic policies. Lead partners were selected based on possession of the administrative and financial capabilities required to deal with the complexities of Cohesion Policy funds. Considerations of administrative capacity were central, due to audit issues that had arisen in the period 2007–13 and the resulting political will to prevent irregularities (SCO3).

In the ERDF programme, urban development projects were funded under thematic Objectives 1 (smart cities) and six (green infrastructure) and were the focus of a strategic intervention named ‘Scotland’s Eighth City: The Smart City’. This was a policy initiative focused on the concept of the smart city, which delivered the Scottish government’s ‘Agenda for Cities’. It entailed a public investment of circa £24.1 million, of which £10 million was from the ERDF, for the seven Scottish cities: Aberdeen, Dundee, Edinburgh, Glasgow, Inverness, Perth and Stirling.³ The lead partner was Glasgow City Council, which was appointed because of its longstanding experience with ERDF regeneration programmes and because it had implemented, since 2013, a precursor initiative to the Eighth City programme: a ‘Future Cities – Glasgow’ project, financed under a domestic UK-wide competitive funding programme (SCO2).

A ‘strategic board’ and an ‘advisory group’ were set-up, comprising the heads of economic development and operational project delivery staff from the seven cities, together with representatives from the Scottish government and the Scottish Cities Alliance (a forum encompassing Scotland’s seven cities and the Scottish government, established in 2011). As lead partner, Glasgow established a programme management office (PMO), which operated as the liaison between the Scottish government and the cities’ administrations (delivery agents), and oversaw general programme management tasks, governance and coordination activities. For the delivery of projects, each city had its own governance, based on its internal organization. ‘Collaboration agreements’ between the lead partner and each other city demarcated roles and responsibilities, and specified the co-financing required. Glasgow, through its PMO, supported the other cities, with a strong focus on ensuring compliance and timely delivery.

The delegation model adopted was two-layered, with different roles assigned to the lead partner and delivery agents. The lead partner was tasked with all programme management functions related to the initiative, while delivery agents had more limited tasks, comprising the choice of investment fields, the selection of operations and the financial management and reporting on operations. While the focus on the smart city concept and the implementation through a strategic intervention model were essentially a political choice by the Scottish government, within this frame each city could target investments choosing freely from the seven operational clusters of the initiative: energy, mobility, waste, public safety, innovation labs, intelligent street lighting and water management. Accountability was assured through the governance framework of the initiative, described in the programme and in the agreements, and through regular meetings between the lead partner and the Scottish government, between the lead partner and the delivery agents, and as part of the wider governance of the Scottish Cities Alliance. There were also stringent reporting obligations, such as monthly ‘project status’ reports, quarterly update newsletters and quarterly feedback to the advisory group. The whole idea behind the decision to implement the programmes through strategic interventions coordinated by lead partners was to place responsibility in the hands of grant holders while making sure that the Scottish government would retain overall control (which the Scottish government were ‘keen to secure because of past failings’; SCO3). Supporting measures were provided by the Scottish government to the lead partner, and from this latter, in a cascading approach, to the other cities. The focus has been on ensuring the timely progression and regularity of expenditure, to prevent automatic decommitments⁴ and avoid auditing problems. The style of delegation has thus been rather hands-on and focused on compliance, through the vehicle of the lead partner. On the whole, therefore, the Scottish delegation model was characterized by a fairly extensive, fairly thick and hands-on style of delegation.

Veneto’s delegation models

The Veneto Region has been home to two different delegation models. One was related to the implementation of the ERDF ROP, for which the Directorate for Unitary Programming of the Veneto Region acted as the managing authority. The other model, involving only the city of Venice, was designed within the framework of the NOP Metro, managed at the national level by the Agency for Cohesion.

To implement the programme priority on sustainable urban development, total funding for which amounted to almost €77 million, the ERDF ROP designated six urban authorities, namely: Montebelluna, Padua, Treviso, Venice, Verona and Vicenza. These urban authorities were selected through a procurement exercise after the European Commission approved the ROP in 2015. Considering the lack of administrative capacity at the local level and the implementation difficulties experienced during previous programming periods, the managing authority opted to delegate only limited tasks to these cities (VEN9). Cities were asked to frame their sustainable urban development strategies within thematic priorities chosen by the managing authority, relating to specific thematic Objectives (2, 4 and 9 – digital agenda, sustainable mobility and social inclusion). With regards to implementation, the responsibilities of the urban authorities included only the selection of operations (in collaboration with the managing authority), assistance to beneficiaries, and reporting on implementation and monitoring, while the financial management of operations was delegated to the regional payment agency (AVEPA). In other words, being aware that most of the cities presented structural human resources deficiencies and a lack of experience in managing EU funds, the ERDF ROP managing authority opted for a narrow delegation model.

The NOP Metro, on the other hand, adopted a totally different delegation model. This programme targeted 14 metropolitan cities selected nationally, including, for the Veneto Region, Venice. The NOP was co-funded by the ERDF and the ESF and focused on the same thematic priorities as the Veneto ERDF ROP (thematic Objectives 2, 4 and 9). Unlike the Veneto ERDF ROP, however, it assigned extensive functions to the metropolitan cities and dedicated funding for capacity-building activities under the technical assistance priority (amounting to circa €1 million for Venice, when the research was conducted). Under the NOP Metro, Venice could define its specific policy priorities within the framework of national guidance, decided on the budget for each priority and was responsible for the design and implementation of all operations. The city was only required to report to the national managing authority on implementation progress every three months and to respect the general management rules established by the national level. The NOP Metro’s delegation model showed an extensive scope of delegation, a significant degree of autonomy and a hands-off delegation style, and can be characterized as, on the whole, rather extensive. Public servants from the Venice urban authority have appreciated the extensive delegation model of the NOP, not least since it contributed to increasing the political ownership of the strategy within the metropolitan city.

ADMINISTRATIVE CAPACITY AND CAPACITY-BUILDING IN SCOTLAND

Administrative capacity gaps and challenges

Scottish cities have been involved in the implementation of EU-funded projects for around three decades. For the implementation of the Eighth City Strategic intervention, the main capacity gaps related to a lack of human resources and of specialist skills in the fields covered by the initiative. While cities employed staff who possessed the baseline competences required to deal with complex EU funding rules, the number of staff devoted to EU funds had fallen over time in all cities. For example, whereas all used to have European funding offices or teams, this was no longer the case when the research was conducted. Budget and organizational cuts and repeated internal reorganizations had inevitable repercussions on the human resources that could be devoted to EU projects. This led local authorities to focus on compliance (‘our main concern is compliance and eligibility … for us administrative capacity is a question of compliance capacity’; SCO4), and to neglect, for example, the strategic seeking of funding opportunities (e.g., from programmes such as Horizon 2020). Glasgow continued to be the most resourced local authority with regards to the personnel devoted to EU funding, and it was for this reason that it was appointed as lead partner for the Eighth City Strategic intervention. Beyond the deficiency of human resources in numbers, a specific administrative capacity issue related to the cutting-edge nature of the investments foreseen. While the generalist competences – for example, on public procurement and eligibility checks – were already present in the cities, the innovative nature of projects, which related to markets that were not yet mature, slowed implementation (SCO2).

Administrative capacity-building measures

A range of capacity-building initiatives were put in place for the Eighth City Strategic intervention. They related to all three dimensions of administrative capacity used in our analytical framework. First, several organizational structures were set up, such as the strategic board and advisory group. The city of Glasgow established a dedicated organizational structure when it agreed to act as lead partner, that is, a PMO comprising four members of staff tasked with functions related to management, provision of support to the delivery agents, and day-to-day activities (such as issuing press releases or ensuring that the cities were using the correct logos; SCO2).

With regards to human resources, the Scottish government entrusted a consultancy firm to support the cities in carrying out a self-assessment exercise, to help identify the most relevant investments associated with them becoming ‘smart cities’ (Scottish Government et al., 2014). Further, to support the implementation process, the Scottish government organized regular meetings with the lead partners of all strategic interventions, in order to brief and train them on procedures and rules, and answer questions. It also held frequent informal meetings with the lead partners. These meetings were attended regularly by the Glasgow officials responsible for the Eighth City. These, in turn, would pass on the knowledge generated from these contacts to the other cities. Most of the capacity-building activities for the cities were delivered through the PMO, which provided training and organized development sessions and workshops. These capacity-building initiatives were ‘part and parcel’ of the governance of the programme (SCO2 and SCO3).

The PMO also prepared capacity-building tools. When Glasgow Council took on responsibility for the management of the programme, it recognized that this would involve a level of risk, so it developed an assurance framework (including briefings, terms of references and collaboration agreements) and ensured that everyone involved was clear about responsibilities and liabilities. PMO staff also spent time with the cities to develop their understanding in relation to these, working with the operational staff involved in delivering projects, explaining ‘here’s what is expected of you as a delivery agent in terms of procurement, budget, financial management, marketing and publicity, monitoring and evaluation’ (SCO2).

In sum, the Eighth City Strategic intervention was accompanied by several dedicated capacity-building initiatives across all three dimensions of our definition. These strengthening measures followed the two-layer model of the intervention itself (from the Scottish government to the Glasgow PMO, and from the latter to the other cities). In addition, these measures were supplemented by other capacity-building activities associated with EU funding more generally. Specifically, both the lead partner and the delivery agents could draw support from the activities of a working group on European funding which had been set up within the Scottish Local Authorities Economic Development Group (SLEAD), under the aegis of the Convention of Scottish Local Authorities (COSLA). This working group was also chaired by an official from Glasgow city administration, which ensured synergy with the work carried out by the PMO.

Improvements realized, spillovers and legacy

The asymmetric approach to the delivery of the strategic intervention determined different administrative capacity-building outcomes. While ‘the Glasgow office are constantly capitalising on what was learnt [including through] detailed project closure reports with capturing of lessons learnt’, this has not been necessarily the case in other cities: ‘Some cities … would listen and say “you guys are getting paid so you can continue to help us”. Some actually withdrew because workload was too much’ (interview code omitted to prevent identification). In practice, the level of engagement was constrained in some cities by the resources available. Historical legacy also played a part. Cities such as Glasgow or Dundee have a past of industrial decline and reconversion, which has arguably made them more entrepreneurial and open to innovation in order to respond to changing economic circumstances (SCO2).

Notwithstanding the above, the Eighth City Strategic intervention was considered to be realizing important gains in terms of open data platforms, which were a useful tool for the cities to assist decision-making. However, more generally, the strong focus on compliance acted as a deterrent to experimentation and constrained the creation of new competences within city administrations. So, while Glasgow is considered to have performed well in ensuring successful delivery, the management support that it received ‘hasn’t trickled down’ to the other cities as much as was hoped (SCO3).

The intended collaborative nature of the initiative did not play out as anticipated either. For example, it proved impossible to develop joint procurement exercises or to achieve shared leaderships for the policy clusters. The asymmetry between lead partner and the other cities had a negative effect on the intended collaborative nature. ‘Collaboration has been happening but has been more on the softer side.  …  Capacity was an issue. Folks had not necessarily the time’ (SCO2). Nevertheless, the necessity to deal with the complex rules of EU Cohesion Policy funding encouraged cities to cooperate, including via SLEAD. Thus, while collaboration under the Eight City was achieved only in part, on the whole cities ‘feel stronger as a group and this is very much an established way for Scottish local authorities’ (SCO1).

ADMINISTRATIVE CAPACITY AND CAPACITY-BUILDING IN VENETO

Administrative capacity gaps and challenges

At the outset of the programming period, most of the urban authorities in Veneto were understaffed and public servants had mostly limited experience in managing EU funds. All municipalities had project management offices which competed for and managed EU direct funding or territorial cooperation projects, but not all of them were sufficiently equipped to deal with the programme design and implementation activities required by the sustainable urban development strategies. Most urban authorities could assign no more than one or two officials to their management. Moreover, staff often divided its time between the activities related to these strategies and other tasks. The cities of Venice and Vicenza represented notable exceptions. In Venice, the administrative unit acting as urban authority comprised already at the beginning of the programming period numerous and highly skilled officials, who had gained significant expertise in managing EU funds during previous periods. In Vicenza, a member of the urban authority staff had previous experience with the ERDF and ESF, having worked on these funds within the regional administration, and thus possessed the necessary knowledge and skills to draft the required documentation, and coordinate the definition of responsibilities and functions for the delivery of the city’s sustainable urban development strategy.

The interviews carried out uncovered that public servants in the cities were well aware of the existing administrative capacity gaps and of the consequences that these had for the management of the urban development strategies. Some interviewees highlighted the organizational dimension of capacity, which is associated with the ability ‘to establish a comprehensive system of functions and tasks that work in a coordinated manner’ (VEN5), others stressed the strategic dimension of capacity: ‘the ability of administrations to support their own strategies with specific capacities … to achieve their objectives’ (VEN1).

Administrative capacity-building measures

Being aware of the above limitations, the managing authority of the ERDF ROP provided city administrations with several dedicated capacity-building initiatives, which embraced the entire spectrum of dimensions covered by our study. First, the managing authority provided the cities with a range of tools, such as dedicated templates and guidelines for the design of the strategies and of the related implementation procedures. To ensure that tasks would be matched by adequate capacities, each urban authority was asked to prepare, based on the templates provided, a clear description of responsibilities and functions and an organizational chart, proving to have sufficient financial and management capacities to perform the delegated functions. Additionally, the managing authority also aimed to strengthen the competences of the human resources responsible for the management and coordination of the strategies in each urban authority. It organized training events, which were delivered with the support of a specialized consultancy, on a range of issues related to the EU funding rules. Additionally, financial management support was provided via the regional agency AVEPA.

Measures related to the organizational structures were adopted, too. First, the managing authority established a dedicated unit, comprising four members of staff, tasked with the vertical and horizontal coordination of all activities relating to the sustainable urban development strategies. The unit was also responsible for the provision of support to the urban authorities, with the involvement, whenever needed, of the sectoral units within the regional administration. Second, the managing authority promoted the establishment of a coordination committee, involving regional and local politicians, the managing authority and the administrative staff from the urban authorities. And, lastly, it also instituted a coordination working group comprising officials from the managing authority and the urban authorities. This latter met regularly, roughly every two to three months, to discuss implementation progress and solve problems on an ongoing basis.

The interviewed urban authorities’ staff showed appreciation for the training modules conducted by the managing authority and by external experts, acknowledging that they were useful for understanding the rules and procedures of EU funds. However, they also pointed out that most of the topics were too general or even obsolete in some cases, and the fact that these initiatives came too late to be deployed in concrete actions. Likewise, the success of the managing authority’s efforts to support local authorities by developing templates for the local plans or delegating the financial management to AVEPA appears to have been limited too, since it could not overcome the staff shortages of many urban authorities (‘Urban authorities do not have sufficient staff, resources and expertise to be able to effectively perform their functions’; VEN2).

Although most interviewees considered the technical support provided by the regional agency AVEPA as essential, since they would not have been able to deal with the complex and time-consuming financial management procedures themselves, they also stressed that the delegation of financial management functions to this agency supplied rather than built capacities. Urban authority officials also highlighted the need to improve existing procedures relating to the management of calls for proposals and applications for funding, in order to avoid duplications and inconsistencies. City officials criticized especially the failure to devolve Technical Assistance resources directly to the urban authorities, which would have helped them strengthen their human resources. At the same time, the urban authorities praised the formal activities organized by the managing authority, through the working group and training events, which were instrumental to establishing a culture of collaboration and mutual support. As a result, city officials collaborated routinely and could rely on each other for support.

The city of Venice has been rather an exception. Not only did it benefit from numerous and skilled staff compared to other city administrations, but it could also use the NOP Metro’s Technical Assistance resources allocated to the city to fund dedicated training projects and recruit qualified personnel to support the implementation of both its urban development strategies. According to the Venice urban authority, dedicated funding and targeted training initiatives, alongside strong political support, were central to the consolidation of its administrative capacity. These resources allowed it to strengthen its organizational structure and adjust it effectively to the new tasks and functions required by the two different delegation models. A comprehensive mapping exercise was carried out in order to accomplish a systematic overview and rearrangement of the internal structures dealing with different strands of project-management activities. The NOP Metro’s framework has also allowed the Venice urban authority to invest significant resources to strengthen the capacities of all actors engaged with the delivery of the programme.

Improvements realized, spillovers and legacy

Notwithstanding the limitations of the ERDF ROP capacity-building activities, urban authorities agreed that the sustainable urban development strategies implemented under the aegis of the programme produced both learning and spillover effects. The implementation of the strategies contributed to increasing the cities’ awareness and knowledge about the functioning of EU funds and the opportunities they provide. Urban authorities’ officials have been unanimous in stating that this process has allowed them to acquire useful, albeit limited, expertise in the field of EU funding, and intensify interactions and communication across sectors and administrations. As reported by one interviewee, ‘formal and informal interactions around the sustainable urban development strategy have helped us establish new collaborative relations within and outside the administration, increasing our ability to work in team and project-management capacities’ (VEN5). However, this process strongly depended on the individual initiative and on the commitment of staff (VEN7). Interviewees emphasized that ‘continuity and further efforts are required in order to consolidate local know-how and boost the capacities that have so far been created through learning-by-doing’ (VEN2). In addition, regional and local civil servants alike highlighted the importance of horizontal interactions and collaborative initiatives that involved the urban authorities and, in the Venice metropolitan area, also the smaller municipalities that benefited from funding under the NOP Metro. However, most interviewees from the urban authorities stressed that stronger political support would have been needed in order to ensure the diffusion and consolidation of participatory practices and local partnerships.

DISCUSSION

The cases examined show significantly diverse approaches to delegation with regards to scope, depth and style. The NOP Metro, as implemented in the city of Venice, is an example of an extensive delegation model; the Veneto ERDF ROP illustrates a case of narrow delegation, while the Scottish two-tiered model sits somewhere in between.

Our first hypothesis related to the cities’ response to delegation, which we expected to be more compliance oriented and less conducive to administrative capacity-building in narrow delegation models. Our evidence confirms this hypothesis. In the narrow delegation model of the Veneto ERDF ROP, city administrations engaged in a limited range of management activities, coherent with their (also limited) human and financial resources. Learning was constrained by an approach to capacity-building that has been top-down, focused primarily on spend and compliance, and not sufficiently targeted. It was confined to the understanding of the rules necessary to the fulfilment of delegated tasks. The provision of support for specific activities through the regional agency AVEPA was essential to allow cities to implement their urban strategies. However, it merely supplied rather than built capacity. Notwithstanding this, the implementation of the Veneto ROP’s sustainable urban development strategies enabled the creation of a self-sustaining informal network among the cities involved, which allows officials to draw on respective strengths to support each other. Intra-municipal cooperation, on the other hand, which was required by the cities’ dual role of designated ‘urban authorities’ and beneficiaries of support, was more challenging, particularly in less-resourced municipalities. This has had an important constraining effect on the degree to which innovative EU-driven practices could permeate within city administrations, beyond the coordinating structures. At the opposite end of the spectrum, Venice’s involvement in the NOP Metro provides proof that a more extensive delegation model – with more responsibilities, autonomy and dedicated technical assistance resources – can lead to positive administrative capacity-building outcomes. In this case, not only was the city administration’s structure in charge of the coordination of the strategy able to effectively design and implement it, it could also act as a stimulus for the introduction of novel administrative practices and procedures in other departments within the city administration. The city of Venice was also able to exercise leadership towards the smaller municipalities of the metropolitan area, acting as a catalyst for improved cross-sectoral and multilevel dialogue. A less hands-on approach by the national managing authority and the city’s deployment of technical assistance resources had a positive impact on the administrative capacity outcomes achieved: capacity-building initiatives were tailored to need and there was also a motivational effect. That granting a degree of autonomy and dedicated financial resources to the cities can yield positive capacity-building dividends seems to be confirmed by the Scottish case. In this context, capacity-building outcomes have been particularly visible within the city of Glasgow, the lead partner of the Eighth City Strategic intervention, but have been variable in the other cities. Some of the Scottish cities disengaged, exactly because of a lack of human resources.

Our second hypothesis was that the implementation of urban development strategies would determine higher capacity-building outcomes in those city administrations which were already better endowed with administrative capacity to start with. This hypothesis is also confirmed. Those cities that already had more, and more professionalized, human resources were better able to strengthen their procedures, practices and skills (e.g., Glasgow, Venice and Vicenza). Three further conclusions qualify this finding: the first is the negative impact of the protracted austerity. City administrations, in both Scotland and Veneto, experienced personnel cuts and these appear to have had more severe effects in the smaller municipalities. A second conclusion is that the capacity-building measures supplied to implement the urban strategies were essential to enable the cities to deliver these, but more support would have been needed in most cases. One last finding relates to the important role played by political leaders. Where the commitment of city mayors and local governments was more prominent and tangible – such as in the Venice, Vicenza and Montebelluna in Veneto or Glasgow in Scotland – this encouraged the administrations to be more engaged and to pursue innovation and learning. Vice versa, where the political leadership did not particularly value the opportunities offered by the urban strategies, the officials in charge of their implementation had a tougher ride to engage colleagues, and tended to focus on ensuring spending and compliance.

CONCLUSIONS

Our research has evidenced that delegation models can have important implications not only for the effectiveness of policy initiatives per se, but also for their capacity-building outcomes. In other words, effective delegation is not only key to policy success, but it can also be a foundation for the success of future policy initiatives, by enabling the development of lasting capacities that transcend the given delegated policy. Our findings also show that adopting a rational-choice institutionalist theoretical lens to policy analysis and a principal-agent approach focused on delegation models can be fruitful when examining policy aspects that so crucially rely on the engagement of actors, and that this type of framework can be applied usefully also beyond the confines of regulatory policies. The appreciation of agency and of actors’ motivations can provide important insights to grasp the concrete bottlenecks of policymaking and allow for a more in-depth understanding of how policy processes play out in practice, why ambitions might not always be fulfilled, and what can be done to improve this. This is particularly crucial for capacity-building initiatives, which rely intrinsically on actors’ willingness to learn.

This investigation also provides some significant insights for the design of future urban development initiatives. Our case studies have shown that even though policymakers will naturally engage with any policy that provides them with financial resources to fulfil some of their constituencies’ development needs, it is useful to afford them sufficient leeway to tailor policy responses to local ambitions and potentials. While the stringent timing of Cohesion Policy constrained managing authorities in their decisions about which functions to delegate to the cities, some of the choices that have been imposed on urban authorities would have been more effective if a more consensual decision-making style had been adopted. This has been particularly evident in the Veneto ERDF ROP, where urban authorities would have welcomed more freedom in the selection of thematic objectives, in itself not neutral for administrative capacity building, and capacity-building solutions. Both could have been better targeted to the specific needs of each city. These considerations should be borne in mind in designing future urban strategies and capacity-building initiatives, to avoid the perpetuation of an administrative capacity paradox, whereby the adoption of more or less extensive delegation models is linked to the consideration of existing capacity gaps, more so than by a desire to overcome those gaps through dedicated (and potentially asymmetric) policy initiatives. Our conclusion that narrow delegation models tend to incentivize compliance and reinforce such an administrative capacity paradox should be taken as a warning for the planning of future urban strategies, if the ambition is to overcome the kind of vicious circle evidenced by our findings.

We are aware that our research presents some limitations. There is certainly a wider array of factors related to the politics, polities and domestic policy frameworks of the cases examined that should be investigated to appreciate more fully the reasons behind the delegation choices adopted and the impact of these choices on the administrative capacity of cities. These factors include the roles and responsibilities assigned to cities in domestic policies, and the availability of domestic funding streams for the cities, which were more present in Scotland than in Veneto, as well as the expectation about future EU funding for urban development. This latter was null in Scotland, given Brexit, and significant in Veneto, linked to the further strengthening of urban development in the 2021–27 regulations through an increased earmarking threshold of 8% and the framing of urban development strategies within a new Policy Objective dedicated to ‘A Europe closer to citizens’ (Mendez et al., 2021). This expectation has subsequently further increased in Veneto, following the launch of the Recovery and Resilience Facility and the related National Recovery and Resilience Plan. Political commitment and preferences, at both meso and urban levels, have emerged as particularly salient in both cases examined, and appear to have determined a markedly different engagement in the management and implementation of the urban development strategies by similar-sized cities in both contexts. Moreover, while austerity-driven cuts to local government have impacted significantly on the human resources available to the cities, clear asymmetries between these persist in both contexts, due to path dependence and to the historical legacy of previous engagement with EU funds. In-depth case studies are not intended to yield generalizable conclusions, nor was this our ambition. Our investigation aimed to be exploratory. As such, it has provided important novel insights: first, it has contributed an original analytical framework for the study of the administrative capacity building potential of different delegation models, paving the way for more research on delegation models and their effects; second, it has provided new knowledge about the capacity-building potential of urban development strategies, which so far has been an under-researched topic; and, last, it has supplied further evidence on the utility of adopting an in-depth qualitative perspective to the study of administrative capacity, which complements existing quantitative studies. Part and parcel of this approach is the acknowledgement of the important role that agency plays in the success of capacity-building initiatives.

ACKNOWLEDGEMENTS

We thank all interviewees for their time and valuable insights, and the participants in the Special Session 15 III: ‘Empowering Actors in Local Dimensions of Cohesion Policy’ at the Regional Studies Association (RSA) Annual Conference (Santiago de Compostela, Spain, 4–7 June 2019) and in the RSA Cohesion Policy Research Network Conference ‘Building Administrative Capacities’ (Delft, the Netherlands, 21–22 November 2019) for their valuable feedback on early drafts of this paper. Our grateful thanks also to the three anonymous referees: we found their comments and suggestions extremely insightful, constructive and helpful.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Additional information

Funding

The research was supported by the CARIPARO Foundation [grant number 2018.0517]. Open Access was supported by the University of Padua’s Department of Political Science, Law and International Studies.

Notes

1. See https://www.sgi-network.org/2020/; and http://info.worldbank.org/governance/wgi/.

2. We developed our assumptions focusing on Scotland and Veneto, which are meso-level authorities. Nevertheless, where applicable, for national operational programmes the same assumption would apply to national authorities, namely ministries or agencies.

3. Internal document obtained during fieldwork, dated 16 October 2018.

4. Under the automatic decommitment rule, a portion of the budgetary commitment is automatically decommitted by the European Commission if it remains unused or if no payment application has been received by the end of the third year following the year of the budgetary commitment (see https://ec.europa.eu/regional_policy/index.cfm/en/funding/financial-management/).