This chapter provides key statistics on demographic change – referring to population decline and ageing – in OECD countries at different geographical scales. It analyses the drivers of demographic change by distinguishing between natural change and net migration. It also explores different classifications or demographic change and approaches to define shrinking places.
Shrinking Smartly and Sustainably
Strategies for Action
Abstract
Defining depopulation is not a trivial task, as it refers to a gradual and substantial population decrease over a significant period. Thus, it will be warranted to identify a threshold of average annual population decline over a timeframe sufficiently long to account for a persistent decline. It is also crucial to note that a threshold on annual population change for municipalities and local areas may not translate well to a broader geographical level such as TL3 regions, where population changes tend to be smoother than at a more localised level. While efforts have been made to define these various thresholds (European Commission, 2021[1]), this paper does not aim at defining them explicitly. These considerations will be addressed in a forthcoming technical paper.
“Shrinking smartly” refers to the strategic approach local governments take to adapt to and manage population decline in a way that ensures quality of life, service delivery, economic stability, and social cohesion. It involves resizing infrastructures, adapting services and housing to meet the population needs in a cost-effective manner. Bozhidar (2022[2]) developed a benchmarking strategy providing a hands-on methodology that integrates demographic shifts and sustainability within various planning contexts, offering a practical framework for effective urban planning.
“Shrinking sustainably” consists of addressing population decline while ensuring environmental sustainability. The environmental impacts of depopulation can be both positive and negative. On the positive side, depopulation implies reduced building energy consumption, lower demand on resources, lower levels of waste production, leading to lower greenhouse gas emissions and environmental pressures in these sectors. However, negative impacts may include the neglect of infrastructure and buildings, which could lead to pollution. Additionally, depopulation can lead to underutilised public transport systems and increase the reliance on personal vehicles leading to higher greenhouse gas emissions in the transport sector. Shrinking sustainably consequently aims for long-term resilience by reducing environmental impacts.
The analysis in this chapter examines demographic trends at the level of small regions (TL3), municipalities or local areas, and Functional Urban Areas (FUAs). This analysis also leverages different territorial typologies:
The TL3 regional typology, which classifies TL3 regions based on their accessibility to cities (large metropolitan regions, midsize metropolitan regions, regions near a midsize/large FUA, regions near a small FUA, and remote regions) (Fadic et al., 2019[3]).
The degree of urbanisation (DEGURBA), which classifies municipalities and local areas as cities, towns and suburbs or rural areas based on a combination of geographical contiguity and population density (UN-Habitat, 2021[4]).
Many OECD countries are now experiencing population stagnation or decline, alongside ageing. In broad lines, the population in OECD countries grew in the last 20 years to reach 1.4 billion people in 2022 – up by 15% since 2000. However, since 2001, seven OECD countries have seen their populations decline. Lithuania and Latvia experienced the most significant drops, each losing around one-fifth (20%) of their populations. Hungary, Estonia, Greece, Japan, and Poland, also saw decreases (Figure 2.1). In Latvia and Lithuania, this decline resulted from both a negative natural change, with more deaths than births, and a negative net migration, where more people left than arrived. Conversely, Luxembourg and Israel saw the largest population increases, growing up to 50% from 2001 to 2022, due mostly to substantial positive net migration in the case of Luxembourg and natural change in the case of Israel.
Population change and components of change (natural change and apparent migration) in OECD countries, 2001-22
Note: Apparent migration is defined as the difference between population change and natural change. It approximates the difference between the number of people entering a country and the number of persons leaving it. Natural change refers to the difference between the number of births and the number of deaths.
Source: OECD Regional Database (OECD, 2023[5]).
By 2060, the population is expected to further decline in 14 OECD countries. Greece, Japan, Latvia and Lithuania are expected to lose more than one-fifth of their population between 2022 and 2060 (Figure 2.2). Other countries, including Czechia, Finland, Hungary, Italy, Korea, Poland, Portugal, Slovak Republic, Estonia, and Slovenia will also experience population decline, although at lower rates.
Projected population change in OECD countries, 2022 to 2060
Source: OECD Population Projection Statistics.
While metropolitan regions experienced the strongest population increase over the last two decades (+13%), other regions experienced either smaller increases or lost population (+5%). One third (29%) OECD small regions (TL3) saw their population decline and more than half (57%) of these regions were regions far from a midsize or large FUA.
Population trends reveal stark disparities across diverse OECD country groups and types of regions (Figure 2.3). Both regions far from and regions near a midsize or large FUA in Korea and Japan have been particularly impacted by population decline, losing 12% and 9% of their population, respectively. This decline can be explained by both a negative natural change and a negative net migration.
Population change and components of change in OECD small regions by access to city typology and by macro-regions, 1st January 2001 to 31st December 2020.
Note: Apparent migration is defined as the difference between population change and natural change. Due to data availability for number of births and deaths, 9 OECD countries are not covered in this chart: Ireland, Denmark, Türkiye, Slovenia, Chile, Colombia, Costa Rica, Israel, and the United States.
Source: OECD Regional Database (OECD, 2023[5]).
In the last twenty years, an increasing share of the OECD population has moved to cities (OECD, 2020[6]), (OECD, 2023[7]), (OECD, 2022[8]). In 2020, about half (49%) of the OECD population lived in cities, while three out of ten (28%) lived in in towns and semi-dense areas and two out of ten (23%) in rural areas (see Annex A for territorial classification). Over 2000-20, the population in cities rose by 24%, while it increased by 8% in towns and semi-dense areas, and by 6% in rural areas (Figure 2.4). In many OECD countries, the population living in rural areas and towns and semi-dense areas had already started to stagnate or even to decline during this period. For example, in 16 OECD countries, the population living in rural areas decreased between 2000 and 2020 (Figure 2.5). This trend is likely to continue over the next decade. Projections show a 5% increase in the urban population by 2030 compared to 2020. Conversely, the population in towns and semi-dense areas is expected to decrease slightly by 1%, and in rural areas, it is projected to grow by 1%.
Population trend from 2000 to 2030, OECD by degree of urbanisation.
Source: Global Human Settlement Population layer GHS-POP R2023A (Schiavina et al., 2023[9])
Population change over 2000-20 by degree of urbanisation in OECD countries.
Source: Global Human Settlement Population layer GHS-POP R2023A (Schiavina et al., 2023[9])
On average, the population in OECD FUAs grew by 15% over the period 2001-21, outpacing the growth rate in the rest of their respective countries (6%). However, not all types of cities have witnessed the same population growth rate in recent years. More than one in five metropolitan areas across OECD countries even experienced population decline. The OECD has seen its most significant population increases in larger metropolitan areas. Over 2000-20, FUAs with over 1 million inhabitants experienced an 18% population rise. In comparison, smaller FUAs with fewer than 250 000 inhabitants and those with 250 000 to 1 million inhabitants saw more modest growth rates of 7% and 14%, respectively. This trend is expected to continue, though at a slower rate, over the next decade. The population in FUAs with 1 to 5 million inhabitants is projected to grow by 4%. Meanwhile, in areas with less than 250 000 inhabitants, the population is expected to peak in 2025 and then start to decline (Figure 2.6).
Population trend from 2000 to 2030, OECD FUAs by size
Source: Global Human Settlement Population layer GHS-POP R2023A (Schiavina et al., 2023[9])
International migrants concentrate in cities and certain regions
While migrants represent an important and growing share of the OECD population, they tend to concentrate in cities and regions that are economically thriving. In 2022, migrants constituted 14% of the population in OECD countries, with figures ranging from 48% in Luxembourg to less than 3% in Japan, Mexico, or Poland. However, they settle unevenly within countries. Migrants often choose to live in cities or regions with higher wages and better job opportunities (OECD, 2022[10]). For instance, eight out of ten migrants reside in metropolitan regions, compared to seven out of ten native-born individuals. In contrast, only 15% of migrants live in non-metropolitan regions, while nearly 26% of the native-born population does (Figure 2.7). This pattern indicates that migrants tend to move to economically thriving areas, suggesting that it is unlikely that international migration can help reverse population decline in less populated areas. However, some policy makers have leveraged immigration, including refugee resettlement, as a strategy to mitigate demographic challenges. For example, the small German city of Altena welcomed refugees to stabilise its population and help fill labour shortages (OECD, 2018[11]). Similarly, Canada’s Atlantic Immigration Program, which is unique to the Atlantic region, provides a pathway to permanent residence for skilled foreign workers to support economic growth and demographic sustainability in ageing and declining territories (OECD, 2022[12]).
Concentration of native (left)- and foreign-born (right) by local access to functional urban areas regional typology, 2010, 2015, and 2020
Note: Concentration of native- (left) and foreign-born (right) by local access to functional urban areas regional typology over the native or foreign population. Data refer to 2010, 2015, and 2022. Only countries with available data in all three years are considered (Austria, Belgium, Denmark, Finland, Germany, Italy, Japan, Netherlands, Norway, Portugal, South Korea, Spain, Sweden, Switzerland, the United Kingdom, and the United States). For Italy, Korea, and Spain, the data refer to the foreign population rather than the foreign-born population.
Over the past decades, many OECD countries have experienced a rapid ageing of their populations, posing substantial challenges to their economic structures and their fiscal systems. Ageing is a direct consequence of a high life expectancy combined with a low fertility rate. Life expectancy has increased in most countries over the past decades, but the COVID-19 pandemic reversed this trend (OECD, 2023[13]). Across OECD countries, Japan has the highest proportion of elderly people, where one in three (29%) persons is aged 65 and over. Italy, Portugal, Finland, and Greece follow, each with around 23-24% (Figure 2.8). This trend is expected to persist in the coming decades. By 2060, close to 44% of the population in Korea will be older than 65 years old, followed by Japan (38%), Lithuania (35%) and Greece (35%) (Figure 2.9). Moreover, Korea, Japan, Italy, and Portugal face the challenge of a shrinking young population, essential for balancing the pressures of an ageing demographic. In 2022, less than 13% of the population in these four countries was under 14 years old.
Ageing will also impact labour markets and pension systems, as a smaller working age population will have to support an increasingly larger elderly population. In 2022, Japan had the highest old-age to working-age population ratio at 49%, followed closely by Italy, Finland, and Portugal, each with a ratio of 37%. By 2060, this ratio is projected to soar to 90% in Korea, 74% in Japan and 66% in Greece and Lithuania posing important challenges for the sustainability of the pension systems. However, not all OECD members currently face these challenges. For example, Mexico has the smallest elderly population in the OECD at 8%, and one of the highest percentages of young people at 25%, second only to Israel (Figure 2.8).
Share of population by age group (0-14, 15-64 and 65+ years old) ordered by the share of population older than 65, OECD countries, 2022.
Source: OECD Population Statistics.
Projected share of population by age group (0-14, 15-64 and 65+ years old) ordered by the share of population older than 65, OECD countries, 2060.
Source: OECD Population Projection Statistics
In 2022, OECD metropolitan regions had a slightly lower share of elderly population (18%) than other types of regions. Yet, this figure masks notable differences among country groups. For instance, in Japan and Korea, regions far from a midsize or large FUA had a significantly higher elderly population, at 35%. This is ten percentage points higher than in metropolitan regions. On the contrary, in OECD North and South America, regions far from a midsize or large FUA showed a higher percentage of young people compared to other regions (Figure 2.10).
Share of population by age group (0-14, 15-64 and 65+ years old), OECD small regions by access to city typology and by macro-regions, 2022.
Note: Costa Rica and Israel are not covered in this chart. The figures refer to 2022 for all countries, except for the United Kingdom (2021) and the United States (2021).
Source: OECD Regional Database (OECD, 2023[5]).
Almost all OECD FUAs have seen the share of people aged 65 years and more increase over 2006-18. In 2006, the median metropolitan area within OECD countries had 14.5 older adults for every 100 inhabitants, compared to 17.6 in 2018. On average, across OECD countries, FUAs that lost population over 2008-18 experienced an increase of 4% in the share of population aged 65 years and over, compared to an increase of 2.7% in metropolitan areas with a growing population. Across G7 countries, the share of older adults increased by 4.2% in metropolitan areas that lost population, compared to 2.9% in metropolitan areas with a growing population. The combination of population decline and ageing gives rise to compounding and multidimensional challenges for shrinking cities, which face the risk of entering a downward spiral in which demographic change undermines both agglomeration benefits and quality of life. For instance, population decline weakens the local tax base and reduces revenues for public administrations, making it more difficult to maintain essential infrastructure and public services. At the same time, population ageing further intensifies these challenges by increasing demand for healthcare, social services, and public transport, contributing to a growing mismatch between service needs and available resources. Lower revenues also hinder the maintenance of public spaces and green areas, reducing overall quality of life and further diminishing the attractiveness of depopulating cities (Burgalassi and Matsumoto, 2024[14]).
Demographic change brings also differentiated challenges within cities, and can exacerbate several forms of spatial clustering of different groups of people based on certain demographic or social characteristics (e.g. age, income, ethnic background) in neighbourhoods. While spatial clustering might provide benefits for social and demographic groups with shared needs and preferences, it might also amplify inequalities within cities (OECD, 2018[15]). A dynamic of population decline can begin in specific neighbourhoods following economic or social shocks such as the closure of a relevant employer. This could result in out-migration from the neighbourhood and subsequent urban decay, and a potential in-migration of less affluent or more marginalised households (Haase et al., 2014[16]). Similarly, ageing can affect differently parts of a city, with different challenges. For instance, within FUAs, ageing tends to be faster in suburban areas than inner cities (Burgalassi and Matsumoto, 2024[14]). Between 2008 and 2018, on average, the share of people aged 65 years or more over total population increased faster in commuting zones than in inner cities in all G7 countries (Figure 2.11).
Growth in the share of people aged 65 years or more over the total population in cities and commuting zones, G7 countries, 2008-2018, in percentage points
Note: Share of people aged 65 years and more. Average values (unweighted averages).
Source: OECD (2023), OECD City Statistics (database), OECD Publishing, Paris, http://dx.doi.org/10.1787/region-data-en
Spatial segregation isolates social groups, curtailing interaction and heightening the risk of “social exclusion” – the inability to engage in economic, social, and cultural life (Duffy, 1995[17]), it undermines community development, and weakens social cohesion. It can also entrench prejudice and discrimination, leading to stigma that exacerbates neighbourhood decline, reduces social cohesion and impede community-building (European Commission, 2019[18]). Affecting settlement trends of households, spatial segregation amplifies housing quality disparities: affluent areas boast well-maintained homes, whereas shrinking neighbourhoods can face prevalent dilapidation. This disparity skews affordability, inflating prices in wealthier areas and entrenching a cycle that deepens community divides. Particularly for minorities, like migrants, it poses a significant barrier to integration. Furthermore, the challenges related to spatial segregation are beyond housing since it amplifies disparities in access to services and urban activities.
Age segregation can challenge cities, too, since it reduces opportunities for cross-generational interaction and can lead to social isolation of groups, especially for aged people, with increasing risks of loneliness and challenges for mental health (Burgalassi and Matsumoto, 2024[14]). This is challenging for both young and aged populations: for young people, higher isolation risks severe long-term consequences, such as dropping out of school (OECD, 2021[19]), while for aged populations, social isolation is compounded with deteriorating physical health, the shift from employment to retirement, and disruptive events like losing family and friends. In the European Union, it is estimated that half of the population aged 60 and older is at risk of depression (OECD/European Union, 2022[20]). The challenge is particularly present in the suburban areas of large metropolitan areas in OECD countries. In many countries, “new towns” were constructed during the 1960s and 1970s in suburban areas accommodated a massive influx of population into metropolitan areas. The people who moved to these new town settlements during their working years are now 65 years and older and constitute the majority of residents in many of these new towns (OECD, 2015[21]). While suburban areas are ageing fast (see Figure 2.11 above), they are likely not to be designed to be “age-friendly”, in particular with reference to housing, infrastructure, and public spaces to support the independent living of older adults, or “ageing-in-place”, as well as to promote social cohesion and fight social isolation (Burgalassi and Matsumoto, 2024[14]). This calls for actions embracing both urban design and community-building initiatives targeted at elderly populations to foster a sense of belonging and community engagement.
Creating a classification of small regions, municipalities and local areas based on their population trend and characteristics helps to better understand the geography of demographic change. This approach also makes it easier to plan and provide services and infrastructures more efficiently and to create policies that meet the specific needs of people living in those areas.
However, creating such a classification is a complex task, as it must account for various local factors influencing demographic change. These include:
1. Population change. “Depopulating regions” are characterised by a sustained decrease in population, which can result from a combination of factors: a lower birth rate, a higher mortality, and a negative net migration. To measure a sustained decrease in population, it is therefore necessary to look at the average annual population change over an extended period. This paper does not define explicitly how many years and which average population decline threshold are necessary to define this sustained decline in population. These considerations will be explored in detail in the forthcoming technical paper. The direction and intensity of population change, as well as the sequence of annual population change serve as crucial indicators to characterise demographic shifts.
2. Natural change. Places with a lower birth rate and a high mortality related to an ageing population are prone to population decline. Understanding these drivers by analysing the number of births and number of deaths is central to characterise demographic change.
3. Migration: Places with a consistent negative net migration, i.e. where more people are leaving than arriving, are also prone to depopulation, especially when coupled with a negative natural change. This trend is in general indicative of economic challenges, lack of opportunities and other adverse conditions. Net migration can be measured by both internal mobility, which involves people moving between regions within the same country, as well as international migration, which corresponds to people moving between a region and another country.
Some methodological considerations are important:
Temporal resolution. Demographic change should be analysed over an extended period to differentiate between short-term fluctuations and sustained population change. This demographic change can be characterised based on either historical data or modelled projections. This analysis only considers historical data.
Spatial resolution. Changes in population differ widely from one place to another. When looking at population change at the country or large region levels, important differences across areas may be hidden. This requires digging deeper into small regions, municipalities, or local areas to better understand these changes.
Studies such as (González-Leonardo, Newsham and Rowe, 2023[22]) and (Newsham and Rowe, 2023[23]) propose to classify respectively Spanish municipalities and European small regions based on their sequence of average annual population change using clustering methods. Looking at the average annual population change offers a more nuanced understanding of demographic trends compared to overall population change, as it allows to identify sustained demographic shifts, such as continuous population decline. However, this classification doesn't differentiate whether the decline is due to higher mortality or net migration losses. Population change, natural change and net migration may indeed show very different patterns across places, as highlighted for the case of Italian municipalities in Box 2.1.
Figure 2.12 Panel A, Figure 2.12 Panel B, and Figure 2.13 Panel A show respectively the population growth rate, natural change rate and net migration rate over the last two decades (from 1 January 2002 to 31st December 2021) in Italian municipalities. During this period, 55% of municipalities experienced population decline. Most municipalities (77%) witnessed a negative natural change, while 29% had a negative net migration rate. Migration patterns reveal a North-South gradient, indicating a movement of people from the South to the North of Italy, with a concentration of inflow in metropolitan areas. Figure 2.13 Panel B shows the direction of change (growth and decline) and the main component of change (positive or negative natural change or net migration). A quarter of Italian municipalities, mostly located in the South, experienced both negative natural change and negative net migration, signalling a robust declining pattern. In 19% of municipalities, both the natural change and net migration were positive, with these municipalities predominantly located in the North and on the outskirts of metropolitan areas such as Rome or Milan. Most municipalities (56%) showed opposing directions for natural change and net migration. A demographic classification of Italian municipalities would need to reflect these different trends.
Source: OECD Municipal and Local Area Database
Source: OECD Municipal and Local Area Database
References
[2] Bozhidar, I. (2022), Shrinking Smart: Identifying policy and planning alternatives for addressing urban shrinkage and population decline through cross-national comparative research on approaches from Spain, Germany and the Netherlands, Technische Universität Kaiserslautern, https://doi.org/10.26204/KLUEDO/6897.
[14] Burgalassi, D. and T. Matsumoto (2024), “Demographic change in cities: Trends, challenges and insights from G7 economies”, OECD Regional Development Papers, No. 67, OECD Publishing, Paris, https://doi.org/10.1787/f2aec988-en.
[17] Duffy, K. (1995), Social Exclusion and Human Dignity in Europe, Counci of Europe.
[1] European Commission (2021), “A long-term Vision for the EU’s Rural Areas - Towards stronger, connected, resilient and prosperous rural areas by 2040”, Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52021SC0166.
[18] European Commission (2019), Social segregation, https://publications.jrc.ec.europa.eu/repository/handle/JRC116711.
[3] Fadic, M. et al. (2019), “Classifying small (TL3) regions based on metropolitan population, low density and remoteness”, OECD Regional Development Working Papers, No. 2019/06, OECD Publishing, Paris, https://doi.org/10.1787/b902cc00-en.
[22] González-Leonardo, M., N. Newsham and F. Rowe (2023), “Understanding Population Decline Trajectories in Spain using Sequence Analysis”, Geographical Analysis, Vol. 55/4, pp. 495-516, https://doi.org/10.1111/gean.12357.
[16] Haase, A. et al. (2014), “Conceptualizing Urban Shrinkage”, Environment and Planning A: Economy and Space, Vol. 46/7, pp. 1519-1534, https://doi.org/10.1068/a46269.
[23] Newsham, N. and F. Rowe (2023), “Understanding trajectories of population decline across rural and urban Europe: A sequence analysis”, Population, Space and Place, Vol. 29/3, p. e2630, https://doi.org/10.1002/psp.2630.
[13] OECD (2023), Health at a Glance 2023: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/7a7afb35-en.
[5] OECD (2023), “OECD Regional Database”, OECD, https://stats.oecd.org/Index.aspx?DataSetCode=REGION_DEMOGR.
[7] OECD (2023), OECD Regional Outlook 2023: The Longstanding Geography of Inequalities, OECD Publishing, Paris, https://doi.org/10.1787/92cd40a0-en.
[8] OECD (2022), Addressing territorial disparities in future infrastructure needs in the wake of the COVID-19 crisis : A G20 perspective, OECD Publishing, Paris, https://doi.org/10.1787/e246f50f-en.
[12] OECD (2022), “Multi-level governance for migrant integration: Policy instruments from Austria, Canada, France, Germany and Italy”, OECD Regional Development Papers, No. 24, OECD Publishing, Paris, https://doi.org/10.1787/efeeda9d-en.
[10] OECD (2022), The Contribution of Migration to Regional Development, OECD Regional Development Studies, OECD Publishing, Paris, https://doi.org/10.1787/57046df4-en.
[19] OECD (2021), Mental health and young people, https://www.oecd.org/coronavirus/en/data-insights/mental-health-and-young-people.
[6] OECD (2020), Rural Well-being : Geography of Opportunities, OECD Rural Studies, OECD Publishing, Paris, https://doi.org/10.1787/d25cef80-en.
[15] OECD (2018), Divided Cities: Understanding Intra-urban Inequalities, OECD Publishing, Paris, https://doi.org/10.1787/9789264300385-en.
[11] OECD (2018), Working Together for Local Integration of Migrants and Refugees in Altena, OECD Regional Development Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264299320-en.
[21] OECD (2015), Ageing in Cities, OECD Publishing, Paris, https://doi.org/10.1787/9789264231160-en.
[20] OECD/European Union (2022), Health at a Glance: Europe 2022: State of Health in the EU Cycle, OECD Publishing, Paris, https://doi.org/10.1787/507433b0-en.
[9] Schiavina, M. et al. (2023), GHS-POP R2023A - GHS population grid multitemporal (1975-2030).
[4] UN-Habitat (2021), Applying the Degree of Urbanisation — A methodological manual to define cities, towns and rural areas for international comparisons, UN-Habitat | European Union | FAO | OECD | The World Bank, https://unhabitat.org/applying-the-degree-of-urbanisation-a-methodological-manual-to-define-cities-towns-and-rural-areas.