3. Measuring what works for regional attractiveness

Abstract

This chapter provides an in-depth introduction to the OECD’s novel quantitative measurement framework for assessing regional attractiveness. This innovative approach considers global engagement beyond international connections and financial drivers alone. In total, the methodology considers more than 50 indicators, covering 14 dimensions, across 6 domains of attractiveness (economic attraction, connectedness, visitor appeal, natural environment, resident well-being, and land use and housing). The chapter also introduces a diagnostic tool in the form of a regional “attractiveness compass” to identify regional strengths and gaps and to inform the development and monitoring of regional attractiveness policies. Finally, it identifies a range of drivers of regional attractiveness to the key international target groups of investors, talent and visitors.

Key messages

A multi-dimensional regional attractiveness compass can guide decision makers

In this report, the OECD proposes an innovative and multi-dimensional approach to assessing regional attractiveness: the regional attractiveness compass. The approach considers global engagement beyond international connections and financial drivers alone. In total, the methodology considers more than 50 indicators to develop regional attractiveness profiles, covering 14 dimensions of attractiveness, across 6 domains (economic attraction, connectedness, visitor appeal, natural environment, resident well-being, land use and housing).
The use of an evidence-based attractiveness compass can help identify regional strengths and gaps as well as consider trade-offs and synergies related to regional development strategies to attract investors, talent and visitors.
For policy makers, the compass can be seen as an off-the-shelf tool for prioritising, monitoring, adapting and evaluating regional attractiveness policies, providing a holistic approach instead of siloed implementation.

Investment attraction is motivated by physical and digital infrastructure as well as universities

Foreign direct investment (FDI) is analysed in terms of two outcome variables: the total number of projects and the total capital expenditure.
Greenfield FDI, in terms of the number of projects, exhibits a significant and positive relationship with rail performance, flight accessibility and the presence of a top 500 university.
When measured by total capital expenditure, Internet download speeds and top universities matter in the context of OECD regions.

Housing and broadband are of utmost importance to talent

The location decisions of international talent – defined as people with skills corresponding to the needs of public and private, place-specific economic and social development strategies – appear to be motivated above all by Internet quality and affordable housing. These trends are likely to rise with the potential continued adoption of hybrid and remote working schemes.

A range of factors influence the attractiveness of regions to visitors

Visitors are enticed by regions with a cleaner environment, paving the way for more sustainable tourism development.
Small and medium-sized enterprises (SMEs) are the backbone of the tourism sector and those regions with high levels of entrepreneurship are likely to be more attractive to visitors.

Some factors are effective in attracting multiple target groups

Regions attractive to talent, visitors and investment appear to either host a higher share of international students or a top 500 university, suggesting a potential policy synergy between the internationalisation of education and attractiveness.
For both talent and FDI, Internet access and quality are prominent drivers, underscoring the importance of upgrading digital infrastructure.
Future work will build out both the TL2 and TL3 databases to better serve policy makers at more local levels, in addition to being applied in more international contexts (e.g. OECD regions in Asia, North America).

Setting the context

This chapter introduces and applies a novel measurement framework for assessing regional attractiveness. It was first published as a methodological working paper and applied in a regional context across France (OECD, 2021[1]; 2022[2]; 2022[3]). Six domains of attractiveness broadly encompass the potential assets of territories:

Economic attractiveness (economy, innovation, labour market).
Connectedness (transportation, digitalisation).
Visitor appeal (tourism, cultural capital).
Natural environment (environment, natural capital).
Resident well-being (social cohesion, health, education).
Land use and housing (land, housing).

Going beyond traditional economic factors, this chapter seeks to explore non-financial and non-market drivers of attractiveness to determine what contributes to drawing investors, talent and visitors to a region. The quantitative analyses focus on actionable regional development policy tools such as infrastructure, skills, environment and well-being, while considering the scope (i.e. breadth of indicators, availability, missing indicators), scale (geographic, time-bound) and robustness of the database. It concludes with a selection of indicators to support the design, monitoring and evaluation of attractiveness policies as they pertain to the three targets covered under this work: talent, investors (considering also export development) and visitors.

While the determinants of attractiveness are context-dependent and vary from country to country, looking at a broad sample of regions across the OECD provides important insights into the common variables that define regions which have proved attractive to talent, investors and visitors on the move. The chapter presents statistically robust, time-bound and large geographic samples of data helping policy makers unpack what is important for regional attractiveness. Given the choice of outcome variables, it is likely that a whole host of factors not shown in these statistical results may also matter. The purpose of this chapter is not to exhaust all significant drivers of regional attraction but to highlight the common denominators across places. Doing so can guide policy makers and regional actors to dive deeper into the results and the performance of their territory in relation to the findings. For example, education – measured by both the number of top institutions and international student shares – appears to matter for talent, investment and visitor attraction. Does your region perform well across this dimension? If not, what other assets can explain regional attractiveness in your territory? The chapter aims to stimulate these types of questions, which can ultimately inform regional attractiveness strategies. Opting for such an approach makes attractiveness strategies available to all regions, enabling them to identify and strategically build on their assets.

These questions are an essential ingredient to reducing territorial asymmetries made clear by recent shocks and crises. The impact of the COVID-19 crisis, compounded by the consequences of Russia’s war of aggression against Ukraine and existing megatrends (e.g. climate change, globalisation, digitalisation and demographic change), continue to produce asymmetric effects within and between regions, with the scope and scale depending on their unique characteristics. For example, regions already found to be caught in or at risk of falling into a regional development trap – regions that are unable to both maintain their past dynamism and keep up with their peers in terms of income, employment and productivity growth – are found to have a smaller share of manufacturing, a higher old-age dependency ratio and a lower share of the population with secondary education. These regions – which span low, medium and high incomes and exist in most developed, emerging and developing countries – will be less resilient to external shocks if preventative measures are not implemented to mitigate against further entrapment (Diemer et al., 2022[4]). These differences manifest also in terms of human development and remain apparent when looking at a global sample of nations: the subnational results of the United Nations Human Development Index (HDI) show a roughly equivalent score difference between the highest-scoring state in the United States (US) – Connecticut – and the lowest-scoring – Mississippi – as between the United States and Sao Tome and Principe, which rank 3^rd and 156^th respectively in terms of HDI.

To respond to these regional disparities, the OECD has developed this innovative and multi-dimensional approach to assessing regional attractiveness to help regional and national policy makers better understand the position of regions in this evolving global context. Moreover, the framework provides a view into emerging challenges and opportunities and supports the identification of policy levers to enhance the attractiveness of regions to the key international target groups of investors, talent and visitors. In doing so, it seeks to support regions’ transition towards new territorial development policies that promote inclusive, sustainable and resilient development, while enhancing regional attractiveness.

The aim of this work is not to rank regions but rather to provide an analysis of how they perform against each dimension. Moreover, it is to improve policy makers’ understanding of how this performance translates into the attraction of investors, talent and visitors (OECD, 2022[3]). In this sense, the framework is more of a diagnosis of where a region stands relative to other regions in its country and international counterparts (e.g. OECD or EU) and a tool that allows regions to prioritise efforts to enhance the attractiveness of their territory to the key target groups, by building on strengths and closing the gaps.

Above all, the imperative is to measure the assets – the natural, human, social, manufactured and economic capital – of regions and to situate these within a broader global context influenced by recent economic shocks, existing megatrends and current shifts in policy priorities. In doing so, policy makers can better assess their strengths and gaps, which can be leveraged or addressed to improve regional resilience. The levers of regional attractiveness identified by this approach are not those of the macroeconomy, of financial allocations or of sectoral policies. They are part of a full array of regional development policy strategies. In addition, each dimension is as actionable as it is measurable. The rationale to include a large selection of indicators and dimensions is based on the cross-cutting nature of the regional attractiveness paradigm. In other words, no one variable suffices to entice key international target groups to a region and no one policy alone will address a region’s attractiveness challenges. Taken together, however, trade-offs and synergies can be identified that stimulate strategic thinking about what works for regional attractiveness in the new global environment.

In the first iteration of the methodology, three potential challenges to regional attractiveness to investors, talent and visitors were considered including, above all: i) citizens’ increasing questioning of globalisation; ii) the risks posed by climate change, demographic change and digitalisation; and iii) global shocks including COVID-19 and the war in Ukraine (OECD, 2022[3]). Regarding globalisation, not only citizens but firms, regions and nations share this sense of insecurity. Since the outbreak of COVID-19, the consideration of potential opportunities related to reshoring, onshoring and nearshoring has increased tenfold in the reports to investors required of publicly listed companies (IMF, 2023[5]). Meanwhile, the climate change rhetoric as outlined in the US Inflation Reduction Act, for example, is very much about enhancing domestic energy security and attracting foreign investment thanks to competitive energy prices. A similar dynamic can be seen in Europe where domestic energy subsidies have surged following the outbreak of the war (Kiel Institute, 2023[6]). Meanwhile, across the OECD, active industrial policies are being explored to identify how the production of strategic assets – from face masks to banking – can be conducted domestically, a policy known as “strategic autonomy” (Criscuolo et al., 2022[7]). As the frequency and gravity of shocks grows, the asymmetric impact within countries becomes more apparent, furthering the need to adopt a regional lens to support attractive policies that lead to more inclusive and resilient regional development.

The extent to which the COVID-19 crisis impacted regional economies depended heavily on their positioning within global value chains and the type of FDI they attract. Disruptions to value chains and the slowdown in FDI were universal in 2020 but, in countries ranging from Brazil to the United States, the subnational impacts demonstrated strong heterogeneity. In the United States, for example, states, where FDI relies on the movement of people, declined (e.g. Nevada, with its strong tourism sector) while foreign investments increased in the Southeastern states where international investment is concentrated in acquisition/market-seeking activity. Across Brazil, while the largest states with the most globally-embedded supply chains suffered negative economic shocks brought on by the pandemic, the effects were more severe in backwardly-integrated states whose core function is the supply of key inputs to the larger states (Wright and Wu, 2022[8]). The same can be said about the impacts of the war in Ukraine, as regions with stronger trade linkages with Belarus, Russia and Ukraine are now needing to rethink their positioning in the global fold (OECD, 2022[9]).

Finally, the changing global climate continues to force regions to rethink their territorial assets. Europe’s ski regions saw some of the highest temperatures and lowest snowfall by January of this year (OECD, 2023[10]). More and more, climate considerations are an integral part of people’s location decisions. A survey conducted in the United States found that roughly one-third of respondents between 18-41 years of age cited climate change as a reason to relocate in 2022 (Forbes, 2023[11]). Increasingly so, moving is less of a decision. While climate migrants are often depicted as refugees from less developed nations, research shows that between 2008 and 2019, nearly 700 000 people on the European continent were internally displaced due to weather events (IDMC, 2020[12]). While many are able to return to their home region, the rate at which disruptive weather events are increasing slowly degrades the liveability of places (Khanna, 2022[13]). More to the point, the environmental quality of a region is of utmost importance to talent (see Annex Table A A.2). People – especially talent and visitors – want to know that the air is clean, nature is accessible and good environmental practices – from recycling to a clean supply of energy – are being implemented.

In an increasingly uncertain global environment, a compass is needed to support decision makers identify the levers at their disposal to take advantage of the opportunities associated with globalisation and its evolution and to make their regions more attractive and economically dynamic, sustainable and resilient, as discussed in Chapter 2.

A regional attractiveness methodology

The intention behind rethinking what a framework for attractiveness entails is about going beyond traditional cost- and market size-related drivers to incorporate less conventional cost and non-cost assets that are core levers of a region’s appeal to investors, talent and visitors. For example, the importance of housing affordability – an issue afflicting regions and cities across the OECD, especially in the wake of the COVID-19 crisis – is a fundamental tenant of a place’s appeal to potential talent and an important indicator to monitor for regional decision makers hoping to attract and retain new residents. The employment rate of migrants is an important measure of integration that, again, regional decision makers should be concerned with in the face of demographic decline, as it demonstrates the absorptive capacity of the region’s economy to new talent. Employment in cultural and creative industries is an indication of the cultural assets of a region to be enjoyed by potential visitors. On the investment front, a robust talent pipeline – illustrated by education rates and top universities – assures investors of access to human capital.

All in all, the dimensions cover policy areas that contribute to a region’s capacity to attract international investors, talent and visitors according to the following taxonomy, which reflects the relationship between attractiveness and regional development:

International investors are international firms, entities or individuals willing and able to make capital investments in regional economies.
International talent are people with skills corresponding to the needs of public and private, place-specific economic, environmental and social development strategies.
International visitors are people travelling inbound from a region outside the region and the country of reference.

While this illustrates that the focus of this exercise is to conduct statistical and policy analysis on the international aspect of regional attractiveness and development, the methodology can be equally applied in a domestic context where a more equal spread of within-country target groups across places can generate more balanced regional development. Indeed, applying this approach in a domestic context can serve to widen the discussion from attraction to retention, focusing on the policies and indicators required to implement and monitor whether attraction policies translate into medium- and long-term retention – especially in the context of talent and investment. The database can be extended to look at a more localised understanding of attractiveness by exploiting available small region data (Territorial Level 3 – TL3) – also allowing for an analysis of disparities within medium-sized regions (Territorial Level 2 – TL2). A preview of the dimensions below forms the foundation of the regional attractiveness methodology (OECD, 2022[3]). Each cover between three and six indicators that are deemed essential ingredients to a region’s attractiveness (Figure 3.1):

Economic attractiveness: Ten of the indicators cover regional economic performance, including yet reaching beyond (regional) gross domestic product (GDP). For example, the framework looks at Patent Cooperation Treaty (PCT) patent applications per million inhabitants, given that intellectual property is a key determinant of a region’s innovation ecosystem and has been found to not only attract but retain foreign investment at the subnational level (Tang and Beer, 2022[14]). Additional indicators cover employment, research and development (R&D), productivity and entrepreneurship.
Connectedness: This domain considers both physical and digital infrastructure, from roads to the Internet, which are fundamental pieces of any region’s development path. In exploring the drivers of attractiveness below, these infrastructure variables show salient results given their strong relationship with greenfield FDI projects and expenditure in regions (see Annex Table A A.1).
Visitor appeal: On the visitor front, indicators assessing the tourism performance of regions are included to take stock of both visitor numbers and tourism capacity across regions. Cultural capital is also considered, one measurement being the share of employment in cultural and creative industries which shows significant correlations with foreign visitor numbers (see Chapter 6).
Natural environment: The natural environment varies significantly depending on the type of region, with rural regions in particular facing vulnerabilities in light of both climate change and the climate transition given their sectoral composition which relies heavily on natural resources (OECD, 2020[15]). Evidence on the connection between talent attractiveness and the natural environment illustrates a negative relationship between deteriorated air quality and talent location decisions (Flood et al., forthcoming[16]). Moreover, outdoor amenities and activities that are part of the local conditions for well-being depend heavily on the quality of and access to the local environment. In the regional case of Nova Scotia, Canada, research on provincial levels of well-being indicate that, on a 7-point scale, a 1-point increase in perceived environmental well-being translates into a 0.59-point increase in overall well-being (Flood and Laurent, 2021[17]).
Resident well-being: The OECD’s work on regional well-being illustrates the importance of looking at multi-dimensional well-being within countries, noting that the within-country differences can sometimes be greater than those across borders. The poverty rate (after taxes and transfers) shows an average gap of 18 percentage points between best- and worst-performing regions within OECD countries, with the most marked differences among European countries observed in Belgium, Italy and Spain (OECD, 2022[18]).
Land use and housing: The importance of housing affordability as a determinant of quality of life cannot be overstated and has worsened in as many OECD countries as it has improved between 2010 and 2020, despite generally increasing incomes (OECD, 2020[19]). This indicator is included as a key measure of regional well-being that varies significantly within countries and can lead to a crowding-out effect for low-income groups together with an agglomeration of high-income groups, thereby exacerbating spatial inequalities (Yang and Pan, 2020[20]). Land prices and land development are part of a complex policy mix whereby governments must balance catering to investors’ needs, providing adequate quality housing for newcomers and residents alike and, above all, protecting the environment through the containment of land conversion. Given the salience of these competencies at the subnational level, it is critical to take stock of their performance across and within countries (OECD, 2022[21]).

Figure 3.1. Dashboard of regional attractiveness indicators

An attractiveness compass to navigate an uncertain global environment

To calculate the scores presented on the OECD regional attractiveness compasses, the data for each indicator are first normalised within a scale of 0 to 10. Then, the normalised values are divided into percentiles on a scale of 0 to 200. For indicators where a negative value signifies higher attractiveness, for instance the air pollution indicator, the percentile scores are adjusted accordingly. As a final step, the averages of all of the indicators are calculated to become composite scores in each dimension (for the full dashboard of indicators and database sources, see Annex Table A B.1). The attractiveness compass serves as a powerful tool for diagnosing the attractiveness of a region to investors, talent and visitors, which can be used to support the design, implementation and evaluation of regional development strategies.

Providing an evidence-based compass to regions gives decision makers a snapshot of their comparative performance that can support regional attractiveness strategies in two ways. First, it demonstrates what advantages the region holds when compared to regional peers at home and abroad. Then, it shows where gaps exist and the scale of those gaps, which can indicate if it aligns with existing regional development priorities and, if not, may indicate potential areas for future development. The compasses are based on databases covering two geographic levels: TL2 (regions) and TL3 (small regions). In the former, an extensive database exists that has now been deployed across the OECD while in the latter case – as depicted in the Swedish example – the database covers a smaller selection of reference indicators in EU countries. The compasses have been put to use across a community of practice now comprising 25 regions in 10 countries, after having been initially applied to the case of French regions (see OECD (2022[2])). Released together with this report, and mainly focused on European countries, are 15 regional cases – each with its own compass and attractiveness diagnostic – including regions in Ireland (3), Italy (4), Portugal (3), Spain (3) and Sweden (2).

Scores in each dimension do not consistently translate into “higher is better” and require regional policy makers to situate their scores within their respective development agenda. For example, the share of land converted to the artificial surface under the Land dimension may indeed signal responsible development and/or investment in needed infrastructure (e.g. roads, housing, rail), while it could also refer to environmental malpractice. In this vein, it is essential to note that trade-offs might exist in one area with respect to another. To take a local-global view, it is a key part of the global sustainable development agenda to protect a minimum of 30% of terrestrial land by 2030 – an indicator which varies significantly across regions within a country while at the same time, land is often needed as a precursor for investment (in particular manufacturing industry) attraction. Similarly, the share of foreign visitors in a region’s tourism mix is encouraging insofar as it signifies the international appeal of the territory; however, the domestic visitor base is a vital driver of the sector’s recovery and long-term resilience (OECD, 2021[22]).

In the case of the Northern and Western region of Ireland (Figure 3.2), the region stands out as a visitor hub exhibiting stronger performance than any region in Ireland and well above that of the majority of EU counterparts. Social cohesion is another strong point, as is the environment and natural capital, particular draws for talent and visitors. At the same time, the region has not benefitted from the same levels of economic growth and productivity that have defined the other Irish regions over the past several decades. Transportation and digital infrastructure remain barriers to the region’s growth and development with limited options in terms of rail and less accessibility to flights and major roads than regions in Ireland and abroad.

In the Swedish county of Norrbotten (TL3 region) (Figure 3.2), an economy based primarily on natural resources (i.e. minerals and forestry resources) combined with critical energy infrastructure help to explain the strong economic performance of the region. Growing demand for tourism in Sweden’s north aligns well with Norrbotten’s existing natural and cultural strengths and provides significant opportunities to leverage the region’s attractiveness to visitors. Yet, Norrbotten has the smallest share of inhabitants with good access to train stations in the country and scores below the European average. Moreover, especially in inland areas, public transport is insufficient, spots without coverage remain and the rail network is not extensive. On the environmental front, Norrbotten produces far more greenhouse gas emissions per capita from the transport sector than the EU and national averages. At the same time, it offers good tree coverage and particularly good air quality compared to the average of EU regions.

Figure 3.2. A regional attractiveness compass to guide decision makers

In each case study, the attractiveness assets and gaps are assessed and aligned with broader regional development objectives and contexts. They are then connected to existing strategies, including how they relate to recent crises, and how multi-actor co-ordination mechanisms can enhance the attractiveness of the regions (explored further in Chapter 7).

An evidence scan – What works for regional attractiveness towards key targets

Given the asymmetric impacts explored above and the need to examine regional attractiveness through a regional lens, the aim of this activity was to develop a methodology to explore drivers of regional attractiveness for the three key international target groups of investors, talent and visitors beyond purely financial factors (Box 3.1). In doing so, this body of work examines some of the links between these target groups and the factors associated with their concentration, including detecting associations that may not have been identified previously in the literature. Moreover, the following chapters situate the findings in regional contexts, illustrating how the evidence can be translated into concrete regional attractiveness strategies. The detailed regression summary and tables from the technical analysis can be found in Annex A (Tables A.1, A.2 and A.3).

There is a particular need to explore these drivers at the subnational level given the significant regional disparities that exist within countries and the policies and tools regions have at their disposal to implement measurable attractiveness strategies (OECD, 2022[3]). The data included look at the TL2 unit of analysis (region-province-state, etc.) to ensure the inclusion of the maximum number of attractiveness indicators and cover a broad sample which encompasses all regions of the European Union and OECD where data are available. Analysis of the data begins with an assessment of the drivers of FDI, followed by talent attraction and, finally, visitors.

Box 3.1. Methods for assessing attractiveness towards the key target groups

Methodology

While the attractiveness database consists of both TL2 and TL3 regional data, the territorial unit of analysis has been chosen as TL2 for better data availability. The results presented hereafter cover a sample of the maximum number of TL2 regions per variable in question, usually consisting of either an EU sample or a larger OECD sample (according to the data accessibility).

While regression analysis has been used for all three key target groups (investors, talent and visitors), different regression methods have been used for each, mainly due to limited data availability. Two-way fixed effects panel regression was conducted for tourism (Figure 3.5), with time and region fixed effects, presented in a stepwise progress. Similarly, time and region fixed effects panel regression was used for talent (Figure 3.4), along with the time-lag model to incorporate some variables that only have two time periods in its time series data. In the case of FDI (Figure 3.3), cross-sectional regression was used with the most recent years chosen for each variable. The dependent variable for FDI was the aggregated sum of greenfield FDI from 2017 to 2022 and some of the main variables that exhibited a strong correlation with the FDI variable were only available for one year of data. For the regressions with talent and tourism, missing values were handled through data imputation so as to avoid the potential loss of data associated with using time series data for a multitude of indicators. The linear interpolation method was chosen to fill in the missing values.

Limitations and future research

To better understand the drivers of FDI attractiveness over time, a panel regression for FDI analysis – as carried out for visitors and talent – should be conducted. Moreover, while this research studies greenfield investment due to data availability, incorporating brownfield data can shed further light on what drives this type of investment, which is crucial for attracting SMEs with less liquidity for capital investments and completing new investors results with the extension of existing FDI in regions.

Finally, while the TL2 (NUTS-2 for European countries) level allows for the testing of a multitude of indicators related to attractiveness, this territorial level can mask within-region disparities in terms of attractiveness. Using a small selection of indicators available at a more local level (i.e. TL3/NUTS-3) can shed light on the attractiveness of sub-regional geographies, which was experimented with in some countries such as France and Sweden.

Note: Technical analysis is included in Annex A of the report.

Foreign direct investment

Attracting foreign investment is typically a major aim for regional economies, which is often reflected in regional development policies. Indeed, FDI can be an important lever to create jobs, foster skills development, increase wages, support productivity and encourage innovation and exports (OECD, 2019[23]). FDI projects tend to concentrate in specific locations, as they benefit from agglomeration effects, which can lead to important imbalances within and between regions. Indeed, regional disparities in FDI are considerably larger than those for GDP and productivity in nearly all OECD countries, with potential implications for inclusiveness. For example, between 2003 and 2021, the top 10% of regions in OECD countries with the highest greenfield FDI attracted on average 700 times more than the bottom 10% of regions (OECD, 2022[24]). Hence, uncovering which factors are driving FDI and how they interact is of capital importance to foster regional development but also to fight regional disparities.

National drivers of FDI have been the subject of significant examination in the past, while regional drivers have been less studied. For example, positive impacts are observed from information and communication technology (ICT) infrastructure development on foreign investment attraction in specific regions of China (Tang and Beer, 2022[14]), while a host of factors including proximity effects, GDP levels, labour abundance and innovation are determinants of FDI attractiveness as evidenced in different country-region contexts across Europe and India (ESPON, 2018[25]; Mukherjee, 2011[26]; Antonescu, 2015[27]). Studies comparing subnational jurisdictions across borders are scarcer, in part due to a relative lack of comparative FDI data at the regional level in different country contexts.

The outcomes of work on regional drivers of FDI tend to support what has been found at the national level. Foreign investors are attracted by regions where exploitable resources are abundant or where there are large consumer markets. They are also increasingly attracted by places where highly skilled workers, research and innovation, but also industry clusters are gathered, while other critical determinants include transport infrastructure and accessibility, as well as the quality of regional governance (OECD, 2022[24]; ESPON, 2018[25]; Amendolagine, Crescenzi and Rabellotti, 2022[28]). This report mainly explores this latter set of non-market drivers to provide support to regional development policies in a more holistic manner.

Methods and definition: Investment

To investigate how these factors relate to foreign investment, this report relies on several models, using two different dependent variables as indicators for foreign investment attractiveness. The three models cover different sub-samples of regions: i) OECD and EU regions; ii) OECD regions; and iii) EU regions. The different sub-samples of regions were constructed, not for comparative purposes, but rather to address the disparate regional coverage of each variable; transport-related indicators, for instance, cover EU regions but not all OECD regions. The two response variables are: i) the number of new greenfield FDI projects in each region over the 2017-22 period; and ii) the sum of foreign capital expenditure received by each region over the same period. There is a rationale to investigate these two indicators which, even if correlated, do not tell the same story. The second one encompasses the number of projects and the size of the projects. A high amount of capital expenditures can come from one or two big multinational firm(s) that make significant investments in a specific region, possibly in establishing a dominant position or from numerous SMEs investing relatively small amounts in a competitive market. In fact, the motivations of smaller investors can be driven by more diverse attractiveness factors as their strategic location decisions may be guided by considerations such as the work of subnational investment promotion agencies (IPAs), local relationships, networks and resources (Crescenzi, Di Cataldo and Giua, 2021[29]).

Summary of regression results

Presented in Figure 3.3, Panels A and B are a summary of regression coefficient results for investment attraction. The detailed regression summary and tables from the technical analysis can be found in Annex A. In Panel A, the number of greenfield FDI projects (the dependent variable) is, above all, concentrated in regions with a (higher number of) top 500 universities, and greater railway accessibility, flight accessibility and digital download speeds. In Panel B, the dependent variable is the total capital expenditure of greenfield FDI, where top 500 universities and digital download speed are the most relevant drivers.

Figure 3.3. Regression coefficients for investment attraction

Key drivers of FDI attraction

Based on the existing literature, a number of variables that have proven to be influential in the location theory behind FDIs are included in the analysis. The results of regression analyses utilising the regional attractiveness database and assessing FDI – in terms of both the number of projects and capital expenditures – are presented in Annex Table A A.1. Above all, the results illustrate the importance of educational institutions as beacons of investment attraction, in addition to good digital infrastructure and well-connected physical infrastructure, notably rail and air. The identified key drivers of FDI to regions are presented as the following:

Education: Top 500 universities – The number of top 500 universities emerges as an important driver of FDI for both the larger subsample and the EU regions’ sample, displaying the highest correlation coefficient with both the number of projects and the amount of capital expenditure. The explanatory link is quite straightforward: Top 500 universities are recognised for their high-level research and their capacity to attract and retain in the region high-skilled workers that are essential parts of the talent pipeline for competitive firms. Additionally, firms can co-operate with universities’ research centres, resulting in collaborative innovations, inventions and knowledge spillovers. They also benefit from a direct pool of skilled talent with managerial, technical and creative skills.
Digitalisation: Access to broadband Internet – Digital connectedness is also identified as a key driver of FDI, as it is increasingly more important for firms because of the digitalisation of their activities and as remote working and online conferences have become commonplace (Dorakh, 2021[30]). One control variable – GDP per capita, purchasing price parity (PPP) – is added to not confuse the influence of digital infrastructure and the presence of universities of excellence with local demand. It matters first and foremost for firms that do not aim only to benefit from efficiency gains and skilled labour of the local environment but that wish to also tap into local consumer markets.
Transportation: Train and flight accessibility – The importance of physical infrastructure can be seen as a means of enticing investment to new subnational locations through providing ready-made logistics solutions for prospective investors while it can also facilitate their plans for exporting to foreign markets. Indeed, this case has been observed in countries such as Portugal, where road expansion led to investment attraction in new localities throughout the 1980s and 1990s. This was particularly the case in places where the distance to the main cities (Lisbon, Porto) previously deterred investment in regional hubs (Guimarães, Figueiredo and Woodward, 2000[31]). Similarly, one of the roles of Italy’s Special Economic Zone programme (discussed in Chapter 4) involves building infrastructure connections (road, rail, freight and air) between lagging regions in the South and national and European transport networks. By connecting relatively remote regions to the existing transport grid, these regions became more attractive to investors and the new infrastructure facilitated the flow of goods (and talent).

Finally, it is important to consider the sectoral differences leading to FDI attraction. The projects with the highest capital expenditures across European regions over the time period studied are made in renewable energy, batteries and semiconductors, manufacturing, oil and gas and electricity. These types of investments are generally driven by access to natural resources – which can render non-metropolitan areas particularly attractive – and rely on a workforce with a unique – yet transferable – skillset. On the contrary, sectors that rely most heavily on human capital, such as business services and R&D sectors are associated with the projects in which lower amounts of money have been invested over the period – but with a larger number of projects.

Talent

Global flows of people are shaped by migration policy regimes, which can both enable and create barriers for those looking to move. With a pool of over 300 million seeking to move and work in other countries, combined with the 750 million who would move if they were able to (Esipova, Pugliese and Ray, 2019[32]), countries and regions have become frequently involved in a “war for talent”. A highly skilled and vibrant migrant population – especially the case in lagging euro regions (OECD, 2022[33]) – contributes significantly to regional economic development matters. In the context of increasing international migrations and the likelihood of climate shocks sending these flows into overdrive, policy makers need to consider how to attract talent to enhance regional development (Khanna, 2022[13]).

There are a number of international indices ranking top destinations for talent, however, these rankings often lack the regional dimension and exclude a more holistic vision of talent attraction that both takes into account a more inclusive definition of talent and less conventional indicators of their attractiveness. One complementary approach to regional attractiveness is the OECD Indicators of Talent Attractiveness, which allow users to assess which country-destination is most attractive to them according to select criteria (quality of life, income and tax, family environment, etc.) (OECD, 2023[34]). Fortunately, some evidence does make headway in exploring the regional dimensions across specific regional contexts. The interregional mobility of skilled labour across European regions may still be somewhat predicated on physical proximity between the host and sending region – often referred to as a gravity model in migration studies. Yet, other important non-cost variables that determine where talent chooses to locate are becoming more evident. For example, while access to amenities is an important factor, the degree to which this favoured metropolitan area decreased over time as non-metropolitan areas of Europe caught up by developing attractive amenities of their own (Miguélez and Moreno, 2013[35]). The results presented hereafter confirm the importance of amenities in the location decisions of talent.

Methods and definition: Talent

To better understand the link between talent attraction and various regional indicators, talent is defined as people with skills corresponding to the needs of public and private, place-specific economic and social development strategies. While traditional literature on talent migration has a narrower definition of talent, mostly as high-skilled people with tertiary education, talent is defined with a broader brush, focusing on the alignment of regional labour markets and the skills offered by talent, whether low- or high-skilled. Reflecting the choice of talent definition, the share of foreign-born employed people in the total working-age population (15-64) is used as the dependent variable for the analysis of talent attraction.

Summary of regression results

Presented is a summary of regression coefficient results for talent attraction. The detailed regression summary and tables from the technical analysis can be found in Annex A. As illustrated in the figure, housing affordability, Internet accessibility and a higher share of international students prove to be strong drivers of international talent attraction.

Figure 3.4. Regression coefficients for talent attraction regression

Key drivers of talent attraction

The results of panel regression analyses with time and region fixed effects as well as time-lag regression models assessing talent attraction are presented in Annex Table A.2. The results demonstrate the crucial role of digitalisation in attracting talent, along with housing affordability and education, notably foreign students in universities. The identified key drivers of regional talent attraction are presented as the following:

Education: International students in higher education – Foreign talent is more open to locating in a region with an already substantial number of international talent (in this case, students). Research shows that the mobility of foreign students contributes to knowledge sharing and creation, underlining the need for collaborations between universities and industry in attracting talent, especially for low-middle-income countries (Sudibor and Ünlü, 2022[36]). A higher share of foreign student population also entails that the region’s social infrastructure and attitudes are receptive to foreign talent, which can facilitate talent’s arrival and integration into the regional economy and society.
Digitalisation: Access to broadband Internet – Talent thrives in regions with fast Internet speed. Higher Internet speeds facilitate teleworking and conducting everyday digital activities. This is indicative of the spreading trend of remote working culture, which slowly began before the pandemic and accelerated with COVID-19. In fact, the filter most chosen by guests on Airbnb, an online house-sharing platform, is Wi-Fi availability, which attests to the importance of digitalisation in regions (OECD, 2023[37]). In this sense, broadband access can serve both the talent and visitor economy – and increasingly so, as firms and places adopt more remote and hybrid-friendly approaches to work and public service delivery.
Housing: Housing affordability – While high housing prices may indicate regional innovation outputs and consequently talent attraction, housing prices start to be negatively associated with innovation outcomes of regions where the housing prices form a bubble and potentially leads to a crowding-out effect (Lin et al., 2020[38]). Many cities across the globe often implement policies that provide financial support for housing to enhance housing affordability, which is one of the crucial components of benefit packages that regions can offer to attract and retain talent, thereby maintaining a stable pool of the working force for the regional economy (UNDP Kolba Lab, 2020[39]).

Visitors

Tourism is a major catalyst for regional development and a policy competency where local and regional policy makers have a significant role to play. It can be a major employment driver in non-urban areas and bring good quality jobs in everything from gastronomy to agrotourism to the creative and cultural domain. In light of the COVID-19 recovery, many countries have laid down strategies for supporting the resurgence of the tourism industry – strategies that require data to measure and monitor success and to design new policies where appropriate (OECD, 2021[22]). In that regard, using data to understand how regional assets translate into visitor attraction is essential.

In terms of what attracts visitors to places, some traditional drivers stand out. In a regional tourism study in Indonesia, evidence points to the importance of regional policy makers’ prioritisation of the tourism sector as a key determinant of destination attraction (Bire, Conterius and Nasar, 2021[40]). In this Indonesian study, natural attractions outrank cultural attractions while infrastructure is considered more important than entertainment options or tour and travel services. While these factors are likely to vary across country and regional contexts, it illustrates the importance of key public services in facilitating the visitor experience. This is represented in the attractiveness methodology, which asserts the importance of the built environment in drawing not just talent and investors with longer-term orientations but visitors too, who may place a premium on the ease of mobility during their visit. A study of domestic tourism across China’s provinces and cities between 2011 and 2018 adds to this, showing that, as the country’s regions rapidly developed, the economy became a lesser predictor of tourism attraction (Ma, Yang and Zheng, 2022[41]). By 2018, the importance of environmental conditions (temperature, sunshine) and transportation (rail, traffic) emerged as complements to a region’s economic development and in driving visitor numbers as well.

Methods and definition: Visitors

To consider both the volume and occupancy of regional tourism, the dependent variable for visitor attraction is constructed as a ratio of the number of overnight stays in tourist accommodation to the number of tourist accommodation beds. This explanation variable was constructed using 2 indicators, which are the number of overnight stays in tourist accommodation per 1 000 inhabitants and the number of tourist accommodation beds per 1 000 inhabitants. Such an approach enables us to consider both foreign and domestic visitors to the region and capture both tourism flows and capacity.

Summary of regression results

Presented in Figure 3.5 is a summary of regression coefficient results for visitor attraction. The detailed regression summary and tables from the technical analysis can be found in Annex A. As illustrated in the figure, the most attractive regions to visitors have a higher firm birth rate (i.e. entrepreneurship), tend to experience lower levels of air pollution (PM 2.5) and have a higher foreign student share, as was the case for the talent regression.

Figure 3.5. Regression coefficients for visitor attraction regression

Key drivers of visitor attraction

The results of panel regression analyses with time and region fixed effects assessing visitor attraction against the selected variables are presented in Annex Table A A.3. The results attest to the significance of entrepreneurship and innovation, a clean environment and education in boosting domestic and international tourism. The identified key drivers of regional visitor attraction are presented as the following:

Innovation and entrepreneurship: Firm birth rate – Innovation and entrepreneurship help regions flourish with increased inflow and outflow of people, resources, knowledge and ideas. Regions with a burgeoning entrepreneurial economy are likely to become competitive destinations not only for talent and investors who are keen to be involved in emerging new firms but also for the general visitor population who are drawn to the region’s vibrant atmosphere. Indeed, firms are an integral part of tourism activity: across OECD regions, SMEs represent 85% of tourism businesses (OECD, 2022[42]).
Health and environment: Air quality and natural environment – Good air quality – measured by levels of particulate matter in the air – and a well-preserved natural environment also attract visitors. It is therefore important to align tourism goals with sustainable regional development. Indeed, natural amenities are some of the key reasons for tourists to visit and explore regions, and several studies find clean air to be a driver for boosting tourism (Eusébio et al., 2021[43]; OECD, 2016[44]). The impact of air quality, and other environmental conditions that directly affect people’s health, on tourism demand, should be taken into consideration by regional policy makers when developing tourism strategies.
Education: International students in higher education – A higher share of foreign students can contribute to stronger tourism as students invite friends and family members to the region where they are studying and as visitors themselves. In fact, educational tourism is an important part of overall tourism, which ultimately leads to local development of the host regions (Tomasi, Paviotti and Cavicchi, 2020[45]). A clear example of this is the Erasmus+ programme which has mobilised over 10 million Europeans over the last 30 years. The presence of foreign students in regions can also be seen as a form of public diplomacy, strengthening the ties between the host and sending regions (Mulvey, 2020[46]). Ultimately, international students create bonds across places that promote intercultural exchange and encourage repeat tourism.

Exploring connections across attractiveness variables and families of regions

Looking beyond the links described above, we can also observe important connections between attractiveness variables. For example, Figure 3.6 shows a linear relationship between the share of employment in cultural and creative industries and overall tourism activity. While this relationship may appear obvious, it has implications for regional visitor attraction strategies, in that supporting firms and employment growth in this sector can have a residual effect of growing the international tourism market.

Figure 3.6. Correlation between international overnight stays and employment in cultural and creative industries

Attractiveness strategies can also prove synergistic with regional development goals, such as innovation strategies. Time series data generated from the PCT reveal a linear correlation between patent applications per million inhabitants and the share of the foreign-born population among the working-age population (Figure 3.7). Attracting international talent can increase the innovation capacity of a region, as migrants are more likely to be entrepreneurs and indeed help foster FDI flows from their countries of origin (OECD, 2022[33]). In turn, it is essential for regional policy makers to embed a talent attraction dimension within their innovation strategies.

Figure 3.7. PCT applications and the share of the working-age foreign-born correlation

Identifying regional attractiveness linkages across families of regions

Further analysis can delve beyond geographical levels (e.g. TL2, TL3, as discussed) to zoom in on various regional classifications according to specific geographic characteristics that may translate into common regional development challenges and advantages. These could include looking at mountainous regions, border regions, island regions, metropolitan regions, capital regions, etc. However, physical characteristics are not a sufficient categorisation of regions, which can be categorised according to the specific structural challenges they face. Regions in a development trap generally struggle to attract interest because they are neither the lagging regions that traditional regional development policies target nor the metropolitan engines of national growth whose innovation dynamics must be preserved (Diemer et al., 2022[4]). Their development path shows a decline, or the fear of future decline, which can generate discontent among residents. Indeed, the European Commission (2023[48]) recently underlined the need to address their specific concerns. The attractiveness approach proposed here can help make these places more visible and respond to the need to rethink regional development policies to adapt them to this “family” of regions and to learn from the challenges they face and the dynamics that led them towards entrapment. A single strategy, even one as powerful as an investment in infrastructure, is not enough. Just as it is not enough to finance the inputs of innovation (i.e. more researchers, more investment in research). The multi-dimensional framework proposed here is a holistic approach with comparable indicators which can be monitored on a regular basis to uncover and adjust priorities where needed. Precisely, it helps these regions to attract what they miss: investors, talent and visitors.

Focusing on such families of regions and conducting quantitative analyses will help to better understand the different context, advantages and challenges they face when pursuing policies to enhance regional attractiveness. For example, early evidence suggests that European mountain regions may face particular regional attractiveness challenges (Flood, Ryu and Camus, 2022[49]):

Mountain regions have on average 2.68 doctors per 1 000 inhabitants, which trails the European regional average of 3.37 doctors per 1 000 population.
Mountain regions in Europe also tend to have relatively slow Internet download speeds, around 20.32% below their national average. The country-wise comparison of average download speed as a deviation from the national average is shown in Figure 3.8.

Figure 3.8. Mountain vs. non-mountain regions in terms of Internet download speeds

From evidence to action

While evidence on “what works” for regional attractiveness is a promising place to start, regions need to integrate this evidence into policy decisions, thereby not only moving from evidence to action but also to monitoring and evaluating attractiveness policies. As demonstrated in the above analysis, important relationships are identified across the target groups with attractiveness assets (Table 3.1).

Table 3.1. Attractiveness drivers at the regional level
Investors	Talent	Visitors
Higher education institutions	International student share	Entrepreneurship
Internet access and speeds	Broadband Internet access	Environmental quality
Rail performance	Affordable housing	International student share
Access to flights
Note: This is a non-exhaustive list of drivers; further analysis is likely to illustrate important relationships at different geographic levels and across families of regions. Descriptions of access to flights and rail performance are provided in Annex A.
Source: See results in Annex A.

Both regional development strategies and target-group location decisions are intertwined with the global environment – the more uncertain it becomes, the more essential it is to adopt a vision for regional attractiveness and tools to adapt strategies to evolving realities.

As explored in the following chapters, different assets matter to different sub-groups of investors, talent and visitors, which could be more precisely categorised as sustainable investors, young entrepreneurs or digital nomads, to name a few. What this evidence does suggest is that broad trends can be demonstrated about what matters most at the subnational level when analysing a large sample of regions, as was explored in this chapter. It shows that even when controlling for economic conditions, a host of other factors play an important part in the attractiveness equation, especially the physical and digital assets that regional decision makers can work to leverage and improve, even in the near term. Moreover, it offers specific indicators (see Annex Table A B.1) that can be embedded into regional development strategies to measure and monitor attractiveness. Some of the indicators were highlighted here through technical analysis and identified as key drivers to each target group of regional attractiveness. Regions with strong digital performance, better rail and air transport, and top universities located in the region appear to have a comparative advantage in attracting FDI. For talent attraction, affordable housing, fast Internet speeds and the share of foreign students are found to have an impact on their location choice. Finally, attracting visitors is shown to be influenced by an entrepreneurial regional economy, a higher share of international students and a cleaner environment as measured by air pollution levels. Unpacking some of the policies and levers will bolster these assets to make regions more attractive to investors, talent and visitors in the focus of Chapters 4, 5 and 6 respectively.

References

[28] Amendolagine, V., R. Crescenzi and R. Rabellotti (2022), “The geography of acquisitions and greenfield investments: Firm heterogeneity and regional institutional conditions”, LSE Research Online Documents on Economics.

[27] Antonescu, D. (2015), “Empirical analysis of foreign direct investments at NUTS 2 region, in European Union and Romania”, Procedia Economics and Finance, pp. 681-689, https://doi.org/10.1016/S2212-5671(15)00284-1.

[40] Bire, R., A. Conterius and A. Nasar (2021), “Drivers of regional destination competitiveness: A DEMATEL – Fuzzy TOPSIS approach”, Indonesian Journal of Geography, Vol. 53/1.

[29] Crescenzi, R., M. Di Cataldo and M. Giua (2021), “FDI inflows in Europe: Does investment promotion work?”, Journal of International Economics, Vol. 132, p. 103497, https://doi.org/10.1016/j.jinteco.2021.103497.

[7] Criscuolo, C. et al. (2022), “An industrial policy framework for OECD countries: Old debates, new perspectives”, OECD Science, Technology and Industry Policy Papers, No. 127, OECD Publishing, Paris, https://doi.org/10.1787/0002217c-en.

[4] Diemer, A. et al. (2022), “The regional development trap in Europe”, Economic Geography, Vol. 98/5, pp. 487-509, https://doi.org/10.1080/00130095.2022.2080655.

[30] Dorakh, A. (2021), “FDI determinants in Europe and Chinese influence”, Serbian Journal of Management, Vol. 16/1, pp. 21-37, https://doi.org/10.5937/sjm16-22279.

[48] EC (2023), Eighth Report on Economic, Social and Territorial Cohesion, European Commission, https://ec.europa.eu/regional_policy/information-sources/cohesion-report_en.

[32] Esipova, N., A. Pugliese and J. Ray (2019), “More than 750 million worldwide would migrate if they could”, https://news.gallup.com/poll/245255/750-million-worldwide-migrate.aspx.

[25] ESPON (2018), The World in Europe: Global FDI Flows towards Europe, https://www.espon.eu/fdi.

[47] Eurostat (2022), Cultural employment by NUTS2 regions (database), https://ec.europa.eu/eurostat/databrowser/view/CULT_EMP_REG/default/table?lang=en (accessed on 1 December 2022).

[43] Eusébio, C. et al. (2021), “The impact of air quality on tourism: A systematic literature review”, Journal of Tourism Futures, Vol. 7, pp. 111-130, https://doi.org/10.1108/JTF-06-2019-0049.

[17] Flood, M. and É. Laurent (2021), “In well-being we trust: The Nova Scotia Quality of Life Initiative”, in The Well-being Transition, Springer International Publishing, Cham, https://doi.org/10.1007/978-3-030-67860-9_10.

[49] Flood, M., I. Ryu and B. Camus (2022), “They’ll move (to) mountains!”, OECD COGITO: Insights on Entrepreneurship, SMEs, Tourism, Regions and Cities, OECD, Paris, https://oecdcogito.blog/2022/12/09/theyll-move-to-mountains/.

[16] Flood, M. et al. (forthcoming), “Addressing the left-behind problem with regional talent attraction and retention”.

[11] Forbes (2023), “30% Of Americans cite climate change as a motivator to move In 2023”, https://www.forbes.com/home-improvement/features/americans-moving-climate-change/.

[31] Guimarães, P., O. Figueiredo and D. Woodward (2000), “Agglomeration and the location of foreign direct investment in Portugal”, Journal of Urban Economics, Vol. 47/1, pp. 115-135, https://doi.org/10.1006/juec.1999.2138.

[12] IDMC (2020), Global Database for Europe, Internal Displacement Monitoring Centre.

[5] IMF (2023), Confronting Fragmentation Where It Matters Most: Trade, Debt, and Climate Action, International Monetary Fund.

[13] Khanna, P. (2022), Move: Where People Are Going for a Better Future, Simon & Schuster, https://www.paragkhanna.com/move/.

[6] Kiel Institute (2023), “One year of support to Ukraine: US leads, EU follows”, https://www.ifw-kiel.de/publications/media-information/2023/one-year-of-support-to-ukraine-us-leads-eu-follows/.

[38] Lin, X. et al. (2020), “Housing price, talent movement, and innovation output: Evidence from Chinese cities”, Review of Development Economics, Vol. 25/1, pp. 76-103, https://doi.org/10.1111/rode.12705.

[41] Ma, X., Z. Yang and J. Zheng (2022), “Analysis of spatial patterns and driving factors of provincial tourism demand in China”, Scientific Reports.

[35] Miguélez, E. and R. Moreno (2013), “What attracts knowledge workers? The role of space and social networks”, Journal of Regional Science, Vol. 54/1, pp. 33-60, https://doi.org/10.1111/jors.12069.

[26] Mukherjee, A. (2011), Regional inequality in foreign direct investment flows to India: The problem and the prospects, pp. 99-127.

[46] Mulvey, B. (2020), “International higher education and public diplomacy: A case study of Ugandan graduates from Chinese universities”, Higher Education Policy, Vol. 33, pp. 459–477, https://doi.org/10.1057/s41307-019-00174-w.

[37] OECD (2023), “Extract from OECD dialogue on the Future of Regions: Insights from Airbnb on the evolving attractiveness of non-metropolitan territories”, OECD, Paris.

[34] OECD (2023), Talent Attractiveness Indicators, OECD, Paris, https://www.oecd.org/migration/talent-attractiveness/.

[10] OECD (2023), “When the white gold thaws”, COGITO: Insights on Entrepreneurship, SMEs, Tourism, Regions and Cities, OECD, Paris, https://oecdcogito.blog/2023/02/02/when-the-white-gold-thaws/.

[21] OECD (2022), 2022 Synthesis Report World Observatory on Subnational Government Finance and Investment, OECD Publishing, Paris, https://doi.org/10.1787/b80a8cdb-en.

[2] OECD (2022), L’internationalisation et l’attractivité des régions françaises, OECD Publishing, Paris, https://doi.org/10.1787/6f04564a-fr.

[3] OECD (2022), “Measuring the attractiveness of regions”, OECD Regional Development Papers, No. 36, OECD Publishing, Paris, https://doi.org/10.1787/fbe44086-en.

[18] OECD (2022), OECD Regions and Cities at a Glance 2022, OECD Publishing, Paris, https://doi.org/10.1787/14108660-en.

[42] OECD (2022), OECD Tourism Trends and Policies 2022, OECD Publishing, Paris, https://doi.org/10.1787/a8dd3019-en.

[33] OECD (2022), The Contribution of Migration to Regional Development, OECD Regional Development Studies, OECD Publishing, Paris, https://doi.org/10.1787/57046df4-en.

[24] OECD (2022), “The geography of foreign investment in OECD member countries: How investment promotion agencies support regional development”, OECD Business and Finance Policy Papers, No. 20, OECD Publishing, Paris, https://doi.org/10.1787/1f293a25-en.

[9] OECD (2022), “The implications for OECD regions of the war in Ukraine: An initial analysis”, OECD Regional Development Papers, No. 34, OECD Publishing, Paris, https://doi.org/10.1787/8e0fcb83-en.

[22] OECD (2021), “Managing tourism development for sustainable and inclusive recovery”, OECD Tourism Papers, No. 2021/01, OECD Publishing, Paris, https://doi.org/10.1787/b062f603-en.

[1] OECD (2021), “Regions and globalisation: An original approach to regional internationalisation and its application to the case of France”, OECD Regional Development Papers, No. 20, OECD Publishing, Paris, https://doi.org/10.1787/75dae685-en.

[19] OECD (2020), How’s Life? 2020: Measuring Well-being, OECD Publishing, Paris, https://doi.org/10.1787/9870c393-en.

[15] OECD (2020), Rural Well-being: Geography of Opportunities, OECD Rural Studies, OECD Publishing, Paris, https://doi.org/10.1787/d25cef80-en.

[23] OECD (2019), FDI Qualities Indicators, OECD, Paris, https://www.oecd.org/fr/investissement/fdi-qualities-indicators.htm.

[44] OECD (2016), The Economic Consequences of Outdoor Air Pollution, OECD Publishing, Paris, https://doi.org/10.1787/9789264257474-en.

[50] Sanguinet, E. et al. (2021), “The subnational supply chain and the COVID‐19 pandemic: Short‐term impacts on the Brazilian regional economy”, Regional Science Policy and Practice, Vol. 13/S1, pp. 158-186, https://doi.org/10.1111/rsp3.12442.

[36] Sudibor, O. and H. Ünlü (2022), “The attraction of talent through international student mobility patterns and triple helix system: Evidence from OECD countries for triple helix system”, Studies in Economics and Econometrics, Vol. 46/2, pp. 149-167, https://doi.org/10.1080/03796205.2022.2127422.

[14] Tang, R. and A. Beer (2022), “Regional innovation and the retention of foreign direct investment: A place-based approach”, Regional Studies, Vol. 56/10, pp. 1757-1770, https://doi.org/10.1080/00343404.2021.2006173.

[45] Tomasi, S., G. Paviotti and A. Cavicchi (2020), “Educational tourism and local development: The role of universities”, Sustainability, Vol. 12/17, p. 6766, https://doi.org/10.3390/su12176766.

[39] UNDP Kolba Lab (2020), “Talent attraction and retention”, United Nations Development Programme.

[8] Wright, R. and C. Wu (2022), “The uneven short-run effects of the COVID-19 pandemic on foreign direct investment”, Journal of Risk and Financial Management, Vol. 15/10, p. 468, https://doi.org/10.3390/jrfm15100468.

[20] Yang, Z. and Y. Pan (2020), “Human capital, housing prices, and regional economic development: Will “vying for talent” through policy succeed?”, Cities, Vol. 98.

╳

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

https://doi.org/10.1787/a9448db4-en

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at https://www.oecd.org/termsandconditions.