This chapter develops and applies an analytical framework to assess the impact of decentralised development co-operation (DDC) on local governance and United Nations Sustainable Development Goal (SDG) outcomes. Building on evidence from OECD DDC impact surveys, case studies and the literature, the analytical framework sets out the expected linkages between financial and non-financial DDC inputs to territorial outcomes, while accounting for the role of local conditions and institutional capacities. It then operationalises this framework by combining surveys and case-study evidence on perceived impacts with quantitative analysis based on localised Creditor Reporting System data and subnational indicators, to examine how DDC contributes to improving local governance and the SDGs and whether DDC official development assistance is statistically associated with measurable outcomes across provider and recipient regions and cities.
The Impact of Decentralised Development Co‑operation
Abstract
While decentralised development co-operation (DDC) is increasingly recognised as a mechanism through which local and regional governments (LRGs) contribute to sustainable development, the evidence base on its impacts remains limited, particularly at the subnational level and for non-financial forms of co‑operation. This gap concerns both sides of DDC partnerships. For partner regions and cities in official development assistance (ODA)-eligible countries, more evidence is needed on whether DDC contributes to improvements in development outcomes at the territorial level and through which governance and institutional channels these effects materialise. For regions and cities in OECD Development Assistance Committee (DAC) countries, the benefits of DDC are also insufficiently measured, particularly those related to non-financial DDC such as institutional learning, governance capacity, policy innovation and multi‑stakeholder engagement. Combining qualitative and quantitative evidence is therefore necessary to capture both measurable changes at the local and regional levels alongside benefits that are more institutional, relational or long-term.
The analysis presented in this chapter draws on complementary sources of evidence to assess DDC impacts from different angles, including self-reported evidence from the 2 OECD surveys on the impact of DDC (hereafter “DDC impact surveys”) and 16 case studies, and DDC ODA data from the OECD Creditor Reporting System (CRS) database (localised at the regional and city levels) combined with local governance and United Nations (UN) Sustainable Development Goal (SDG) outcome indicators. While the surveys and case studies provide insights into the perceived impacts of DDC and the mechanisms and conditions under which these impacts are generated, the analytical framework structures these elements into expected relationships between DDC inputs, enabling conditions and outcomes. The quantitative analysis then examines whether some of these expected relationships can be observed across a wider set of regions and cities by linking localised DDC ODA flows to subnational indicators of governance and SDG outcomes in provider and recipient territories. The quantitative findings, which are exploratory and non-causal, are then considered alongside the perceived benefits reported in the DDC impact surveys and case studies to provide a more complete picture of DDC impacts. The overall analysis provides an initial empirical basis to inform the assessment of DDC impacts and identify priorities for future work as more granular data become available.
Building on the evidence gathered through the DDC impact surveys, the case studies and literature review, this section develops an analytical framework to assess the impact of DDC. The framework examines how DDC inputs, both financial and non-financial, relate to improvements in local governance and SDG outcomes, and the ways in which specific conditions and territorial characteristics, such as institutional capacities and local assets, interact with DDC to shape these outcomes. The proposed analytical framework provides a conceptual basis for assessing the impact of DDC across cities and regions in DAC and partner countries. It also helps identify the main impact pathways, expected outcomes and enabling conditions that guide the analysis and structure the evidence presented in the chapter.
The analytical framework to assess the impact of DDC on local governance and SDG outcomes builds on and expands existing monitoring and evaluation (M&E) tools. It takes as a starting point the OECD-European Commission (EC) M&E framework for city-to-city partnerships to localise the SDGs (Box 4.1), and integrates elements from national and local frameworks, including information collected through the DDC impact surveys and interviews conducted as part of the case studies. These provide valuable conceptual inputs, particularly for identifying enabling conditions, partnership mechanisms and expected areas of impact. National and local M&E tools are often more granular and context-specific, but they typically lack the harmonisation needed for international comparison. Conversely, international frameworks enable comparability but must be adapted to reflect diverse territorial and institutional contexts. To support a robust assessment of DDC, these tools must be expanded and operationalised across both sides of the impact model: inputs (i.e. financial and non‑financial DDC) and outcomes (i.e. changes in local governance capacity and SDG performance). This requires careful attention to indicator selection, territorial alignment and methodological consistency to enable meaningful cross-country analysis at the subnational level.
The OECD and the European Commission have developed an M&E framework to assess how city-to-city partnerships contribute to localising the SDGs. The framework builds on the ten Group of Twenty (G20) High-Level Principles on city-to-city partnerships for localising the SDGs and the four objectives of the EC Partnerships for Sustainable Cities programme: i) strengthen urban governance; ii) ensure social inclusiveness of cities; iii) improve resilience and greening of cities; and iv) promote prosperity and innovation in cities (Figure 4.1).
Source: OECD (2023[1]), City-to-City Partnerships to Localise the Sustainable Development Goals, https://doi.org/10.1787/d2fe7530-en.
The framework combines two complementary components:
A self-assessment framework: A checklist for local governments and territorial stakeholders to assess their alignment with the ten G20 principles. Inspired by other OECD governance checklists (e.g. on water governance, circular economy and policy coherence), it aims to identify framework conditions for effective partnerships and stimulate dialogue on what works, what should be improved and who is responsible. The tool includes three stages (preparation, diagnosis and action) and uses a five-scale evaluation (plus “not applicable”) to capture the level of implementation.
An indicator framework: A set of 59 indicators to measure progress of cities engaged in partnerships towards achieving the SDGs and the EC objectives. Indicators are structured around the 4 objectives of the EC programme and cover all 17 SDGs. The framework draws on existing tools such as the OECD localised SDG indicator framework, the EC Joint Research Centre (JRC) European Handbook for Voluntary Local Reviews, the United Nations Human Settlements Programme (UN-Habitat) Global Urban Monitoring Framework and the European Environment Agency indicators for European cities. Indicators were mapped against the EC Global Europe Results Framework to ensure consistency.
The analytical framework is also informed by primary evidence from the DDC impact surveys and case-study interviews. These sources provide insights into the mechanisms through which DDC generates outcomes and shape the framework in two key ways. First, as highlighted in Chapter 3, responses from cities and regions in DAC countries and in partner countries highlight the central role of non-financial components, including technical assistance, knowledge exchange and peer learning, alongside financial flows, underscoring the need to account for both types of input. Second, the evidence points to positive impacts across partners, with LRGs in DAC countries reporting improvements in local governance, international engagement and policy innovation linked to their participation in DDC. These insights inform the bilateral structure of the framework, which considers outcomes on both sides of the partnership and distinguishes between direct and indirect pathways linking DDC inputs to changes in local governance and SDG performance.
Understanding the direct and indirect pathways through which DDC influences local governance and SDG outcomes is essential for designing a sound empirical strategy to assess impact. The analytical framework should identify both direct and indirect linkages connecting DDC inputs, including financial and non-financial resources, to measurable changes in local governance performance and SDG progress at the subnational level in both DAC and partner regions and cities. These pathways may operate through various mechanisms, such as conducive institutional frameworks, improved administrative capacity, enhanced local leadership and stakeholder participation, or better access to finance, among others. In turn, improvements in local governance or development outcomes may also reinforce the effectiveness of future DDC efforts. Mapping these relationships provides a more nuanced understanding of how DDC functions as a driver of institutional change, and helps define the assumptions and variables needed for to assess the impact of DDC. A structured framework of these linkages forms the basis for the assessment presented in this chapter, guiding the formulation of the quantitative analysis and the interpretation of quantitative and qualitative evidence.
The framework should also recognise that impact pathways are expected to differ across sides of the partnership. Recipient territories are more likely to experience direct effects on local governance and SDG outcomes through financial and technical support. For provider territories, the expected benefits of DDC engagement relate primarily to governance performance, through peer learning, knowledge exchange and multi-stakeholder engagement, with any broader SDG effects likely to materialise indirectly and subsequently as a result of these governance improvements.
Beyond direct pathways between DDC inputs and outcomes, the framework should also consider how improvements in one outcome can enable progress in another, and how broader framework conditions shape these interactions. While some DDC projects are designed to produce direct outcomes, such as improving access to services and opportunities (SDG 11 on sustainable cities and SDG 10 on reduced inequalities), for instance through capacity building and investment in public transport, other indirect benefits, such as reduced emissions (SDG 13 on climate action), may emerge through interlinkages between governance improvements and SDG outcomes, as well as across the SDGs themselves. Figure 4.2 illustrates these dynamics, showing how financial and non-financial DDC inputs can generate direct benefits that can also evolve into indirect, self-reinforcing improvements between governance and SDG outcomes. The analytical framework also underscores the influence of national, regional and local conditions, which mediate how DDC is practised and determine both the extent of its impact and the degree to which synergies between governance and SDG outcomes are achieved. It also recognises that the principles and approaches embedded in DDC partnerships (e.g. feminist development policy, human‑rights-based frameworks, local democracy or peacebuilding) constitute a qualitative dimension of DDC mechanisms that might explain differentiated impacts.
Operationalising the analytical framework: methodological considerations
Operationalising the analytical framework requires combining different sources of evidence to capture perceived impacts in DAC-and-partner-country LRGs, underlying mechanisms and measurable territorial outcomes. While the framework is built on the DDC impact surveys, case studies and the literature, its application in this chapter draws on a broader evidence base. Surveys and case‑study evidence is used to document perceived benefits and the mechanisms through which DDC generates positive benefits, while localised CRS data and subnational outcome indicators are used to examine whether DDC ODA is associated with measurable outcomes across provider and partner territories. This approach makes it possible to connect the perceived impact reported through the surveys, observed territorial patterns and statistical associations, while recognising that not all DDC impacts can be captured through comparable indicators.
Indicators and data sources
Assessing the measurable territorial impacts of DDC needs robust and comparable territorial indicators. These include measures of inputs, i.e. the DDC activities themselves; outcomes, including on local governance and SDG performance; and enabling conditions such as institutional frameworks and local characteristics. While several OECD databases, such as the OECD CRS and the OECD Measuring Distance to the SDGs in Regions and Cities, provided significant support to initiate this work, additional sources will need to be leveraged to fill existing data gaps.
The OECD CRS provides a solid basis to measure the financial component of DDC, but greater efforts are needed to capture its non-financial dimensions. Capturing DDC is particularly complex, as it comprises both financial and non-financial components. The financial dimension can be analysed using the OECD CRS, which offers structured project-level data for DDC ODA flows (OECD, 2025[2]). However, substantial data gaps remain, especially for the non-financial elements of DDC, such as technical assistance, knowledge exchange and peer learning, which are often under-reported and/or poorly documented (OECD, 2023[3]). Targeted self-reported information, including through dedicated OECD surveys such as the DDC impact surveys (see Box 1.3), can help close some of these gaps and generate a more complete picture of DDC activities. As total official support for sustainable development (TOSSD) reporting continues to develop, it could further expand the evidence base on DDC through its broader coverage of providers, non-concessional instruments and support addressing macro regional and global challenges with substantial benefits for developing countries (International Forum on TOSSD, 2026[4]).
Measuring SDG outcomes in regions and cities in ODA-eligible countries remains challenging, although existing international frameworks and databases provide useful initial proxies. The OECD has developed a comprehensive indicator framework for measuring the distance to achieving SDGs in regions and cities, which includes some governance and civic engagement indicators (OECD, 2025[5]). Nonetheless, this framework must be extended to include non-OECD countries, as the majority of partner cities and regions lie outside the OECD area. Complementary sources with global coverage can support this extension, such as the EC-JRC Stats in the City database (Mari Rivero et al., 2025[6]), the Global Data Lab SDG Dashboard (2025[7]) and the ongoing OECD-UN-Habitat Global Stocktake on SDG Localisation, whose first objective is to measure SDG progress in a harmonised manner across cities and regions worldwide (OECD, 2025[8]).
For DAC regions and cities, assessing the domestic benefits of DDC engagement requires complementary sources that capture institutional quality, administrative practices and public trust. These dimensions are only partially covered by the Measuring Distance to SDGs in Regions and Cities and other OECD databases. The European Quality of Government Index, produced by the Quality of Government Institute at the University of Gothenburg, Sweden (Charron, Lapuente and Bauhr, 2024[9]), provides a complementary source by drawing on large-scale citizen surveys to assess perceptions of public sector performance across three pillars: i) quality of public services; ii) impartiality; and iii) corruption. These dimensions align closely with the governance improvements identified in the analytical framework and reported in the OECD DDC impact surveys as key domestic benefits of DDC engagement.
Existing OECD and global databases provide baseline information on territorial characteristics, but dedicated data collection is needed to capture the institutional conditions that shape DDC impact. The impact of DDC efforts depends on a range of national, regional and local factors that determine how inputs translate into governance improvements and SDG outcomes. Standard variables such as population size, urban-rural classification and economic composition can be integrated into quantitative analysis using existing OECD and global datasets, such as the OECD database on regions, cities and local areas (2025[10]) and JRC Global Human Settlement Layer data (Pesaresi et al., 2024[11]). More specific factors, however, require tailored approaches. These factors include national legal frameworks regulating DDC, the degree of political leadership and institutional support for international engagement, the availability of human and technical resources at the regional and local levels, and the quality of multi-level governance frameworks. No harmonised international source currently captures these conditions in a consistent and comparable manner. To remediate this gap, the OECD is using the aforementioned surveys to map these enabling conditions and collect targeted data from a sample of cities and regions engaged in DDC (see Box 1.3).
Selecting the right indicators is a crucial step in assessing the impact of DDC on local governance and SDG outcomes. An overly broad selection of indicators risks overlap and reduces the clarity of the analysis. For DDC, it is essential that outcome indicators are carefully linked to the thematic focus of DDC projects. For example, poverty reduction-related partnerships should be tracked against relevant SDG 1 “No poverty” indicators, while climate-focused initiatives should be connected to indicators for SDG 13 “Climate action”. Streamlining indicators in this way ensures that the framework captures broad progress on governance and sustainable development and also reflects the specific contribution of DDC activities.
Geographical definitions
Measuring the impact of DDC requires identifying the appropriate geographical scale. This is particularly important given the multiplicity of subnational governments involved, each with different responsibilities, capacities and characteristics that shape both how DDC is provided and how it generates outcomes. For cities and regions in DAC countries, the starting point is the administrative boundaries of the subnational government engaged in DDC, typically at the regional or municipal level. For partner cities and regions, the situation can be more complex. Support may target specific neighbourhoods within a city or, conversely, generate spillovers that affect surrounding non-targeted municipalities, which points to the need for a functional approach. Ideally, impact should be measured at the geographic scale corresponding to the DDC commitment itself. That is, the territory where activities are implemented. However, the assessment should also account for potential spillovers, whether positive or negative, in surrounding areas. In practice, however, the absence of systematic data on the precise geographic scope of DDC activities means that analysis must rely on administrative boundaries as a first pragmatic approximation (see Box 4.2 for more details on OECD geographical definitions). Even this represents a significant improvement compared to country-level information, which is too aggregated to yield meaningful insights into the local impacts of DDC.
International comparisons of subnational indicators require consistent definitions. The OECD has developed several territorial classifications and typologies for international comparisons.
Regions and territorial levels
Regions within the 38 OECD countries are classified on 2 territorial levels reflecting the administrative organisation of countries. The 433 OECD large (TL2) regions represent the first administrative tier of subnational government, for example, the province of Ontario in Canada. The 2 414 OECD small (TL3) regions correspond to administrative regions, except for four countries. All of the regions are defined within national borders. This classification – which, for European countries, is largely consistent with the Eurostat NUTS 2021 hierarchical system – facilitates greater comparability of geographic units at the same territorial level.
Functional urban areas and city definitions
The OECD, in co-operation with the European Union, has developed a harmonised definition of functional urban areas (FUAs). Being composed of a city and its commuting zone, FUAs encompass the economic and functional extent of cities based on people’s daily movements. Using this methodology, the OECD has delineated around 1 300 FUAs in 37 OECD countries.
Municipalities and Local areas
The OECD is also maintaining a database and territorial grid of municipalities and local areas for all OECD countries. These geographical levels are chosen based on the following principles:
administrative units corresponding to local governments (lowest administrative level)
units for which data are made available by national statistical offices
units used as building blocks for FUAs and cities
units appropriate for the implementation of the degree of urbanisation classification
subdivisions of other OECD territorial levels (TL2 and TL3 regions).
Sources: OECD (2025[10]), OECD Database on regions, cities and local areas, http://oe.cd/geostats. OECD (2025[12]), OECD Geographical Definitions, https://www.oecd.org/en/data/datasets/oecd-geographical-definitions.html.
Developing an empirical model provides a first step towards quantitatively assessing the relationship between DDC and both local governance and SDG outcomes. In this framework, territorial indicators on local governance (e.g. civic engagement or trust in government) and on the SDGs (e.g. air quality, educational attainment or gender equality) serve as dependent variables. The main explanatory variable is the intensity of DDC ODA (e.g. DDC ODA per capita) flows received or provided by subnational units, which can be measured in total, per capita or by sector. Control variables capture structural and contextual characteristics, including population size, socio-economic composition, institutional frameworks and territorial typologies, to account for differences across places. Such a model should allow for the measurement of the magnitude and direction of associations between DDC inputs and local outcomes as well as for the assessment of mediating factors (interaction effects) that shape how DDC translates into benefits. Where feasible, the framework should move beyond identifying correlations to apply strategies that strengthen causal inference.
The quantitative component provides an exploratory assessment of whether DDC ODA is associated with measurable territorial outcomes. In practice, the analysis links localised DDC ODA flows to subnational indicators of local governance and SDG outcomes, while accounting for structural and contextual characteristics such as population size, socio-economic conditions, institutional frameworks and territorial typologies. The objective is not to establish causal effects, but to assess whether the relationships suggested by the analytical framework can be observed across a wider set of territories on both sides of the partnerships. The detailed empirical specifications, variable definitions and regression outputs are presented in Annex D.
Guided by the analytical framework developed in the previous section, this section presents a quantitative approach to assess the associations between DDC ODA and both local governance and SDG outcomes. It begins by describing the data underpinning the analysis, including the localisation of the CRS to identify subnational providers and recipients and the compilation of indicators on local governance and SDG outcomes for cities and regions in both DAC and partner countries. The chapter then outlines the empirical specifications used to estimate the relationship between DDC ODA and development outcomes.
Localising CRS data: identifying subnational providers and recipients
A robust empirical strategy for analysing DDC impacts requires going beyond country-level averages. National aggregates are too coarse to capture where aid is delivered and the local outcomes it generates. This calls for the localisation of both DDC variables and outcome indicators related to SDGs and local governance. This section outlines the ongoing effort to localise the CRS, identifying OECD and partner cities and regions in the 15 DAC countries for which DDC ODA data are available.
For most countries, the OECD CRS database does not provide structured data on the LRGs engaged in DDC. Provider subnational-level information is available in a standardised format for only three countries: Germany (16 TL2 regions), Spain (17) and Belgium (3). For the United Kingdom, disaggregated data are available only for 2 out of the 12 TL2 regions. While disaggregated data for Belgium and the United Kingdom are available from the beginning of the study period in 2013, the series start in 2014 for Germany and in 2017 for Spain. For other providers, DDC data are reported only at the national level, without identifying the specific subnational entities involved. On the ODA-eligible side, no DAC country systematically reports the cities or regions that benefit from DDC. This limits the ability to assess the contribution of individual actors and the geographic distribution of support.
Recent efforts have begun to improve the localisation of DDC ODA providers in the CRS, though important gaps remain. Initiatives led by the Basque Agency for Development Cooperation and consulting team ECOPER (2024[13]), through a network of offices responsible for CRS reporting, have gathered information on the specific regions and cities providing DDC ODA flows for 2022 and 2023. This represents a significant step towards improving the visibility of subnational providers and strengthening the territorial granularity of DDC data. However, these efforts remain limited in temporal coverage and do not yet extend to longer time series, constraining the ability to conduct longitudinal analysis. In addition, further work is needed to systematically localise cities and regions in partner countries, which continues to represent a key limitation for fully assessing the spatial distribution and benefits of DDC.
To address remaining gaps in the localisation of recipient regions and cities, the OECD has developed a methodology to identify the localisation of CRS project-level data. While recent initiatives, such as those mentioned above, have helped improve explicit references to subnational providers in CRS reporting, systematic information on recipient territories remains largely unavailable. In order to provide these data, the OECD has developed an approach to identify recipient regions and cities using text analysis and artificial intelligence (AI) techniques applied to CRS project descriptions; referred to as “localisation” in the remainder of this report. By identifying the specific LRGs involved, the methodology strengthens the analytical basis for assessing associations between DDC and outcomes at the subnational level, both in DAC and partner countries. The methodology can be described in five main steps (for more details on the methodology, see Annex B):
1. Build a global dictionary: Compile a reference list of city and region names from global geographic sources (e.g. Natural Earth, GeoNames and Database of Global Administrative Areas, see Box A B.1), including variants and alternate spellings. All three sources use Latin-script place names and cover a range of spellings found in CRS project descriptions, which are primarily reported in English, French, German and Spanish. The use of three complementary sources reduces the risk of missing names due to linguistic variation. For instance, “Bombay” is captured by GeoNames, whereas Natural Earth includes only “Mumbai”, helping to identify alternative spellings used across provider languages. However, coverage of minority- and regional-language names remains a limitation.
2. Preprocess CRS descriptions: Clean and standardise project descriptions to remove formatting inconsistencies and harmonise linguistic inputs. This step addresses common issues in the raw CRS data, including inconsistent use of accented characters, mixed-language descriptions, extraneous punctuation and encoding artefacts, which could otherwise hinder accurate string matching against the geographic dictionary.
3. Match names using country filters: Perform string matching between project descriptions and the global dictionary using a combination of exact and fuzzy matching to capture minor spelling variations. Matching is constrained by a country filter, ensuring that place names are matched only against territories plausibly associated with each project. This is particularly important for common place names that occur in multiple countries, as an unconstrained search would generate a high number of false positives.
4. Validate the results using AI: Use AI-assisted review (OpenAI API, GPT 5.2) to confirm identified matches and minimise false positives. For each candidate match, the model is provided with the original project description and the proposed location and asked to assess whether the place name genuinely refers to that territory in context, distinguishing, for instance, between a geographic reference and an organisation or personal name sharing the same string. Where a candidate is confirmed as a subnational place, the model is also asked to infer whether it refers to a city, a region or an unknown territorial type, based on the information available in the project description and independently of the territorial type inherited from the underlying geographic dictionary.
5. Assess the performance of the localisation approach. Conduct independent human verification of both the dictionary-based matching and the AI-assisted validation. For the dictionary step, manually review a random sample of projects for which no candidate location was identified to assess whether recipient cities or regions were nevertheless mentioned in the project description. For the AI step, manually review a random sample of projects with dictionary-generated candidates and compare the human and AI classifications to assess the accuracy with which the model identifies genuine recipient locations and their territorial type. The review also assesses localisation at project level, accounting for cases in which several candidate names or territorial references may correspond to the same recipient location. Together, these verification exercises provide an assessment of the reliability of the overall localisation process and help identify remaining sources of error and areas for further improvement.
Overview of the localised provider regions and cities
The localisation of DDC ODA providers builds on the CRS database and complementary information from DAC countries. To complement the CRS database, and through their Decentralised Co-operation Reports, the Basque Agency for Development Co-operation and ECOPER have collected disaggregated data on provider regions and cities in DAC countries (eLankidetza, 2026[14]; ECOPER, 2025[15]; ECOPER, 2024[13]). Combining these sources allows the identification of LRGs providing DDC ODA in 13 DAC countries. Among them, 11 countries report contributions from regions (TL2 or TL3) and/or cities, while in 2 countries reporting is limited to cities only. Coverage varies significantly across countries and levels of government. Austria, Germany and Spain show the highest reporting rates at the TL2 level, with more than 80% of regions reporting DDC ODA. Switzerland stands out for its strong coverage at the TL3 level, with around 94% of TL3 regions reporting. France presents a more balanced pattern, with 74% of TL2 regions and 31% of TL3 entities included. By contrast, Latvia and Portugal report only at the city level, with relatively low coverage across municipalities. Belgium records the highest share of municipalities reporting DDC ODA, with slightly more than one-quarter of local governments included (Table 4.1). Differences in reporting coverage across levels of government and provider countries should be taken into account when making cross-country comparisons of DDC ODA contributions.
Sources: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds]; ECOPER (2024[13]), Decentralised Cooperation Report: Mapping Donor Cities, https://www.elankidetza.euskadi.eus/contenidos/documentacion/publicaciones_descentralizada/es_def/INFORME-2024-ENG-.pdf.
Cross-border DDC ODA contributions vary widely across regions in DAC countries even within the same country. Across DAC countries, regional contributions range from less than USD 40 000 to more than USD 9 million in average annual disbursements. The largest contributions come from Comunidad Valenciana (USD 66 million), Catalonia (USD 53 million) and the Basque Country (USD 49 million) in Spain, Flanders in Belgium (USD 54 million), Geneva in Switzerland (USD 44 million) and Scotland in the United Kingdom (USD 21 million). Across regions in DAC countries, cross-border DDC ODA is highly associated with population size. However, within the same country, the scale of DDC engagement at the regional level is not only driven by population size, as the largest region or the capital region of the country is often not the main donor. Similarly, variation is not only explained by country differences, as several countries display substantial disparities across their regions. This is notably the case in Canada, Germany and Switzerland (Figure 4.3, Figure 4.4, Figure 4.5).
Note: Based on 11 DAC countries with available data at the regional level.
Sources: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds]; ECOPER (2024[13]), Decentralised Cooperation Report: Mapping Donor Cities, https://www.elankidetza.euskadi.eus/contenidos/documentacion/publicaciones_descentralizada/es_def/INFORME-2024-ENG-.pdf.
Note: The capital region refers to the region containing the country’s capital city.
Sources: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds]; ECOPER (2024[13]), Decentralised Cooperation Report: Mapping Donor Cities, https://www.elankidetza.euskadi.eus/contenidos/documentacion/publicaciones_descentralizada/es_def/INFORME-2024-ENG-.pdf.
Note: Darker blue points indicate higher observations density.
Sources: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds]; ECOPER (2024[13]), Decentralised Cooperation Report: Mapping Donor Cities, https://www.elankidetza.euskadi.eus/contenidos/documentacion/publicaciones_descentralizada/es_def/INFORME-2024-ENG-.pdf; OECD (2025[10]), OECD Database on regions, cities and local areas, http://oe.cd/geostats.
The evolution of cross-border DDC ODA over time also varies considerably across regions within the same country. In countries with consistent regional data availability over time, such as Germany and Spain, contributions have not increased systematically. In Germany, some regions have increased their engagement. This is notably the case for Bavaria and North Rhine-Westphalia, while others, such as Hamburg and Hesse, show declining contributions over the same period. Spain shows a similar pattern. On the one hand, Catalonia and Comunidad Valenciana (Valencia) have significantly increased their disbursements despite already high levels in 2017. On the other, Andalusia and the Basque Country have reduced their contributions since 2017, yet they remain among the top contributors in Spain (Figure 4.6).
Among the sample of reporting cities, cross-border DDC ODA contributions are highly concentrated in a small number of leading cities, with variation both across and within countries. Among cities with available provider-level data for 2022-2023, the largest contributions come from and Spain (Bilbao with USD 11.3 million, Barcelona with USD 8.4 million and Madrid with USD 8.4 million) and Switzerland (Geneva with USD 9.7 million and Zurich with USD 7.1 million), as well as Paris in France (USD 5.9 million). This distribution reflects patterns observed at the regional level, while also highlighting the prominence of specific cities as key actors in DDC. Around one-third of cities (220 out of 640) report relatively small ODA volumes, often below USD 10 000 annually (Figure 4.7, Figure 4.8).
Sources: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds]; ECOPER (2024[13]), Decentralised Cooperation Report: Mapping Donor Cities, https://www.elankidetza.euskadi.eus/contenidos/documentacion/publicaciones_descentralizada/es_def/INFORME-2024-ENG-.pdf.
Note: Based on nine DAC countries with available data at the city level.
Sources: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds]; ECOPER (2024[13]), Decentralised Cooperation Report: Mapping Donor Cities, https://www.elankidetza.euskadi.eus/contenidos/documentacion/publicaciones_descentralizada/es_def/INFORME-2024-ENG-.pdf.
Sources: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds]; ECOPER (2024[13]), Decentralised Cooperation Report: Mapping Donor Cities, https://www.elankidetza.euskadi.eus/contenidos/documentacion/publicaciones_descentralizada/es_def/INFORME-2024-ENG-.pdf.
While larger cities tend to provide more cross-border DDC ODA, population size alone does not explain the level of engagement. Across 793 provider city observations, there is a clear positive relationship between population and cross-border DDC ODA disbursements, indicating that more populated municipalities generally contribute higher volumes of cross-border DDC ODA. However, there is considerable variation. Some cities, such as Bilbao and Donostia/San Sebastián in Spain and Geneva in Switzerland, contribute substantially more than would be expected based on their population size, pointing to factors such as institutional capacity, political commitment and established co-operation practices. In contrast, several large cities, including Madrid in Spain and Paris in France, contribute amounts that are more closely aligned with what their population size would suggest (Figure 4.9).
Note: Darker blue points indicate higher observations density.
Sources: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds]; ECOPER (2024[13]), Decentralised Cooperation Report: Mapping Donor Cities, https://www.elankidetza.euskadi.eus/contenidos/documentacion/publicaciones_descentralizada/es_def/INFORME-2024-ENG-.pdf; OECD (2025[10]), OECD Database on regions, cities and local areas, http://oe.cd/geostats.
Overview of the localised recipient regions and cities
A substantial share of DDC projects can be geographically localised at the level of recipient regions and cities, and this share has increased over time. Around 36 000 of the nearly 98 000 cross-border DDC ODA projects recorded in the OECD CRS from 2013 to 2023 contain sufficient geographic information to identify recipient regions or cities. The share of projects that can be localised has increased steadily over the past decade, rising from 34% in 2013 to 44% in 2023, possibly reflecting gradual improvements in reporting practices and the growing inclusion of geographic references in project descriptions. Localisation occurs both at the regional and city levels, expanding the analytical potential to examine how DDC flows relate to local governance and SDG outcomes (Figure 4.10).
Source: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds].
The share of DDC projects in the OECD CRS for which the localisation methodology identifies recipient regions and cities varies significantly across DAC countries. Among countries reporting a large number of projects to the CRS, Italy and Spain display relatively high localisation rates, with the localisation methodology identifying recipient regions or cities for more than two-thirds of projects. By contrast, France and Switzerland show much lower localisation rates, with fewer than 5% of projects containing geographically identifiable recipient territories. These differences reflect a considerable variation in how provider countries report geographic information in project descriptions submitted to the CRS. This variation affects the ability to systematically localise DDC flows across territories and the comparability of results across donor countries. However, they may also reflect particular challenges that the identification mechanism has for some languages, and further work will be needed to examine this (Figure 4.11).
Note: Number of total cross-border DDC ODA projects in parentheses.
Source: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds].
The localisation of recipient regions and cities provides partial but analytically useful coverage of DDC projects in the CRS. The localised dataset does not capture the full set of projects, primarily because many project descriptions do not include references to specific subnational locations. The results should therefore be interpreted with caution. This missing geographic information may introduce selection bias if projects that mention specific places differ systematically from those that do not, for example by project size, sector, SDG focus or type of recipient territory. As such, the localised dataset is not intended for reporting official statistics on localised DDC ODA flows. Nevertheless, it represents a significant step in improving the territorial granularity of DDC data and enables some exploratory analysis of how cross-border DDC ODA (for simplicity, cross-border DDC ODA flows localised in the territories of recipient regions and cities can also be referred to as DDC ODA) relates to development outcomes across the available sample.
The localisation process identified close to 6 000 recipient regions and cities across more than 100 countries worldwide. In total, the dataset covers 3 028 cities in 113 countries and 2 873 regions in 107 countries,1 based on 2 605 projects involving cities and 3 295 projects involving regions. More than one‑fifth of the projects involve multiple partner regions or cities. On average, regions received slightly higher amounts than cities, with mean DDC ODA disbursements of approximately USD 39 000 per region during the period, about 18% higher than the average amount received by cities. Disbursement amounts are highly skewed in both the regional and city samples, with the mean around four times higher than the median, indicating that a small number of territories receive large volumes of aid while most receive relatively modest amounts (Table 4.2).
Notes: Regions include the first and second administrative level of subnational governments. A project-location corresponds to one project matched to one recipient location; the same project may therefore appear multiple times when it covers several locations. When a project covers multiple places, the amount of aid is split equally across the identified locations.
Source: Based OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds].
DDC ODA disbursements across recipient regions and cities show a relatively strong geographic concentration within continents. DDC ODA reaches subnational governments across most continents, but localised flows are particularly concentrated in specific areas within Central and South America and the Caribbean, Sub-Saharan Africa, as well as the Middle East (Figure 4.12Figure 4.12-Figure 4.13). This pattern should be interpreted considering provider-country composition (Figure 4.11) and the historical, linguistic and institutional ties that shape many DDC projects. As shown in Chapter 2, Spain accounts for a large share of DDC activities conducted with partners in Latin America and the Caribbean, while Belgium have relatively strong links with Sub-Saharan Africa, contributing to the observed geography of localised flows.
Note: Regions correspond to the first or second administrative level of subnational governments.
Source: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds].
Source: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds].
DDC ODA disbursements are highly unequal across subnational territories within the same recipient country. The distribution of DDC ODA across partner regions and cities varies considerably. For example, among the 25 localised territories in Peru, Cusco, Huancavelica and Piura received over USD 11 million in the past decade, while one-third of the regions received less than USD 500 000. Similar differences are visible across cities. In El Salvador (188 cities) and Mozambique (79), San Salvador and Maputo received over USD 15 million, while over 80% of the remaining cities received less than USD 1 million in the same period, reflecting in part differences in local needs, population size (as many of the top recipients seem to be among the largest cities), economic context and the type and scope of projects implemented (Figure 4.14-Figure 4.15).
Note: Regions correspond to the first administrative level of subnational governments.
Source: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds].
Source: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds].
Measuring local governance and SDG outcomes in DAC and partner countries
The selection of outcome indicators for recipient and provider regions and cities is guided by analytical relevance and data availability. Evidence from the DDC impact surveys and case studies helps identify the outcome areas most closely linked to DDC activities on both sides of the partnerships. Recipient regions and cities are expected to experience more direct improvements in certain aspects of local governance and certain SDG outcomes. For provider regions and cities, benefits are more likely to materialise first through indirect channels, including peer learning, knowledge exchange, strengthened multi-stakeholder engagement and policy innovation. These channels may primarily affect governance performance in the short to medium term, while broader SDG outcome improvements may emerge over longer time horizons, as discussed in the analytical framework (Figure 4.2). At the same time, important data limitations remain for regions and cities in both DAC and ODA-eligible countries, where the availability of comparable subnational indicators is still limited across local governance and SDGs. The empirical analysis therefore focuses on indicators that are broadly aligned with the expected channels of DDC impact and sufficiently available across territories and over time. The available datasets provide a set of potential indicators, which are assessed according to their thematic relevance, territorial and temporal coverage, quality and suitability for the empirical specifications. In particular, the available governance indicators for DAC LRGs remain broad proxies and do not directly measure the specific administrative, institutional or policy-learning benefits that may result from DDC engagement. The empirical analysis for DAC LRGs should therefore be understood as a first step to examine whether DDC engagement is associated with wider governance outcomes, rather than as a direct or comprehensive measure of the impact generated by DDC for provider territories. Further work would be needed to develop more specific and comparable measures of administrative capacity, institutional learning, international engagement and policy innovation, which could better capture the full benefits of DDC for provider regions and cities.
Two complementary global datasets are used to measure SDG outcomes at the subnational level in ODA‑eligible countries.
At the regional level, the Subnational Human Development Index (SHDI) provides annual indicators of social development outcomes, including income, health and education, across more than 1 800 regions in over 160 countries. By providing harmonised indexes and indicators on regional income over time, for most countries in the world, SHDI data allow for a first exploration of the relationship of targeted cross-border DDC ODA with key SDGs such as poverty reduction (SDG 1) and decent work and economic growth (SDG 8) (see Annex C for more details on the SHDI).
At the city level, the Global Human Settlement Urban Centre Database (GHS-UCDB) offers harmonised data on environmental and climate-related indicators for over 10 000 urban centres worldwide. It provides harmonised indicators on green space and emissions, which allow to explore whether cities that receive targeted environmental DDC ODA experience measurable improvements in these areas. The data are derived from satellite observations and emissions modelling to ensure consistency and comparability across cities worldwide, including in contexts where ground-based monitoring systems are limited or unavailable. This makes them particularly valuable for the analysis environmental outcomes in partner cities and regions, where data gaps are often significant.
Together, these datasets are among the most comprehensive sources currently available for tracking SDG outcomes (or proxies of them) at the subnational scale in ODA-eligible countries. While they remain subject to limitations, including differences in spatial and temporal coverage and the use of modelled or proxy indicators for some dimensions, they provide a basis for capturing some aspects of both social and environmental dimensions of development and assessing how DDC is associated with outcomes across different territorial levels. However, not all candidate indicators are retained in the main regression analysis, as some outcomes are more constrained by data quality, temporal dynamics or interpretation. The final set of indicators used in the regressions is therefore selected according to both thematic relevance to DDC activities and the suitability of the available data for empirical analysis.
SDGs and local governance outcomes in DAC regions and cities can be measured using indicators from the OECD Measuring Distance to the SDGs in Regions and Cities and the European Quality of Government Index (EQI). The selection of candidate indicators is guided by the analytical framework (Figure 4.2) and evidence from the qualitative analysis (DDC impact surveys and case studies), which suggest that the benefits of DDC for provider regions are most likely to materialise first through governance improvements (before they may translate into broader SDG progress in a later stage), data availability and the extent to which each indicator can be linked to the expected channels through which DDC may generate for provider regions. Given the nature of the available data, these indicators should be interpreted as broad proxies of governance outcomes, rather than direct measures of the institutional benefits generated by DDC engagement. The OECD dataset underpinning the measurement of the distance to achieving the SDGs in regions and cities provides comparable indicators on the SDGs at the subnational level. The EQI dataset, produced by the Quality of Government Institute at the University of Gothenburg, Sweden, draws on large-scale citizen surveys to capture perceptions and experiences of public sector performance (Charron, Lapuente and Bauhr, 2024[9]). Together, these sources provide a broadly relevant but partial set of SDG and governance indicators, combining internationally standardised SDG measures with perception-based insights on the quality of public services, impartiality and corruption at the regional level. Future work would be needed to develop more specific and comparable measures of local governance benefits linked to DDC.
The indicators from the OECD Measuring Distance to the SDGs in Regions and Cities database cover key dimensions of SDG 16 “Peace, justice and strong institutions” related to governance, institutional quality and public trust. These include measures of corruption in government, trust in public institutions such as the police and government, civic participation through voter turnout, as well as safety-related outcomes such as perceptions of safety when walking at night.
The EQI enhances the analysis by providing perception-based measures of governance that complement OECD indicators. In addition to the overall EQI score, the dataset includes separate pillars on quality, impartiality and corruption, allowing for a more granular assessment of governance outcomes. These indicators are harmonised with CRS provider-region identifiers through region-name standardisation and manual matching, enabling their integration into the empirical analysis. Combined with OECD territorial SDG indicators, the EQI strengthens the measurement of local governance outcomes and allows for a more comprehensive assessment of how DDC engagement relates to institutional performance at the regional level.
Empirical specifications and regression results
The quantitative analysis assesses whether the relationships suggested by the analytical framework can be observed across a wider set of partner regions and cities. It uses localised CRS data to identify where cross-border DDC ODA is provided from and allocated to, and links these flows to comparable subnational indicators of local governance and SDG outcomes. The analysis focuses on two policy-relevant questions: whether cross-border DDC ODA is targeted towards territories with greater development needs, and whether partner regions and cities engaged in cross-border DDC ODA projects experience measurable changes in relevant and available local governance and SDG outcome indicators. In a final step, correlations between DDC engagement and local governance proxies in DAC LRGs are also explored and documented.
The approach is exploratory and should be interpreted as evidence of statistical associations rather than causality. Differences across territories are taken into account, including population size, socio-economic conditions, territorial characteristics and broader country-level factors, to better isolate the relationship between cross-border DDC ODA and local outcomes. However, the results remain constrained by the partial localisation of CRS projects, uneven territorial coverage across countries, small samples for some indicators and the fact that many DDC impacts – particularly those related to peer learning, institutional capacity and partnership quality – are not fully captured by available quantitative indicators. In particular, reverse causality remains a concern, as places with worse initial conditions, or those affected by crises, may receive more aid, making it difficult to isolate the causal effect of aid on subsequent outcomes. Although some steps have been taken to mitigate this concern – such as excluding projects labelled as emergency response, testing several specifications and applying placebo tests – the quantitative evidence remains non-causal and indicative of plausible associations.
The detailed variable definitions, empirical specifications and full regression outputs are presented in Annex D. The following section reports the findings in policy terms and interprets them alongside the survey and case-study evidence to provide a broader assessment of DDC impacts on local governance and SDG outcomes.
This section brings together the evidence on the impact of DDC on local governance and SDG outcomes. By combining evidence from the DDC impact surveys and case studies with quantitative analysis based on localised CRS data and subnational indicators, it assesses how the impact pathways identified in the analytical framework are reflected across different sources of evidence, including perceived impacts reported by LRGs, concrete examples from DDC projects and measurable territorial outcomes.
The section is structured around the two sides of DDC partnerships. It first examines impacts in cities and regions in partner countries, where the primary development objective of DDC is expected to materialise. It then discusses benefits in cities and regions in DAC countries, reflecting the role of DDC in supporting mutual learning, exchange of knowledge in strategic sectors for service delivery and institutional innovation. The final parts draw policy implications from the quantitative evidence and identify the main limitations and priorities for future research.
Impacts of DDC in partner regions and cities
Survey evidence indicates that partner LRGs perceive strongest DDC impacts in SDG 11 on sustainable cities and SDG 13 on climate action, while also reporting important effects in gender equality (SDG 5), good health (SDG 3) and clean water (SDG 6). In addition to SDG 17 on partnerships for the goals (which intrinsically includes DDC, Box 4.3), partner LRGs report the strongest SDG impacts in sustainable cities (4.1, on a scale of 1 to 5), climate action (4.0), gender equality (3.8), good health (3.8) and clean water (3.7) (Figure 4.16). Except for SDG 14 “Life below water” (2.8), partner LRGs rated the impact of DDC activities on SDG outcomes in their region or city with a score above 3 (the middle point in the rating scale). Overall, the highest scores among partner LRGs cluster mainly in the environmental dimension, followed closely by parts of the social dimension, while the economic dimension (SDG 8 “Decent work and economic growth” and SDG 9 “Industry, innovation and infrastructure”) tends to score slightly lower.
SDG 17 “Partnerships for the goals” calls for strengthening the means of implementation and revitalising the global partnership for sustainable development. Its 19 targets span finance, technology transfer, capacity building, trade and systemic issues such as policy coherence and multi-stakeholder partnerships. At its core, SDG 17 recognises that achieving the other 16 goals requires co-ordinated action across levels of government, sectors and borders and that no single local or national actor can deliver the UN 2030 Agenda alone.
LRGs have a distinct role to play in the achievement of SDG 17. As the level of government closest to citizens and communities, they are well placed to translate global commitments into locally relevant action, mobilise local resources and engage a wide range of territorial stakeholders including businesses, civil society and knowledge institutions. DDC contributes directly to SDG 17 by building structured partnerships between cities and regions across countries, fostering peer-to-peer knowledge exchange, strengthening the capacity of local governments in partner countries to design and implement sustainable development policies, and connecting local action to broader national and global frameworks. In doing so, DDC operationalises the core of SDG 17 at the subnational level by turning the global partnership for sustainable development into a network of concrete, place-based collaborations.
Partner LRGs report the strongest perceived local governance impacts in inclusiveness and participation, as well as accountability, transparency and public sector integrity. Overall, partner LRGs perceive a strong impact of DDC activities on local governance in their region or city (score of 4.4, on a scale of 1 to 5). The local governance dimensions with the strongest reported impacts are inclusiveness and participation (4.2), accountability, transparency and public sector integrity (4.2), strategic planning (4.1) and management and innovation (4.0). Administrative and technical capacity of public servants as well as multi-level and cross-sectoral co‑ordination (3.9) are also among those local governance dimensions that DDC supports the most in partner LRGs (Figure 4.16). The high ratings across inclusiveness, accountability, strategic planning and administrative capacity suggest that DDC partnerships contribute to strengthening core governance functions in partner LRGs. Beyond supporting specific sectoral interventions, DDC appears to enhance institutional performance by improving planning processes, management practices and accountability mechanisms.
Notes: Number of respondents in parentheses. Rating on a scale of 1 to 5 (where 5 = very high impact).
Source: OECD (2025[16]), “OECD Surveys on the Impact of Decentralised Development Co-operation on Local Governance and Sustainable Development Goal Outcomes”, Unpublished, OECD, Paris.
Case-study evidence shows how DDC impacts materialise in partner cities and regions through concrete projects and institutional mechanisms across several policy areas. These include waste and water management, for example through improved waste management infrastructure, such as the waste sorting facility established in El Guettar (Tunisia) in partnership with Böblingen (Germany). Other areas of impact include environmental sustainability and climate action, local economic development and job creation, education, gender equality and safety. In Gießen’s (Germany) partnership with Mubende (Uganda), for example, the installation of street lighting improved safety, enabled women to work at night and supported local businesses, while solar-powered lamps for schoolchildren helped improve literacy (for more examples, see Box 4.4).
City of Lublin, Poland – City of Chisinau (Moldova)
In Lublin’s (Poland) partnership with Chisinau (Moldova), which focuses on the redesign of the local public transport system, the establishment of a traffic monitor centre to provide data-driven analysis on road traffic flows and infrastructure projects, resulted in the identification of pedestrian risk hotspots, especially near schools and kindergartens. It also led to the installation of the first smart traffic sensors at one of the city’s busiest intersections, generating real-time data to inform a digital model for traffic optimisation. Furthermore, the partnership resulted in the production of technical analyses for major road rehabilitation projects, running 18 traffic simulation scenarios to guide future investments.
City of Fredericton, Canada – Port St. Johns, South Africa
The city of Fredericton in Canada dispatched specialists from their municipal water utility to support the city of Port St. Johns in South Africa to assess and manage the climate risks of their water system assets. Using local data, they helped to map and visualise the data using geographic information system technology, to better communicate the information with local authorities and politicians to make the case for investments in the city’s ailing water infrastructure. Through their engagement with local authorities and the water utility, they also helped to break down silos between different divisions in the municipality. After their intervention, specialists in Port St. Johns were equipped with the capacity to independently manage and visualise their data and maintained an improved working relationship with other parts of the administration to ensure climate resilience of the local water system.
Municipality of Baruth/Mark, Germany – City of Mörön, Mongolia
The partnership between Baruth/Mark (Germany) and Mörön (Mongolia) led to the establishment of a low-energy centre for sustainability education co-created with local partners and built entirely by the local workforce using Mongolian materials to showcase local feasibility and capability. Baruth also supported a community-driven waste management project, developed in close co-operation with schools, civil society and municipal actors. The initiative combined infrastructure with awareness raising: 35 large waste containers were designed and produced locally by the vocational school’s metalworking class, providing students with hands-on training and visible results in their community. Households and schools received training on waste separation, including the use of organic waste for composting and the recovery of recyclables. The project also promoted composting to support local food production and conducted public sensitisation campaigns, including the distribution of thousands of reusable bags. One of the tangible outcomes of the waste management project is the emergence of a local women-led enterprise.
Municipality of Lahr/Schwarzwald (Germany) – City of Alajuela, Costa Rica
The partnership between Lahr/Schwarzwald (Germany) with the city of Alajuela (Costa Rica) resulted in the protection of drinking water sources and improved water quality in Alajuela through the rehabilitation of its central wastewater treatment plant, upgrades to two decentralised treatment facilities and the setting-up of physical barriers and public awareness campaigns to safeguard water sources. These measures also triggered an institutional shift with wastewater and drinking water management being reorganised, gaining a stronger position and visibility within the municipal administration. This has attracted greater political attention to water issues.
Sources: OECD (2025[16]), “OECD Surveys on the Impact of Decentralised Development Co-operation on Local Governance and Sustainable Development Goal Outcomes”, Unpublished, OECD, Paris; bilateral interviews conducted with case-study representatives.
Is higher cross-border DDC ODA associated with improvements in SDG outcomes?
Within the same country, cross-border DDC ODA is targeted towards partner regions and cities with greater development needs in terms of income per capita. Localised CRS data show that, when comparing regions within the same country, poorer regions tend to receive higher levels of cross-border DDC ODA per capita. A 1% higher level of gross national income per capita is associated with around 0.57‑0.64% lower cross-border DDC ODA received, indicating that resources are directed towards less developed regions in terms of income per capita within recipient countries. In the available sample, the relationship becomes visible only through subnational analysis, as cross-country comparisons may mask territorial allocation patterns due to differences in provider-country priorities, country contexts, historical ties and country-specific co-operation frameworks.
Quantitative evidence points to positive associations between cross-border DDC ODA and certain social and environmental SDG-related indicators. For recipient regions, cross-border DDC ODA targeting poverty reduction (SDG 1) is associated with modest but positive improvements in regional income growth even when country-level factors are taken into account. For cities, increases in cross-border DDC ODA related to urban sustainability and climate action (SDGs 11 and 13) are associated with increases in urban green areas and reductions in per capita transport carbon dioxide emissions. These indicators were used because they are conceptually aligned with the analytical framework and available at the relevant territorial scale, with sufficient comparability for quantitative analysis. Further details on indicator selection, data sources and empirical specifications are provided in Annex D. These and the following quantitative results of this chapter should be interpreted as indicative associations rather than causal effects, but they suggest that cross-border DDC ODA can be linked to measurable territorial changes in areas connected to the thematic focus of DDC activities.
Taken together, the qualitative and quantitative evidence from regions and cities in partner countries suggests that DDC can contribute to both SDG-related outcomes and local governance improvements through a combination of financial support, technical assistance, capacity building and peer-to-peer exchange. Quantitative results point to positive associations with available and policy-relevant subnational indicators, including income, urban greenness and transport emissions, while survey and case-study evidence highlight perceived impacts on a range of environmental and social SDGs, as well as on local governance dimensions such as inclusiveness and participation, accountability, strategic planning and administrative capacity.
Benefits of DDC engagement in DAC regions and cities
Perceived SDG impacts in DAC territories are centred on partnership building, sustainable cities and institutional strengthening, with additional benefits reported in education, responsible consumption and production and climate action. SDG 17 “Partnerships for the goals” stands out with the highest average rating (4.1), reflecting the partnership-based nature of DDC itself. It is followed by SDG 11 on sustainable cities (3.5), reflecting benefits on urban sustainability and local policy learning, while SDG 16 related to peace, justice and strong institutions is also among the highest-rated SDGs (3.2), in line with the governance-related benefits identified in the analytical framework. For most other SDGs, impacts in DAC territories are likely to materialise over longer timeframes and mainly through non-financial components, such as peer learning, knowledge exchange, stakeholder mobilisation and policy innovation. These include perceived positive contributions to SDG 4 for quality education (3.2), SDG 12 on responsible consumption and production (3.1) and SDG 13 on climate action (3.0) (Figure 4.17).
DAC LRG respondents also report positive perceived impacts of DDC on local governance, particularly on multi-level governance and inclusiveness and participation. On a scale of 1 to 5, LRGs in DAC countries rated the impact of their DDC projects on overall local governance in their own region and city at 3. The highest-rated governance dimensions were multi-level and cross-sectoral co‑ordination, including co-ordination with national government, other subnational authorities, departments and stakeholders within the same region or city (3.3), inclusiveness and participation, including civic engagement and public participation in decision making (3.3), and administrative and technical capacity of public servants (3.0). Other impacts, although more moderate and around the midpoint of the rating scale, were also reported for strategic planning, management and innovation, and accountability, transparency and public sector integrity. These results suggest that, for DAC cities and regions, DDC is most often perceived as strengthening collaboration, participation and policy learning, rather than directly changing regulatory or legal frameworks (Figure 4.17Figure 4.17).
Notes: Number of respondents in parentheses. Rating on a scale of 1 to 5 (where 5 = very high impact).
Source: OECD (2025[16]), “OECD Surveys on the Impact of Decentralised Development Co-operation on Local Governance and Sustainable Development Goal Outcomes”, Unpublished, OECD, Paris.
Case studies illustrate how these benefits can materialise through institutional learning, stakeholder engagement and policy innovation in DAC cities and regions. In Lahr/Schwarzwald (Germany), an education for sustainable development project with Alajuela (Costa Rica) led to the development of bilingual educational materials in German and Spanish used in both cities, while also increasing the visibility of Lahr’s DDC engagement among other German municipalities. In Zurich (Switzerland), engagement in DDC helped foster horizontal co-ordination within the administration and strengthened co-operation with a broad range of national and international stakeholders across multiple areas of expertise, for example with the Federal Institute of Technology Zurich on urban food systems. In Châtellerault (France), knowledge sharing with Kaya (Burkina Faso) supported innovation in education for sustainable development and contributed to raising awareness of the SDGs and international solidarity among municipal staff, elected officials and local stakeholders. Glasgow’s (the United Kingdom) participation in the United Nations Environment Programme (UNEP) Generation Restoration project also illustrates how DAC cities can share expertise on nature-based solutions while gaining access to international networks and potential funding opportunities (Box 4.5).
Municipality of Châtellerault, France
In the municipality of Châtellerault, knowledge sharing on environmental issues with its partner municipality Kaya in Burkina Faso has fostered innovations for both partners, particularly in the area of low-technology solutions in education for sustainable development, for example by adapting waste management education methods for children developed in Burkina Faso, in schools in Châtellerault using an educational game format (contributing to SDG 12” Responsible consumption and production”, and SDG 4 “Quality education”). The co-operation has also raised awareness of the SDGs and international solidarity within the municipality, including among staff and elected officials, and led to the establishment of an annual SDG Week with exhibitions and events to showcase local contributions to the 2030 Agenda.
City of Glasgow, the United Kingdom
Glasgow participates as a lighthouse city in the UNEP Generation Restoration project led by UNEP, Local Governments for Sustainability and the World Bank as part of the UN Decade on Ecosystem Restoration. The project connects Glasgow with pilot cities in Africa, India and the Philippines, where Glasgow contributes technical expertise on nature-based solutions for urban development. As such, Glasgow shares knowledge, best practices and real-world experiences on drainage, flooding and soil health, while also gaining access to potential funding streams from partners such as the World Bank to implement and expand its own local restoration and nature-based solutions projects.
City of Fredericton, Canada
The partnership between Fredericton (Canada) and Port St. Johns (South Africa), presented in the previous box in relation to its impacts in the partner municipality, also generated benefits for government officials and experts from Fredericton. First, they had the opportunity to collaborate closely with colleagues from different agencies and departments, fostering cross-institutional networks. These relationships were maintained even after their assignment on the development project ended, helping to break down silos in their home administration. Second, they developed a broader strategic perspective by observing and analysing climate-related challenges that the cities in Canada have not yet encountered but may need to anticipate in future planning and resilience strategies.
Sources: OECD (2025[16]), “OECD Surveys on the Impact of Decentralised Development Co-operation on Local Governance and Sustainable Development Goal Outcomes”, Unpublished, OECD, Paris; bilateral interviews conducted with case-study representatives.
Is higher DDC engagement associated with stronger local governance?
Higher DDC engagement, measured as cross-border DDC ODA per capita, is associated with stronger governance performance in DAC regions. This relationship is visible across regions within the same country, although it does not hold for all DAC countries with available data (Figure 4.18, see Annex E for the association with the indicator of trust in local police, a proxy of trust in local institutions). Quantitative analysis shows that higher levels of cross-border DDC ODA per capita are positively and statistically significantly associated with the EQI, after accounting for regional gross domestic product per capita, population and provider-country differences. Positive and statistically significant associations are also observed for the indicators of impartiality, integrity (lower perceived corruption) and quality of services-related dimensions of governance, suggesting that at the regional level, DDC engagement and governance performance tend to move in the same direction (detailed specifications and regression outputs are presented in Annex D).
Note: n denotes the number of observations.
Sources: Based on OECD (2025[2]), CRS: Creditor Reporting System (flows), https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds]; ECOPER (2024[13]), Decentralised Cooperation Report: Mapping Donor Cities, https://www.elankidetza.euskadi.eus/contenidos/documentacion/publicaciones_descentralizada/es_def/INFORME-2024-ENG-.pdf; Charron, N., V. Lapuente and M. Bauhr (2024[9]), “The Geography of Quality of Government in Europe: Subnational Variations in the 2024 European Quality of Government Index and Comparisons with Previous Rounds”, https://www.gu.se/en/quality-government/qog-data/data-downloads/european-quality-of-government-index.
Overall, the qualitative and quantitative evidence points to a distinct but complementary set of benefits for DAC regions and cities engaged in DDC. While the primary development objective of DDC remains focused on partner territories, provider regions and cities can also derive institutional value from participation in DDC. Survey and case-study evidence highlights perceived benefits related to co‑ordination, participation, innovation and new knowledge for service delivery. In addition, the statistical analysis suggests that cross-border DDC ODA per capita and governance performance are positively associated in DAC provider regions. While these associations may primarily reflect underlying factors that are linked to both stronger governance and higher levels of DDC engagement, it is also consistent with evidence from the DDC impact surveys and case studies showing that non-financial DDC activities, including peer learning and technical assistance, can provide DAC regions and cities with access to new knowledge and innovation, while strengthening institutional capacity and co-ordination. These results point to DDC as a mechanism that can generate value on both sides of the partnerships.
Policy implications of the quantitative results
The quantitative evidence presented in this chapter forms part of a mixed-sources approach to assessing the impact of DDC. The empirical results are complemented by the DDC impact surveys, covering over 180 LRGs across 43 DAC and partner countries, and the evidence from the 16 in-depth case studies conducted across 9 DAC countries. Together, these sources provide a more comprehensive understanding of how DDC generates outcomes than any single method alone.
Quantitative results should be interpreted in light of the characteristics of the indicators analysed, as well as the data and methodological constraints. Estimated relationships, whether statistically significant or not, depend on the indicators used, sample sizes and time horizons covered. Some results point to statistically significant associations but, beyond not being causal, they should not be interpreted as universal, as they apply only to the observed sample in a specific timeframe. Conversely, the absence of measurable effects for some SDG dimensions likely reflects the long timeframes over which these outcomes materialise, rather than a lack of impact.
With these considerations in mind, the quantitative evidence presented in this analysis points to three main policy-relevant insights for DDC:
1. Within countries, cross-border DDC ODA appears to be broadly targeted towards regions with greater development needs in terms of income per capita. The allocation analysis suggests that subnational actors tend to direct resources to less developed regions within countries. This pattern is not observed across countries in the sample, potentially as some middle-income countries receive levels of cross-border DDC ODA similar to those in low-income countries, likely reflecting the role of shared history, sectoral priorities and country-specific co-operation frameworks. This finding also highlights the value of subnational analysis, as national-level comparisons can mask important territorial differences in the allocation of cross-border DDC ODA.
2. Cross-border DDC ODA shows indications of positive local effects in partner regions and cities, although these effects are only partially captured by available quantitative indicators. The empirical results indicate that cross-border DDC ODA is associated with improvements in available and policy-relevant SDG-related indicators, including income growth in recipient regions and environmental indicators such as urban greenness and transport emissions in cities. While these associations are modest, they are detectable despite the relatively small scale of DDC compared to wider structural drivers of development. At the same time, important dimensions of DDC – such as institutional learning, governance improvements and capacity building – are not fully captured in available DDC data, suggesting that measured impacts may only reflect part of the overall contribution. Survey and case-study evidence complements these quantitative findings by showing that partner LRGs perceive impacts across all SDGs and several dimensions of local governance, including participation, accountability and administrative capacity.
3. DDC engagement and governance performance are associated in provider DAC regions. These are simple associations and should not be interpreted as evidence of causal effects in either direction. However, they are compatible with the types of non-financial channels through which DDC engagement may be linked to governance performance, such as peer learning, knowledge exchange, policy innovation and strengthened institutional capacity and co-ordination. Survey responses and case studies show that DAC LRGs perceive benefits from DDC engagement in multi-level and cross-sectoral co-ordination, inclusiveness and participation (including civic engagement and public participation in decision making), and administrative and technical capacity of public servants.
Limitations and future research agenda
The analysis highlights four main areas where further work is needed to strengthen the evidence base on DDC impacts.
First, while the localisation of the CRS represents a significant methodological advance, it remains partial and does not yet cover the full universe of DDC projects. The current approach enables the construction of a sufficiently large sample for empirical analysis, but further efforts are required to systematise localisation across all CRS entries and over longer time periods. The localised sample may also be affected by selection bias if projects with certain characteristics are easier to detect than others. For example, larger or more visible projects may be more likely to include specific geographic references in CRS descriptions and therefore be identified by the localisation tool. Improving the consistency and completeness of geographic information in project reporting would enhance both coverage and comparability.
Second, important data gaps remain in the measurement of SDG and local governance outcomes at the subnational level in both DAC and ODA-eligible countries. Despite recent progress, available indicators are still limited in both spatial and temporal coverage. Expanding time series is especially important to capture long-term effects and to analyse outcomes that evolve slowly, such as education, health or institutional quality. Strengthening the availability of harmonised, longitudinal subnational indicators would allow for more robust and comprehensive impact assessments.
Third, future research should develop identification strategies to move from statistical associations towards causal analysis. The quantitative analysis presented in this chapter controls for observable territorial characteristics and country-level differences, but it cannot fully account for all factors that may shape both DDC allocation and local outcomes, such as historical partnerships, political priorities, local institutional capacity or external shocks. As longer time series and more complete territorial data become available, future work could explore quasi-experimental approaches, event-study designs or other identification strategies suited to the structure of DDC data. This would help assess more precisely whether changes in local governance and SDG outcomes can be attributed to DDC engagement, rather than to broader contextual dynamics.
Fourth, the non-financial dimension of DDC remains largely unmeasured in quantitative terms. While the DDC impact surveys conducted as part of this project provides initial insights, there is currently no systematic set of indicators capturing key channels such as peer learning, capacity building or institutional exchange. These dimensions are central to the value proposition of DDC, particularly for regions and cities from DAC countries, and are likely to shape both governance outcomes and longer-term development trajectories. Developing robust metrics in this area is essential to better understand how, and under what conditions, DDC generates mutual benefits across partners and to inform the design and scaling of future co-operation initiatives.
References
[9] Charron, N., V. Lapuente and M. Bauhr (2024), “The Geography of Quality of Government in Europe: Subnational Variations in the 2024 European Quality of Government Index and Comparisons with Previous Rounds”, QoG Working Paper Series, No. 2024:2, Department of Political Science, University of Gothenburg, https://www.gu.se/en/quality-government/qog-data/data-downloads/european-quality-of-government-index.
[15] ECOPER (2025), Decentralised Cooperation Report 2025, https://www.elankidetza.euskadi.eus/contenidos/documentacion/publicaciones_descentralizada/es_def/adjuntos/2025-DODA_ENG.pdf.
[13] ECOPER (2024), Decentralised Cooperation Report: Mapping Donor Cities, https://www.elankidetza.euskadi.eus/contenidos/documentacion/publicaciones_descentralizada/es_def/INFORME-2024-ENG-.pdf.
[14] eLankidetza (2026), Development Co-operation Reports, Basque Agency for Development Co-operation, https://www.elankidetza.euskadi.eus/documentacion/publicaciones-sobre-cooperacion-descentralizada/webela01-eduki/es/.
[7] Global Data Lab (2025), “SDG Dashboard”, https://globaldatalab.org/sdgs/.
[4] International Forum on TOSSD (2026), About TOSSD, https://www.tossd.org/en/about-tossd.html.
[6] Mari Rivero, I. et al. (2025), Urban Centre Database (UCDB) R2024A, European Union, https://human-settlement.emergency.copernicus.eu/ucdb2024Overview.php.
[2] OECD (2025), CRS: Creditor Reporting System (flows), OECD, Paris, https://data-explorer.oecd.org/vis?lc=en&pg=0&snb=18&df[ds].
[8] OECD (2025), “Integrated urban policy to achieve the SDGs”, OECD Regional Development Papers, No. 179, OECD Publishing, Paris, https://doi.org/10.1787/ada07311-en.
[5] OECD (2025), Measuring the distance to the SDGs in regions and cities, OECD, Paris, https://www.oecd-local-sdgs.org/.
[10] OECD (2025), OECD Database on regions, cities and local areas, OECD, Paris, http://oe.cd/geostats.
[12] OECD (2025), “OECD Geographical Definitions”, OECD, Paris, https://www.oecd.org/en/data/datasets/oecd-geographical-definitions.html.
[16] OECD (2025), “OECD Surveys on the Impact of Decentralised Development Co-operation on Local Governance and Sustainable Development Goal Outcomes”, Unpublished, OECD, Paris.
[1] OECD (2023), City-to-City Partnerships to Localise the Sustainable Development Goals, OECD Urban Studies, OECD Publishing, Paris, https://doi.org/10.1787/d2fe7530-en.
[3] OECD (2023), “Decentralised development co-operation: A global policy toolkit and guidance for practitioners”, OECD Regional Development Papers, No. 46, OECD Publishing, Paris, https://doi.org/10.1787/3cb22851-en.
[11] Pesaresi, M. et al. (2024), “Advances on the Global Human Settlement Layer by joint assessment of Earth Observation and population survey data”, International Journal of Digital Earth, Vol. 17/1, https://doi.org/10.1080/17538947.2024.2390454.
← 1. Where possible, homonymous regions and cities are disambiguated using the information available in project descriptions. However, as some cases cannot be fully disambiguated, particularly where a region has the same name as its main city, region-level and city-level analyses are always reported separately to avoid double counting.