This chapter examines population trends, innovation performance, and labour force dynamics across Japan’s TL3 regions, with a particular focus on rural areas where demographic decline is most acute. It underscores the importance of fostering not only the diffusion of advanced technologies but also non-technological and public sector innovation. It does so through three lenses: rural places and innovation, the structure of the economy, and the dynamics of the labour supply.
Enhancing Rural Innovation in Japan
Abstract
Japan is one of the leading high-tech innovators among OECD countries. In terms of high-tech innovation and patenting activities, from the period of 2016 to 2020Japan had a significantly higher number of patent applications (approximately 222 000) than all other OECD countries except the US. The average patent intensity or rate of patent applications per 1 000 individuals is the highest among OECD countries in Japan, at 0.35 patents per 1 000 individuals between the period of 2016 and 2020. This is double the average in OECD countries (0.14) and is also higher than that observed in other high-patenting countries such as Norway (0.14) and the United States (0.16). Furthermore, high-tech innovation intensity has increased in Japan in the decade leading up to 2020, as it has in 24 of 36 OECD countries with available statistics. From 2011 to 2015, the patenting intensity in Japan was 0.30 per 1 000 individuals, while it increased in the period from 2016 to 2020 to 0.35 patents per 1 000 individuals.
Nevertheless, high-tech patenting activities are concentrated in metropolitan regions of Japan. The difference in the average patent application intensity between Japan's metropolitan and non-metropolitan regions is 0.32 patents per 1,000 individuals (in levels) over the 2016-2020 period. This represents the greatest disparity observed in all countries for which data were available. Furthermore, the gap has increased from the period 2011-2015 by 0.05 patents per 1 000 individuals (0.27 patents per 1 000 individuals in the 2011-2015 period).
Other measures of a healthy, vibrant and innovative economy that go beyond science and technology statistics suggest that more can be done to support rural innovation and development. For instance, although the rate of entrepreneurship is declining in Japan, both in metropolitan and non-metropolitan regions, they are higher in non-metropolitan regions. In 2007, the rate of entrepreneurship was 89.5 per 1 000 workers in metropolitan regions, and slightly higher at 90.1 per 1 000 workers in non-metropolitan regions. The numbers in 2018 declined to 71.5 entrepreneurs per 1 000 workers in metropolitan regions and 74.8 entrepreneurs per 1 000 workers in non-metropolitan regions. Other high-innovation countries with data available on entrepreneurship have either increased the number of entrepreneurs, or stayed the same, around the same time. For instance, in Canada between 2007 and 2018, there was a 17% increase in the number of (non-Indigenous) businesses, while in the US from 2013 to 2016 the growth rate of the number of firms stayed constant.
Over the past two decades from 2000 to 2021, the population in Japan has declined, primarily led by declines in non-metropolitan areas. The compound annual growth rate declined by 0.32%. Japan was one of 9 OECD countries to have a decline in the overall population during this period. In comparison, the average weighted compound annual population growth in OECD countries was 0.45%. However, the population of Japan did not decline in all regions, rather the population declined only outside of large metropolitan regions. There was a 0.09% growth in large metropolitan regions of Japan and substantial declines in all other regions. For example, in midsized metropolitan regions observed a 0.30% decline in Japan. In comparison, similar OECD midsized metropolitan regions observed a 0.55% increase over the same period. In comparison, on average in OECD countries, non-metropolitan regions near a midsized or large functional urban areas saw a 0.36% annual increase in population over the same period. Lastly, non-metropolitan regions near a small functional urban area saw a decline of 0.61% in Japan, while in comparison on average in OECD countries, non-metropolitan regions near a small function urban area saw an increase of 0.38%.
Rural areas with low population density in Japan, are often considered to be primarily agricultural. However, agriculture is not the largest sector of activity in rural areas. The largest sector for employment in rural Japan is in the trade and services sector (38%), followed by public sector (25%) and the manufacturing sector (19%). This largely follows similar observations in other OECD countries. While there is a higher share of agriculture in rural areas than in urban areas, the share is still small. In 2020, 9% of employment in rural areas was in the agricultural sector, in comparison to 1% in urban areas.
The structure of employment in Japan and rural areas of Japan is undergoing a period of transition. Between 2010 and 2020, the rural economy lost 1.8 million jobs, while urban areas lost only 200,000 jobs. The majority of job losses were concentrated in the trade and service industries in urban areas, while the agricultural sector in rural areas experienced the greatest decline. The only areas of employment growth were in the public works and services sector, which expanded by 3.6% in urban areas and by 2.8% in rural areas.
Limited access to resources that can enable innovation such as human and financial capital and digital infrastructure can hinder firm-based innovation in rural areas and this is observed in Japan.
In Japan, inward foreign direct investment (FDI) is primarily concentrated in metropolitan areas. Between 2003 and 2022, 10% of all FDI investments were directed towards non-metropolitan firms in Japan, which was lower than the OECD average of 31%. In comparison, other high-innovation countries saw relatively larger shares of total FDI investment in non-metropolitan regions. Furthermore, the share of FDI has increased over time in Japan. From 2003 to 2022 the share of total FDI investment in Japan increased from 0.1% in 2003 to 5% in 2022 in Japan. The highest point was in 2006 and 2020, when the share of FDI to non-metropolitan regions was close to 20% of total FDI. Furthermore, the proportion of foreign direct investment (FDI) in non-metropolitan regions allocated to innovation activities through research and development is low in Japan. In OECD countries, 15% of all FDI in R&D activities was directed towards non-metropolitan regions between the years of 2003 and 2022. In Japan, the share was close to 2%.
Access to skills in non-urban areas can hinder innovation. For rural areas with ageing populations, this comes as a larger challenge, as there is a higher rate of skills decline among older populations and difficulty with digital skills among older workers can compound challenges for innovation in rural areas. The available data on reading scores find that, on average, in the OECD there is a 14-point difference in reading scores between cities and non-cities in OECD countries. In Japan, this difference is larger, at a 22-point difference between cities and non-cities in OECD countries. However, much of this difference is due to socio-economic characteristics, and as such, is not only addressed through education and skills policies.
Non-metropolitan regions of Japan are not as well digitally connected as metropolitan regions. Internet speeds are 21 percentage points higher in metropolitan regions than in regions far from a metropolitan area, and similarly 19 percentage points higher in metropolitan regions near a metropolitan area. However, this dispersion is not as large as on average in the OECD. On average in OECD countries, there are 25 percentage points higher speeds in metropolitan regions as compared to non-metropolitan regions near a metropolitan area, and 20 percentage points higher speeds in non-metropolitan regions far from a metropolitan area.
It is crucial to acknowledge that firms and communities are confronted with difficulties associated with depopulation, ageing, and a decline in the national labour force. These challenges impede the conditions under which innovation and innovation adoption can occur in rural areas.
The number of workers in employment declined from 2010 to 2020. From 2010 to 2020, the overall decline in the number of workers in Japan was 3.3%, or 0.3% in terms of compound annual growth rates (CAGR). Nevertheless, the decline was more pronounced in rural areas, where the total number of employed individuals fell by 8.9% (aggregate change) or 0.9% annually (CAGR) between 2010 and 2020, in comparison to a decline of approximately 0.49% (aggregate change) or 0.05% annually (CAGR) in urban areas. In aggregate, the proportion of the total workforce in rural and urban areas remained relatively stable, despite a 2% decline in the proportion of the total workforce in rural Japan.
Young workers as a share of total workers in rural areas of Japan is low (24 younger workers per 100 employed individuals). The rural share of younger workers is 4 individuals fewer than the same aged workers in urban areas (28 younger workers per 100 individuals). Meanwhile, the number of rural youth not in employment, education, or training (NEET) has risen. In 2020, there were 18 NEET youth per 100 individuals in rural Japan, up by nearly 5 since 2010 (12.8 per 100). In urban areas, the NEET rate rose even more sharply, with 22 NEET youth per 100 in 2020—an increase of 7 since 2010 (when it was 15 per 100).
There are changes in the youth employment trends suggest that youth are increasingly disengaged. In 2010, 27 youth per 100 individuals in the labour force (between the ages of 15 and 64) were employed. In 2020, this rate fell by 3.2, to 24 young workers per 100 individuals. At the same time, the ratio of youth not in employment, nor in education and training to the rest of the population increased by 5 youth per 100 individuals. The trends were similar in urban areas. In urban areas, youth employment fell by 2.8 young workers per 100 individuals. In urban areas, the rise in inactive individuals (NEETs) was higher than in rural areas at 7.3 more youth per 100 individuals in 2020 as compared to 2010.
The share of older workers in rural areas is relatively higher than those in urban areas. In Japan, rural areas have a larger proportion of older workers (ages 45-64) compared to urban areas among the working age population (between 15-64 years of age). In 2020, 53 out of every 100 employed individuals in rural areas were between the ages of 45 and 64. In urban areas, this number was 49 per 100. Additionally, rural areas had a higher ratio of older NEET individuals in 2020 (68 per 100) compared to urban areas (65 per 100).
Employment ratios changed for older workers in rural areas, as it did in urban areas, but to a lesser extent. In 2010, the employment-to-population rate for older workers (45-64 years of age) in rural Japan was 49 per 100 individuals in the labour force (15-64 years of age). It grew to 53 individuals per 100 in the labour force in 2020 (approximately 3 more individuals per 100). This growth was accompanied by a rise in unemployment and a fall in NEET. In 2010, 42 older workers per 100 individuals were unemployed in rural Japan, which grew by 3 individuals, to 45 in 2020. At the same time, NEET rates for older individuals fell by close to 9 individuals per 100 individuals. In comparison, in urban areas, employment increased from 42 to 49 individuals per 100 among the older age category, while unemployment increased from 38 to 44 individuals per 100, and NEET decreased from 75 to 64 individuals per 100 from 2010 to 2020.
Across geographical regions, the proportion of men and women participating in the labour force is becoming more equal, despite national level evidence that women’s wages and part-time and part-year work patterns still differ significantly from men’s. This is evidenced by the observed improvement in the female share of female employment rates and total employment between 2010 and 2020. Among the active labour force (those in employment or unemployment), female employment rates (the share of women employed out of all who are active in the labour market) were higher than men’s in both urban and rural areas, suggesting that once women search work, they tend to find it at a higher rate than men. This trend has been increasing over time. In 2010 the employment rate for women was 96% while the rate in 2020 grew to 97%. In 2010, 43% of total employment was women with a higher share present in rural (43.3%) than in urban (42.5%). By 2020, this rate had increased to 45% (45.5% in urban and 45% in rural), with similar improvements observed in both rural and urban areas. A similar pattern was observed in OECD metropolitan and non-metropolitan regions. Better engaging with the female labour force can provide the critical labour mass needed to ensure that firms can have the right conditions for innovation.
In Japan, the proportion of foreign-born workers was relatively low, and even lower in rural areas. In the most recent year for which data are available, 2015, the proportion of foreign-born workers in Japan was 1.4%, with the majority of these workers hailing from other Asian countries. In rural areas, the proportion of foreign-born workers was lower, at 1.05%, while in urban areas the proportion was higher, at 1.53%. These numbers, while still small, have shown improvements from 2010 with 0.97% in rural areas and 1.43% in urban areas. In most OECD countries, there is a higher proportion of foreign-born workers in metropolitan regions. In 2020, there was a 2.1 percentage point higher share of foreign-born workers in metropolitan regions in comparison to non-metropolitan regions. Foreign-born workers are a critical resource that can provide one of the resources for overcoming labour supply challenges in rural Japan.
Innovation can be labour augmenting or labour replacing in rural areas. Innovation can be encouraged when labour and firm resources are more easily available and can accommodate the changing demand of the economy. This can be done with better utilisation of pre-existing local skills and resources. In addition, innovation can help solve challenges associated with demographic change, by reducing demand for labour resources and encouraging the adoption of new business models that are better suited for rural and sparsely populated areas.
With close to 125 million inhabitants in 2022, Japan is the third most populous country in the OECD, following Mexico (126 million), and the United States (332 million) (OECD, 2024[1]). This number, while still large, has fallen by close to 2.5 million (2% aggregate decline) from two decades ago in 2002 (127 million) when it was the second most population country in the OECD and went against the average trend in OECD countries that saw an increase in the average population (13% aggregate increased). Only 9 out of 38 OECD countries observing a decline in population.
The large population of Japan is also highly concentrated. In 2022, Japan was the 5th most densely populated country, with 334.5 individuals per km2, following Korea, the Netherlands, Israel, and Belgium. Of a total of 2530 small regions (TL3), 2 regions among those identified as being in the top 1% of all densest regions in the OECD, were located in Japan (according to estimates from the most recent comparative data in 2018). This includes Tokyo and Osaka (7252 and 5776 population per km2, respectively). Altogether over half of the population (55%) of Japan live within large metropolitan regions, composing 13% of the national territory. In comparison, large metropolitan regions in Korea accounted for 68% of the population on 14% of total landmass, while 60% of the population of the United States was in large metropolitan regions that comprises of 25% of the landmass, and 59% of the population of Australia in large metropolitan region, that comprises of 1% of the landmass (OECD, 2024[2]).
In high-tech innovation, Japan is a leader (EC-OECD, 2024[3]), but innovation is also highly concentrated in Japan. In a context of an ageing and declining working age population, particularly in rural areas, the diffusion of labour-saving technologies becomes crucial, if not critical to the continued welfare of the economy. While we see that already in Japanese firms, the government and private sector are taking a proactive approach to the AI technology, often with direct worker consultation (OECD, 2023[4]), there is still room to support non-technological innovation and public sector innovation that may be more relevant for innovation in rural areas with less high-tech and manufacturing firms and a higher share of the public sector.
The demographic context of Japan has an impact on its innovation potential. Considering this, the following 3 sections describe rural places and innovation; the structure of the economy; and the structure of the labour supply as it relates to understanding opportunities for innovation in rural places.
Defining rural places in Japan and population trends
Non-metropolitan regions of Japan are home to 21 million individuals, in 2021, according to OECD regional typologies (Annex 2.A). With 125 million individuals in Japan in 2021, non-metropolitan Japan represents 17% of the population and 32% of total landmass. Non-metropolitan regions far from a midsized or small functional urban area have 2.5 million individuals in Japan in 2021, or 2% of the population and 4% of the total landmass.
How rural is defined can substantially change statistics. According to estimates using the national Densely Inhabited Districts (Annex 2.A), rural Japan composes of 38 million individuals or 30% of the total 126 million individuals in 2020, and 96% of the total surface of Japan. Furthermore, for the purpose of policy analysis, rural areas in Japan often to refer to areas through a land-use lens that include relatively high agricultural and forestry economic activities as well as places where agricultural workers and farm owners live.1 This land-use definition and its implications for policy analysis are covered further in chapter 3. The data in this chapter uses these two main types of territorial definitions to provide analysis for innovation in rural areas.
The first definition is based on the OECD classification of small regions (territorial level 3, or TL3). These classification uses driving distance to urban areas and density as classifying factors that are applied to all OECD countries and are referred in this report as sub-groups of metropolitan and non-metropolitan regions, illustrated in Figure 1.1 and Annex 2.A. In Japan, the administrative unit used by the OECD to classify small regions are prefectures. The OECD classification are useful for comparative purposes, but they refer to relatively large administrative units (prefectures) rather than the national Japanese definitions based on a smaller, local administrative units (district). Japan has no prefecture (TL3 region) classified as a non-metropolitan region that is remote (NMR-R), and as such the category used to capture relatively rural regions only refer to non-metropolitan regions near a small functional urban area (NMR-S), alternatively referred to as “Far from a midsize/large FUA.”
Source: Author’s own elaboration
The second definition is based on national definitions that focus primarily on density where non-dense areas refer to rural areas. Statistics Bureau of Japan identifies Densely Inhabited Districts (DID) as districts (local administrative units) that have 4,000 or more inhabitants per square kilometre, and consisting of adjacent basic unit blocks, etc. whose population is 5,000 or more at the time of the Population Census of Japan (for more information, see Annex 2.A). A map of densely inhabited districts for the majority of mainland Japan is provided in Annex Figure 1.A.1. In this definition, rural areas are considered those that are not Densely Inhabited Districts (DID).
Definitions of rural are critical to understanding rural places, and any policy related outputs that may come from placed-based analysis, including on innovation and entrepreneurship. In Japan, the DID definition provides a grid-based analysis that does not depend on administrative boundaries. In other commonly used definitions in Japan, land-usage is a prominent categorization factor. For example, for identifying rural places for land-use analysis, such as those in chapter 2, rural areas are also identified as mountainous farming areas, flat farming areas and hilly farming areas. The fourth category in the classification are urban areas.2 The lack of a systematic approach to defining rural areas is not uncommon in OECD countries, however, access to harmonized, updated and reliable statistics with rural classifications are critical to provide evidence-based analysis for rural policy making.
Population trends in rural areas
Between 2000 to 2021, the population3 in Japan has declined by an average of 0.32% per year across its 47 TL3 regions and 0.05% overall. Japan was one of 9 OECD countries to have a decline in the overall population during this period. In comparison, the average weighted compound annual population growth in OECD countries was 0.45% (Figure 1.2).
However, the population of Japan did not decline in all regions, rather the population declined only outside of large metropolitan regions. There was a 0.09% growth in large metropolitan regions of Japan. In comparison, in OECD countries large metropolitan regions observed a 0.83% growth in non-metropolitan regions (Figure 1.2). On the other hand, Japan experienced substantial declines in all other regions. For example, in midsized metropolitan regions observed a 0.30% decline in Japan. In comparison, similar OECD midsized metropolitan regions observed a 0.55% increase over the same period. Non-metropolitan regions of Japan saw the largest declines. Non-metropolitan regions near a midsized or large functional urban area saw a 0.64% annual decline in population. In comparison, on average in OECD countries, non-metropolitan regions near a midsized or large functional urban areas saw a 0.36% annual increase in population over the same period. Lastly, non-metropolitan regions near a small functional urban area saw a decline of 0.61% in Japan, while in comparison on average in OECD countries, non-metropolitan regions near a small function urban area saw an increase of 0.38%.
Annual average growth rates across TL3 regions, 2000-2021
Note: MR-L corresponds to large metropolitan regions, MR-M to metropolitan regions, NMR-M to regions near a Functional Urban Area (FUA) larger than 250K, NMR-S to regions near a FUA smaller than 250k and NMR-R to remote regions. Japan does not have any TL3 remote regions.
Source: OECD Regional Indicators
The size of the Japanese population is, in part, impacting the relatively large changes observed in regional statistics. Relative statistics on the share of the population in metropolitan and non-metropolitan regions in 2001 and 2021 demonstrate that while growth in large metropolitan areas and relative declines in all other types of regions is still persistent, even in relative terms, population changes are more mitigated when taking into account the large size of the Japanese population (Figure 1.3). For example, the percentage point difference from 2001 and 2021, between the shares of the population in large metropolitan areas and in non-metropolitan regions near a midsized or large functional urban area was 0.05 percentage points (0.03 percentage point growth in MR-L and 0.02 percentage point fall in NMR-S). Similarly, the percentage point difference from 2001 and 2021, between the shares of the population in large metropolitan areas and in non-metropolitan regions near a small functional urban area was 0.04 percentage points. In comparison, this difference is smaller than 11 countries. Nevertheless, it is larger than most other highly populated OECD countries such as the United States and Mexico.
Percentage point differences between total shares of the population, 2000 to 2021
Note: Figures are sorted based on largest gaps between highest and lowest values in each regional category. MR-L corresponds to large metropolitan regions, MR-M to metropolitan regions, NMR-M to regions near a Functional Urban Area (FUA) larger than 250K, NMR-S to regions near a FUA smaller than 250k and NMR-R to remote regions. Japan does not have any TL3 regions defined as remote regions.
Source: OECD Regional Indicators
Defining innovation in rural Japan
Innovation, for policy purposes, is not commonly measured in the same way, across agencies within countries, and much less between agencies in countries. However, taking the broadest definition of innovation is critical when understanding non-urban economies, in part because of unequal distribution of economic activities across places.
The OECD regularly publishes, the Oslo manual, that defines innovation as the development of products and processes that provide new or a significant improvement to the firm (OECD/Eurostat, 2018[5])4 with some degree of novelty which is new to the firm, new to the market, and new to the world. While the perimeter of the “market” is still not well defined, nor adapted to mobility patterns in rural markets, the definition at least provides a harmonized approach that is flexible enough to be relevant in many countries (see Box 1.1 for further information). Several statistical agencies have started adopting the definition within firm surveys on innovation, such as Eurostat’s Community Innovation Survey5, or the United Kingdom’s Innovation Survey6. However, often the low-density nature of many rural places means that without an over-sampling of surveys for rural analysis, survey-based analysis overlooks rural places. This is due to the low number of observations in the sample selection, that lead to a lack of statistical strength.
What is the Oslo Manual?
The Oslo manual is a publication that outlines a commonly agreed upon approach to measure and report statistics on innovations. Started in the early 1990’s, the Oslo Manual was elaborated through the consensus of the OECD Working Party of National Experts on Science and Technology Indicators (NESTI) and has been adopted by over 80 countries. The guidance outlined in the manual is used by major international organisations, and researchers worldwide. Its revision was conducted through consultation of both the NESTI and Eurostat’s Community Innovation Survey (CIS) Task force.
Defining Innovation
The 4th edition of the Oslo Manual distinguishes between innovation as an outcome (an innovation) and the activities by which innovations come about (innovation activities). It defines an innovation as “a new or improved product or process (or combination thereof) that differs significantly from the unit’s previous products or processes and that has been made available to potential users (product) or brought into use by the unit (process)”.
The major additions to the previous versions include measuring innovation not only from businesses but also other organisations and individuals; updates to improve harmonization between core definitions and taxation; better accounting of globalisation, digitalisation, and trends in investment in intangible assets; guidance on measuring internal and external factors influencing business innovation; prioritization of the measurements of government policies on innovation; expansion on methodological guidelines; guidance on the use of innovation data and a new glossary.
However, often statistics available on innovation consider only some forms of innovation, such as those that require heavy research and development investments, or that result in patents or registered trademarks. This type of innovation is often the type of innovation that is targeted in innovation policies (see chapter 3), but also simultaneously is more commonly observed in cities because of agglomeration factors that are associated with easier access to finance, skills and infrastructure, and the location of firms that participate in high-tech innovation. Innovations in firm processes, in the public sector and those that serve a non-financial purpose (social innovation) are not captured in such statistics but often are important to solve challenges in rural areas that are not observed to the same degree as in urban areas.
Innovation in rural areas
Japan is among one of the top high-tech innovation economies in the OECD. In terms of high-tech innovation and patenting activities, Japan had more patent applications (approximately 222 000) than all other OECD countries from the period of 2016 to 2020 (Figure 1.4 and Annex Figure 1.C.1), with the only exception being the United States with close to 269 000 patent applicants between 2016 and 2020. The average patent intensity or rate of patent applications per 1000 individuals is the highest among OECD countries in Japan, at 0.35 patents per 1000 individuals between the period of 2016 and 2020. The patent intensity in Japan is double the average in OECD countries (0.14), and other high-patenting countries such as Norway (0.14) and the United States (0.16). Similarly high patent intensity rates are observed in Sweden (0.33), Switzerland (0.31) and Korea (0.31).
High-tech innovation intensity has increased in Japan in the decade leading up to 2020, as it has in 24 of 36 OECD countries, with available statistics. From 2011-2015, the patenting intensity in Japan was 0.30 per 1000 individuals, while it increase in the period from 2016-2020 to 0.35 patents per 1 000 individuals (Figure 1.4 and Annex Figure 1.C.1). Japan had the second highest increase in terms of patent intensity with 0.05 more patent applicants per 1000 in the last period (2016-2020), as compared to the first period (2011-2015). This was only preceded by Korea with 0.08 more patents per 1 000 individuals in the last period (2016-2020), as compared to the first period (2011-2015). Other high patent intensity countries such as Sweden and Switzerland observed comparatively smaller increases. In Sweden, the number of patents per 1 000 individuals increased by 0.001, while in Switzerland it increased by 0.014.
Total applications and patent intensity (applications per 1000 individuals) in 5-year periods
Note: Patent applicants are calculated as the fractional count of locations of applicants for each patent in the year of application. Data on the rest of the OECD countries is available in Annex Figure 1.C.1.
Source: Authors’ own elaboration and OECD (2023[6]) Regional Innovation, OECD Regional Statistics (database), https://doi.org/10.1787/1c89e05a-en.
Even though Japan is a strong in national innovation performance, there are large gaps between the performance of metropolitan and non-metropolitan regions (Figure 1.5). The difference between the average patent application intensity is 0.32 patents per 1000 individuals (in levels) in Japan, from the 2016 to 2020 period. This is the highest gap in all countries for which data was available. In addition, the gap increased from the period of 2011-2015 by 0.05 patents per 1000 individuals (0.27 patents per 1000 individuals in 2011-2015 period). In contrast, the average gap in OECD countries was 0.15 in the period of 2016-2020, only experiencing a marginal increase of .01 from the 2011-2015 period.
High-tech innovation is densely concentrated in metropolitan regions of Japan. As compared to other OECD countries, patenting application intensity is higher in metropolitan regions of Japan than the OECD averages and among the 4 highest following Finland, Sweden and Norway. Patenting application intensity is also substantially lower in non-metropolitan regions of Japan than in many other places. In the period from 2016 to 2020, non-metropolitan regions far from a metropolitan area had a 0.32 lower average patent application intensity than metropolitan regions (Figure 1.5). In comparison, the average gap in patent intensity from metropolitan regions and non-metropolitan regions near a metropolitan area in OECD countries was less than half of the gap of Japan, at 0.14 lower average patent applications per 1 000 individuals. The gap most closely resembled that of Korea’s non-metropolitan regions near metropolitan areas at 0.25 less patent intensity than in metropolitan regions, and the difference between metropolitan regions and non-metropolitan regions near to a metropolitan area in Sweden (0.25) and Finland (0.22).
Patent applications per 1000 individuals, in 5-year periods (levels)
Note: OECD values refer to un-weighted country averages. Only countries with regional data were included. Patent applications that were not assigned to a location were not included in the analysis. Small regional classification system (TL3) is based on the description found in Annex 2.A. The category metro includes large metropolitan regions and standard metropolitan regions; the category non-metro refers to non-metropolitan regions close to metropolitan regions, non-metropolitan regions close to small and medium sized cities and non-metropolitan remote regions. For the 3-tiered classification in panel B, metropolitan regions refer to large metropolitan and metropolitan regions; regions near a metro refer to non-metropolitan regions near a functional urban area of 250K or over; regions far from a metro refer to regions far from a functional urban area of over 250K. Further information on classifications is available in Annex Table 1.A.2. Deviations of patent intensity refer to level differences between metropolitan and non-metropolitan region patent intensity from national means. Metro/non-metro gaps refer to level differences between highest and lowest patent applicant intensity rates, based on the 3 categories previously described.
Source: Authors’ own elaboration and OECD (2023[6]) Regional Innovation, OECD Regional Statistics (database), https://doi.org/10.1787/1c89e05a-en.
While innovation performance may seem low in rural areas based on patent statistics, this type of indicator is only one form of measurement. The regional disparity and concentration in innovation is often associated with composition bias and challenges associated with access to resources (such as investment, and legal services) (OECD, 2022[7]). As such, high-tech innovation is not necessarily the most relevant form of activity that creates new products or processes to provide solutions to ongoing challenges and capture new opportunities in rural areas. However, without further access to data on innovation as measured by the Oslo manual described in Box 2.1, it is not possible to further understand challenges to innovation for rural firms.
Innovation in Japan often focuses heavily on innovation in the high-tech firms or manufacturing industry, with a definition that leans towards the development of high-tech products rather than the more general development of products and processes that are either new to the firm, or new to the market. However, some government initiatives go beyond a Sciences and Technology Innovation (STI) focused view. For example, the Project for Implementation of Advanced Technologies in Regional Society, implemented by the Cabinet Office, is one of the initiatives of the government which can contribute to the innovation, and regional revitalisation in regions and rural areas. The project aims to promote regional revitalization by encouraging the adoption and diffusion of near-future technologies. Several ministries gather to support the selection of proposals from regions that are innovative in how they carry-out regional revitalization, have strong leadership, and implement the project in a horizontal manner. Through these projects, the government builds local support systems (regional implementation councils) and provides comprehensive support through relevant ministries and agencies, aiming to realize the social implementation of future technologies in the regions. Beyond in-kind support, the government also provides various grants and subsidies to such initiatives. When selecting which initiatives to fund, no special consideration is given to rural areas, but the selection is made from the viewpoint of "contributing to the development of local areas."
Innovation in rural areas can be heavily influenced by structural aspects of rural firms and economy. The following section will highlight trends in entrepreneurship activity rates, sectors and foreign investment as key features that determine innovation capacity and opportunity in rural economies. It primarily uses the internationally comparative OECD classification of small regions (TL3) as described in Annex 2.A, with the exception of the sectoral analysis, which uses the Japanese national DID definition of geography.
Entrepreneurship rates in Japan
In the context of rural Japan, demographic change may also affect opportunities for entrepreneurship and innovation. In a previous study on territorial development in Japan, ageing and population trends were suggested as hindering factors for a dynamic economy, in particular when it came to innovation and entrepreneurship (OECD, 2016[8]). On the demand side, the ageing population trend builds new opportunities for innovators and entrepreneurs in the silver economy, however on the other hand, older individuals are less likely to become entrepreneurs, despite the fact that they may be more successful, with higher wealth, networks and experience.
The churning of businesses (the creation of new businesses and the closing of old ones) is critical to the generation of new activities in both rural and urban areas. Highly churning economies are important for promoting dynamism and are associated with high levels of innovation adoption. While it is difficult to capture exact measures of business churning, changes in the rate of business ownership can give a proxy for new firm activity. The following section uses the internationally comparative OECD classification of small administrative regions (TL3).
The rate of entrepreneurship in Japan is falling on a national level, and in particular in metropolitan regions. The latest available data in Japan, suggests that over the span of a decade (2007-2017), the rate of entrepreneurship decreased from close to 90 entrepreneurs per 1 000 workers, to 72 entrepreneurs per 1000 workers (Figure 1.6, panel A). The change was most pronounced in metropolitan regions, where there were 18 fewer entrepreneurs per 1,000 workers in 2017 than there were 10 years earlier (in 2007). The largest fall was in large metropolitan regions (20 less entrepreneurs, per 1 000 workers), and the smallest losses were in non-metropolitan regions near a small functional urban area (15 less entrepreneurs, per 1 000 workers). Other high-innovation countries with data on a sub-national level on entrepreneurship have increased the number of entrepreneurs, or stayed the same, around the same time. For instance, in Canada between 2007 and 2018, there was a 17% increase in the number of (non-Indigenous) businesses, while in the US from 2013 to 2016 the growth rate of the number of firms stayed constant ( (OECD, 2024[9]) and (OECD, 2023[10])).
The decline of entrepreneurship rates in non-metropolitan regions was primarily due to a more pronounced reduction in the number of business owners (Figure 1.6, Panel B). On the other hand, in metropolitan areas, the decline in the entrepreneurship rates in metropolitan areas was accompanied by an increase in the number of employed individuals. From 2007 to 2017, the number of business owners in metropolitan regions declined by 19%, while in non-metropolitan regions the decline was more pronounced at 22%. The greatest decline was observed in non-metropolitan regions close to small functional urban areas, with a 23% reduction in the number of entrepreneurs between 2007 and 2017. A similar decline was observed in non-metropolitan regions near midsized or large cities, with a 22% reduction in the number of entrepreneurs between 2007 and 2017. In contrast, in large metropolitan regions, the relatively low decline in the number of entrepreneurs was accompanied by a 3.9% increase in the number of individuals employed in large metropolitan areas.
Without further information on the dynamics of firm activities, it is challenging to identify concrete evidence of a decline in economic activities, as some firms may be undergoing a period of consolidation. Nevertheless, in the context of both falling employment and a declining number of business owners, the evidence suggests a need to enhance our understanding of the factors influencing firm formation, with a particular focus on non-metropolitan regions.
Number of business owners in Japan, by OECD Classification of small (TL3) regions (2007 and 2017)
Note: Entrepreneurship rates refer to as the number of entrepreneurs over the total population, per 1000 workers. The change in the entrepreneurship and employment rates refer to the average aggregate change from first year to last year, of total numbers of entrepreneurs and total employed workers. MR-L corresponds to large metropolitan regions, MR-M to metropolitan regions, NMR-M to regions near a Functional Urban Area (FUA) larger than 250K, NMR-S to regions near a FUA smaller than 250k and NMR-R to remote regions. Japan does not have any TL3 regions defined as remote regions.
Source: Japan National Statistics
Productivity in rural Japan
Innovation absorption, often proxied with productivity statistics, can provide further information on the well-being of communities and firms. With low regional disparities in productivity, innovation adoption may be more equally distributed than high-tech innovation. Japan has the 3rd highest productivity in metropolitan regions and in non-metropolitan regions (both far and near) (Figure 1.7). While the absolute gap between high and low productivity regions is relatively high (17th out of 26 countries with available data), the percentage gap between places is relatively low. From example, in 2020, the gap between metropolitan regions and regions far from metropolitan areas, was 11 000 USD per worker in Japan, which was equivalent to 10.5% less productivity in regions far from metropolitan areas, as compared to metropolitan regions. This was the 9th lowest gap (out of 26 countries).
Gross value added per employed worker (2020 or most recent year)
Note: Data refers to 2020 or latest available year: Data for Japan are for 2016. Data refers to 2019 for Austria, Switzerland, Germany, Spain, Finland, Greece, Italy, Lithuania, Latvia, the Netherlands, Norway, Portugal, Sweden and the United States. Productivity measures use equal weights for each TL3 region. Countries are sorted in ascending order of average productivity for regions far from metropolitan areas. Ireland is not included in the analysis because it has the starkest productivity differences across regions, reflecting in large part the relatively high share of multinational companies with significant intellectual property assets with headquarters in (including those that have redomiciled to) Ireland, and Dublin in particular (Central Statistics Office of Ireland). Metropolitan regions refer to large metropolitan and metropolitan regions; regions near a metro refer to non-metropolitan regions near a functional urban area of 250K or over; regions far from a metro refer to regions far from a functional urban area of over 250K. Further information on classifications is available in Annex Table 2.A.2
Source: OECD (2022[11]) and OECD Regional Statistics (database), http://dx.doi.org/10.1787/region-data-en.
Sectoral composition of rural areas
Innovation activities occur in all sectors. The Oslo definition of innovation is specifically sector and tech agnostic as innovation can occur in any activity (Box 2.1). However, some sectoral activities, such as manufacturing sectors tends be associated with more innovation activities, in part because innovation outputs are more easily statistically measurable. In addition to the importance of sectoral component of measuring innovation, the diversification of the economy is an important aspect of creating conditions for innovation within a region or place. This section will look specifically at sector composition and trends to create the setting for rural innovation. It uses the national Japanese DID definition for geography.
Rural areas, as defined by the DID classification, are characterized by a high share of jobs in the trade and services sector (38%) and in the public sector (25%)7 in 2020 (Figure 1.8, Panel A). This is followed by the manufacturing sector (19%), agriculture (9%), construction (9%) and a very small amount of mining and quarrying. In 2020, of the 38% of the rural economy working in trade and services sector8, the wholesale retail and trade sector accounted for the largest share (35% of trade and services sector) and other unclassified services (15%) sector.
Urban areas have the similar trends in sectoral distribution of jobs as in rural areas but tend to have higher shares in all sectors except notably the agricultural sector. The largest share of job creating activities in urban areas belong to the trade and services sector (52%)9, followed by the public sector (26%) (Figure 1.8, Panel A). These sectors are followed by the manufacturing (14%) and construction sector (7%). Like rural areas, there is also a very low share of jobs in the mining and quarrying sector. However unlike in urban areas, in 2020, there was a very low share of jobs in the agricultural sector. In fact, the quasi-totality of all agricultural jobs (85%) was in rural areas of Japan.
Over the period of 2010-2020, the structure of the rural economy demonstrated change in Japan (Figure 1.8, Panel B), as observed in other OECD countries. However, unlike other OECD countries that are on average showing transition into the trade and services sectors, in Japan, the growth of these sectors are slow, in both urban and rural areas. This is not explained by the impact of the covid-19 pandemic trends, as the changes are also prevalent prior to the crisis (Annex Figure 1.C.2). However, it may be explained by aggregate population losses combined with urbanisation trends.
From 2010 to 2020, the rural economy lost 1.8 million jobs, while urban areas lost only 200 000 jobs (Figure 1.8, Panel D), based on the use of DID classification. The primarily losses in jobs came from the trade and service industry in urban areas, while most losses came from the agricultural sector in rural areas. The only growth in jobs were in the public sector that grew by 3.6% in urban areas, and by 2.8% in rural areas. The largest fall in rural areas have been in the agricultural sector that observed a 1.4% decline from 2010 to 2020. While trade and services also fell in rural areas (0.8%), it did not fall to the same extent as in urban areas (3.1%). Furthermore, the fall of the trade and services sector in rural areas was stronger prior to the covid-19 pandemic (Annex Figure 1.C.2), suggesting that the impact on the trade and services sector in rural areas may have been relatively insulated as compared to urban areas.
Sectoral shares of jobs (2020) and changes in shares (2010-2020), by DID typology
Note: Sector groupings are further described in Annex 2.A. Changes refer to aggregate composite changes from first year to last year. DID correspond to Densely Inhabited Districts defined by the Statistics Bureau of Japan that have 4,000 or more inhabitants per square kilometre, and consisting of adjacent basic unit blocks, etc. whose population is 5,000 or more at the time of the Population Census of Japan (for more information, see Annex 2.A).
Source: Statistics Bureau, Ministry of Internal Affairs and Communications, Japan.
Outside of the public sector, based on the current sectoral structure of the rural economy, there is an important share of the rural economy that is in the trade and services, or manufacturing. The fact that the trade and services is the largest sector in both rural and urban areas is in line with the general trend in OECD countries. However, a large share of manufacturing in rural areas than in urban areas is not as common of a trend in OECD economies.
With increasing risk of automation of manufacturing jobs in Japan (OECD, 2021[12]), innovation in the manufacturing sector, would necessarily imply a reduction of jobs, yet the relative fall in manufacturing in rural areas was very low. This can imply that the sector is more resilient to the demographic and employment declines than the agricultural sector, or that the type of manufacturing in rural areas may require different non-automated skills. For instance, while the car manufacturing industry often located closer to metropolitan areas, may be easier to automate, other types of manufacturing may value more individual know-how and re-enforce the need for process innovation as much as product innovation (OECD, 2022[7]). For example, process and product innovations that do not necessarily require as heavy R&D or patenting include bespoke or design furniture manufacturing that may change from client to client, or culturally relevant manufacturing such as in the food (rice) and drink (sake, wine and other alcohols) manufacturing industry. Some of these innovations, would be better captured in trademarking practices or rather direct surveys on innovation in firms.
Lastly, the public sector is a relatively large employer in rural areas, as in urban areas. As a relative share of total jobs, it is the second highest in both rural areas (25%) and in urban areas (26%), which is in line with national efforts to uphold service provision in areas outside of large metropolitan areas. Given the large share of the public sector, and the higher costs of delivering services to rural places (OECD/EC-JRC, 2021[13]), focusing on bringing innovation to the public sector, that can either mean providing services more efficiently, or more targeted to rural areas, is increasingly important. In part, this means thinking about providing public services at the right scale that can build mass in low density areas (OECD, Forthcoming[14]). Innovation in the public sector can be supplemented by support for social innovation, or innovation with a social purpose often delivered by non-governmental organisations such as cooperations or communities. This is already adopted as an important part of the regional and rural innovation strategy in Scotland and many other countries (OECD, 2023[15]; OECD, 2024[16]).
The statistics on sector classifications are gathered using the Japanese classification system. They large correspond to International Standard Industrial Classifications (ISIC 4th revision) with a few exceptions in ordering and numbering. In the report, the sectoral classifications are grouped into the following 6 major sectoral groups to facilitate readability of the analysis:
1. Agriculture consisting of agriculture and forestry; and fishery (A and B)
2. Mining consisting of mining, quarrying and gravel extraction (C)
3. Construction (D)
4. Manufacturing (E)
5. Trade and services (private), consisting of information and communication (G); transportation and postal services (H); wholesale and retail trade (I); finance and insurance (J); real estate and goods rental (K); academic research, professional and technical services (L); accommodation and food services (M); complex service businesses (Q); services (R); and industries not elsewhere classified (T).
6. Public sector, consisting of electricity, gas, heat and water supply (F), education and learning support (O); medical care and welfare (P); public utilities, except elsewhere classified (S); and personal and recreation services (N).
The 6th Sector
In line with policy discourse in Japan, the rise of the importance of value-added supply chains has taken a particular place in Japan. Reference to a “Sixth-order industrialisation” involves the creation of integrated value chains encompassing production, processing, distribution and sales activities by linking producers in agriculture, forestry and fisheries with partners who have expertise in the secondary and tertiary sectors. The name reflects the fact that rural producers of (primary) agricultural commodities are engaged in processing (secondary) activities and distribution/marketing (tertiary) operations (OECD, 2016[8]).
Source: Authors’ own elaboration
Enabling factors for innovation
There are several factors that can impact opportunities for innovation. Some firms may choose to locate in places that minimize the cost of accessing services and resources that are needed for innovation and growth, while others may benefit from local conditions, despite lack of similar opportunities. The following section explores some of the different framework conditions that support innovation and innovation diffusion, and adoption as related to access to finance, skills and digital infrastructure.
Access to finance and foreign investment in rural regions
Access to finance can be an important enabler of innovation. It provides resources for firms to conduct research, development and upscaling, as well as the access to networks of investors that creates new monetary and non-monetary resources through national and international networks. In Japan, access to finance in rural areas is further challenged by modest business dynamism nationally, including low firm entry rates and an undersized venture capital market (OECD, 2024[17]).Recent work in agricultural innovation in Japan, identified access to local banks for financial resources as a critical aspect of encouraging innovation (OECD, 2019[18]), despite some misalignment of incentives for financial institutions. Even with a well-developed banking and financial market, commercial banks still play a small role in agricultural finance. Other disincentives for the development of commercial banks in rural areas, such as high levels of guaranteed credit with public institutions, are likely to hinder entrance of more commercial bank support in agricultural financing, in particular. Little is known about funding in rural areas outside of agricultural sector.
In addition to domestic sources of financing, foreign sources of financing can be an important enabler of innovation, however, it is unclear if foreign investment is conducive towards the type of business and innovation rural areas. For example, evidence from a similar study in Canada suggested that foreign ownership (as a proxy for investment) is not equally as likely to lead to innovation in rural areas as it is in urban areas (OECD, 2024[9]).
In Japan, non-metropolitan regions have attracted only a small share of inward foreign direct investment (FDI). In contrast it has been highly concentrated in metropolitan regions. Using data from the FDI markets database that covers a period of close to 2 decades (2003 to 2022), FDI directed towards non-metropolitan regions of Japan was only on par with OECD averages in 2004 (Figure 1.9). In all other years, the share of FDI in non-metropolitan areas were lower than the OECD average and often much lower than other high innovation OECD countries such as the United States, and Norway. Between 2003 and 2022, 10% of all FDI Investments were directed towards non-metropolitan firms, which was lower than the OECD average of 31%. Overtime, this grew from 0.1% in 2003 to 5% in 2022 in Japan. The highest share of investment in non-metropolitan firms was in 2006 and 2020, when the share of FDI to non-metropolitan regions was close to 20% of total FDI. In comparison, other high-innovation countries observe, on average, larger shares in non-metropolitan regions. For example, in Norway 51% of all inward investments were to non-metropolitan areas, while the United States and the United Kingdom both had close to the OECD average of FDI investments in non-metropolitan areas (31 and 32%, respectively).
In addition to lower shares of total FDI, the share of FDI in non-metropolitan regions that were specifically directed towards research and development activities is very low, as compared to other high-innovation partners (Figure 1.9). In OECD countries, 15% of all FDI in R&D activities was directed towards non-metropolitan regions between the years of 2003 and 2022. In Japan the share was close to 2%. In other high-innovation countries like Norway and Germany, this share was much larger, at 28% and 13%, respectively. Low R&D focused-FDI is rural Japan signals a significant hurdle for fostering firm-based innovation and necessitates strategic interventions to attract investments to non-metropolitan regions.
Note: Panel A only includes countries that are relatively high in terms of innovation performance as identified in Figure 2.4. The OECD average category refers to simple country averages for Australia, Austria, Belgium, Bulgaria, Canada, Chile, Colombia, Croatia Czechia, Germany, Denmark, Estonia, Finland, France, Greece, Hungary, Ireland, Israel, Italy, Japan, Korea, Lithuania, Luxembourg, Latvia, Malta, Mexico, the Netherlands, Norway, New Zealand, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Switzerland, Sweden, Turkey, the United Kingdom, and the United States. R&D shares include 0 values in averages. Estimates generated from the FDI Markets Database capture all major investments reported in newspapers. Other small-scale investments are not captured. Furthermore, reported estimates only include funds that are associated with a specific location. Additional FDI flows exist that were not allocated to a specific location within the country. FDI markets databases give information on the largest publicly traded firms that have announcements of FDIs in financial news outlets. They do not encompass all investments and therefore do not provide a complete picture of all FDI investments. The share of R&D -specific FDI refers to FDI investments that were tagged specifically for use in research and development. Nevertheless, other sources of funding could be related to R&D investments, contributing to R&D through expansion of activities or acquisition of new offices. These are not included in the R&D-specific funding figures.
Source: FDI Markets Database.
Access to skills outside of cities
With a declining population and falling employment, access to skilled labour can become and increasing challenge for rural places in Japan. The increase in employment of women, older workers and the elderly has contributed to more recent improvements in reducing skills gaps the Japanese labour market (OECD, 2021[12]). Yet, a large share of jobs in Japan are at risk of automation and there is a larger a skills deficit observed in non-metropolitan in Japan (OECD, 2016[19]). The skills mismatch exists in particular in digital problem-solving skills, despite high-level of skills of adults in Japan (OECD, 2021[12]), and skills hit women, young people and those with low skills during the covid period leading to a continuation of the rise of skills imbalances (OECD, 2022[20]). Furthermore, for older individuals in work, there are low levels of upskilling at mid and older ages (OECD, 2019[21]).
Information on skills imbalances and mismatches across places is not available as widely as national statistics. However new data based on the OECD PISA scores in cities and those not in cities, has helped shed some light on territorial inequalities as related in skills. Indeed, in most OECD countries, the performance of individuals in reading tests suggest that reading skills are higher in cities than in other areas. On average, in the OECD this amounts to 14-point difference in scores between cities and non-cities in OECD countries. In Japan, this difference is larger, at a 22-point difference between cities and non-cities in OECD countries (Figure 1.10, Panel A).
There are also challenges in digital skills, and among ageing populations in Japan (OECD, 2019[21]). Given that there are the nationally documented challenges in digital problem-solving skills in particular among older populations, and that such populations make a larger share of rural workers, as compared to urban worker, the skills in digital problem-solving are potentially a larger issue in rural areas, as compared to urban.
The skill level of workers is determined by different opportunities for education and skills training. These are impacted by supply of education institutions as well as the socio-economic background of parents that can induce a demand for skills upgrading. In Japan, controlling for the socio-economic background for individuals explains the majority of the difference in the testing scores for reading (Figure 1.10, Panel B). Once controlling for socio-economic backgrounds through proxies such as parental occupation, education and income levels, the difference between cities and non-cities is not significantly different from 0. This implies that much of the differences in reading skills between cities and places outside of cities are due to underlying differences in the socio-economic welfare of individuals, rather than capacity of individuals per se.
Note: Panel B includes a control for socio-economic status that is created as an index. The PISA of economic, social and cultural status (ESCS) is a composite score derived, as in previous assessments, from three indicators related to family background: parents’ highest education, in years (PAREDINT), parents’ highest occupational status (HISEI) and home possessions (HOMEPOS). Further information on the generation of the socio-economic indicator can be found in OECD (2023, p. Annex A1[22]).
Source: 2022 Database - PISA (oecd.org) ; https://www.education.gouv.fr/pisa-programme-international-pour-le-suivi-des-acquis-des-eleves-41558, accessed 11 June 2024; OECD (2023[22]), PISA 2022 Results (Volume II) : Learning During – and From – disruption. Annex A1. Construction of indices, https://doi.org/10.1787/a97db61c-en.
Access to quality digital infrastructure
Conditions for innovation can be facilitated or hindered by the availability and quality of public services, and infrastructure in rural places. Given the remote nature of rural regions, access to quality digital infrastructure can facilitate developing networks beyond the physical network. While roads, shipping and air transport still remain a critical dimension of auxiliary services that support the general framework conditions for innovation, digital infrastructure plays a particular role in encouraging innovation, as it helps overcome many of the challenges faced by rural firms through the capacity to go through more seamless digital transition, access to innovation networks and solutions to labour restrictions through remote work. However, often rural remote regions often have lower access to high-speed internet and policies are not well adapted to assure last mile-service provision (OECD, 2021[23]).
In Japan, access to high-speed internet is lower outside of metropolitan regions, as in the case in most OECD countries. Internet speeds are 19 percentage points higher in metropolitan regions than in regions far from a metropolitan area, and 21 percentage points higher in metropolitan regions near a metropolitan area (Figure 1.11). However, this dispersion is not as large as on average in the OECD. On average in OECD countries, there are 25 percentage points higher speeds in metropolitan regions as compared to non-metropolitan regions far from a metropolitan area, and 20 percentage points higher speeds in non-metropolitan regions near a metropolitan area. Other high-innovation countries, such as the United States, Sweden, Switzerland and Korea all have relatively lower differences in access to internet speeds between metropolitan regions and non-metropolitan regions.
Dispersion in mean speeds from national means, Q4 2023
Note: Metropolitan regions refer to large metropolitan and metropolitan regions; regions near a metro refer to non-metropolitan regions near a functional urban area of 250K or over; regions far from a metro refer to regions far from a functional urban area of over 250K. Further information on classifications is available in Annex Table 2.A.2
Source: OECD (forthcoming), Bridging Connectivity Divides. OECD Publishing: Paris.
Because of its relative size, population decline is often a phenomenon that impacts non-metropolitan areas (and regions) more than metropolitan regions, creating additional challenges for innovation as it limits the access to a larger pool of workers. In Japan, the challenge is particularly stark, with the highest old-age dependency ratio of all OECD countries and projections that predict the continuation of the increasing trend until 2050 (OECD, 2023[24]). However, the labour supply can be replenished, through better engagement with rural women and elderly populations.
In rural areas, while communities have long understood challenges in adapting markets to older population, strategies to attract of youth and skilled workers, feature much more strongly, as part of local innovation strategies. However, these strategies often overlook the importance of a.) activating older workers and the silver economy, b.) women as an economic agent for innovation, as well as a part of the larger labour supply, and finally, c.) the engagement with foreign labour to fill gaps in the skills and labour gaps in the economy.
Employment trends and ageing in rural Japan
The number of workers in employment declined from 2010 to 2020. From 2010 to 2020, the overall aggregate decline in the number of workers in Japan was 3% or 0.9% in terms of compound annual growth rates (Figure 1.12, Panel A). However, using the DID classification, this fall was relatively higher in rural areas, that experienced an 8.9% fall in total aggregate employment, or 0.9% annually (CAGR) from 2010 to 2020, as compared to a 0.49% aggregate fall, or 0.05% fall in the CAGR for employment in urban areas. In rural areas, the fall in the employment numbers is also impacted by a 2% overall reduction in the share of total workers in rural areas of Japan (Annex Figure 1.C.3). However, employment rates grew over the same period in both rural and urban areas. Employment rates grew in rural areas from 90 to 92% between 2010 and 2020, and in urban areas from 90 to 93% in urban areas.
Despite different levels of severity in population and employment changes, rural and urban areas had similar labour force characteristics, with similarly rising employment rates (Figure 1.12, Panel B), and shares of individuals in employment, unemployment and not employed, nor in education or training (NEET) (Figure 1.12, Panel C). Both rural and urban areas experienced a fall in unemployment and rise in employment in line with a tight labour force market that may have been due to relative population declines. The only marginal difference is a 1% higher share of individuals in NEET in rural areas, suggesting that even with tight labour markets, that opportunities for employment, education and training are still a challenge for (potential) workers, and skills-matching may also be a challenge for employers.
Note: Employment, Not Employed, in Education, or Training (NEET) and unemployment rates are calculated as a share of the total labour force (15-64). Unknown category is excluded from the analysis. DID area Densely Inhabited Districts defined by the Statistics Bureau of Japan that have 4,000 or more inhabitants per square kilometre, and consisting of adjacent basic unit blocks, etc. whose population is 5,000 or more at the time of the Population Census of Japan (for more information, see Annex 2.A).
Source: Japan National Office for Statistics
With demographic change in rural Japan, ageing workers in rural areas can increasingly become a challenge for innovation, and increasingly, lack of opportunities for youth can create additional impetus for youth to move out of rural areas. Without lifelong training and specific strategies to upscale skills among rural workers, the ageing work force may have a harder time adapting to new forms of work, while the youth workforce may not find the right opportunities for career paths (OECD, 2018[25]).
A few trends are apparent for younger worker (15-34) from analysis based on data using the DID typology, in Japan:
The share of young workers as a share of total workers in rural areas of Japan is low (24 younger workers per 100 individuals) (Figure 1.13, Panel A). The rural share of younger workers is 4 individuals fewer than the same aged workers in urban areas (28 younger workers per 100 individuals). Meanwhile, the number of rural youth not in employment, education, or training (NEET) has risen. In 2020, there were 18 NEET youth per 100 individuals in rural Japan, up by nearly 5 since 2010 (12.8 per 100). In urban areas, the NEET rate rose even more sharply, with 22 NEET youth per 100 in 2020—an increase of 7 since 2010 (when it was 15 per 100).
There are changes in the youth employment trends suggest that youth are increasingly disengaged. In 2010, 27 youth per 100 individuals in the labour force (between the ages of 15 and 64) were employed. In 2020, this rate fell by 3.2, to 24 young workers per 100 individuals. At the same time, the ratio of youth not in employment, nor in education and training to the rest of the population increased by 5 youth per 100 individuals. The trends were similar in urban areas. In urban areas, youth employment fell by 2.8 young workers per 100 individuals. In urban areas, the rise in inactive individuals (NEETs) was higher than in rural areas at 7.3 more youth per 100 individuals in 2020 as compared to 2010 (Figure 1.13, Panel B).
Youth are increasingly disengaged in both rural and urban areas of Japan. In line with aggregate decreases in employment over the last 10 years in both urban and rural areas, there is a decreasing rate of employment and unemployment for youth in Japan. The joint occurrence of decreasing employment and unemployment is explained by increasing shares of workers that are not in employment, education, or training (NEET). In rural areas, these challenges are accentuated with stronger losses in the rate of employment and unemployment and a rise in NEET rates.
Similarly, some trends were observed for older workers (45-64) in Japan:
In Japan, rural areas have a larger proportion of older workers (ages 45-64) compared to urban areas among the working age population (between 15-64 years of age). In 2020, 53 out of every 100 employed individuals in rural areas were between the ages of 45 and 64. In urban areas, this number was 49 per 100. Additionally, rural areas had a higher ratio of older NEET individuals in 2020 (68 per 100) compared to urban areas (65 per 100) (Figure 1.13, Panel A).
Employment ratios changed for older workers in rural areas, as it did in urban areas, but to a lesser extent. In 2010, the employment-to-population rate for older workers (45-64 years of age) in rural Japan was 49 per 100 individuals in the labour force (15-64 years of age). It grew to 53 individuals per 100 in the labour force in 2020 (approximately 3 more individuals per 100). This growth was accompanied by a rise in unemployment and a fall in NEET. In 2010, 42 older workers per 100 individuals were unemployed in rural Japan, which grew by 3 individuals, to 45 in 2020. At the same time, NEET rates for older individuals fell by close to 9 individuals per 100 individuals (Figure 1.13, Panel B). In comparison, in urban areas, employment increased from 42 to 49 individuals per 100 among the older age category, while unemployment increased from 38 to 44 individuals per 100, and NEET decreased from 75 to 64 individuals per 100 from 2010 to 2020.
Older workers are more prevalent in rural areas than in urban areas, yet there may be challenges in engaging with them in the labour market, or in labour reskilling opportunities. With a fall in the source for population in rural areas of Japan over the past decade, and many efforts to digitalize rural economies, bringing opportunities for re-skilling and engaging with older workers needs to be a critical part of encouraging innovation and innovation adoption in rural areas.
Note: Shares of age groups are calculated as the number of each category of labour force status over the total labour force between 15-64 years of age in the same labour force status. The age groups between 35 and 54 years are not presented in the figures. Changes in shares are the difference in shares from first year to the second year (the share in 2020 minus the share in 2010). DID area Densely Inhabited Districts defined by the Statistics Bureau of Japan that have 4,000 or more inhabitants per square kilometre, and consisting of adjacent basic unit blocks, etc. whose population is 5,000 or more at the time of the Population Census of Japan (for more information, see Annex 2.A).
Source: Statistics Bureau, Ministry of Internal Affairs and Communications, Japan
Women in rural Japan
Over the past decade, the participation of women in the labour markets have improved in Japan. Japan’s female employment rate rose from 60.7% in 2012 to 72.4% in 2022, surpassing the OECD average of 66.5% (OECD, 2024, p. 95[26]). However, the type of jobs and compensation for employment still demonstrates room for improvement, with a larger share of women in non-regular work contracts (not full time or not full year) than men in 2022 (53.4% of employed women as compared to 22% of men) (OECD, 2024, pp. 95-96[26]). The gender wage gap was also the fourth highest in OECD countries, with men’s wages close to 21 percent higher than women’s, at almost the double of the OECD average wage gap (at median earnings) (OECD, 2024, p. 95[26]).
In OECD countries, as in Japan, the differences between men and women’s employment rates have steadily declined (OECD, 2023[27]). In OECD countries, the gender gap in employment rates fell by 5.4 percentage points between 2000 and 2010 and continued to decline by another 2.2 percentage points between 2010 and 2021 (OECD, 2023, pp. 146-7[27]). In Japan, the difference between male and female employment rates fell 4.3 percentage points in 2010 as compared to 2000 and continued to decline by 7.3 percentage points from 2010 to 2020. In Japan, there is close to a 20-percentage point gap in 2010 and 12.6-percentage point gap in 2020.
In part, the rise of women’s employment rates (in absolute terms on a country level), has been aligned with an increase in childcare capacity over a similar period. Enrolment in childcare rose close to a quarter from the fiscal year of 2013 to 2021 despite the falling number of children overall, while at the same time, the employment rate for women aged 25 to 44 rose from 68% in 2013 to 79% in 2022, in a context of tight labour markets and short labour supplies (OECD, 2024, p. 85[26]). Despite excess capacity, 14% of Japan’s 1741 municipalities have waiting lists for childcare, although approximately 60% of the children on waiting lists were in urban areas (Nikkei, 2021[28]).
Across geographies, the share of men and women’s participation in employment is becoming more equal, with improvements in the female share of total employment (as a share of male and female totals) (Figure 1.14, Panel A) and the share of employed among women from 2010 and 2020 (Figure 1.14, Panel B). This trend has often been suggested as being due to access to older family members who may provide childcare support in rural places. In 2010, 43 % of total employment was women (42.5% in urban and 43.3 % in rural), while in 2020, this rate increased to 45% among which 45.5% was in urban and 45% in rural (Figure 1.14, Panel A). The share of females in unemployment is much lower than men’s in both rural and urban areas and still increased suggesting some tightening in the labour force. In 2020, the share of women among individuals who were unemployed was close to 40% in urban areas while this share was close to 35% in rural areas.
Among the active labour force (those in employment or unemployment), female employment was higher than men’s in both urban and rural areas, suggesting that once women search work, they tend to find it at a higher rate than men. This trend has been increasing over time. In 2010, the share of employed women among those that are active (employed and unemployed) was 96% while in 2020 the share grew to 97% (Figure 1.14, Panel B). In comparison, the share of male employment among the active labour force was 93% in 2010 and grew to 96% in 2020. The employment share of women in rural areas is higher than those in urban areas. In 2020, the employment share was 96% for urban women, while it was 97% for rural women.
Jointly, the female shares in total employment and female employment rate trends suggests that labour markets may be tightening for female workers. However, without further information about inactivity rates and hourly work (part-time versus full-time), it is unclear if low female shares of unemployment are due to inactivity, non-standard work (part-time, or part-year) or a higher demand for female workers.
Note: In panel A, the female share in total employment and unemployment refers to the share of women (as compared to men) in total employment and unemployment in urban and rural areas separately. In panel B, the active labour market does not include individuals not in employment, education or in training. In panel C, female share in total employment refers to the share of women (as compared to men) between the age of 15 to 64 in total employment in metropolitan and non-metropolitan areas separately. DID area Densely Inhabited Districts defined by the Statistics Bureau of Japan that have 4,000 or more inhabitants per square kilometre, and consisting of adjacent basic unit blocks, etc. whose population is 5,000 or more at the time of the Population Census of Japan (for more information, see Annex 2.A).
Source: Data on OECD countries from the OECD Regional Labour Database (Accessed May 2023). Data on Japan from the National Statistics Database.
Foreign labour in rural Japan
In many OECD countries, the immigrant labour force is considered an important aspect of a labour market that can enable innovation through access to different sets of skills. In Japan, the share of foreign-born workers was relatively low, and even lower in rural areas (Figure 1.15, Panel A). In the most recent year for which data is available, 2015, the proportion of foreign-born workers in Japan was 1.4%, with most of these workers coming from other Asian countries. In rural areas, the proportion of foreign-born workers was lower, at 1.05%, while in urban areas the proportion was higher, at 1.53%. These numbers, while still small, have shown improvements from 2010 with 0.97% in rural areas and 1.43% in urban areas.
The majority of foreign-born workers come from other Asian countries, with nearly 45% employed in the three most populus prefectures: Tokyo, Aichi (including Nagoya) and Osaka. More than half of the foreign workers are in the service sector, with only 8% of workers in the knowledge-intensive sector of Information, communication technology. Most foreign workers also work in relatively small firms, with 55% employed in firms with less than 100 workers (OECD, 2024[26]).
In OECD countries, the average share of foreign-born workers is 9% (Figure 1.15, Panel B).10 It ranges from the lowest, 1.3% in Mexico, to the highest at 47% in Luxembourg and 28% in Switzerland. Other high innovation countries such as Australia (19.5%), Canada (17.6%) and the United States (12.1%), have relatively high levels of foreign-born workers. In most OECD countries, there are higher shares of foreign-born workers in metropolitan regions (Figure 1.15, Panel B). In 2020, there was 2.1 percentage points more foreign-born workers in metropolitan regions (including the 2 categories of large metropolitan as well as metropolitan regions) as compared to non-metropolitan regions (all 3 categories of non-metropolitan regions) in OECD countries.
Over the past few years, the number of participants in temporary foreign worker programme have increased in Japan. Japan admitted over 20 000 foreigners for employment under the Specified Skilled Worker Programme (SSW) after 2019. The SSW, designed to address labour shortages in 12 eligible industries, was introduced in 2019 but border closure and slow roll-out of SSW testing in origin countries due to the COVID-19 pandemic limited arrivals from abroad (OECD, 2023[29]). This programme, however, is not specific to rural places, but rather targeted at industries. Examples of specific skills-based immigration programmes focusing on skills needs in rural places include those in the United States (OECD, 2024[9]), Canada (OECD, 2023[10]), and in Scotland, United Kingdom (OECD, 2023[30]) as well as a job-matching platform as recommended in (OECD, 2024[26]).
Note: For the Japanese statistics, foreign born shares are calculated as total foreign-born workers over the share of workers over the age of 15. The rest of the OECD statistics, foreign-born shares refer to foreign-born individuals over the total foreign-born resident population across all ages. DID area Densely Inhabited Districts defined by the Statistics Bureau of Japan that have 4,000 or more inhabitants per square kilometre, and consisting of adjacent basic unit blocks, etc. whose population is 5,000 or more at the time of the Population Census of Japan (for more information, see Annex 2.A).
Source: Author’s elaboration using Japan National Statistics Office and OECD (2022[31]) The Contribution of Migration to Regional Development. Accessed 02 August 2023.
Trends in population and innovation has long been an occupation in several OECD countries that in many cases can be addressed with innovation diffusion, as in Japan. However, unlike many countries, Japan actively monitors and creates strategies on bringing opportunities and innovation to rural places. It commonly participates in foresight projects for rural development. The foresight exercises are coordinated by the Council for Science, Technology and Innovation (CSTI), previously called the Council for Scientific and Technology Policy, that is a centre of government consultative body composed of the Prime Minister, relevant Ministers and experts. In the past, the council focused on long-term trends, promising technologies and the monitoring of the state of rural development (Yokoo and Okuwada, 2012[32]). Currently, the council itself is still active and met in 25 March, 2024 to discuss the topic, which was the evaluation of research and development trends and incentives, however, it is unclear whether they still meet to discuss challenges specifically for rural places.
Secondly, in November 2021, the Japanese Prime Minister published “The Vision for a Digital Garden City Nation” (further covered in Chapter 3) that aims to solve rural issues and improve rural attractiveness through digital technologies. The centre of government vision and associated strategy in 2022, aims to use digital technology to solve rural issues such as population decline, declining birth-rate and ageing population, and hollowing out of regional industries for example by promoting remote work and relocation to rural areas, establishing satellite offices and implementing remote medicine, distance education, automated driving and drones (OECD, 2023[33]). A similar initiative is also apparent in Korea that faces similar population decline and concentration challenges and has established a monitoring and foresight initiative to address the changing opportunities in rural places (OECD, 2022[34]).
In the vision and strategy in Japan, innovation adoption is the main driving force, however, innovation that goes beyond science and technology is not as apparent. Such insights are critical to understanding and preparing for demographic change.
The use of statistical definitions for rural places can vary substantially. When possible, the report prioritises consistency, however, in many cases, rural statistics may not be available either in a comparable way or for information needed. As such, this section outlines different statistical definitions used throughout the report.
Densely Inhabited Districts are designated in units of census basic unit blocks, and census enumeration districts if there are several census enumeration districts in a census basic unit block (hereinafter referred to as "basic unit blocks, etc."), and should meet the following criteria, in principle.
1. A district containing basic unit blocks, etc. with a population density of 4,000 or more per square kilometre, such districts being adjacent to each other in a municipality.
2. A district consisting of the above adjacent basic unit blocks, etc. whose population is 5,000 or more at the time of the Population Census of Japan.
With the idea that Densely Inhabited Districts represent urban areas, a basic unit blocks, etc. which has educational, cultural, and recreational facilities (e.g. schools, laboratories, shrines, temples, athletic fields), industrial facilities (e.g. factories, warehouses, business offices), and communal and social welfare facilities (e.g. public offices, hospitals, sanatoriums), and which is adjacent to the basic unit blocks, etc. mentioned under criteria 1), is also regarded as a district that meets the criteria 1). In this regard, however, population is concentrated in the remaining part excluding the area occupied by those facilities, or those facilities occupy more than half of the area of the whole district. The population census of Japan is conducted every 5-years, and the DID designated are updated as a result of the aggregation.11
Note: The map was created and provided by the Japanese Ministry for Agriculture, Farming and Forestry (MAFF) and only represents the Honshu, Shikoku, and Kyushu parts of Japan.
Source: 2020 Census ; (stat.go.jp)
The OECD classifies regions within the 38 member countries into 2 territorial levels that reflect the administrative organisation of countries. The 433 OECD large regions (TL2) represent the first administrative tier of subnational government, for example, the Ontario provinces of Canada. The 2 414 OECD small regions (TL3) correspond to administrative regions, with the exception of Australia, Canada and the United States. These TL3 regions are contained in a TL2 region, with the exception of the United States, for which the economic areas cross the states’ borders. For Costa Rica, Israel and New Zealand, TL2 and TL3 levels are equivalent (OECD, 2022[35]). See further description in the OECD Territorial grid.12
In 2019, the OECD published a new classification that is based on functional urban areas (FUA) within TL3s that incorporates density and the driving estimations for the time it takes to access dense metropolitan areas. The classifications are as the following:
Metropolitan regions are defined as having 50% or more of the regional population that lives in a functional urban area. This classification includes large metropolitan regions with a functional urban area that has over 1.5 million inhabitants, or a mid-sized metropolitan region that has a functional urban area with between 250k and 1.5 million inhabitants.
Non-metropolitan regions are defined as having less than 50% of the population living in an FUA with a population larger than 250 000 inhabitants (otherwise referred to as a metro region). The three types of non-metropolitan units include regions with access to a metropolitan region, non‑metropolitan areas with access to a small or medium-sized city (referred to as a non-metro near region) and a non-metropolitan region in remote areas (referred to as a non-metro far region).
Non-metropolitan regions near a mid-sized or large functional urban: These regions have 50% or more of the regional population that live within a 60-minute drive to a functional urban area of at least 250 000 inhabitants. An example of such regions includes Tyrolean Oberland in Austria (AT334), Montmagny in Quebec, Canada (CA2418), Jura in France (FRC22) and Nagasaki in Japan (JPJ42).
Non-metropolitan regions near a small functional urban area: These regions are regions with 50% or more of the regional population living within a 60-minute drive from a small or medium-sized city, defined as a functional urban area of between 50 000 and 250 000 inhabitants. Examples of these types of regions include the administrative district of Neufchâteau in Belgium (BE344), San Antonio in Chile (CL056), South Bohemia in the Czech Republic (CZ031), East Lancashire in the United Kingdom (UKD46) or Springfield in Illinois, United States (US158).
Remote rural, non-metropolitan regions: These regions have 50% of more of the regional population without access to a functional urban area within a 60-minute drive. Examples of these types of regions include Dordogne in France (FRI11), Yukon in Canada (CA6001) and Honolulu, Hawaii in the United States (US074).
The schematic breakdown is available in the table below.
Note : See further description in the OECD Territorial correspondence (http://stats.oecd.org/wbos/fileview2.aspx?IDFile=db68c5c3-5fd5-465c-b25b-b50aa14c2da1, accessed 24 May, 2024).
Source: (Fadic et al., 2019[36]), “Classifying small (TL3) regions based on metropolitan population, low density and remoteness”, https://doi.org/10.1787/b902cc00-en; and OECD (2024[2]) Regional Typologies. https://doi.org/10.1787/region-data-en, accessed 24 May, 2024.
In Japan, the unit of classification on the level of small regions (TL3) are prefectures. The classification of small regions (TL3) consists of 11 large metropolitan regions, 20 midsized metropolitan regions, 13 non-metropolitan regions new a midsized/large functional urban area with a population of over 250 000, and 3 non-metropolitan areas far from a mid-sized/large functional urban area with less than 250 000 individuals. It has no small regions classified as non-metropolitan remote regions. The classification for each of the 47 prefectures of Japan are further elaborated in Annex Table 1.A.2.
Source: OECD (2024[2]) Regional Typologies. https://doi.org/10.1787/region-data-en, accessed 24 May, 2024.
Using national statistics and the DID definitions, the total population in Japan increased in the post war period with close to 93 million inhabitants in 1960 to 126 million in 2020, an increase of 33 million inhabitants (Annex Figure 1.B.1, Panel A). The years of 1960 to 1975 demonstrated a peak of compound annual population growth rates. However, shortly after 1975, population growth started declining. Starting from 2010, the population of Japan started retracting in terms of absolute growth in numbers (Annex Figure 1.B.1, Panel B). Between 2010 and 2020, there was an absolute decline of approximately 1.9 million individuals in Japan, based on national statistics. The compound annual growth rate between 2010 and 2020 was negative (-0.1%) whereas the 5 previous decades showed was positive compound annual growth in population (0.5% from 1960 to 2010).
Despite a national level slow down, such a decline was not observed equally across Japan. Urban areas continued to experience population growth, and increasingly accounted for the largest share of the population (Annex Figure 1.B.1, Panel B). On the other hand, rural areas experienced a decline in both total population and population growth from the 1960’s onward. In 1960, the population in rural areas was 53 million individuals, or over half (56%) of the total population. From 1960 to 2020, the rural population fell by 15 million individuals, with a -0.5% compound annual growth rate from the initial year of 1960. In 2020, rural areas accounted for less than a third of the total population (30%). In comparison, urban population grew by 47 million or 1.3% in terms of compound annual growth rates over the same period of time.
In more recent years, the trend of absolute and relative decline in rural areas has remained on the same trajectory. From 2000 to 2020, the rural population was in decline, with an absolute fall of 5.6 million individuals, amounting to a -0.76% compound annual growth rate, and a fall in the share of the total population in rural areas by 5 percentage points (35% in 2000 and 30% in 2020). At the same time the population in urban areas continued to grow, with an absolute increase of 4 million individuals amounting to 1.3% compound annual growth rate, and an increase in the share of the total population in urban areas from 65% in 2000 to 70% in 2020.
Within the 2 decades from 2000 to 2020, we observe a slowdown of population growth at a national level and a decline in absolute numbers from 2010. The compound annual growth rate from 2000 to 2010 was 0.09%, while it was -.015% from 2010 to 2020. On a sub-national level, we observe a slowdown of population increases for urban areas, but an acceleration of population decline in rural areas. In rural areas, the compound annual growth rate over the period of 2010 to 2020 became increasingly negative (-1.02% from 2010 to 2020, as compared to -0.51% in the period of 2000 to 2010). In comparison, the growth of the compound annual growth rate over the period of 2010 to 2020 in urban areas (0.25%) slowed down as compared to the previous decade from 2000 to 2010 (0.39%).
Note: Figure A reports total population numbers for rural, urban and the total population. Figure B reports the distribution of total population in rural and urban areas (left-axis), as well as the year-on-year growth of population (right-axis). The dotted red line is a reference line for visual ease in comparing to changes in compound annual growth rates. DID area Densely Inhabited Districts defined by the Statistics Bureau of Japan that have 4,000 or more inhabitants per square kilometre, and consisting of adjacent basic unit blocks, etc. whose population is 5,000 or more at the time of the Population Census of Japan (for more information, see Annex 2.A).
Source: Statistics Bureau, Ministry of Internal Affairs and Communications, Japan
Source: Authors’ own elaboration and OECD (2023[6]) Regional Innovation, OECD Regional Statistics (database), https://doi.org/10.1787/1c89e05a-en.
Source: Statistics Bureau, Ministry of Internal Affairs and Communications, Japan
Source: Statistics Bureau, Ministry of Internal Affairs and Communications, Japan
Labour force characteristics, by age group (2010, 2015, and 2020)
Note: Employment, Not Employed, in Education, or Training (NEET) and Unemployment rates are calculated as a share of the total labour force (15-64).
Source: Statistics Bureau, Ministry of Internal Affairs and Communications, Japan
References
[3] EC-OECD (2024), STIP Compass: International Database on Science, Technology and Innovation Policy (STIP), https://stip.oecd.org (accessed on 12 June 2024).
[36] 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.
[28] Nikkei (2021), Vacancies at daycare centers increase in rural areas as new occupancies in 2021 fall below previous year’s figure for first time, https://www.nikkei.com/article/DGXZQOCC02CF90S1A201C2000000/ (accessed on 27 May 2024).
[16] OECD (2024), “Assessing the framework conditions for social innovation in rural areas”, OECD Local Economic and Employment Development (LEED) Papers, No. 2024/4, OECD Publishing, Paris, https://doi.org/10.1787/74367d76-en.
[9] OECD (2024), Enhancing Rural Innovation in Canada, OECD Rural Studies, OECD Publishing, Paris, https://doi.org/10.1787/a9919c66-en.
[26] OECD (2024), OECD Economic Surveys: Japan 2024, OECD Publishing, Paris, https://doi.org/10.1787/41e807f9-en.
[17] OECD (2024), OECD Economic Surveys: Japan 2024, OECD Publishing, Paris, https://doi.org/10.1787/41e807f9-en.
[2] OECD (2024), OECD Regions and Cities databases, http://oe.cd/geostats (accessed on 21 May 2024).
[1] OECD (2024), Population (indicator), https://doi.org/10.1787/d434f82b-en (accessed on 21 May 2024).
[15] OECD (2023), Enhancing Rural Innovation in Scotland, United Kingdom, OECD Rural Studies, OECD Publishing, Paris, https://doi.org/10.1787/33b8c803-en.
[30] OECD (2023), Enhancing Rural Innovation in Scotland, United Kingdom, OECD Rural Studies, OECD Publishing, Paris, https://doi.org/10.1787/33b8c803-en.
[10] OECD (2023), Enhancing Rural Innovation in the United States, OECD Rural Studies, OECD Publishing, Paris, https://doi.org/10.1787/22a8261b-en.
[29] OECD (2023), International Migration Outlook 2023, OECD Publishing, Paris, https://doi.org/10.1787/b0f40584-en.
[4] OECD (2023), “Japan”, in OECD Employment Outlook 2023: Artificial Intelligence and the Labour Market, OECD Publishing, Paris, https://doi.org/10.1787/f6705504-en.
[27] OECD (2023), Joining Forces for Gender Equality: What is Holding us Back?, OECD Publishing, Paris, https://doi.org/10.1787/67d48024-en.
[33] OECD (2023), OECD Regional Outlook 2023: The Longstanding Geography of Inequalities, OECD Publishing, Paris, p. Japan Annex, https://doi.org/10.1787/92cd40a0-en.
[24] OECD (2023), Old-age dependency ratio (indicator), https://doi.org/10.1787/e0255c98-en (accessed on 21 July 2023).
[22] OECD (2023), PISA 2022 Results (Volume II): Learning During – and From – Disruption, PISA, OECD Publishing, Paris, https://doi.org/10.1787/a97db61c-en (accessed on 11 June 2024).
[6] OECD (2023), “Regional innovation”, OECD Regional Statistics (database), https://doi.org/10.1787/1c89e05a-en (accessed on 18 July 2023).
[34] OECD (2022), Adapting Regional Policy in Korea: Preparing Regions for Demographic Change, OECD Rural Studies, OECD Publishing, Paris, https://doi.org/10.1787/6108b2a1-en.
[20] OECD (2022), “How the labour market and skills needs in Japan are changing during the COVID‑19 crisis”, in The New Workplace in Japan: Skills for a Strong Recovery, OECD Publishing, Paris, https://doi.org/10.1787/25350bd4-en.
[11] OECD (2022), OECD Regions and Cities at a Glance 2022, OECD Publishing, Paris, https://doi.org/10.1787/14108660-en.
[35] OECD (2022), OECD Territorial Grids, https://www.oecd.org/cfe/regionaldevelopment/territorial-grid.pdf. (accessed on 24 May 2024).
[31] OECD (2022), The Contribution of Migration to Regional Development, OECD Regional Development Studies, OECD Publishing, Paris, https://doi.org/10.1787/57046df4-en.
[7] OECD (2022), Unlocking Rural Innovation, OECD Rural Studies, OECD Publishing, Paris, https://doi.org/10.1787/9044a961-en.
[23] OECD (2021), Bridging digital divides in G20 countries, OECD Publishing, Paris, https://doi.org/10.1787/35c1d850-en.
[12] OECD (2021), “Changing skill needs in the Japanese labour market”, in Creating Responsive Adult Learning Opportunities in Japan, OECD Publishing, Paris, https://doi.org/10.1787/8f7fecd9-en.
[18] OECD (2019), Innovation, Agricultural Productivity and Sustainability in Japan, OECD Food and Agricultural Reviews, OECD Publishing, Paris, https://doi.org/10.1787/92b8dff7-en.
[21] OECD (2019), “Skills development and activation in Japan”, in Working Better with Age: Japan, OECD Publishing, Paris, https://doi.org/10.1787/9789264201996-7-en.
[25] OECD (2018), Working Better with Age: Japan, Ageing and Employment Policies, OECD Publishing, Paris, https://doi.org/10.1787/9789264201996-en.
[19] OECD (2016), Job Creation and Local Economic Development 2016, OECD Publishing, Paris, https://doi.org/10.1787/9789264261976-en.
[8] OECD (2016), OECD Territorial Reviews: Japan 2016, OECD Territorial Reviews, OECD Publishing, Paris, https://doi.org/10.1787/9789264250543-en.
[14] OECD (Forthcoming), Building the Conditions for Rural Innovation, OECD Publishing.
[13] OECD/EC-JRC (2021), Access and Cost of Education and Health Services: Preparing Regions for Demographic Change, OECD Rural Studies, OECD Publishing, Paris, https://doi.org/10.1787/4ab69cf3-en.
[5] OECD/Eurostat (2018), Oslo Manual 2018: Guidelines for Collecting, Reporting and Using Data on Innovation, 4th Edition, The Measurement of Scientific, Technological and Innovation Activities, OECD Publishing, Paris/Eurostat, Luxembourg, https://doi.org/10.1787/9789264304604-en.
[32] Yokoo, Y. and K. Okuwada (2012), “Validity of foresight derived from the evaluation of past activities in Japan”, International Journal of Foresight and Innovation Policy, Vol. 8/4, p. 296, https://doi.org/10.1504/ijfip.2012.049776.
← 1. This is according to the OECD Enhancing Rural Innovation Survey responses by Japanese government officials.
← 2. Japan Ministry of Agriculture, Fisheries and Forestry [no date]. Agricultural type typology. (maff.go.jp), accessed 28 May 2024.
← 3. The average regional population refers to the average of all regions in Japan with each region represents one sample. In contrast, populated weighted averages represent the whole population of Japan.
← 4. “An innovation is a new or improved product or process (or combination thereof) that differs significantly from the unit’s previous products or processes and that has been made available to potential users (product) or brought into use by the unit (process)… A business innovation is a new or improved product or business process (or combination thereof) that differs significantly from the firm's previous products or business processes and that has been introduced on the market or brought into use by the firm.”
← 7. Within the public works and services sector, over half (54%) of all jobs in rural areas were in the medical care and social welfare sectors, followed by the education sector (20%) in 2020.
← 9. Within the trade and services sector, that accounts for the majority of jobs in both rural and urban areas, there were some variations between places, but trends were similar. In rural areas, wholesale retail and trade accounted for 35% of all trade and services, and other unclassified services accounted for the 15% of jobs in 2020 (Figure 2.8, Panel A). Similarly, in urban areas, wholesale retail and trade accounted for 31% of the trade and services, and other unclassified services accounted for 13% of the sector. The largest differences was in the information and communication sector that accounted for 9% of trade and services jobs in urban areas, but only 2% of trade and services jobs in rural areas.
← 10. This is the average for statistics available within small regions (TL3).
← 11. Statistics Bureau Home Page/What is a Densely Inhabited District?, accessed 23 May 2024; and https://www.e-stat.go.jp/en/stat-search?page=1&toukei=00200521 , accessed 23 May 2024.
← 12. http://stats.oecd.org/wbos/fileview2.aspx?IDFile=cebce94d-9474-4ffc-b72a-d731fbdb75b9 , accessed 24 May, 2024