The NSPA regions are experiencing growth and leading in innovation and the green transition, despite challenges like population decline and ageing demographics. Key strengths include a rising younger, educated workforce, growing GDP per capita, and improved labour productivity in sectors like agriculture and professional services. Innovation and exports outperform OECD benchmarks, with trade surpluses and advancements in digital accessibility. The green transition is strong, with low GHG emissions and renewable energy dominance. However, challenges remain, including slow overall productivity growth, low firm density, and uneven digital progress. Addressing these requires targeted, place-based policies to enhance competitiveness, manage demographic shifts, and ensure equitable resource access.
Navigating Global Transitions in European Arctic Regions
Abstract
The Northern Sparsely Populated Areas (NSPA), 14 regions located in northern Finland, Sweden and Norway, is a collaborative Network established in 2008 with the aim to raise awareness within European Union (EU) institutions regarding the shared challenges and circumstances encountered by these regions (Figure 2.1).1 The NSPA Network is composed of two key bodies: the Steering Committee and the Co‑ordination Group. The Steering Committee includes the political leaders (chairs) of the regions within the network. The Co‑ordination Group is made up of three European Offices based in Brussels: the East and North Finland EU Office, North Norway European Office, and North Sweden European Office.
Note: The NSPA regions are indicated in grey in the map.
Source: Author’s elaboration.
The Northern Sparsely Populated Areas (NSPA) includes:
the seven northernmost and eastern regions of Finland (Lapland, Northern Ostrobothnia, Central Ostrobothnia, Kainuu, North Karelia, Pohjois-Savo and South-Savo),
the four northernmost regions of Sweden (Norrbotten, Västerbotten, Jämtland Härjedalen, and Västernorrland), and
the three northernmost regions of Norway (Nordland, Troms and Finnmark).
The OECD regional typology streamlines the comparability of regional data across OECD countries. It categorises geographic units into two levels within each member country: i) large regions (TL2), typically representing the primary administrative tier of subnational government; and ii) small regions (TL3), which aggregate local administrative units. TL3 regions are further divided into regions with varying accessibility according to the functional urban areas (FUA) framework. They include Metropolitan Regions (MR) and Non-metropolitan Regions (NMR) that are further classified into 5 sub-categories:
Large metropolitan regions (MR-L) with a FUA that is greater than 1.5 million inhabitants.
Metropolitan mid-sized regions (MR-M) with a FUA that is between 250 000 and 1.5 million inhabitants.
Non-metropolitan regions near a mid-sized or large FUA (NMR-M) with more than 250 000 inhabitants.
Non-metropolitan regions near a small FUA (NMR-S) with between 50 000 and 250 000 inhabitants.
Non-metropolitan remote regions that are far from a FUA (NMR-R).
According to this definition, of the 14 NSPA regions, eleven are defined as NMR-R (Lapland, Central Ostrobothnia, Kainuu, North Karelia, South-Savo, Norrbotten, Jämtland Härjedalen, Västernorrland, Nordland, Troms and Finnmark), while three are defined as NMR-S (Northern Ostrobothnia, Pohjois-Savo, Västerbotten).
Source: Brezzi, M., L. Dijkstra and V. Ruiz (2011), "OECD Extended Regional Typology: The Economic Performance of Remote Rural Regions", OECD Regional Development Working Papers, No. 2011/06, OECD Publishing, Paris, https://doi.org/10.1787/5kg6z83tw7f4-en. 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
In addition to the 14 NSPA regions, there are 38 non-NSPA regions in Finland, Sweden and Norway (excluding Svalbard and Jan Mayen). NSPA are distinct from other regions in Finland, Norway and Sweden (i.e. non-NSPA) and OECD regions with similar regional characteristics. On average, NSPA regions are larger than non-NSPA regions and similar to non-metropolitan OECD regions. They also tend to have a relatively lower average population.2 For instance, the average land cover surface area of the NSPA regions is close to 41 000 0003, as compared to close to 32 000 000 for the average non-metropolitan region that are far from a function urban area, often considered a remote rural area (NMR-R), and substantially larger than 9 000 000 for the average non-metropolitan region near a small functional urban area (NMR-S) (Table 2.1). In 2022, the NSPA had a lower density than the average OECD NMR-R and OECD NMR-S region, and in terms of total population, close to 2.5 times lower population than non-NSPA regions and a third less population than the average OECD NMR-S region, despite a marginally larger average population than the most remote regions of the OECD.
NSPA regions face an acute ageing challenge. On average, the NSPA regions have an older population than non-NSPA regions and OECD regions with similar characteristics based on statistics. In the NSPA, the average regional elderly dependency ratio (EDR), or the share of the elderly population over 64 years of age to those between 15-64 years of age, was 42.09 (or 39.71 weighted average) in 2022. This is higher than the EDR of non-NSPA regions of 36.63 (or 31.71 weighted average). The NSPA’s average EDR was close to 10 units higher than both the OECD NMR-R regions (32.08 regional average and 34.29 weighted average) and OECD NMR-S regions (31.59 regional average or 32.13 weighted average).
At the same time, there are slightly lower shares of very young people, aged less than 15, in NSPA regions than in non-NSPA and in other comparable OECD regions. For instance, youth dependency is on average lower than in the non-NSPA regions of Finland, Norway and Sweden. In 2022, the youth dependency ratio (YDR) was 26.85 (or 26.94 weighted average) in non-NSPA regions. This was higher than the NSPA youth dependency ratio regional average of 26.02 (or 26.24 weighted average), similar to the OECD NMR-S regional average of 26.24 (or 24.94 weighted average), but lower than the regional and weighted averages of the OECD NMR-R regions (30.73 regional average and 28.17 weighted average).
Note: All statistics are from 2022, except for GDP per capita which is are from 2020. GDP per capita are measured in 2015 USD PPPs (purchasing price parities). Values in parenthesis are weighted averages. In the case of surface, total land mass surface is reported and therefore territory that includes water mass is not included. The elderly dependency ratio is the number of individuals above the age of 65 over the population between 15-64 years of age, and the youth dependency ratio is the number of individuals below the age of 15 over the population between 15-64 years of age. For a discussion on the use of weighted versus unweighted statistics for use in regional comparisons, please refer to the Box below (Box 1.2.). NSPA regions are defined in Box 1.1. Non-NSPA region include all other TL3 regions within the 3 NSPA countries of Finland, Norway and Sweden that are not one of the 14 TL3 NSPA regions. For consistency and temporal harmonisation challenges in Troms and Finnmark due to the recent administrative changes, the two regions are combined as one region when calculating NSPA regional averages, resulting in 13 units for calculating NSPA regional averages. The NMR-S classification refers to non-metropolitan regions (TL3) near a small FUA (NMR-S) with between 50 000 and 250 000 inhabitants. The NMR-R classification refers to non-metropolitan remote regions that are far from a FUA (NMR-R).
Source: OECD Regional Statistics
In general, there are two main methods of calculating regional averages: one gives the same weight to each region (regional averages, or unweighted averages), while the other takes into account the relative size, e.g. weighted by population, of each region (weighted averages).
Regional averages or unweighted averages are calculated by simply averaging the values from different regions, without considering the size or importance of each region. This method gives equal treatment (or weight) to each region, regardless of its population or economic size. Weighted averages, on the other hand, take into account the size or importance of each unit (in this case, each region) by assigning different levels of importance (weights) to the values from different regions.
The choice of method can lead to different estimates. Benchmarking using weighted averages, results in a comparison that is skewed towards the more densely populated or significant areas, which are often capital regions. For instance, in Finland, Helsinki-Uusimaa is the region that includes the capital city and accounts for over 30% of the national population and close to 40% of the national GDP. The regional average of GDP per capita Finland is USD 39 532, while the weighted average is USD 44 671. The upward bias of the weighted estimate is due to the strong performance of the capital region (which can be considered as an outlier, beyond the 99th percent confidence interval.) For example, the GDP per capita of Helsinki-Uusimaa is USD 57 607. The second highest performing region is Kymenlaakso at USD 42 955 GDP per capita. The second highest performing region is below the weighted average estimate, whereas the unweighted average is closer to the median value. The use of an unweighted regional average allows a comparison to be made between regions rather than with the more densely populated regions, which tend to have higher levels of production and population.
As such, unweighted averages can often offer a more relevant view of regional comparisons, as they are not influenced by the larger size or economic output of major urban areas.
Both statistical choices are complementary and provide useful information. However, for the purpose of the focus on regional development, a priority has been placed on regional comparisons. Weighted statistics are provided only in the introductory table for reference.
The larger distances and lower density of NSPA regions, relative to similar regions, also comes with opportunities. For instance, economic activity per capita seems to be remarkable high as compared to similar OECD regions. The GDP per capita, is a little over USD 41 000 (USD 42 000 in terms of weighted averages). It is higher in NSPA regions than on regional averages in OECD NMR-S and OECD NMR-R. Furthermore, based on regional analysis, natural resource-based activities and transition towards the green transition are prominent in the areas. Indeed, in most cases land-use planning prioritises nature-based activities including nature conservation and traditional husbandry activities, while indicators of the green transition such as the emissions of greenhouse gases (GHG) are declining in NSPA regions, more than in non-NSPA regions. There are also notable opportunities in the growing services sectors in several of the NSPA regions, despite only a handful of NSPA regions attaining top levels of high-tech innovation performance in the region.
This chapter considers the NSPA as a single unit, while drawing on examples of specificities in the 14 regions. All references to non-NSPA regions refer to the rest of the TL3 regions in the NSPA countries of Finland, Norway and Sweden, that are not included as the 14 designated NSPA regions. The rest of this chapter will focus on thematic areas including demographic, economic, competitiveness, social indicators, accessibility and the green transition.
The NSPA region has a relatively low population. In 2022, the population in the NSPA regions was on average 2.5 times lower than in non-NSPA regions, and a third lower than OECD NMR-S regions (Figure 2.1). In addition to a relatively low population, over the last two decades, the NSPA region is also declining in population. From 2001 to 2022, the NSPA population observed an aggregate decline of close to 2%, or a 0.01% decline in compound annual growth rates (Figure 2.2, panel A). In comparison, all other benchmark regions increased in population size. The non-NSPA regions observed an aggregate growth rate of close to 12% (or a 0.76% increase in terms of compound annual growth rate, CAGR), while OECD NMR-S regions grew by 10% (or a 0.31% CAGR), and the OECD NMR-R regions grew by 8% (0.43% CAGR).
The population decline in the region coincided with changes in the age demographics in the NSPA region. In 2003, the prime working age population (35 to 49 years of age) accounted for a third of the population (33%), while the older group of working age individuals (50 to 64 years of age) 4 accounted for 30% of the population (Figure 2.2. , panel B). The youngest, post-secondary school, working age population (20 to 34 years of age) accounted for 27% of the population. The remaining secondary school and early post-secondary aged group (15-19 years of age) accounted for the remaining 10% (not pictured).5
Over the course of close to two decades, from 2003 to 2022, the NSPA observed a quasi-reversal of the age hierarchy. This emerged primarily through a relatively large loss of the prime age working population and a relatively steady share, despite some limited growth, of older working age individuals. By the end of 2022, the prime working age population, 35 to 49 years of age (29%), accounted for the lowest share of the age-based distribution. The prime working age population dropped 3 percentage points from 2003 to 2022. This age group also observed the highest fall in terms of annual changes, reaching 0.1% decline in the compound annual growth rate (CAGR). The age group declined over the majority of the period from 2003 to 2022 but demonstrated a more recent reduction in the negative growth starting 2018, leading to positive growth only in the 2021 to 2022 period. While this is the largest change to the age composition of NSPA, this trend was also occurring in other regions. The prime age working group (35 to 49 years of age) saw declines in most of the benchmarked regional groups. For instance, in the non-NSPA regions, the share of the prime aged working group (35-49) decreased from 32% to 31%, in the OECD NMR-R regions the share decreased from 33% to 30%, whereas the share remained the same in the OECD NMR-S regions (32%) (Figure 2.2. , panel B). This demographic shift also coincided with a sharp rise in the Elderly Dependency Ratio (EDR). In 2002, the NSPA and non-NSPA regions had nearly identical EDRs, with around 26 people aged 65+ per 100 working-age individuals (15-64). Over time, however, the NSPA aged more rapidly, with an average CAGR of 2.4%, reaching 42 in 2022, while non-NSPA regions saw a slower increase (1.7% CAGR), ending at around 37 (Figure 2.2, panel C).
Unlike in 2003, the largest share of the working-age population in 2022 was the older group (50–64 years), making up 32% (Figure 2.2, panel B). This group became the largest after its share increased by 2 percentage points since 2003. It occurred despite a negative annual compound growth of 0.03% of the older aged working population (Figure 2.2. , panel C), and can be explained in part by a relatively smaller fall in the total number of older aged individuals than on average in the rest of the population. The trend of increasing shares of older working age individuals was also observed in OECD NMR-S, NMR-R regions as well as to a smaller extent in the non-NSPA regions. In the OECD NMR-S regions, the share of older working age individuals increased from 25% in 2003 to 31% in 2022. In the OECD NMR-R regions, the share of the older working age individuals increased from 26% in 2003 to 32% in 2022. However, the share of the older working age individuals in non-NSPA regions increased only by 0.5% (from 28.9% to 29.4%) (Figure 2.2. , panel B). The NSPA also experienced faster population aging compared to its benchmark regions. In 2002, OECD NMR-S and NMR-R regions had lower EDRs, starting at roughly 22 and 21, respectively. By 2022, these had risen to around 32 in both regions, still trailing behind both the NSPA and non-NSPA regions (Figure 2.2, panel C).
Contrary to the falling growth rates of the oldest and prime age populations, the share of the younger post-secondary school working age individuals increased from 2003 to 2022 by 3 percentage points, from 27% to 30% (Figure 2.2. , panel B). This change amounted to the only absolute increase in the number of individuals in the age group observed among the NSPA age groups. The number of 20-34 year olds increased by 0.15% annually, amounting to an aggregate period increase of 3% (Figure 2.2. , panels B and C). The positive growth in terms of year-on-year changes in this age group occurred only between the years of 2009 and 2017, with relatively smaller losses that the other age groups in the remaining years in the period between 2003-2022. This increase was stronger than the increase observed in non-NSPA regions (1 percentage point, from 30% in 2003 to 31% in 2022 or 0.78% annual increase). The NSPA and non-NSPA regions increases came in opposition to trends in the wider OECD non-metropolitan regions. The OECD NMR-S and NMR-R regions observed a decline in the share of this age group, from 32% to 28%, or -0.73% annually in NMR-S and from 30% to 28%, or -0.21% annually in NMR-R.
Average population growth (Base=2001) and shares of age groups among working age population (15-64)
Note: Panel A in the figure above shows the aggregate annual population growth of regions from 2001-2022, from the base year of 2001. The OECD average (36 countries) is presented and broken down by regional typology. All averages give equal weight to each individual region, regardless of the country in which it is located. In addition, Svalbard and Jan Mayen in Norway were excluded from the analysis. Panel B shows the share of the different working age groups, including the oldest (50-64 years of age), the middle (35-49 of age) and the youngest (20-34 years of age) age groups as a portion of the total population in the working age group (between 15-64 years of age) in the region. The youngest group (15-19) is excluded due to the fact that a larger share of this age population is still in schooling years in many OECD countries. Totals are used. Panel C shows the Elderly Dependency Ratio (EDR) for NSPA and benchmark regions.
Source: OECD Regional Indicators
Most non-NSPA regions are observing annual population increases over the period of 2003 to 2022, along with an increase in the share of elderly population (65+, relative to the working age population 15-64 years of age), as illustrated in Figure 2.3. . The relationship between an increase in the annual population change and an increase in the elderly population rate is negative, based on the weighted linear regression fitted to regional observations in Figure 2.3. . This means that population decline is also associated with an ageing population. The lower regional population growth rates are associated with higher regional ageing trends.6
The double challenge of a decreasing and ageing population is also observed in NSPA regions. As population declines, the population gets older, or in other terms, has relatively higher shares of individuals aged 65 and over. However, regional performance in the NSPA demonstrates additional challenges. NSPA regions have i.) lower population levels (size of weighted co‑ordinates in Figure 2.3. ), ii.) population growth that is more often negative or at lower levels than in non-NSPA regions (Figure 2.3. , quadrant II)7, and iii.) increases in elderly population ratios is often more positive than in non-NSPA regions (Figure 2.3. , quadrant I and II).
Only 6 NSPA regions our of 14 (46%) are growing in terms of population over the period of 2003 to 2022, where are there are 34 out of 38 (89%) non-NSPA regions are growing in terms of population. South Savo experienced the biggest decrease in population with a compound annual growth rate of -0.81% from 2003-2022, while the region with the highest population growth in the NSPA was Northern Ostrobothnia with a compound annual growth rate of 0.49% from 2003 to 2022. In comparison, the largest decline in the population growth was in Kymenlaakso with 0.51% decline from 2002 to 2022, and the highest growth was in Oslo with a 1.52% increase in population over the same period.
In terms of ageing trends, on average both NSPA and non-NSPA regions saw an increase in the elderly age population (65+), however, NSPA regions saw a larger increase of the ageing population. In NSPA regions, the largest increase in the elderly age population was in Kainuu with a 3.41% increase over the period of 2003-2022, while the lowest was in Jämtland Härjedalen that saw an increase of the elderly age population of 1.21% (which was close to the median increase of regional EDR rates of non-NSPA regions). On the other hand, in non-NSPA regions, one region, Oslo, showed a 0.38% decline in the elderly age population rate (jointly with the highest positive population growth), while the median region observed an increase of the elderly age population of 1.45%, and the highest regional increase was in Päijät-Häme (3.30%).
Weighted regional correlation between annual changes in population and older-to-younger age ratios (TL3)
Note: Data points on the TL3 level are represented in either NSPA or non-NSPA groups. Fitted lines and circle size are weighted by population in 2022. The y-axis represents the compound annual growth rate (CAGR) of population from 2003 to 2022. Values above the 0 line on the y-axis reflect an increase in the annual population growth of TL3 region in the period. The x-axis represents the compound annual growth rate (CAGR) of the elderly dependency ratio from 2003 to 2022. The elderly dependency ratio is the number of individuals above the age of 65 over the population between 15-64 years of age.
Source: OECD Regional Statistics
In sum, there are 5 distinct patterns in the shift of demographics that are notably unique to the NSPA regions as compared to other regional groupings.
First, the NSPA region has a relatively low population, and is declining while other regions such as non-NSPA, OECD NMR-S and OECD NMR-R regions are increasing.
Second, there was a relatively strong fall in the prime working age group (35-49 years of age) in the NSPA regions. However, a similar trend was also observed in non-NSPA and OECD NMR-R regions, while the share of the prime working age group in the OECD NMR-S regions remained the same. The age group declined over the majority of the period from 2003 to 2022, but showed progress with an upward trend starting with reduction in the negative growth in 2018 resulting in positive growth only in the 2021 to 2022 period.
Third, the share of the oldest working age group (50-64 years of age) is the highest and increasing relative to other age groups in the NSPA regions, making it the group representing the largest share of working age individuals in 2022. This increase is similar to trends in non-NSPA regions and OECD NMR-S regions, albeit to a larger extent, and lower than increases in OECD NMR-R regions.
Fourth, contrary to trends in other benchmarked regional groups, the younger working age group (20-34 years of age) is increasing in the NSPA regions relative to other age groups. This is stronger than increases in the share of the age group in non-NSPA regions, and in other OECD NMR-S and NMR-R regions. It is due to an absolute increase in the number of individuals in the age group, which amounted to a 0.15% annual or 3% aggregate increase from 2003 to 2022. This was the only increase in the number and share of an age group observed in the NSPA.
Lastly, with the exception of Oslo, all regions in NSPA and non-NSPA are facing demographic challenges related to an ageing population. However, more NSPA regions are also facing population decline, lower or negative population growth and stronger ageing trends than non-NSPA regions. In NSPA regions, a higher share of regions are jointly losing population and ageing underling that the double challenge of demographic change is more critical for NSPA regions than non-NSPA regions.
The analysis suggests that population decline and ageing is a challenge for NSPA regions. Based on age-based analysis, the fall in the population is primarily due to a loss in prime working age individuals over the last 2 decades in both NSPA and non-NSPA regions, but more so in NSPA regions. For non-NSPA regions as the workforce ages, it can more easily replenish itself through a higher level of population growth.
Yet there are positive signs of growth in the NSPA population over the last few years. In particular, this is due to increases in the upturn in the prime working age in the last few years, and increases in the young working age groups, which outperforms non-NSPA and other OECD regions. As such, despite the fact that the NSPA is still in population decline, relatively old, and has higher (and increasing) elderly dependency ratios (beyond the older age working categories), the working age population can potentially rejuvenate itself if it provides enough opportunities for the younger working age groups in the region. While there are signs of growth of prime and young working age groups, the NSPA regions still need to consider demographic changes and challenges associated to delivering services in the context of a shrinking population with higher levels of older workers and elderly dependency ratios.
The NSPA region, as a whole, has a relatively strong economic performance. The regional average GDP per capita in the NSPA amounted to USD 41 284 in 2020, an increase of over USD 10 000 from the regional GDP per capita average in 2001 (Figure 2.4, panel A). The GDP per capita in the NSPA was higher than those in similar OECD NMR-S and NMR-R regions, despite remaining lower that the non-NSPA average. The regions within the NSPA with the higher GDP per capita were Norrbotten, Troms and Finnmark and Nordland. The regions within the NSPA with the lowest GDP per capita were South Savo, North Karelia and Kainuu.
In addition to strong GDP per capita in 2020, the NSPA also observed higher growth per capita than other comparable OECD and non-NSPA regional averages. From 2001 to 2020, the NSPA region grew an aggregate 30%, or a 1.35% annual growth from 2001 to 2020 (Figure 2.4, panel B). This was a higher growth than any other regional average, including those from non-NSPA regions. The non-NSPA region grew an aggregate 18%, or 0.85% annual growth from 2001 to 2020, the OECD NMR-S grew an aggregate 25%, or 1.12% annual growth, and the OECD NMR-R grew an aggregate 15%, or 0.69% annual growth.
Note: Panel A shows the total regional GDP per capita (USD per head, constant prices, constant PPP, base year 2015) from 2001 to 2020 for all NSPA regions as well as the average for NSPA, non-NSPA, OECD NMR-S and OECD NMR-R. The OECD average includes over 25 countries, broken down by regional typology. All averages give equal weight to each individual region, regardless of the country in which it is located. In addition, Svalbard and Jan Mayen in Norway were excluded from the analysis. Panel B shows the trend over the whole period for the selected benchmarks.
Source: OECD Regional Indicators
In addition to aggregate increases, most NSPA regions saw an increase in their national ranking in terms of GDP per capita (Figure 2.5. , panel A). As compared to GDP per capita in 2001, GDP per capita in 2020 in all but 2 NSPA regions increased relative to national regional averages. This suggests that the relative inequalities between regions within countries, in terms of regional GDP per capita, has decreased for NSPA regions. Only Jämtland Härjedalen and Västermorrland saw a decrease in their relative national ranking (despite having higher levels in 2020, than 2001 as depicted in Figure 2.5, panel A). However, in the case of Västermorrland, the region was over-performing as compared to the national regional average in 2001 by 10%, and in 2020 the region is now at the average regional GDP per capita in 2020. In the case of Jämtland Härjedalen, a strong tourism region, the year of 2020, when travel and tourism became restricted because of the COVID-19 pandemic may have impacted the performance substantially that year. A few regions over-performed in terms of national regional averages. Lapland, Central Ostrobothnia and Vasterbottens moved from having a GDP per capita that was lower than the national regional average GDP per capita by between 10-16% in 2001, to a GDP per capita that was higher than the national regional average, by 1-2.5% in 2020.
While changes in regional average differences suggest a positive outlook for NSPA regions, comparisons with the national weighted averages reflect substantial differences between NSPA country TL3 regions and capital regions (Figure 2.5. , panel B). In part, this is due to substantially larger economies in capital regions. As many NSPA and non-NSPA regions still remain below national weighted regional averages, there is still room to reduce inequalities within NSPA countries (for both NSPA and non-NSPA regions).
Note: Panel A shows the deviation in GDP per capita of NSPA regions from the respective national average in 2001 and 2020. Panel B presents the regional values for 2020 across all regions in Norway, Sweden, and Finland, with the weighted national average highlighted.
Source: OECD Regional Indicators
Along with growth in GDP, labour productivity is increasing over time in NSPA regions and non-NSPA regions. Aggregate average regional productivity grew by 0.46% in NSPA regions from approximately USD 74 000 in 2008 to USD 79 000 in 2020 (Figure 2.6. , Panel A). In comparison, aggregate regional productivity grew more in non-NSPA regions, by 0.56% between 2008 and 2020, from approximately USD 76 600 in 2008 to USD 82 000 in 2020. Productivity gains between 2008 and 2020 in both the NSPA and non-NSPA regions, were primarily due to upgrading of products and processes within sectors rather than increasing reallocation of resources (such as labour and capital) between sectors. The upgrading and upskilling of resources within sectors is depicted by the positive increases associated with the within factor of the productivity growth decomposition, found in Figure 2.6. (Panel B) . This growth is in opposition to the losses associated with the between factor of the productivity growth decomposition. In the NSPA region, the relocation of resources between sectors has a relatively stronger downward pressure on productivity than in the non-NSPA regions.
Labour productivity (in USD) and Labour productivity growth (in %)
Note: The figure illustrates labour productivity increases and the contribution of within-sector versus between-sector shifts to overall productivity gains in NSPA and non-NSPA regions.
Source: OECD Regional Indicators
Productivity growth trends grew differently by sector in NSPA and non-NSPA regions. The agricultural (A), real estate (L), professional services (M-N), mining (B, D, E), information and communication (J), construction (F) and financial services (K) sectors all showed positive productivity growth (Figure 2.7, panel A). On the other hand, other services (R-U) sector observed no growth, while the trade and accommodation sector (G, H and I), public administration (O-Q) and the manufacturing sector (C) all contributed negatively to productivity growth. In comparison to non-NSPA regions, the directional (positive and negative) trends were relatively similar, albeit at different levels of growth (Figure 2.7, panel B).
The highest increases in overall productivity were due to improvements in the upgrading and upskilling of resources in the agricultural, forestry and fishing sector (Figure 2.7, panel A). This sector, alone contributed to a +50% improvement of productivity from 2008 to 2020. However, these productivity growth increases in gross value added happened at the same time as decreases in employment, suggesting that productivity increases in this sector were likely in part due to labour-substitution. In comparison, in non-NSPA regions, the agriculture, forestry and fishing sector contributed positively to productivity growth, but to a lesser extent (+6%) (Figure 2.7, panel B). Likewise, productivity growth was likely due to labour-substitution effects as labour decreased with increases in gross value added.
The professional services sector was the second highest positive contributing sector to productivity growth (+42%), if we exclude the real estate sector (Figure 2.7, panel A)8. Most of this growth was due to increasing reallocation of resources, such as capital investments or labour to sector, however, additional resources into the sector were also productivity increasing. Indeed, productivity growth was labour-inducing and output enhancing for this sector between 2008 and 2020 and largely aligns with the increasing trend of tertiarisation of OECD regions. In comparison, in non-NSPA regions, the professional services sector also grew due to similar reasons, but to a lesser extent than in NSPA regions (+30) (Figure 2.7, panel B).
The mining9 sector is a relatively important industry in the NSPA regions that contributed to a +37% improvement in the overall productivity growth of the NSPA region from 2008 to 2020 (Figure 2.7, panel A). This sector’s growth was primarily due to reallocation of resources to the sector which aligned with an increase in both employment and gross value added. The mining sector similarly increased in non-NSPA regions, but to a larger extent (57%), and demonstrated that more of the productivity growth was due to upgrading and upskilling within the sector, at the same time as employment and gross value added increased (Figure 2.7, panel B).
The other notable sectors of growth include the information and communication sector, that observed a +28% aggregate growth from 2008 to 2020 (Figure 2.7, panel A). This sectors growth was primarily due to increases in the upgrading of resources or upskilling of labour, and coincided with a decline in employment and an increase in gross value added, suggesting labour-replacing processes leading to productivity gains in the sector. In comparison, in non-NSPA regions, the information and communication sector contributed by +44% to overall productivity growth, primarily driven by upgrading and upskilling within the sector (Figure 2.7, panel B).
There were two notably strong losses in terms of labour productivity in NSPA regions. The strongest was in the manufacturing industry. The manufacturing industry contributed negatively (-73%) to productivity growth between 2008 and 2020 (Figure 2.7, panel A). This decline occurred because of a strong loss of resources in the sector (between component). The decrease in productivity came with strong losses in employment and strong losses in gross value added in the sector, aligning with a transition away from the manufacturing sector. In the non-NSPA region, manufacturing similarly contributed negatively to productivity growth (-69%) and occurred similarly due to strong losses of resources in the sector, and a drop in employment and gross value added (Figure 2.7, panel B).
The second notable sector that contributed negatively is the public sector, with a -56% contribution to total productivity growth (Figure 2.7, panel A). The sector itself is a particularly relevant sector for NSPA regions because of the large number of workers it employs and the fact that delivering public services often costs more in remote and sparsely populated areas. However, gross value added in the sector is calculated differently than the private sector (based on public expenditures or wage bills), and as such it does not respond directly to market forces. The increased cost of providing services to sparsely populated regions also explains the high costs and the large number of workers. Over the period of 2008 to 2020, the decline in the ratio of public expenditures to workers was due primarily to less allocation of resources to the sector, and coincided with less expenditure (lower value added) but more employment. This may suggest that some cost cutting measures (or privatisation) may have taken place either in terms of infrastructure or in terms of personnel costs to compensate for the large increase in employment. These cost cutting measures could have compensated for increasing costs in new hires. In comparison, the non-NSPA regions, also saw a decrease in public sector productivity, but to a much lower extent (-9%). This was also primarily due to losses of efficiency within the sector, and coincided with a large increase in employment and a much smaller increase in value added, or expenditures (Figure 2.7, panel B).
Note: The figure illustrates the contribution of within-sector versus between-sector shifts to overall productivity gains in 11 sectors in the NSPA and non-NSPA regions.
Source: OECD Regional Indicators
In sum, there are 3 main trends observed in the NSPA regions when it comes to general and sectoral economic trends:
GDP per capita is high and is growing strongly in the NSPA regions. The region performs better than other OECD benchmark regions, in both levels and growth from 2003 to 2020.
Aggregate productivity increased in NSPA regions between 2008 to 2020, but less than the increase in non-NSPA regions. Between 2008 to 2020, productivity increases were due in a larger part to improvements associated with upgrading or upskilling within the NSPA region, rather than additional reallocation of resources between sectors in the NSPA region.
On a sectoral level, the strongest improvements came from the agricultural, forestry and fishery sector. However, productivity growth in this sector was labour-substituting. This trend was stronger than in non-NSPA regions. There were also strong improvements in the professional services sector. On the contrary to the improvements in the agricultural sector, improvements in the professional services sector were labour-inducing, meaning that it came along with an expansion of employment. Lastly, the manufacturing sector observed a decline in productivity that was due to losses in resources, as well as efficiency. It coincided losses in employment and gross value added. As a note, the public sector saw a drop in productivity, however its output is only measured by its increases in wage bills, and reflects lower expenses despite higher employment. These sectoral trends went in a similar direction as those in non-NSPA regions —meaning positive growth in agriculture and professional services, and negative in manufacturing— despite differences in terms of levels of change.
The analysis suggests that the NSPA region is growing both in terms of GDP per capita and in productivity. However, the productivity growth in the region is associated more with advances within the sectors, than the introduction of new resources in the region. This is the case for both NSPA and non-NSPA regions, but to a greater extent in NSPA regions. It implies that productivity growth so far is driven by upskilling and upgrading resources within the region and its sectors, given the amount of resources that are already available to the region.
On the other hand, the decline of labour resources and other resources in the NSPA is also contributing negatively to productivity growth, to a larger extent than in non-NSPA regions. This jointly suggests that the NSPA is going through a catching up effect, but would benefit from additional labour and capital resources to the region, in particularly to the sectors that are growing in value added, such as the professional services and agricultural sectors. Furthermore, the sectoral analysis suggests a hollowing out of the manufacturing sector towards the agricultural and services sectors but with different driving factors. In the agricultural sector, increases in productivity are associated with labour-substituting advances, which are useful given the demographic decline the region is facing. On the other hand, in the services sector, increasing resources and increasing efficiency suggests that the regions is going through tertiarisation rather than a transition or development of the manufacturing sector.
Firms in the NSPA region are faced with competition on both national and international markets. For the most part, NSPA regional governments have strategies to support innovation and entrepreneurship that work on promoting the competitiveness of firms. Most of these strategies have a specific focus on supporting small and medium sized firms, which characterise the larger share of firms in NSPA regions. This section provides descriptive information on the relevance of small firms in NSPA regions, and trends in innovation and exports, two indicators of competitiveness of regions.
Firm size
Smaller firms (1-9 employees) account for a larger share of the economy than firms that employ 10 or more workers. In the NSPA regions, they account for 85% of all firms, with employees (Figure 2.8, panel C). In comparison, firms with between 1-9 employees account for 86% of firms in non-NSPA regions, and 87% in OECD NMR-R regions, and 89% of OECD NMR-S regions.
Note: Norwegian firms are not included due to lack of data. Density is per 1000 population.
Source: OECD Regional Indicators.
While the largest share of firms are smaller firms (1-9 employees), these firms employ only 28% of total employment in the NSPA regions (Figure 2.8, panel B). The low share of employment for such a large share of firms is also commonly observed in OECD NMR-S and NMR-R regions, however, NSPA regions seem to have a larger share than non-NSPA regions, whose smaller firms only account for 21% of total employment. Despite containing a larger share of employment, the share of employment for firms with 1-9 employees in NSPA regions is still behind those in OECD NMR-S and NMR-R regions, suggesting that there maybe a stronger distribution of firms with a lower average number of workers in NSPA regions as compared to the OECD regional equivalents.
Smaller firms (1-9 employees) are relatively more prevalent than firms with more than 10 employees in NSPA and non-NSPA regions, when firm density is measured as a ratio of number of firms to the population (Figure 2.8, panel C). On average, NSPA regions have a smaller firm (1-9 employees) density of 21 firms per 1 000 individuals, as compared to 3.6 larger firms (10 employees and over) per 1 000 individuals. This trend is also observable in non-NSPA and OECD metropolitan and non-metropolitan regions.
However, there are less firms per inhabitant in NSPA regions as compared to non-NSPA regions for both firm size categories (Figure 2.8, panel C). The NSPA density of smaller firms is less than the non-NSPA’s density by close to 3 firms per 1 000individuals (24 firms per 1 000 in non-NSPA, versus 21 firms per 1 000 in NSPA). Similarly, the NSPA has a lower share of larger firms than in non-NSPA regions, however this discrepancy is much smaller (0.3 firms per 1 000 individuals). Similarly, OECD NMR-S and NMR-R regions tend to have a lower density of both types of firms than other region types.
Jointly, the analysis on firm size suggests first that while smaller firms are more prevalent than larger firms in NSPA regions, there are less firms (of any size) per individuals in NSPA regions than in non-NSPA regions. At a first level, it underlines the importance of continuing to support small firms within the NSPA regions. On a second level, the relatively lower density of both sizes of firms suggests that there are still barriers to both small and larger firms in NSPA regions. Lastly, it is likely that competitive forces between firms within the NSPA region may be more limited than in non-NSPA regions, however, as firms do not only compete locally, but a more global analysis should also be taken into consideration. At the bare minimum, governments should not only continue to support competitiveness of small firms, but also continue to increase support for start-ups and small firm growth.
Innovation
The NSPA region is relatively innovative, despite its lower performance compared to other non-NSPA regions. In 2020, the NSPA had an average of 106 patent applicants per 1 million individuals (Figure 2.9. , panel A). This average was lower than the non-NSPA regional average of 159 patent applicants, but substantially higher than the OECD regional averages for the NMR-S (46) and NMR-R (21) regions. This suggests that the high-tech innovation, as proxied by this metrics, is on a whole higher in NSPA countries, and, despite the fact that NSPA regions are lagging behind non-NSPA regions, their performance is better than regions with similar characteristics across OECD countries.
Over the last decade, the innovation performance of NSPA regions increased marginally, while it fell in non-NSPA regions, and OECD NMR-R regions, but remained the same in OECD NMR-S regions (Figure 2.9. , panel B). For instance, the average patent intensity for the NSPA regions in the 5-year period from 2016-20 was 104 patent applicants per 1 million individuals, a small increase from the average patent intensity of 102 patent applicants in the 2011-15 period. In comparison, non-NSPA regions saw a fall from 205 patent applicants in the 2011-15 period to 188 patent applicants in the 2016-20 period. In comparison, the OECD NMR-S regions had an average of 50 patent applicants per 1 million individuals in both periods, while the OECD NMR-R regions saw a decline between the periods, from 26 patents per 1 million individuals in the 2011-15 period to 24 patents per 1 million individuals in the 2016-20 period.
Patent applicants per million individuals.
Source: OECD Regional indicators.
The implication of the analysis suggests that the high-tech innovation performance of the NSPA region is relatively strong as compared to similar OECD regions, despite being behind national non-NSPA averages. Furthermore, this particular innovation performance is growing despite a stagnant or backward trend in benchmark regions from the beginning to the end of the 2010’s. However, analysis from regional profiles suggests that there are still quite a bit of differences within the NSPA region, and as such innovation diffusion is still an important policy goal to pursue.
Innovation is often both an output and an input of competitive environments and an analysis of the innovation performance of firms in regions should ideally look at both dimensions. For regions within OECD countries, often innovation analysis falls short as it fails to either capture innovation process of firms due to challenges in measurement, or overlooks the relative nature of innovation when using evaluation metrics that are relevant to the structure of the economies, which often have different trends in firm characteristics based on size, sector and access to externa resources. While this analysis provides some information on innovation outputs, further work on innovation inputs, in particular, on innovation that goes beyond the high-tech sectors is relevant to supporting innovation, competition and regional development. For example, in the Swedish NSPA regions, there is a divide between the more urban, research-driven innovation near universities and the more sparsely populated inland areas, where innovation is driven by incremental entrepreneurship. It would be beneficial to better integrate and cross-fertilise these innovation ecosystems to scale up research and innovation closer to the market across all these regions. However, there is a lack of reliable statistics to capture the actual innovation outcomes needed to build on. This issue affects large parts of the NSPA.
Trade
One of the NSPA’s defining characteristic is sparsity, and with it, often distance to networks for trade. In many cases, maritime trade routes are considered major trade ports, but within region physical access to trade routes may sometimes be a challenge, as in many remote rural areas.
The value of total exports from the NSPA region is relatively low as compared to non-NSPA regional averages. In 2020, export values reached close to USD 2 billion (Figure 2.10, panel A). This was similar to export values in OECD NMR-R regions, but lower than the average of OECD NMR-S regions (USD 2.6 billion), and less than half of the average of non-NSPA regions (USD 5.5 billion).
Despite a relatively weak export performance, exports are increasing on average in the NSPA region. Increases are growing at a faster aggregate growth rate than any other benchmark group. From 2015 to 2020, aggregate growth in export values increased by 21% in the NSPA (Figure 2.10, panel B). This growth was higher than the increase in the regional averages of non-NSPA (10%), OECD NMR-S (9%), and OECD NMR-R (4%). In terms of compound annual growth rates, the NSPA grew 3.3%, almost twice the rate of non-NSPA regions (1.7%), and over two times the rate of OECD NMR-S and NMR-R regions (1.5% and 0.7%, respectively).
Source: OECD Regional Indicators. Trade balance figures refer to exports (X) minus imports (M). Numbers for NSPA regional growth in trade surplus were inverted to reflect negative trends. This was chosen to facilitate interpretation and avoid logarithmic transformation which would be inconsistent with the approaches for all positive value changes in the rest of the report.
Despite having lower export values, NSPA regions have higher trade surplus than non-NSPA regions, and are relatively higher than similar OECD regional averages in the NMR-S and NMR-R regions (Figure 2.10, panel C and D). This means that the value of exports were substantially higher than the value of inputs to the region. While some values may not adequately reflect regional outputs, due to standard issues with regional analysis, including the headquarter bias, as well as challenges related to intermediary trade companies, the trends nevertheless also demonstrate a trade surplus for NSPA regions, while trends (Figure 2.10, panel C and D) seem to indicate, that despite these measurement challenges, there is an upward trend with more value being exported than values imported. More granular and harmonised data using trade-in-value added would be a more targeted approach used in further research on trade balances.
In sum, there are 3 main observations related to the context of the competitiveness of the NSPA regions:
Firm intensity is low in NSPA regions. While there are more firms that are smaller (1-9 employee) in NSPA regions than other sized firms, the region could use support in promoting more start-up and scale up activities in the region to boost competition.
High-tech innovation is relatively well performing in the NSPA regions, despite a relatively weak performance as compared to non-NSPA averages, and variations between regions within the NSPA. The NSPA region is outperforming OECD regions with similar characteristics, and is showing signs of improvement while all other benchmarks are either declining or remaining the same. However, additional analysis on other forms of innovation is also needed to support competitiveness in the region.
Exports are relatively low in the NSPA region, but there is a trade surplus. Export values in the NSPA region are similar to export values in the average OECD NMR-R region, but lower than the average non-NSPA region, and OECD NMR-S region. Nevertheless, the export values and trade surplus is increasing in the NSPA region at a faster rate than in other benchmark regions.
The analysis jointly suggests that the context for competitiveness of the NSPA region has some strengths and weaknesses. With low firm intensity, strong high-tech innovation performance, it’s likely that competitive pressures between firms are not driving innovation performance as is often the case in more dense regions. Competitiveness of firms in the NSPA are more likely driven by different factors that may allow for experimentation and innovation despite low firm competition within the same region. The fact that exports are low but growing, jointly with high-innovation performance and low firm intensity, suggests that firms may be increasingly innovating to improve exports or increasingly upgrading production of goods to more high-value goods, however further analysis on the drivers of export growth are needed to establish the causal direction of these trends. While support for more start-ups and scale-ups is still needed in the region, the competitiveness of firms the region, based on this analysis, comes from its capacity to innovate and increase exports despite in a low regional competition environment.
The NSPA region jointly face challenges related to a large older and elderly age population, and in many cases, challenges related to delivering public services in places with low density and long distances to urbanised areas. While further analysis is needed on a range of social indicators, one example is the provision of health-related services.
Across the NSPA and in non-NSPA regions, trends in the availability of public health facilities and practitioners are changing. In NSPA regions, as in non-NSPA regions, access to urgent care, as proxied by hospital beds per person is falling. In 2012, there were 12 hospital beds per person, whereas in 2019, there was 3.7 beds per person (Figure 2.11, panel A). However, there are still more hospital beds per person in NSPA regions than there are in non-NSPA regions, and non-NSPA regions also saw a fall in the number of hospital beds per person. In 2012, there were 3.8 hospital beds per person in non-NSPA regions and in 2019, this ratio went down to 2.7 hospital beds per person. This may indeed be explained by the higher demand due to a relatively larger share of older aged workers and elderly. Further analysis with NSPA harmonised statistics on age-adjusted older age dependency could help determine whether it is the demand of hospital beds, that is leading to higher rates, or whether there is another reason why there are more hospital beds per person in NSPA regions.
Trends in NSPA and non-NSPA were also similar in terms of active physicians per person (Figure 2.11, panel B). In 2012, NSPA regions had 3.5 physicians per person, slightly fewer than 3.6 in non-NSPA regions. In 2019, the active physical rate increased relative to 2012, and was slightly higher in NSPA regions with 4.1 active physicians per person in NSPA regions, versus 4.0 active physicians per person in non-NSPA regions. Despite similar trends, and slightly higher rates at the end of the period (2019) in NSPA regions, given the relatively higher rate of expended demand from older populations in NSPA regions, it is necessary to do further analysis to better accounts for ageing trends in the provision of physician care services.
Source: OECD Regional Indicators.
In summary, access to social services like health care suggests similar changes in the system across NSPA and non-NSPA regions. While the regional profile reports demonstrate some heterogeneity between regions, the average regional trends suggest that it may rather be national trends may be a stronger determinant of regional trends, and that at least in terms of access to urgent care some adjustment has been made for the relatively older population. More information on age adjusted statistics would help to support further analysis on whether the supply of urgent care facilities provided are at least remaining constant with ageing trends. On the other hand, the increase in active physician rates is occurring across the board in NSPA and non-NSPA regions, and it is if anything, relatively more concerning that there is not a higher rate in NSPA areas, that are likely to have a higher demand for physician services with an ageing population.
The vast geography and dispersed population of the NSPA region present distinct challenges in terms of ensuring physical accessibility. With residents spread out over large distances, often far from urban centres, it’s vital to assess how easily they can access essential services. A useful metric in this regard is the drive time to the nearest large city, which can provide valuable insights into how easily residents can reach employment hubs, educational institutions, healthcare facilities, and other vital services.10 Extended drive times can reflect substantial barriers, potentially limiting economic opportunities and diminishing the overall quality of life.
When comparing the NSPA to non-NSPA regions, a clear pattern emerges: while southern Norway, Sweden, and Finland are home to the majority of large cities, the northern parts of these countries – where the NSPA is located – have far fewer urban centres. This gap makes drives to the nearest large city longer in NSPA than in non-NSPA regions (Figure 2.12. ). The increased travel distances in the North highlight the logistical challenges faced by these communities, reflecting the geographic and infrastructural differences that characterise the NSPA region.
These challenges are further compounded by the lack of comprehensive railway connections across the NSPA, which forces a heavy reliance on road transport. However, road networks in the region are often narrow, of low standard, and subject to frequent bottlenecks. These limitations restrict both passenger and freight movement, leading to inefficiencies that hinder economic growth and increase environmental impacts. Moreover, while the drive time data reflects car travel, it does not capture the full scope of transportation dynamics in the region. In many NSPA areas, air travel plays a critical role in connecting remote communities, therefore collecting more data on this could support more comprehensive and accurate analyses.
Average drive time (in hours) from settlements to the next city (with at least 50 000 people)
Note: The map in the figure above represents the average drive time, in hours, of individuals from a population settlement to the next city with at least 50 000 people with a population density threshold of 1 500 per squared kilometre, within the same country. The figure is population weighted for each settlement. It is based on DEGURBA definition of settlements. In Finland, the cities are identified as Oulu, Jyväskylä, Tampere, Lahti, Helsinki and Turku. The delineation of regional boundaries are based on boundaries from 2021.
Source: OECD Regional Statistics.
In NSPA regions, geographical distances present significant challenges for connectivity. However, advancements in digital infrastructure can help mitigate these barriers by improving access to high-quality internet. Despite ongoing progress, many rural areas within the NSPA continue to face limited internet availability. Even among those with access, residents in more remote locations tend to experience lower-quality connections.
According to OECD estimates from Q4 2023, national average fixed broadband download speeds stood at approximately 140 Mbps in Norway and Sweden and 105 Mbps in Finland.11 Progress across NSPA regions has been uneven, with disparities narrowing in most, though not all, regions (Figure 2.13, panel B). Among the 14 NSPA regions, seven reported user-reported internet speeds exceeding their respective national averages in Q3 2023. The highest relative exceedances were recorded in Central Ostrobothnia (13.3%), Northern Ostrobothnia (13.2%), the regions of Troms and Finnmark (11.4%), Lapland (8.5%), Kainuu (5.9%), and Pohjois-Savo (5.6%).Conversely, several regions reported below-average speeds, including Nordland (-1.1%), North Karelia (-7.0%), Västernorrland (-8.9%), Jämtland Härjedalen (-14.1%), Norrbotten (-16.6%), Västerbotten (-27.5%), and South Savo (-29.8%). While disparities have generally decreased over time, four of the six regions with below-average speeds in Q4 2021—South Savo (7.7 percentage points), Västerbotten (2.6 p.p.), Norrbotten (3.0 p.p.), and North Karelia (4.7 p.p.)—saw their gaps widen further.
In areas with at least basic internet coverage, fixed broadband download speeds have improved across OECD, NSPA, and non-NSPA regions. Between Q4 2021 and Q3 2023, speeds increased by 15% in NSPA regions, 17% in non-NSPA regions, 35% in OECD NMR-R regions, and 40% in OECD NMR-S regions (Figure 2.13, panel A). However, NSPA regions experienced comparatively slower growth. This is partly because regions with lower initial speeds, such as OECD NMR-R and NMR-S, had more potential for improvement, leading to higher percentage gains.
A similar trend is evident at the national level. In Finland, Sweden, and Norway, where initial internet speeds were already relatively high—particularly in Sweden and Norway—growth has been more limited. Between Q4 2021 and Q3 2023, fixed broadband speeds in Finland increased by 30%, while Sweden and Norway recorded growth rates of 12% and 11%, respectively—both below the OECD average (Figure 2.13, panel A).
These trends highlight the importance of continued investment in digital infrastructure to reduce regional disparities and ensure equitable access to high-speed internet across NSPA regions.
Source: OECD Regional Indicators.
The NSPA regions, with a relative advantage in natural endowments are uniquely positioned to invigorate initiatives in the green transition. With nature conservation as an issue that may NSPA regions value, concrete advancements in emission reductions and transition to renewable energy sources are vital for NSPA regions.
On this front, production-based statistics on reaching net-zero targets in the NSPA regions are advancing substantially. In terms of greenhouse gas (GHG) emissions reductions12, the NSPA regions on average observed a fall in emissions reductions over the past 5 decades (Figure 2.14, panel A). From 1970 to 2022, greenhouse gas emissions per capita in terms of CO2 equivalent fell from 14 to 10 tons of emissions per capita. This is substantial progress, as estimates from comparable OECD regions were less promising. In OECD NMR-S regions, the emissions remained at close to 12 tons from 1970 to 2022, while in OECD NMR-R regions, GHG emissions fell from 19 to 18 tons per capita. However, improvements in the NSPA region were behind, both in terms of levels, and progress, of non-NSPA regions, suggesting that progress within Finland, Norway and Sweden were occurring on as well. In the non-NSPA regions, GHG emissions halved over the same period of time, from 12 tons per capita in 1970 to 6 tons per capita in 2022.13
Despite relatively higher per capita emissions, overall emissions are lower in the NSPA regions than in non-NSPA regions. In 2022, the NSPA regional average emissions was 3.26 tons of CO2 equivalent. On the other hand, in the non-NSPA regions, the regional average emissions amounted to 4.32 tons of CO2 equivalent. While the NSPA has lower levels of average emissions per region, average emissions increased in the NSPA region, whereas it fell in non-NSPA regions. In 1970, the NSPA region emitted 2.89 tons of CO2 equivalent, whereas the non-NSPA regions emitted 4.93 tons of CO2 equivalent. In 2022, emissions in the NSPA regions grew to a regional average of 3.26 tons of CO2 equivalent, while they fell to 4.32 tons of CO2 equivalent in non-NSPA regions.
Source: OECD Regional Indicators.
In addition to improvements in the reduction of GHG emissions, the NSPA region is exemplary in its use of renewable energy overall and as compared to other benchmark regions. In 2019, the NSPA’s share of renewable energy in electricity production was at 99% (Figure 2.14, panel B). In comparison, the share of renewable energy in electricity production was 77% in non-NSPA regions, and even lower in OECD NMR-S (72%) and NMR-R (70%) regions.
In sum, there has been much progress and increasing potential for the green transition to continue in the NSPA regions. In particular, GHG emissions are falling in NSPA regions, and are lower on aggregate than in non-NSPA regions. Secondly, the NSPA region is exemplary for its pivot towards renewable energy sources and could be considered an example for non-NSPA regions to better learn from. Further analysis, based on energy consumption data could help direct policy makers in understanding whether to better target consumption reduction measures of households and firms, or continue to focus on reduction of emissions due to production statistics that may be more concentrated in oil and natural resource extraction regions.
Greenhouse gas (GHG) emissions at the subnational level were estimated using the using the Emissions Database for Global Atmospheric Research (EDGAR) version 8.0 developed by the EC JRC and IEA (EC JRC and IEA, 20231). EDGAR provides annual sector-specific grid maps for the four GHGs (CO2, CH4, N2O and F-gases) at a 0.1° spatial resolution (~11 km). The different sectors and subsectors covered are:
Energy: Power generation.
Industry: Combustion in manufacturing industry, oil refineries and transformation industry, chemical processes, fuel exploitation, iron and steel production, non-energy use of fuels, non-ferrous metals and non-metallic minerals production, solvents and products use.
Transport: Ground transport: road, trains and off-road transport. Shipping and aviation are excluded in the subnational GHG estimates for the transport sector.
Building: energy for buildings
Agriculture: Agricultural soils, agricultural waste burning, enteric fermentation, manure management, indirect N2O emissions from agriculture
Waste: Solid waste incineration, landfills, waste water handling
Emissions from Land Use and Land Cover Change (LULCC) are not included. National GHG emissions are disaggregated by using subsector-specific geospatial proxies. GHG emissions are expressed in CO2 equivalents using 100-year global warming potential from the IPCC 5th Assessment Report (AR5), i.e. 28 for CH4, and 265 for N2O.
While emissions data provides valuable insight to production-based activities contributing to climate change, it does not account for the usage or consumption of goods or services that may occur outside of where the goods are produced. However, emissions-based data is widely available both geographically and historically and nevertheless provide insight on attainment of climate change reduction goals.
Note: 1 EC JRC and IEA (2023), “EDGAR (Emissions Database for Global Atmospheric Research) Community GHG Database: IEA-EDGAR CO2, EDGAR CH4, EDGAR N2O, EDGAR F-GASES version 8.0”, European Commission, JRC (Datasets).
Note: 2 Land Use and Land Cover Change (LULCC) refers to the transformation of the Earth's surface due to human activities and natural processes. It includes changes in how land is used (e.g. agriculture, urban development) and changes in land cover (e.g. deforestation, desertification).
The statistical diagnostic of the NSPA regions reveals a picture of a region that is growing despite challenges in terms of population, and furthermore, providing important contributions both to innovation and to the green transition. Table 2.2 provides a summary of strengths and areas of policy attention from the NSPA regions.
Source: Author’s elaboration
In conclusion, the NSPA face significant challenges, including declining populations, labour shortages, low firm density, and regional disparities in digital accessibility. Economic activity is hindered by a shrinking working-age population, low innovation diffusion outside high-tech sectors, and infrastructure limitations that increase logistics costs and limit trade expansion.
However, the region has substantial opportunities for growth. High GDP per capita, increasing labour productivity, and a growing younger working-age population provide a foundation for economic expansion. Innovation potential is strong, particularly in emerging green and digital sectors, offering growth prospects for SMEs. The region also leads in renewable energy production, which can be leveraged to strengthen its competitive edge in the green transition. Expanding eco-tourism and cultural tourism could drive further economic benefits, provided tourism infrastructure improves.
To address these challenges and capitalise on opportunities, the NSPA must enhance digital connectivity, invest in skill development, and support start-ups and scale-ups. Strengthening regional co‑operation within the NSPA platform can accelerate best practice sharing, particularly in green and digital transitions. By strategically investing in key sectors and infrastructure improvements, the region can mitigate demographic and economic constraints while reinforcing its position in global trade and sustainable industries.
← 1. Small administrative regions refer to the classification of territories based on Territorial Level 3 (TL3) regions elaborated by the OECD that is aligned with the Nomenclature for territorial units of statistics 3 used by Eurostat. Further information on the classification of small administrative regions is available in Box 1.1.
← 2. Depending on the region, the NSPA regions can either be classified as an NMR-S or an NMR-R region. More information on this classification is available in Box 1.1.
← 3. The NSPA regions have a total land area of close to 533 000 km2 and non-NSPA regions have close to 483 000 km2. For consistency and temporal harmonisation challenges in Troms and Finnmark due to the recent administrative changes, the two regions are combined as one region when calculating NSPA regional averages, resulting in 13 units for calculating NSPA regional averages.
← 4. The older worker category is reflective of the general retirement age trends in OECD countries, but the actual retirement age varies from country to country with the NSPA (and often by occupation and sector).
← 5. While this is relatively low as compared to the rest of the age groups, it also accounts for individuals in 5 years of age, whereas the other comparison groups, included 15 years.
← 6. The figure shows trends, but the correlation is relatively low and spurious, or statistically weak and noisy. It provides general trends, but does not account for many other factors that impact population and aging trends.
← 7. The quadrant numbering refers to the placement on the x-y coordinate plane in Figure 1.3. Positive values for both the x and y axis refer to quadrant I. Positive values for the y, and negative values of x axis refer to quadrant II. Negative values for both x and y axis refer to quadrant III. Positive values for x, and negative values of the y axis refer to quadrant IV.
← 8. The second highest sector was the real estate sector. However, the real estate sector is excluded from this analysis. Even though real estate activities contribute substantially to positive increases in productivity, this sector often consists of lumpy profits that reflect multiple years rather than one, and the high-speculation dimension of the sector also suggests that it does not function in the same way as activities in other sectors, making the consideration of it within the context of productivity analysis less relevant. For instance, the real estate activities sector often involve heavy investment in a specific year and can be used as a financial instrument that is highly sensitive to national financial markets. Outside of this argument, it is also a sector that functions through local market population pressures that are not a reflection of higher efficiency in the production or processes within the sector.
← 9. While the mining sector’s profits and value added may indeed be a result of improvements in extraction processes and equipment, the price of most extractive resources related to energy, such as fossils fuels are set internationally, which do not reflect regular market mechanisms for setting prices based on market demand and supply. The interpretation of productivity in this sector, therefore, does not have the same implications as they would in other manufacturing and services sectors.
← 10. The data used are based on the DEGURBA definition, which identifies settlements from clusters of adjacent 1 square kilometre (km2) grid cells with medium or high population density. These clusters are defined as a city if they have a minimum population density of 1 500 per km2 and a minimum population in the cluster of 50 000. An advantage of using the DEGURBA definition is that it avoids the identification of multiple urban centres for a single city and helps with international comparability. However, DEGURBA defines settlements such as cities in terms of their population density, excluding the surrounding commuting areas. For furthermore information please see OECD (2024), "Services in towns and villages", in Getting to Services in Towns and Villages: Preparing Regions for Demographic Change, OECD Publishing, Paris, https://doi.org/10.1787/48fe743e-en.
← 11. OECD (forthcoming), Bridging connectivity divides, OECD Publishing, Paris.
← 12. GHG emissions are one form of measuring the progress towards net-zero but has many flaws. While it is possible to do harmonized analysis with this data, when it is regionalised it suffers both from headquarter bias, as well as only measures a production-based approach, which does not take into account where goods produced that emit GHG are consumed. While harmonized consumption-based indicators are currently in development, emissions-based measures nevertheless provides some understanding of general trends.
← 13. The relatively stronger growth of the population in non-NSPA regions and decline in the NSPA regions impacts the indicator.