Estonia’s experiment with pilot regional councils marks a step towards the more place-based approach to regional development discussed in Chapters 1 and 2. Introduced in Central and South Estonia, the pilot councils bring together representatives from national government, local governments and other regional stakeholders to chart a path forward for development in the two regions. This chapter presents the motivations to establish the councils, and distils lessons from the design, implementation and first results of the two pilots. It also explores how the Estonian Government can evaluate the experiment and consider potential pathways for scaling up regional governance.
Strengthening Place-Based Regional Development Policy in Estonia
Abstract
In 2023, the Estonian Government introduced two pilot regional councils to test a new approach to establishing and implementing regional development objectives. The government recognised that a new approach was needed to address what it called “a general dissatisfaction with the country’s current limited regional perspective” (Estonian Ministry of Regional Affairs and Agriculture, 2025[1]). After exploring different options (Box 3.1), the government established two pilot regional councils – one in Central Estonia and one in South Estonia1 – to create regional development agreements defining the development priorities and activities of the region. Bringing together national government, local government and other regional stakeholders, these councils also promised to serve as a “systematic two-way feedback loop” between plans and policy at the national and local levels.
Estonia’s pilot regional councils represent a step towards laying the foundations for a place-based approach to regional development (Chapter 2), adapting the multi-level governance structure to favour cross-cutting responses to regional needs and opportunities. Place-based policy for regional development requires co-ordination and alignment across different policy areas and among levels of governments. Chapter 2 highlights that Estonia lacks both: i) co-ordination for regional development across different ministries on the national level; and ii) strong mechanisms to bridge national and local planning and policy making for regional development. The pilot regional councils help to address these two gaps by offering a structure to create common ground among national, regional and local actors. Ultimately, this approach promises to favour a co-ordinated regional approach among all of these actors to best address regional needs and take advantage of regional opportunities.
This chapter distils lessons from the experiment with regional councils. It traces the journey of the pilot councils through four phases: idea, implementation, results and future. It starts with the idea, exploring the Estonian Government’s motivations for the councils. It then considers implementation, identifying lessons learnt from how the pilot councils were put into practice, including the need to broaden ministerial representation, clarify decision-making processes and harness stakeholder engagement. It continues with the results, assessing the agreements themselves to identify success factors, such as avoiding “shopping lists” of local demands, embedding enforcement mechanisms and building the capacity building of implementers. Finally, the chapter looks ahead to the future of the councils, exploring how the Estonian Government can evaluate and consider whether and how to scale up the pilots. In sum, this chapter can help the Estonian Government build on the experience of pilot councils to refine its approach, ensuring that future iterations live up to the promise of serving as a tool to advance place-based regional development.
Important note: At time of writing, the draft agreements have been prepared but not yet signed.
The chapter focuses on the text of the draft agreements available in Autumn 2024, which may change before they become finalised. This chapter is also unable to comment with certainty about who will sign the agreements, although the draft agreements provide an indicative list of draft signatories.
In 2020, an influential paper written by Estonian academics and experts made the case for regional governance in Estonia. The authors made some of the same observations as Chapter 1: growth in Estonia is concentrated in Harju County while other areas struggle to attract investment and skilled labour. As discussed in Chapter 2, local governments tend to lack sufficient influence over national strategic development planning and policy and sufficient capacity to plan and implement for development, which has limited the ability of Estonia to respond to regional challenges and take advantage of regional opportunities for development. Their answer to these issues was regional governance, and the authors proposed a range of alternative governance arrangements that would facilitate better co-ordination among national ministries, regional actors and local authorities.
Governance models proposed by the academics involved lesser or greater degrees of decentralisation and regional autonomy. One option was to strengthen county-level co-operation, requiring local governments to co-operate more systematically through formalised agreements and consolidating the various regional development and regional service institutions under the county-level associations of local governments (co-operative bodies that bring together local governments to promote balanced and sustainable development (Riigi Teataja, 2002[2])). Another model suggested the creation of regional councils to facilitate co-operation among local governments, the national government and other stakeholders, helping to ensure that regional priorities would be considered in national policy making. A more ambitious alternative was the establishment of regional self‑governments, granting certain regions greater fiscal and legislative autonomy to tailor development policies to their specific needs. The authors also put forward a hybrid model, incorporating elements of both regional councils and co-ordination mechanisms, to balance local input with national oversight. The paper did not recommend a specific model above all others, but did propose a pilot project to begin testing a model (in addition to stakeholder engagement and political discussions).
Source: Kattai et al. (2020[3]), Analysis of Estonian Regional Development: Proposals for Shaping Regional Governance Models – Final Report (Eesti regionaaltasandi arengu analüüs: Regionaalse valitsemise mudelite kujundamise ettepanekud – Lõpparuanne), https://skytte.ut.ee/sites/default/files/2022-05/1.%20Regionaalne%20valitsemine%20lopparuanne_final.pdf.
A trio of motivations drove the Estonian Government to pilot the councils: i) strengthening engagement between levels of government; ii) helping local governments achieve economies of scale by working regionally; and iii) tapping into the perspectives and resources of businesses and research institutions for regional development (Figure 3.1). First, the pilot regional councils were envisioned as a platform to strengthen engagement between local and national governments, a step towards more alignment between local-level planning and national-level policy and activities. Second, they aimed to help municipalities plan and work together more effectively on a larger scale. Finally, the councils sought to bring other stakeholders to the table, with the initial idea of deepening co-operation among government, businesses and researchers to harness their expertise and resources to drive regional development. This section explores these three drivers in detail, examining how each shaped the design of the pilot councils and how they were intended to address Estonia’s regional development challenges.
The councils were intended to support collaborative planning and collaboration among municipalities
In general, Estonian local authorities are small and shrinking in terms of their populations, which can limit their ability to attain economies of scale for service delivery and investment. Municipal mergers in 2017 decreased the number of municipalities from 213 to 79, but local governments remain small in terms of population size. Following the mergers, the average municipal population is 16 653, which puts it well below the average of OECD countries (40 107 inhabitants). Estonia’s municipalities are small even when compared with its small nation peers: the average of other small unitary OECD and European Union (EU) countries is 30 804 inhabitants. In addition, Estonian municipalities are getting smaller: nearly 60% of its municipalities are shrinking (OECD, 2022[4]). Small municipalities can struggle to deliver public services and invest in the development due to their limited size and resources. As discussed in Chapter 2, the 2017 Administrative-Territorial Reform reduced but did not eliminate the number of municipalities deemed too small to function independently. For small municipalities, co-operation with their neighbours can be a way to deliver public services more efficiently and to invest at the right scale (Allain-Dupré, 2018[5]).
The pilot regional councils were intended to encourage Estonia’s local governments to plan together, which can help them find opportunities to work together on a larger scale. One justification for putting in place regional governance structures is achieving economies of scale (along with a host of other justifications, such as achieving strategic co-ordination, gaining stronger administrative resources and capacities, and internalising cross-jurisdictional externalities). Regional co-operation can provide economies of scale2 and lower transaction costs by encouraging subnational governments to operate on a larger scale (OECD, 2020[6]). Co-operative regions – the category of regional governance that best describes the pilot councils (Table 3.2) – can favour efficiencies by creating a structure within which municipalities collaborate rather than work in isolation (OECD, 2022[4]). The process the pilot regional councils were to undertake – which involved identifying shared regional priorities and aligning municipal strategies to create a regional development plan – promised to help municipalities find new areas for co-operation. Finding new areas for co-operation at the regional level would mean dealing with heterogeneity within regions (including, for example, an urban-rural mix in South Estonia as shown in Table 1.1) to find a common strategic direction. However, the Estonian Government did not designate specific areas for co-operation beyond planning. This meant that any opportunities to achieve greater scale for activities would need to be identified through the planning process itself.
Note: NUTS 2 regions are basic regions established within the EU's Nomenclature of Territorial Units for Statistics (NUTS) (Eurostat, n.d.[7]).
Source: OECD (2022[4])Regional Governance in OECD Countries: Trends, Typology and Tools, OECD Publishing, https://doi.org/10.1787/4d7c6483-en.
Co-operative regions, like the pilot regions, can help build economies of scale, not just in planning but in service delivery and administration. In general, this type of region forms through a collaboration among existing local governments to manage functions that benefit from co-ordination at a higher, regional level. The potential for economies of scale depends on the function in question, but common responsibilities include regional development, spatial planning and the management of public investment funds, including EU funding. Sometimes they have functions in public service provision. For example, Iceland’s regional associations of municipalities are charged not only with preparing and implementing regional development plans for their regions, but also tasks delegated by municipalities like waste collection and the administration of school offices (OECD, 2022[8]). The scope of the work of co-operative regions usually reflects areas where regional-level operations add the greatest value (OECD, 2022[4]).
However, it would be a mistake to assume that the pilot regional councils alone would automatically translate to any greater co-ordination than joint planning. Shared regional plans can certainly support joint decision making and policy implementation among local authorities. At the very least, they create a shared vision of a development future to which constituent municipalities can align. Ideally, they detail actions that require municipal co-operation in order to implement. This is true in both regional development agreements, which unite local authorities to, for example, consolidate the organisation of public transport in South Estonia and create new vocational education opportunities in Central Estonia (Central Estonian Pilot Regional Council, 2024[9]; South Estonian Pilot Regional Council, 2024[10]). However, given that co-operation among municipalities is rare in Estonia, the country would need to address barriers that have historically limited co-operation to allow inter-municipal co‑operation to flourish, including through regional councils and regional development agreements (Box 3.2).
Inter-municipal co-operation offers a flexible and pragmatic tool for helping Estonia’s municipalities improve service delivery, address the challenges of population decline and overcome capacity constraints. Inter-municipal co-operation could help Estonia’s municipalities improve service efficiency and address challenges related to population decline. Economies of scale are particularly important for capital-intensive services such as education, infrastructure maintenance and waste management. By pooling resources, municipalities could provide higher-quality services at a lower cost. Additionally, inter-municipal co-operation can help maintain local democracy by enabling small municipalities to retain some decision-making power while benefitting from shared services.
Inter-municipal co-operation in Estonia remains limited despite the potential benefits it could offer. Following the 2017 Administrative-Territorial reform, some co-operative structures were maintained at the county level, particularly for public transport and spatial planning. However, compared to other OECD countries, Estonian municipalities have been slow to embrace inter-municipal co-operation. One key reason is that Estonian municipalities rely heavily on central government transfers, which weakens financial incentives for municipalities to collaborate voluntarily.
Focus group interviews suggested that, even when the will to collaborate is there, the practicalities often prevent civil servants from doing so. Local officials, already overburdened with their core tasks, often struggle to carve out the time and secure the necessary resources to engage in co-operation with other municipalities. For this reason, voluntary co-operation tends to occur only when an urgent need or a dedicated funding stream exists.
For inter-municipal co-operation to be more widely adopted in Estonia, the OECD has identified several barriers to be overcome. First, the legal framework for voluntary inter-municipal co-operation should be clarified to make agreements easier to establish and exit. Second, the transfer system could be adjusted to provide financial incentives for municipalities to co-operate. Third, national and countywide spatial plans should outline clearer guidelines for municipal collaboration in areas like education and infrastructure. Finally, the central government could play a more active role by supporting inter‑municipal co-operation pilot projects, building local government administrative capacity and facilitating knowledge-sharing among municipalities.
Sources: OECD (2022[4]), Shrinking Smartly in Estonia: Preparing Regions for Demographic Change, https://doi.org/10.1787/77cfe25e-en; OECD (2024[11]), “Stakeholder interviews”, OECD, Paris.
The pilot regional councils also promised to create a stronger connection among levels of government on regional development issues
County governments once served as a link between national and local government interests related to regional development, although they faced criticism for over-representing national needs. When county governments were still in place, state-appointed county governors were charged with ensuring the development of the county. Their county administrations developed county plans in co-operation with the national and municipal levels of government, including, for example, through negotiations with both national and local governments. They thus played a role in searching for common ground in the interests expressed by different stakeholders, ensuring that the county development plans included the different inputs in a complementary way. That is not to say that the county administrations performed this role as well as local actors would have liked; in fact, county plans at the time were subject to the criticism that national government needs were too prominent and also that these plans were not attached to state budget lines (Arenguseire Keskus, 2017[12]; OECD, 2011[13]). With the dissolution of county governments in 2018,3 local authorities lost this process of negotiation between national and local interests. One group of experts went so far as to say: “with the abolition of county governments, the previous coherence between the central government and local governments at the county level disappeared” (Kattai et al., 2020, p. 16[3]).
Their replacements – the county development organisations – do not fill the same role of co-ordinator and mediator between levels of government. The legal forms of county development organisations vary: they may be municipal associations, county development agencies or simply a local government itself (Kattai et al., 2020[3]) – but they look very different from the previous county governments. Importantly, they lack the built-in national perspective that a state-appointed county governor provided. Thus, while there is a requirement for county plans to reflect national objectives (see Chapter 2), there is no longer a built-in mechanism that brings national and local governments together to consider how they can find common ground on and then advance the development of the territory.
The Estonian Government hoped the pilot regional councils would bring back one of the advantages of the county governments: strengthening the connection between levels of government on regional development issues. The pilot councils would serve as a mechanism to bring together national and local governments in a deliberative discussion, where each brings their own perspective, and seek to find areas of mutual benefit to carry forward together. This, in theory, could have two effects. First, bringing representatives from different sectors of the national government around the table to develop county plans could favour national support and guards against local government “shopping lists” (a complaint about today’s county plans, discussed further in Chapter 2). Second, it would elevate local voices to national-level decision making and encourage ministries to consider regional perspectives in policy making, something that is currently not routinely done. With national government council members coming from different ministries, the way national government is represented on the councils was also meant to advance greater cross-sector co‑ordination (Chapter 2) (Estonian Government, 2023[14]).
Ultimately, the improved connection between levels of government offered by the pilot regional councils was meant to help both levels of government better address regional development issues. Even simple dialogue and information sharing with subnational governments can help national governments ensure that national policy and investment can better meet local needs, as discussed in the previous chapter (OECD, 2019[15]). The councils’ negotiation process also promised to encourage central and subnational levels of government align their objectives and activities, which also would help them avoid conflicting activities or misallocated resources. As the following sections will show, however, the way the regional councils and their agreements are designed and executed will determine whether they deliver fully on the benefits of better engagement between the two levels of government.
The councils were designed to bring other stakeholders to the table
One of the motivations behind the pilot regional councils was to foster collaboration among government, businesses and research institutions, with a view towards harnessing their perspectives and resources to promote specialisation and competitiveness. The pilot regional councils aimed to integrate the voices of companies and research institutions into regional planning and set the stage for co-operation among all participating actors towards the agreements’ objectives. Doing so, the Estonian Government believed, would offer two benefits:
Promote innovation in the two regions: The Estonian Government’s initial focus on bringing private and knowledge actors to the table was born of an early focus on innovation-led regional growth. At the beginning, the government envisioned a focus on research and development (R&D) and innovation as a way to promote regional competitiveness. To best identify and then act on opportunities for regional R&D and innovation, the initiators of the councils believed having the businesses and research institutions around the table was a must (Estonian Government, 2023[14]). This idea reflects the triple helix model of innovation, which emphasises close collaboration between among government, universities and the private sector. Collaboration among the three groups can create alignment between government priorities, academic research and private sector activity, so that, for example, the private sector is better able to turn scientific knowledge into commercial applications to drive economic growth (Etzkowitz, 2008[16]).
Ensure workforce needs for regional development goals are met: The Estonian Government had asked pilot regional councils to define focus areas for regional development and wanted to ensure that there would be a workforce with the appropriate skills to take these areas forward (Estonian Government, 2023[14]). Thus, including educational institutions would encourage them to adjust their approach to prepare a workforce that could drive the regional development vision of the councils.
While the councils’ focus quickly evolved to consider a broader range of regional development drivers, the core aim of bringing stakeholders to the table remained unchanged. The substantive focus of the pilot regional councils broadened from the initial focus on competitiveness and specialisation: the government and the council members recognised that the regions could benefit from considering a broader set of regional development drivers, including, for example, infrastructure and service delivery. Despite the shift in focus, the core principle of bringing stakeholders to the table remained, albeit with a different aim, no longer seeking opportunities for innovation but rather encouraging networked co-ordination to address complex regional development challenges.
This section examines how the government’s ambitions for the councils were translated into practice through the institutional design of the councils, and what design choices meant for the councils’ ability to function effectively. It discusses three institutional design choices that shaped the pilots: who participated, how decisions were made and how stakeholders were engaged throughout the process. First, in terms of participation, while the councils successfully established a tripartite structure – including representatives from national government, local government and business and academia – narrow ministerial participation constrained their ability to influence national planning and policy. Next, the section explores decision making within the councils: during council negotiations, members struggled to balance their roles as independent advisors and representatives of their institutions, generating questions about on whose behalf they were participating: their institutions’ or their own. Finally, the section looks at stakeholder engagement, which evolved over time to include a greater range of voices, granting the councils a broader perspective on regional needs and ambitions. The way each of these design choices played out in practice provides lessons on how to shape future councils.
Lesson 1: Broaden national engagement to support the agreements
The pilot regional councils were designed as a multi-level governance tool to unite national government, local government and stakeholders in driving the development of the two regions. Their tripartite structure reflects the Estonian Government’s recognition that, while the government views addressing regional disparities as “primarily the responsibility of the central government”, success depends on close collaboration with local governments and key stakeholders (Estonian Government, 2023[14]). Each group would play a different role:
National government: The intention was that council members from the national government – nominated by relevant line ministries – would play a leading role in integrating regional needs into national policy frameworks and ensuring inter-ministerial co-operation to do so. They would ensure alignment between regional development agreements and national strategies, including sector strategies across government, adjusting their activities (including investment) where relevant to meet their commitments in the regional development agreements (Estonian Government, 2023[14]).
Local government: The role of local government members was to reflect the priorities and needs of the municipalities in the region. Members, which may come from county development organisations or from larger local governments in the region (e.g. Tartu), are nominated by the development organisations of the counties in the region (Estonian Government, 2023[14]). In practice, representatives from local governments tended to be top leadership of county development organisations or elected officials (when it was established, the South Estonian Pilot Council included four mayors) (Estonian Ministry of Public Administration, 2023[17]).
Other key stakeholders: This category includes other parties, such as regional businesses, universities and researchers, which should make up at least one-third of the council. As introduced above, the Estonian Government placed a strong focus on including members from businesses and universities in the pilots to favour regional development agreements that reflect the needs of local business and opportunities for regional innovation (Estonian Government, 2023[14]).
The pilot regional councils succeeded in gathering all three categories of participant, but national representation was limited. The government hoped that the pilot regional councils would unite a broad range of ministries with local governments, businesses and universities. Its concept note, sketching the contours of the councils, included a list of potential member organisations selected to ensure representation across the most critical issues of the councils’ focus areas (originally regional specialisation and competitiveness, which evolved to consider regional development more broadly as described above). The Estonian Government confirmed the final composition of each council through a decree that named the individual members. While the final councils had roughly equal numbers from each of the three key participant categories, the group of national government representatives was narrower than outlined in the concept note (Table 3.3). Beyond the Ministries of Finance and of Agriculture,4 both councils included representatives from only two additional ministries: the Ministry of Education and Research, and the Ministry of Economic Affairs and Communications (Estonian Ministry of Public Administration, 2023[17]; 2023[18]). One additional ministry participated in discussions, even though it was not a formal council member: the Ministry of Climate. This highlights the difficulty of securing engagement from line ministries in regional development, a challenge explored in detail in Chapter 2.
Note: Those categories of participant that were not represented on the councils are listed in italics.
LEADER action groups are groups of people and local organisations that form within the European LEADER programme, designed to involve local actors in rural areas in the development of their regions (European Leader Association for Rural Development, 2025[19]).
Source: Estonian Government (2023[14]), Pilot Project on Regional Development Plans and Councils.
For future regional councils, the Estonian Government can take steps to broaden ministerial representation. It can target ministries whose involvement would have strengthened past councils, such as those responsible for environment, finance and social affairs. As the pilot regional councils begin to show the results of their work, more ministries may naturally wish to be represented on the councils. To encourage wider participation, the Ministry of Regional Affairs and Agriculture can proactively communicate the results of the experiment to target line ministries. Its pitch can highlight how adopting a regional lens can help these ministries advance their sectoral objectives more effectively (discussed in Chapter 2). As shown in the example from Iceland (Box 3.3), the councils can also show their value as a ready-made “bridge” for engagement between national and local governments. Additionally, securing leadership-level attention for the councils, e.g. by ensuring regional council representation at the Economic Cabinet (discussed in Chapter 2), could help to attract stronger high-level commitment that would encourage line ministries to actively participate in regional development efforts.
Iceland’s Regional Plans of Action are five-year agreements developed for each of the country’s eight statistical regions. Introduced in 2012, they are the primary legal tool through which municipalities co-operate regionally and co-ordinate development priorities. The plans guide investments in priority areas related to regional development, cultural initiatives, and social progress. These include projects designed to strengthen local communities, support economic diversification, and enhance quality of life across Iceland’s regions.
Ministries contribute funding to the implementation of the Regional Plans of Action, and the plans have become a mechanism for connecting national policy goals with local action. The Regional Plans of Actions started with two participating ministries, the Ministry of Infrastructure and the Ministry of Culture, Innovation and Higher Education, both of which contribute funding for the implementation of the plans. As time went on, the Ministry of the Environment, Energy and Climate decided to participate too, recognising the value of Regional Plans of Action as a bridge between national and local governments and a way to meet national objectives at the regional level. There is an expectation that the Ministry of Industries will take on a direct role in the agreements in the near future.
Over time, the plans have strengthened the regions' ability to develop and implement a shared strategic vision. At the end of the first five-year period, for example, an independent report on the results of the action plans reported that “the regional associations are now well equipped to take on increased responsibility and manage additional funding.”
Sources: Iceland Ministry of Infrastructure (2023[20]), “Interview with the OECD”, OECD, Paris; Icelandic Government Steering Group on Regional Affairs (2020[21]), Sóknaráætlanir landshluta, https://www.stjornarradid.is/verkefni/fleiri-rit/rit/2020/11/23/Greinargerd-um-soknaraaetlanir-landshluta-2015-2019/.
Lesson 2: Clarify council member roles to avoid an uneasy relationship between “council members as expert advisors” and “council members as signatories”
During the negotiation of the regional development agreements, council members were expected to act as expert advisors. The government specified that council members should not be considered direct representatives of their organisations (Estonian Government, 2023[14]). They were asked to make decisions beyond the narrow interest of their own organisations, instead focusing on big-picture regional needs. This meant applying their own sector, business, organisational lens to the work of the council and, for non-ministry council members, thinking as residents or interested community members of the region. This approach was intended, in part, to prevent the need for participants to align positions with their home institutions before meetings, allowing for more dynamic discussions. The council members were asked to contribute their expertise, rather than their representative voice (Estonian Ministry of Regional Affairs and Agriculture, 2023[22]). The government also acknowledged that some key stakeholders (e.g. non-profit organisations, businesses and local authorities not directly represented on the councils) would not have a seat at the table and tasked council members with considering their perspectives as well. By asking the councils to speak for them too, they were meant to function as more than just the sum of their parts (Estonian Government, 2023[14]).
At the same time, there was always an expectation that signatories would make concrete commitments for activities that would bring about agreed-upon regional development goals. While a group of expert advisors might be expected to conclude with a set of recommendations, the councils were asked to conclude with an agreement (Estonian Government, 2023[14]). The regional development agreements were envisioned as contracts, which requires that signatories commit to a set of mutual obligations (OECD, 2007[23]). The below excerpt from the minutes of one of the pilot regional councils shows how council members were asked to consider themselves both expert advisors and institutional representatives.
“[The chair of the council] notes that one of the questions that arises is who this council is and who the council members represent, what is the substantive and legal competence of the council.
[The Ministry of Finance council support member] suggested that in this case the council members should be treated primarily as a body of experts with a broader perspective, whose task in preparing a regional development agreement is not to represent the interests of just one organization, ministry or county. If the development agreement is ever signed, it is important who exactly the signatories of the agreement represent in doing so.”
Excerpt from the minutes of the South Estonian Pilot Regional Council (Estonian Ministry of Regional Affairs and Agriculture, 2023[22])
The unclear distinction between personal input and institutional representation created challenges for participants in decision making. Local government representatives sometimes hesitated to make decisions in meetings, uncertain whether they were expected to speak independently or as formal representatives of their organisations. Meeting minutes, for example, show council members reiterating that they lack a mandate to formally agree by signing onto – or commit funding to – the councils’ goals (Estonian Ministry of Regional Affairs and Agriculture, 2024[24]; 2023[22]). Similarly, national government representatives often sought prior confirmation from their ministries before committing to decisions. This may be partly due to the level of officials sitting on the councils; they may not have felt sufficiently empowered to make commitments on behalf of their institutions. Whatever the root cause, questions about decision making and accountability reduced the ability of council members to make decisions and to commit to the resulting agreement.
For future regional councils, the Ministry of Regional Affairs and Agriculture could begin by establishing guidance on the role of council members. These guidelines should specify expectations around council members’ formally representing their organisations (discussed in greater depth in the following paragraphs). These guidelines should also include information on decision making. For example, they can stipulate how council members will report back to their home institution, and when and how council members will have the option of validating their positions with their home institutions. These guidelines should be sent to potential council members with a request to share them with other decision makers in their organisation. The goal is to ensure that council members and their home institutions are aware of the expectations for council members so that neither the council member nor their colleagues balk at these demands during the process of developing the agreement.
If the government wants council members to serve both as expert advisors and as institutional signatories, it can clearly separate these functions within the agreement process. To sharpen the line between the two roles of council members – expert advisor and representative signatory – one option is to structure the agreement process in two distinct phases, making a clear distinction between consultation and negotiation (Box 3.4). Participants in each phase can have a different, clearly defined role:
1. Problem and solution scoping phase: During this stage, council members would act as independent experts, focusing on identifying regional challenges and opportunities beyond their own organisational interests. This phase would correspond to the consultation phase in Box 3.4.
2. Agreement negotiation phase: During this stage, council members would cease to serve as advisors, and instead step into their representative role to negotiate a final agreement among signatories. This phase would be dedicated to arriving at a final agreement that aligns with the institutional priorities of signatories, one they are comfortable signing. Participation in this phase may differ slightly from the previous phase. First, some council members may not plan on signing the agreements (this holds true even for the current pilots: the list of expected signatories on the two draft agreements, for example, does not include any of the entrepreneurs who participated) (Central and South Estonian Pilot Regional Councils, 2024[25]). At the same time, some other bodies may wish to sign the agreement, even if they did not participate as council members (for example, local authorities without direct representation on the councils or other relevant ministries).
During this stage, signatories are acting as representatives of their organisations, with the goal of negotiating strong and formal organisational commitments in the final agreement. To make sure that signatories are speaking with the voice of their organisations, councils could work into the schedule opportunities for signatories to consult with their home institutions in advance of key decisions. There is precedent for this already: the South Estonian Pilot Council solicited written responses by council members to documents produced by a working group, including official response letters with ministers’ signatures (Estonian Ministry of Regional Affairs and Agriculture, 2024[26]).
Contracts among levels of government typically follow a multi-stage process to arrive at an agreement, consisting of:
1. Consultation: This initial phase involves identifying shared priorities, assessing regional needs and engaging a broad range of stakeholders, from public authorities to civil society. This phase may be lengthy and involved: for example, in France, a two-year consultation process leading to State-Region Contracts brought together national and regional actors to define investment priorities collaboratively. Designing shared objectives is facilitated by the use of indicators agreed upon by all so priorities are clearly evidence-based.
2. Negotiation and signature: In this stage, formal agreements are drafted and negotiated between legitimate representatives of central and subnational governments. This stage sees parties defining the roles, responsibilities and funding mechanisms that will shape the activities under the contract. In order to not forget any key stakeholder and formalise the system for negotiation, it is useful to produce an “institutional mapping” clarifying “who does what and with whom.” During the negotiation of the French State-Region Contracts, parties pay special consideration to sequencing and financial contributions (which may, although not systematically, include transfers from the national government).
3. Implementation: This phase translates commitments into action. Depending on the contents of the agreement, it may involve not only the contracting governments but also private actors, associations and other local institutions.
4. Enforcement: Contracts should have mechanisms in place to ensure that both parties meet their obligations and to resolve conflicts if they arise, as described in greater depth in the section. For example, Italy’s Pacts for the South between Italian regions or cities and the national government included provisions for regular six-month reviews and allow for sanctions if milestones are not met.
5. Evaluation: Ex post evaluation helps assess performance, learn from experience and adjust future contracts. Carefully defined indicators will help parties know whether the contract has met objectives, whether new funding is needed, whether it should be renewed and, if this features in the agreement, whether rewards or sanctions should be given to parties.
Sources: Charbit, C. and O. Romano (2017[27]), “Governing together: An international review of contracts across levels of government for regional development”, https://doi.org/10.1787/ff7c8ac4-en; Charbit, C. (2025[28]), “Slides: State-Regions Contracts to support administrative reforms and regional development”, OECD Webinar on State-Regions Contracts - 11 April 2025.
An alternative approach would be to separate advisory and signatory roles within the regional governance model. The councils could be split into a dual structure: one body focused on strategic guidance and expertise (including private sector, civil society and academia) and another tasked with decision making and implementation (comprising institutional signatories, such as municipalities or representatives nominated by municipalities and relevant national ministries). In this model, the central government could engage primarily with the signatory group through formal negotiations and oversight mechanisms, while the agreements still benefit from the contributions of the advisory group in shaping priorities and assessing progress.
Future regional councils could be encouraged to include relational aspects to their contracts (Box 3.5) when parties are able to agree to objectives but not yet define the steps and actions necessary to reach them. For example, take the proposed objective in the draft South Estonian Regional Development Agreement to ensure that the region’s enterprises can access renewable energy at an economically favourable price (South Estonian Pilot Regional Council, 2024[29]): instead of defining the tasks that this objective would entail in a set of transactional commitments, the agreement could include a relational commitment that details how parties will work together to bring about the commitment. Signatories could agree to work together with local government to meet the shared objective of ensuring new businesses have access to the energy they need without excessive upfront costs, within a specified co-operation and decision-making structure. Doing so can still lay a strong foundation for implementation, with signatories committing to working together within a specified framework and towards a specified objective, while allowing the steps and activities to be defined as they become clear.
Experts make a distinction between transactional contracts, which rigidly define obligations in advance, and relational contracts, which establish a framework for ongoing co-operation.
Transactional contracts resemble market exchanges, relying on formal rules and legal enforcement to manage well-defined tasks and contingencies. They are best suited for simple co‑ordination problems, where all relevant outcomes and obligations can be anticipated.
Relational contracts are designed for long-term co-operation under conditions of uncertainty. Rather than detailing fixed actions, these contracts establish governance frameworks that facilitate trust, learning and adaptive collaboration. They are especially effective where goals cannot be fully specified ex ante.
In practice, many contracts combine both: for example, agreements that require signatories to achieve specific targets may also include open-ended provisions that are defined during implementation.
Source: OECD (2007[23]), Linking Regions and Central Governments: Contracts for Regional Development, https://doi.org/10.1787/9789264008755-en.
Lesson 3: Design stakeholder engagement for well-rounded agreements
With a limited membership, future regional councils must look beyond their ranks to best represent regional interests in the agreements. Regional development involves complex, multi-dimensional issues, as described in Chapter 2. A relatively narrow representation of public sector, private sector and academic actors does not provide a comprehensive view of regional dynamics, needs and opportunities. Without broader engagement, the councils risk having blind spots in their final agreements. This is one reason why stakeholder engagement should play an important role in regional development agreement processes: it allows the councils to hear from a broader range of voices, including non-profit organisations and communities who have a direct stake in their region’s development (OECD, 2023[30]). This engagement is crucial for gathering and co-producing the knowledge needed to identify regional needs and leverage regional assets (OECD, forthcoming[31]; 2018[32]).
Beyond informing decision making, stakeholder engagement can also play a role in maintaining the legitimacy of the pilot regional councils as well as future regional councils. The pilot councils are not elected bodies but rather appointed by the central government, although the government is supposed to take into account proposals from county development organisations and universities when appointing these members (Estonian Government, 2023[14]). Regional bodies without directly elected leaders can face criticism for lacking democratic legitimacy. Critics argue that the absence of direct elections can create a disconnect between regional governance bodies and citizens, as the leadership may not reflect the diverse interests and needs of the population they serve. Additionally, these bodies lack direct accountability to voters, who cannot remove representatives through elections (OECD, 2022[33]). Keeping stakeholders engaged helps to provide oversight and accountability, ensuring that councils act in the best interest of their regions.
To develop their agreements, the pilot regional councils drew on input from multiple stakeholder groups. At the outset, the planned stakeholder engagement prioritised businesses, with focus areas for development to be identified using regional data analysis and entrepreneur interviews (Estonian Government, 2023[14]). The Estonian Government recognised that relying solely on quantitative data and interviews with entrepreneurs would not provide a full picture of regional challenges and assets. It worked with the OECD to broaden input into the agreements by involving a wider range of stakeholders in foresight/futures and vision-setting workshops in each region. These workshops – designed to explore potential futures and begin to develop a shared vision for regional development in each region (described in Chapter 1) – brought together a diverse group of regional stakeholders. Participants included a range of local government officials, national government ministries, county development organisation representatives, tourism and cultural sector representatives, a representative from a local hospital, even a member of the Estonian parliament from one of the regions, as well as a range of business and industry representatives (OECD, 2023[34]). Broadening engagement through these workshops helped capture a fuller picture of regional strengths, challenges and aspirations.
Future regional councils should preserve an emphasis on broad engagement. They may consider starting with inclusive foresight/futures and vision-setting workshops that bring a wide range of stakeholders together to collectively define regional challenges, opportunities and aspirations. Starting with a collective exploration of the future of the regions gives regional councils something to aim towards, laying the foundation for an agreement that can best serve the regions. Following this broad visioning phase, councils can then narrow their engagement to target groups during the drafting phase, for example, entrepreneurs, industry groups and sector-specific experts. At this stage, the focus of the engagement shifts from developing a collective foundation for the regional development agreements to refining the “how”, i.e. the activities needed to bring about the ambitions of the regions.
They should also make explicit how they will continue to engage with stakeholders during the implementation of the regional development agreements. Stakeholder engagement does not end with the signing of the agreements but will continue during implementation. Future regional councils should make explicit how they will keep stakeholders informed and/or involved during the implementation phase (while it is signatories that will implement the agreements – discussed in the following section – it is the councils who will monitor them). To keep up momentum, councils should establish regular engagement mechanisms and/or status updates with relevant parties, e.g. ministry officials, local authorities and advocacy groups, who play important roles in translating agreements into action. The Central Estonian Pilot Council has taken a first step in this direction by foreseeing in its agreement the organisation of regular conferences to share progress towards objectives (Central Estonian Pilot Regional Council, 2024[9]). These touchpoints can help to maintain support and commitment of relevant parties to the implementation of the agreements, which will be particularly important if changes arise during the implementation process that require additional funding, staff or time. It also requires the councils to be attuned to changes in priorities or leadership that could affect the ability of parties to implement the initiatives in the agreement (Michels and Wu, forthcoming[35]). Additionally, offering opportunities to scrutinise the decisions and outcomes of regional bodies through regular and transparent reporting can help mitigate the risk of decisions that counter the region’s interests (OECD, 2022[33]).
To broaden national government representation in the pilot regional councils, Estonia can:
1. Share the pilot councils’ results with target line ministries, highlighting how engaging with the councils can advance participating ministries’ goals. Cultivating greater political support for taking a regional approach – see recommendations in Chapter 2 – can encourage more ministries to seek representation on future regional councils.
To help overcome discomfort of regional council members to make decisions, Estonia can:
2. Establish clear guidelines on when members act as independent experts versus institutional representatives, potentially structuring the process into two phases, one for scoping and one for formal agreement negotiation that includes built-in opportunities for consultation with home institutions before decisions.
3. Encourage councils to incorporate relational contract elements into their agreements when they are able to agree to objectives, but not yet define the steps and actions to achieve them.
To ensure that stakeholder engagement captures a broad range of voices, both in shaping the agreement and throughout its implementation, Estonia can:
4. Begin with inclusive foresight and vision-setting workshops, bringing together diverse stakeholders to collectively define regional challenges and opportunities before narrowing engagement to targeted groups during the drafting phase.
5. Establish clear mechanisms for ongoing engagement during the implementation of the agreement to maintain support and resources, and adapt to changing needs.
The work of the pilot regional councils led up to two draft regional development agreements. The previous sections explored the rationale behind the councils and how they functioned in practice. This section examines their first results: the pilot regional development agreements. It uses observations about the first pilot agreements to distil lessons on what future agreements should include to be credible, actionable and enforceable. In particular, it highlights three foundations for future agreements: securing credible commitments, clarifying implementation responsibilities and strengthening enforcement.
Lesson 4: For credible and actional agreements, continue to avoid “shopping lists” of unsupported commitments
Council members recognised that ministerial support was essential to realising the promises of the regional development agreements, both to favour a cross-government approach and to secure the resources needed for implementation. The Ministry of Regional Affairs and Agriculture’s vision for the councils relied on broad ministerial participation, with line ministries expected to make concrete and co-ordinated commitments and provide resources to support implementation of the plans. Other council members from the two regions shared this perspective but were particularly focused on the funding and investment ministries could bring to their regions. They pointed to past experiences, such as the South Estonia Smart Specialisation Strategy, which struggled due to insufficient national backing (OECD, 2024[11]). To make the regional development agreements credible, the councils emphasised the need for each signatory to make clear and balanced commitments rather than producing documents that simply outlined local demands (OECD, 2024[36]). Without commitments from ministries across the national government, council members saw a risk that the agreements would be limited in their ability to promote a regionally co-ordinated approach.
Council members agreed that key to ministerial buy-in was avoiding “shopping lists” of local authority wishes for national government support, which the draft agreements have largely avoided. The fact that county plans sometimes included “shopping list” items – activities that local governments in the county would like the national government to support – is one of the reasons local and national government stakeholders consider them less effective than they could be (Chapter 2). Council members warned of the same thing happening with the regional development agreements (OECD, 2024[36]). As explored in Chapter 2, if an agreement is perceived as a collection of municipal funding requests without clear mutual benefits for national and local authorities, ministries may see little incentive to commit resources or policy support. The two pilot agreements have steered away from this and ministries evoked in the activities of the agreement are, for the most part, those who sat on the councils and helped shape the commitments. This should be maintained in future agreements.5 Local authorities will still have the opportunity to voice their requests for their regions but during the negotiation of the agreement instead of at its release.
Future agreements should concentrate on ensuring “hard” commitments: direct and concrete commitments from signatories. These mutual commitments – a core feature of contracts among levels of government (Box 3.6) – provide the strongest foundation for implementation, ensuring ministries are aware of their commitments and can be held accountable to them. The councils can favour this kind of commitments from a wider group of ministries by broadening ministry representation on the councils (see recommendation on this in Chapter 2). However, signatories to the agreements do not necessarily have to be council members. Future regional councils can consider how to actively engage and negotiate with individual ministries outside the council to secure concrete commitments that align with their regional development objectives. Further, important non-council members could, for example, be integrated into collective negotiations during the negotiation phase of agreement preparation, as suggested above.
A contract represents a mutual agreement in which the parties commit themselves either to take specific actions, or to follow a jointly established decision-making process (depending on whether they are transactional or relational contracts, as explained in Box 3.5). Contracts define the mutual obligations of parties, including:
Authority: who has the rights of decisions among parties.
Mutual duties: the distribution of contributions, including funding, human capital, etc.
Enforcement: mechanisms to enforce mutual contributions.
Clarifying responsibilities and making mutual commitments explicit makes contracts actionable and, combined with mechanisms for enforcement, make sure that signatories are held to their commitments.
Lesson 5: Clarify who is doing what to implement the agreements and with what resources
The draft agreements do not always make clear who is doing what. Items in the draft agreements sometimes lack a responsible party (e.g. an activity dedicated to encouraging labour mobility in the Central Estonian agreement, which states simply “[t]he country should develop measures to encourage labour mobility” (Central and South Estonian Pilot Regional Councils, 2024[25]). Some tasks invoke multiple parties without specifying how the activity can be accomplished, like the task in the South Estonian agreement that states that “universities, companies and local governments [will strengthen] co-operation” (Central and South Estonian Pilot Regional Councils, 2024[25]). Without a specified lead, these tasks risk being neglected or poorly co-ordinated. It is important that all activities have a designated responsible party to ensure accountability and effective implementation.
The agreements also do not specify what resources will support implementation. The pilot councils received initial organisational support from the Estonian Government,6 but they were not promised funding for implementation. Instead, they were to reach an agreement among signatories for how the implementation of their agreements would be funded. The agreements do not include a firm plan for implementation funding, although councils have ideas. The Central Estonian agreement includes two proposals for how to fund its implementation: either EUR 5 million per year from the national government, to which local governments would add EUR 1.5 million or a share of value added tax (VAT) (they provide 0.5% as an indicative figure) to fund implementation (Central and South Estonian Pilot Regional Councils, 2024[25]). These are just proposals, however, and there is no confirmed, long-term source of funding for implementing the agreements. Nor are there clear commitments from parties to fund specific initiatives. The most specific is a proposal in the Central Estonian agreement that says the Ministry of Regional Affairs and Agriculture should increase funding for the Central Estonia Scattered Settlement Program by 25%. Otherwise, activities in the agreements do not specify the resources that will be required, nor where they will come from. For example, a goal to develop a Central Estonian Science Park in the Central Estonian agreement only notes that the park will be founded by local governments, universities (Estonian University of Life Sciences) and the Ministry of Regional Affairs and Agriculture with no mention of who will contribute what resources towards this goal (Central Estonian Pilot Regional Council, 2024[9]). A lack of confirmed financial commitments from national and local governments, as well as other stakeholders, leaves the agreement at risk of stagnation.
Future regional councils can start with an institutional map that outlines the roles, responsibilities and linkages between all those involved in implementation, which can be formalised in a founding charter. This map should clearly identify which party is responsible for each task, including for tasks requiring collaboration across multiple entities, such as ministries, regional actors and local actors. By mapping these roles and relationships, the councils ensure that every party understands their responsibilities, and potential gaps in co-ordination can be more easily identified. In addition to showing roles, the maps should highlight the financial contributions from participants. This will help to reassure stakeholders that the agreements are feasible.
The agreement itself should also clearly define roles, responsibilities and resources. Doing so helps to ensure that agreements are realistic, avoiding, for example, unsupported tasks like the “shopping list” items for ministries discussed above. In addition, a well-structured contract creates effective enforcement mechanisms, discussed further below, to resolve conflicts and ensure commitments are upheld. Clarity on roles, responsibilities and resources makes it easier to hold parties accountable: it is simpler to measure contributions and overall progress during implementation when these factors are well-defined.
Lesson 6: Boost potential for success through strong enforcement mechanisms
While the pilot agreements include monitoring provisions, they lack built-in enforcement mechanisms to ensure that parties adhere to their commitments. The agreements do include provisions for monitoring. In Central Estonia, the agreement states that progress will be monitored by the pilot regional council annually, aided by the preparation of a monitoring report. The South Estonian agreement also includes a provision for annual monitoring and adds that the minister responsible for regional development will present an annual update to the Estonian Government. The Central Estonian agreement lacks explicitly defined performance indicators but mentions that they will be developed in the action plan, while the South Estonian agreement includes some broad performance indicators (e.g. regional gross domestic product growth and infrastructure targets). Neither, however, include any specific provisions for enforcement (Central and South Estonian Pilot Regional Councils, 2024[25]).
Without mechanisms to ensure that each party fulfils their commitments, the implementation of an agreement could flounder. One of the foundations of a contract is defining the binding reciprocal commitments of the parties, minimising ambiguities that could hinder implementation or create conflicts (Charbit, 2024[37]). Agreements can reduce the risk of non-co-operative behaviour through monitoring and evaluation systems that will allow stakeholders to know if parties are fulfilling their commitments (Charbit and Romano, 2017[27]). The pilot regional development agreements were right to include provisions for monitoring, even if they lack specific indicators. This monitoring is crucial to gauge whether contracts achieve their intended outcomes. It provides information that can help parties determine if a contract should be renewed, if additional resources are needed and if adjustments are required for future agreements. However, the agreements are missing monitoring’s counterpart: enforcement. Contracts often include enforcement mechanisms such as penalties for non-compliance and provisions for third-party oversight, which ensure that commitments are met. Together, monitoring and enforcement help to ensure that the agreements are meeting signatories’ objectives. Thus, the results of the monitoring exercise may show the councils where signatories are not delivering their contributions, but the councils will have no already-established mechanism to hold them to their commitments.
Like the pilot regional councils, future regional councils should include provisions for monitoring their agreements. Along with provisions for how monitoring should take place, agreements among levels of government should also include performance indicators. There is no one formula for determining the ideal performance indicators for an agreement but future councils can follow three rules of thumb. First, indicators should be measurable and linked to contract objectives. Second, the set of indicators should be broad enough in scope to mitigate the “streetlight effect”, where parties focus on meeting their specific indicators rather than overall policy goals. Finally, they should be used in a way that promotes transparency and accountability, which can provide an incentive for good behaviour (Charbit and Romano, 2017[27]). The combination of these three principles can help create a system that helps parties track progress, identify obstacles early, keep agreements aligned with objectives and reinforce incentives for parties to follow through on their commitments.
Future regional councils should consider which enforcement mechanisms they can use to strengthen their agreements. Without provisions for enforcement, the agreements lack assurance that the commitments made by the different signatories are credible. Like monitoring and evaluation, enforcement can come from parties to the agreement (internal enforcement) or beyond the agreement (external enforcement):
Internal enforcement can take the shape of reputational effects among agreement parties, which can create pressure for self-enforcement by these parties. Yardstick competition, where the relative performance of parties pushes other parties to behave efficiently, may also provide a source of internal enforcement. For agreements where money is on the line, financial awards can be adjusted based on performance.
To create opportunities for internal enforcement within the agreements, arrangements can be made for regular performance reports that present progress and measurable outcomes, creating pressure on parties to perform well. In addition, the agreements could tie agreed-upon financial disbursements to progress on specific targets (e.g. a specific project completed or a predefined number of businesses supported).
External enforcers can include judges or independent agencies. The courts can serve the role of mediator, extracting information and using judicial procedures to solve conflicts. Independent agencies, international experts and media can use “revelation pressure”, exerting pressure on parties by placing their actions or performance under scrutiny. Finally, stakeholders can also provide “social control”, with stakeholders and citizens penalising parties for non-compliance, including at the ballot box.
Future regional councils can put revelation pressure to work for them by creating regular opportunities for external scrutiny. For example, to create opportunities for external scrutiny of the agreements’ implementation, agreements could specify that performance reports would be shared with specified external parties annually (e.g. through annual in-person reports to the Economic Cabinet, the Association of Estonian Cities and Municipalities and/or the network of county development organisations). Doing so would create an added layer of accountability, as these external actors would be able to scrutinise progress and highlight any shortcomings.
These mechanisms help to preserve motivation by creating incentives to deliver on commitments. Incentives can be positive, too: instead of only imposing penalties, agreements can specify rewards for positive contributions and results, for example, by maintaining a performance reserve of funds that are accessible if parties meet targets (Charbit and Romano, 2017[27]).
To help ensure that future agreements are credible, actionable and realistic, Estonia can:
1. Eliminate “shopping list” actions for ministries in the agreements (unsupported requests that lack ministerial backing).
2. Develop an institutional map to clarify responsibilities, especially for multi-party initiatives and specify financial contributions. Ensure the agreements themselves also define roles, responsibilities and resources for implementation.
To hold signatories accountable to their commitments, Estonia can:
3. Include provisions for monitoring the agreements, including performance indicators that are measurable, linked to objectives, broad enough to encourage signatories to keep their eye on overall development objectives and used to enhance transparency and accountability.
4. Specify enforcement mechanisms, such as internal reputational incentives, financial performance-based adjustments or external oversight, in the agreements.
The pilot regional councils have reached the right moment for an evaluation, and this section considers how the Estonian Government can evaluate the pilots. They have reached a natural inflection point, having largely completed their primary task of arriving at the regional development agreements (in itself a monumental task involving identifying regional strengths, challenges and ambitions before negotiating the text of the agreement). Evaluating the effectiveness of the pilot regional councils is essential before making decisions on their future: the Estonian Government will need to know if the experiments have yielded results that justify scaling them up and if any adjustments need to be made. This stocktake will help the Estonian Government understand which elements worked, which did not and what modifications may be required to ensure that any flaws in the first pilots are not replicated. This section draws upon core lessons for policy experiments from other OECD countries (Box 3.7) to consider how the Estonian Government should proceed as its pilot regional councils reach this stage.
Policy experimentation refers to the structured testing of new approaches on a more limited scale – often through pilots – before wider implementation. Governments use policy experiments for all kinds of things: testing new service delivery models, piloting funding schemes, exploring alternative governance arrangements and more. They offer a lower-risk, evidence-informed way to introduce new initiatives, with the possibility of retiring them or refining them based on what actually works in context.
A review of OECD country experiences with policy experiments for regional development can provide some guidance for other countries designing experiments (or, in the case of a country like Estonia, implementing their experiment). These are the ABCDs of experimentation:
A – Articulate the scope
Experimenters should start by defining a clear rationale, goals and scope for the experiment, including the timeline, governance structure and evaluation checkpoints. While clarity upfront helps all participants know what to expect, allowing for flexibility in execution allows for real‑time learning and adjustments during implementation (revisited in “C – Create learning cycles”).
B – Bring stakeholders into the process
Participants in pilots are not passive subjects and experimenters can actively involve stakeholders, such as local governments, civil society and the private sector, in shaping, executing and evaluating the experiment.
C – Create learning cycles
A hallmark of an experimental approach is its ability to learn through action and reflection. Each experiment reaches a point when an evaluation is conducted: the “scale, shape or sunset” stage. This phase of the experiment involves a thorough review of outcomes. At this point, decision makers assess whether the experiment should be broadened (“scaled”), adjusted (“shaped”) or retired (“sunset”). This stocktake is crucial for understanding which elements of the intervention worked, which did not and what modifications may be required. It may be at the “end” of an experiment, or at the intermediate stage between two stages of an experiment.
D – Dare to fall short
The inherent uncertainties and potential for visible failure may deter governments from taking an experimental approach. Countering this risk aversion requires that governments cultivate a safe environment for experimentation, including by shielding experiments with supportive legal frameworks and making clear that undesired results from an experiment will not result in cuts in resources. Governments can also take care to praise and share good experiments, not just good results, to help build a culture of experimentation. Finally, experimenters themselves should seek to recognise missteps and failures, and use them to learn.
Source: Durand, A. and M. Vincent (forthcoming[38]), “The ABCDs of policy experimentation”, OECD Working Paper, OECD Publishing, Paris.
An evaluation starts by looking at an experiment’s motivation and objectives. The Estonian Government can return to its original motivations for the experiment, which were described in the first section of this chapter: enabling economies of scale, fostering engagement among and across levels of government, and involving businesses and research institutions to promote regional competitiveness (as described earlier in the chapter, this has since broadened to involving a range of regional stakeholders for a more well‑rounded approach to promoting regional development). It can also consider the government’s objectives for the councils: what they were meant to do. The Estonian Government’s concept note on the councils envisioned one overarching objective for the councils: to provide regional direction for activities and funding of national government (including the Ministry of Regional Affairs and Agriculture and line ministries) and local government, as well as funding from the European Union and other sources (e.g. the private sector) (Estonian Government, 2023[14]). Assessing these motivations and objectives against the actual outcomes of the experiment will help the Estonian Government determine whether and how well the councils have fulfilled their intended role.
The evaluation can then consider whether the experiment delivered on the ministry’s hopes for it. None of the founding documents for the regional councils – the Estonian Government’s original concept note, the establishing decrees – define how the pilot will be evaluated. The absence of predefined indicators and an evaluation framework makes ex post analysis of the councils challenging. Instead, the Ministry of Regional Affairs and Agriculture will have to define these retrospectively, developing indicators of pilot performance based on the ideas of what the pilots were supposed to do. By drawing a line from the intentions for the pilot regional councils to the actual results, the metrics that will help gauge whether the motivations and objectives behind the councils were satisfied (Table 3.4). Many of these indicators focus on process and immediate outputs of the pilots because the pilot regional councils are just now coming to an end of the drafting of their regional development agreements and these have not yet borne fruit that might be measurable in terms of broader outcomes (e.g. economic or well-being indicators). The ministry can also undertake a comparative analysis between pilot and non-pilot regions, where data are available.
These performance metrics do not have to be quantitative. Indeed, qualitative information plays an important role in evaluation when quantitative measures are not possible, and in order to contextualise quantitative measures (many of the metrics suggested in Table 3.4 are qualitative and rely on stakeholder perspectives, as discussed further below). For example, the Estonian Government may wish to gauge the impact of the pilots by understanding the extent to which the agreements fostered alignment in policy priorities among levels of government. To do so, it may be that the government could say that a certain number of sector plans included new references to agreed-upon regional priorities. It may also involve documenting how council discussions shaped funding decisions, influenced ministerial strategies or led to new mechanisms for alignment among levels of government. In those cases, the Estonian Government can describe the impact in a clear and objective way to allow other evaluators to assess at face value (OECD, 2008[39]). For the pilot regional councils, this can be as simple as including a qualitative description of these impacts (OECD, 2020[40]), explaining why the impact is relevant and the magnitude. While this type of evidence does not offer the rigour of a before-and-after or inter-regional quantitative comparison, it can still help the Ministry of Regional Affairs and Agriculture develop a well-rounded view of the performance and impact of the councils to make decisions about their future.
Soliciting stakeholder perspectives – and considering using an independent evaluator – can ensure that the evaluation benefits from broad and candid feedback. The pilot regional councils were designed to be collaborative, meaning that their effectiveness depends on the engagement and commitment of participants. For this reason, gathering information on the perspectives of participants, for example through a survey and interviews with council members, will be important to provide a fuller picture of the councils’ impact. The Ministry of Regional Affairs and Agriculture can also collect the perspectives of stakeholders that did not sit on the councils but that do have a stake in their activities. These may include, for example, representatives from local authorities without direct representation on the councils, or ministries that had some degree of engagement with the councils even without serving as council members (e.g. the Ministry of Climate). Involving non-participants can provide additional insights into both the relevance and impact of the councils’ work but also highlight any gaps in engagement and barriers to participation. The Ministry of Regional Affairs and Agriculture may choose to enlist the help of an evaluator external to the process to complete the evaluation. Making the evaluation independent can not only help to ensure that it is objective, but may also encourage participants to share more openly about their perspectives on the pilot regional councils.
If the Estonian Government decides to scale up the experiment, it should consider how to establish feedback loops to create new opportunities for learning and capacity building. Scaling up does not have to be an all-or-nothing decision. Rather than immediately scaling up the councils, Estonia could scale up in phases that allow for ongoing adjustments. If the Estonian Government decides it would like to bring additional powers (and resources, as discussed in the next paragraph) to the regional level, this sort of gradual “learning by monitoring” approach, such as that adopted by Sweden (Box 3.8), can also help the regions build capacity (OECD, 2018[32]). To this end, Estonia could take a phased approach to expanding the remit of the councils, where new responsibilities are introduced gradually and only after certain readiness thresholds are met. For instance, regions that meet certain conditions (e.g. demonstrating a strong process, e.g. sufficient stakeholder engagement, or certain desired results) could be the first to take on additional responsibility. By taking on new responsibility incrementally and meeting benchmarks, the regions would gradually increase their capacity, while the national government would gain greater confidence in their ability to manage expanded roles effectively.
Sweden took a phased approach to regionalisation, refining its system through a series of pilot initiatives and evaluations. The process began in 1997 with the Regional Pilot Project, where certain regions were granted new competences, including responsibility for regional development planning, EU funding allocation and transport infrastructure. These pilot regions experimented with different governance structures, such as elected regional assemblies and indirectly elected councils.
A mid-term evaluation of the pilot regions led to the next phase: the introduction of Regional Development Councils in 2003 as a model available to all regions. These councils, formed through municipal co-operation, were given limited responsibilities compared to the pilot regions but still played a role in co-ordinating regional development initiatives.
Finally, in 2007, the Committee on Public Sector Responsibilities issued recommendations for further regional reforms, recommending principles for the division of responsibilities among levels and suggesting a potential restructuring of Sweden’s regional map.
Sweden’s approach demonstrates how iterative evaluation and feedback loops can be used to gradually refine and expand regional governance models.
Sources: McCallion, M. and P. Tallberg (2008[41]), “Regionalisation in Sweden”, https://www.stals.santannapisa.it/sites/default/files/stals_Tallberg.pdf; OECD (2017[42]), Multi-level Governance Reforms: Overview of OECD Country Experiences, http://dx.doi.org/10.1787/9789264272866-en.
The “learning by monitoring” approach can extend to funding. For example, consider a scenario where the Estonian Government empowers the regions to execute their own regional development projects by systematically delivering additional national or EU funds through the regions, an idea that came up repeatedly among pilot regional council members and was alluded to in the draft Central Estonian agreement (OECD, 2024[36]; Central Estonian Pilot Regional Council, 2024[9]). Participating ministries might balk at contributing substantial funds, given the limited capacity of local authorities to plan, execute and invest (see Chapter 2). In this case, the Estonian Government could start by creating a small, time-limited pilot fund, pooling money from involved ministries and testing this arrangement before introducing a larger or longer-lasting arrangement. For example, it could ask relevant ministries to contribute to a fund that the pilot regional councils would administer to implement the agreements. Importantly, an impact evaluation plan should be prepared before the pilot begins. After one funding cycle, an evaluation could be carried out by an external, impartial body to ensure credibility. Based on this assessment, if the results of that exercise show that the councils managed the funds appropriately and with the desired impacts, the ministries could scale up by providing a larger pool of funds or a longer spending horizon. This incremental process – with funding ramping up incrementally as regions show signs of readiness – would allow local authorities to build capacity while the national government builds confidence in the ability of municipalities to handle additional funds.
To capture lessons learned from the pilot regional councils and inform decisions about scaling them up, Estonia can:
1. Evaluate whether the experiment met the Estonian Government’s expectations by linking the motivations and objectives behind the pilot regional councils to the results that they achieved. The evaluation can draw upon both qualitative and quantitative information to create a well-rounded picture of the performance and impact of each of the councils. Ultimately, this evaluation will help the Ministry of Regional Affairs and Agriculture to make decisions about their future.
2. Involve stakeholders in the evaluation of the pilots by soliciting their views, e.g. through a survey or interviews. Pilot regional council members are an important target of this recommendation, but the Ministry of Regional Affairs and Agriculture can also seek the perspectives of those without direct representation e.g. local authorities and ministries, who, while not sitting on the councils themselves, were engaged in the work of preparing the agreements.
To allow regional governance in Estonia to continue to develop in response to the needs of all parties and to the ability of parties to make the most of regional governance, Estonia can:
3. Consider a phased approach to scaling up the pilots, introducing new elements in a limited way to see how they work, and to gradually build the capacity and confidence of participants before committing to a larger-scale rollout.
The pilot regional councils promise to begin to lay the foundations for place-based policy for regional development in Estonia by promoting a cross-cutting approach across the government and place sensitivity to regional needs and opportunities. First, the pilot councils provide a structure that brings line ministries together to exchange on multi-dimensional regional development challenges, which can help to address a lack of systematic co-ordination for regional development across government. Second, the pilot regional councils offer a platform for exchange between the two levels of government (and with other important regional development actors), helping to address a disconnect in vertical co-ordination that restricts bottom-up input from shaping national strategy and policy. In short, Estonia’s pilot regional councils represent a step towards laying the foundations for place-based policy presented in Chapter 2.
The question is whether the pilot regional councils deliver in practice, however, and the moment is right for the Estonian Government to evaluate its pilots to see how they lived up to ambitions. To start its evaluation, the Estonian Government can look back to its original rationale for the councils, to see if and how they have created new economies of scale, fostered engagement between levels of government and involved stakeholders to promote regional competitiveness. This evaluation will determine whether they show sufficient promise to justify scaling them up and provide valuable information about adjustments that should be made before they are scaled up.
The next question – which is beyond the scope of this report – will be how to scale up. Parties to the agreements have already expressed interest in scaling up the pilot regional councils. They have voiced interest in expanding geographically, with new regional councils to cover the other parts of Estonia. There has also been discussion of expanding regional governance by, for example, introducing regional development agencies to help with the implementation of agreements. The results of the evaluation will certainly help inform the decision of how to scale up the councils. At the same time, the Estonian Government can benefit from embracing an experimental approach again by rolling out new regional governance features in an iterative way.
References
[5] Allain-Dupré, D. (2018), “Assigning responsibilities across levels of government: Trends, challenges and guidelines for policy-makers”, OECD Working Papers on Fiscal Federalism, No. 24, OECD Publishing, Paris, https://doi.org/10.1787/f0944eae-en.
[12] Arenguseire Keskus (2017), “Spatial and strategic planning of development, including regional level functions in directing spatial development (rengu ruumiline ja strateegiline kavandamine, sh regionaaltasandi funktsioonid ruumilise arengu suunamisel)”, https://www.etag.ee/wp-content/uploads/2018/04/Arengu-ruumiline-ja-strateegiline-kavandamine_ARUANNE_2018-02-25_RaM.pdf (accessed on 18 March 2025).
[25] Central and South Estonian Pilot Regional Councils (2024), “Draft regional development agreements”, https://www.agri.ee/regionaalareng-uhistransport/regionaalareng-ja-poliitika/regionaalsed-arengulepped.
[9] Central Estonian Pilot Regional Council (2024), “Draft regional development agreement”, Unpublished.
[28] Charbit, C. (2025), “Slides: State-Regions Contracts to support administrative reforms and regional development”, OECD Webinar on State-Regions Contracts - 11 April 2025.
[37] Charbit, C. (2024), “The use of contracts across levels of government”, Unpublished.
[27] Charbit, C. and O. Romano (2017), “Governing together: An international review of contracts across levels of government for regional development”, OECD Regional Development Working Papers, No. 2017/04, OECD Publishing, Paris, https://doi.org/10.1787/ff7c8ac4-en.
[38] Durand, A. and M. Vincent (forthcoming), “The ABCDs of policy experimentation”, OECD Working Paper, OECD Publishing, Paris.
[14] Estonian Government (2023), Pilot Project on Regional Development Plans and Councils.
[44] Estonian Ministry of Finance (2017), “The government sent the county government reform bill to the Riigikogu”, https://fin.ee/uudised/valitsus-saatis-maavalitsuste-reformi-seaduseelnou-riigikokku (accessed on 2 June 2025).
[18] Estonian Ministry of Public Administration (2023), Ministerial Decree on the Establishment of the Kesk-Eesti Regional Council (Ministri käskkiri Kesk-Eesti regionaalse arenguleppe nõukogu moodustamine).
[17] Estonian Ministry of Public Administration (2023), Ministerial Decree on the Establishment of the Lõuna-Eesti Regional Council (Ministri käskkiri Lõuna-Eesti regionaalse nõukogu moodustamine), https://www.agri.ee/sites/default/files/documents/2024-08/arengulepe-le-nk-kk-2023-45-moodustamine.pdf.
[1] Estonian Ministry of Regional Affairs and Agriculture (2025), Regional Development Agreements (Regionaalsed arengulepped), https://www.agri.ee/regionaalareng-planeeringud/kohalikud-omavalitsused/regionaalsed-arengulepped.
[24] Estonian Ministry of Regional Affairs and Agriculture (2024), “Meeting minutes of the South Estonia Development Council, 2 December 2024”, https://www.agri.ee/regionaalareng-planeeringud/kohalikud-omavalitsused/regionaalsed-arengulepped (accessed on 7 March 2025).
[26] Estonian Ministry of Regional Affairs and Agriculture (2024), “Meeting minutes of the South Estonian Regional Council, 20 September 2024”, https://www.agri.ee/regionaalareng-planeeringud/kohalikud-omavalitsused/regionaalsed-arengulepped#louna-eesti-arengule--2 (accessed on 7 March 2025).
[22] Estonian Ministry of Regional Affairs and Agriculture (2023), “Meeting minutes of the South Estonia Regional Council, 22 February 2023”, https://www.agri.ee/regionaalareng-planeeringud/kohalikud-omavalitsused/regionaalsed-arengulepped (accessed on 7 March 2025).
[16] Etzkowitz, H. (2008), The Triple Helix: University-Industry-Government Innovation in Action, https://doi.org/10.4324/9780203929605.
[19] European Leader Association for Rural Development (2025), About LEADER and CLLD, https://elard.eu/leader-clld/.
[7] Eurostat (n.d.), Overview: Nomenclature of Territorial Units for Statistics, https://ec.europa.eu/eurostat/web/nuts.
[20] Iceland Ministry of Infrastructure (2023), “Interview with the OECD”, OECD, Paris.
[21] Icelandic Government Steering Group on Regional Affairs (2020), Sóknaráætlanir landshluta, https://www.stjornarradid.is/verkefni/fleiri-rit/rit/2020/11/23/Greinargerd-um-soknaraaetlanir-landshluta-2015-2019/.
[3] Kattai, K. et al. (2020), Analysis of Estonian Regional Development: Proposals for Shaping Regional Governance Models – Final Report (Eesti regionaaltasandi arengu analüüs: Regionaalse valitsemise mudelite kujundamise ettepanekud – Lõpparuanne), https://skytte.ut.ee/sites/default/files/2022-05/1.%20Regionaalne%20valitsemine%20lopparuanne_final.pdf.
[41] McCallion, M. and P. Tallberg (2008), “Regionalisation in Sweden”, STALS Research Paper, No. 6/2008, https://www.stals.santannapisa.it/sites/default/files/stals_Tallberg.pdf.
[35] Michels, J. and Y. Wu (forthcoming), “Strengthening the impact of regional development strategies in the face of global challenges: A focus on Implementation”.
[36] OECD (2024), “Information gathering interviews”, OECD, Paris.
[11] OECD (2024), “Stakeholder interviews”, OECD, Paris.
[34] OECD (2023), “Foresight and vision-setting workshops”, OECD, Paris.
[30] OECD (2023), Recommendation of the Council on Regional Development Policy, OECD, Paris, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0492.
[33] OECD (2022), Regional Governance in OECD Countries: Trends, Typology and Tools, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://doi.org/10.1787/4d7c6483-en.
[4] OECD (2022), Shrinking Smartly in Estonia: Preparing Regions for Demographic Change, OECD Rural Studies, OECD Publishing, Paris, https://doi.org/10.1787/77cfe25e-en.
[8] OECD (2022), SNGWOFI - Iceland profile, https://www.sng-wofi.org/country_profiles/iceland.html.
[40] OECD (2020), Regulatory Impact Assessment, OECD Best Practice Principles for Regulatory Policy, OECD Publishing, Paris, https://doi.org/10.1787/7a9638cb-en.
[6] OECD (2020), The Future of Regional Development and Public Investment in Wales, United Kingdom, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://doi.org/10.1787/e6f5201d-en.
[15] OECD (2019), Effective Public Investment Across Levels of Government: Implementing the OECD Principles, OECD, Paris, https://www.oecd.org/en/about/projects/effective-public-investment-toolkit.html.
[32] OECD (2018), Rethinking Regional Development Policy-making, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264293014-en.
[42] OECD (2017), Multi-level Governance Reforms: Overview of OECD Country Experiences, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264272866-en.
[13] OECD (2011), Estonia: Towards a Single Government Approach, OECD Public Governance Reviews, OECD Publishing, Paris, https://doi.org/10.1787/9789264104860-en.
[39] OECD (2008), Introductory Handbook for Undertaking Regulatory Impact Analysis (RIA), OECD, Paris, https://www.oecd.org/gov/regulatory-policy/44789472.pdf.
[23] OECD (2007), Linking Regions and Central Governments: Contracts for Regional Development, OECD Multi-level Governance Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264008755-en.
[31] OECD (forthcoming), Place-Based Policies for the Future, OECD Publishing, Paris.
[2] Riigi Teataja (2002), Local Government Associations Act, https://www.riigiteataja.ee/en/eli/506112013002/consolide.
[10] South Estonian Pilot Regional Council (2024), “Draft regional development agreement”, Unpublished.
[29] South Estonian Pilot Regional Council (2024), Draft regional development agreement, https://www.agri.ee/regionaalareng-planeeringud/kohalikud-omavalitsused/regionaalsed-arengulepped#dokumendid-ja-viited (accessed on 7 March 2025).
[43] Statistics Estonia (2021), “Levels of Administrative Units and Spatial Data”, https://rahvaloendus.ee/en/census-2021/methodology-and-quality/levels-of-administrative-units-and-spatial-data.
← 1. Estonia’s two pilot regions differ slightly from its statistical regions. In Estonia, there are five small (NUTS 3) regions under the Nomenclature of Territorial Units for Statistics system: Central Estonia, North Estonia, Northeast Estonia, West Estonia and South Estonia (Statistics Estonia, 2021[43]). For the pilot regions, one county switched regions: the county of Viljandi requested to be included in Central Estonia instead of South Estonia.
← 2. It is important to note that regional governance is not the only way to achieve economies of scale (others include inter-municipal co-operation, outsourcing or centralisation). Indeed, in some cases, it may be more effective to let municipalities collaborate flexibly on a case-by-case basis, because a fixed regional body risks imposing a one-size-fits-all approach, which may not lead to efficient service provision.
← 3. A decision justified as a measure to make more efficient and rational the division of tasks among government bodies, in light of the dwindling responsibilities of county governments (Estonian Ministry of Finance, 2017[44]).
← 4. The representatives of which were both transferred to a new Ministry of Regional Affairs and Agriculture with a reorganisation in the government.
← 5. One way to reduce the risk of “shopping list” agreements is to anchor regional development agreements in a clearly defined national strategy for regional development (Charbit and Romano, 2017[27]). In countries where such strategies are in place (for example, Poland), contracts are typically limited to activities that contribute to both national and regional priorities.
← 6. The relevant ministry (first the Ministry of Finance, then the Ministry of Regional Affairs and Agriculture) appointed a project manager from the Local Government Policy Department for each pilot regional council and an assistant project manager from the Regional Development Department. The project managers provide technical and logistical support for procedures like consultations, publicity and communications with ministries and other partners. EUR 30 000 was earmarked for 2023 for each pilot project, funding the drafting of a development agreement to external experts, the provision of expert advice and operational support for meetings (such as room hire, catering, technical services, interpretation services).