Understanding the Scottish rural economy: research paper

Annex 2: Urban and Rural Definitions

There are a range of different ways that Scottish Government classifies Urban and Rural differences. Primarily, they are based upon the threefold, sixfold or eightfold classification; though for economic data the Randall definition is also used.

Threefold Urban Rural definition

Classification	Definition
1 Rest of Scotland	Settlements of 3,000 or more people
2 Accessible Rural	Settlements of less than 3,000 people and within 30 minutes drive of a settlement of 10,000 or more.
3 Remote Rural	Settlements of less than 3,000 people and with a drive time of over 30 minutes to a settlement of 10,000 or more.

Sixfold Urban Rural definition

Classification	Definition
1 Large Urban Areas	Settlements of 125,000 or more people.
2 Other Urban Areas	Settlements of 10,000 to 124,999 people.
3 Accessible Small Towns	Settlements of 3,000 to 9,999 people and within 30 minutes drive of a settlement of 10,000 or more.
4 Remote Small Towns	Settlements of 3,000 to 9,999 people and with a drive time of over 30 minutes to a settlement of 10,000 or more.
5 Accessible Rural	Areas with a population of less than 3,000 people, and within a 30 minute drive time of a settlement of 10,000 or more.
6 Remote Rural	Areas with a population of less than 3,000 people, and with a drive time of over 30 minutes to a settlement of 10,000 or more.

Eightfold Urban Rural Definition

Classification	Definition
1 Large Urban Areas	Settlements of 125,000 or more people.
2 Other Urban Areas	Settlements of 10,000 to 124,999 people.
3 Accessible small Towns	Settlements of 3,000 and 9,999 people and within 30 minutes drive of a settlement of 10,000 or more.
4 Remote Small Towns	Settlements of between 3,000 and 9,999 people and with a drive time of over 30 minutes to a settlement of 10,000 or more.
5 Very Remote Small Towns	Settlements of 3,000 and 9,999 people and with a drive time of over 60 minutes to a settlement of 10,000 or more.
6 Accessible Rural	Areas with a population of less than 3,000 people, and within a 30 minute drive time of a settlement of 10,000 or more.
7 Remote Rural	Areas with a population of less than 3,000 people, and with a drive time of over 30 minutes but less than 60 minutes to a settlement of 10,000 or more.
8 Very Remote Rural	Areas with a population of less than 3,000 people, and with a drive time of over 60 minutes to a settlement of 10,000 or more.

Randall Definition rebased-2017

Originally produced in 1985 for the Scottish Economic bulletin ( SEB) as a means of profiling economic trends and indicating need for support in rural Scotland. The system is based upon population density within a unitary authority. Where a unitary authority has a population density of less than one person per hectare it is considered Rural. On this basis there are 14 rural unitary authorities. These are:

• Aberdeenshire
• Angus
• Argyll and Bute
• Dumfries and Galloway
• East Ayrshire
• Highland
• Moray
• Na h-Eileanan Siar
• Orkney Islands
• Perth and Kinross
• Scottish Borders
• Shetland Islands
• South Ayrshire
• Stirling

According to the Randall definition of rurality, 89% of Scotland's landmass and 29% of its population is classified as rural (defined in 1995, and rebased in 2017).

Benefits:

Unitary Authority data is readily available and it is therefore very easy to use this system for classification.

Limitations:

Since the classification system is Unitary Authority based, some urban pockets including Stirling and Inverness, are classified as rural.

Fourfold RESAS Classification

The fourfold RESAS Classification distinguishes local authorities according to their level of rurality. In order to define rurality from a conceptual and methodological point of view, a literature review on existing urban: rural classification at an aggregate data level was conducted.

• If data is available at a data zone level, the Scottish Government most commonly applies its Scottish Executive Rural Urban Classification that is based both on population levels and distance from urban areas with more than 10,000 inhabitants.
• However, because data is not always available at the data zone level, the Scottish Government uses the Randall definition that is based solely on population density aggregated at the local authority level. Thus, it tends to over-state the rural population ( e.g. the populations of Stirling Perth and Inverness are considered 'rural) and understate the rural area ( e.g. East Lothian and South Lanarkshire are considered 'urban').
• The OECD classifies regions as predominantly urban, intermediate or predominantly rural based on the percentage of population living in local rural units and divides Scotland into 23 regions of which nine are considered to be predominantly rural accounting for 75% of the landmass and 17% of the population ( OECD, 2008).
• Scholars such as Cloke (1977) however suggest that a definition of rurality cannot only be based on population density and distance to urban settlements, but should include factors such as demographics, occupational structure, household amenities and migration.

Based on the literature and classifications defined thus far, but also on suggestions of various stakeholders, an extensive range of demographic, economic, social and geographic indicators was collected. Data included:

• Persons; Km ²; Persons/Km ²; % Population not in localities - classed as Rural;
• Share of 15-45 year olds; Share of 15-64 year olds; Percentage population change mid-2006 to mid-2016; % pensionable age;
• Local government employment by local authority ( FTE); Net Revenue Expenditure per capita (£); Workforce from the LA;
• Road Length; % of premises unable to receive 10Mbit/s
• Local Share of Access to Services for most/least deprived areas

To identify which variables correlate and thus help to explain groupings of local authorities, a principal component analysis was conducted. The analysis generated the following four different components:

1. Population; population per km ²; Share of 15-64 year olds; Local government employment by local authority ( FTE); share of population in pensionable age; Share of population that not assigned to either a settlement or locality; share of premises unable to receive 10Mbit/s; population in settlements under 10,000; and access to Services 40% most deprived areas in LA.

2. Road length, km ²; workforce stemming from the local authority

3. Share of population in pensionable age; hare of population that not assigned to either a settlement or locality; share of premises unable to receive 10Mbit/s; population in settlements under 10,000; and access to Services 40% most deprived areas; workforce stemming from the local authority; Net Revenue Expenditure per capita (£)

4. Share of 16-64 year olds; share of population in pensionable age; Percentage population change mid-2006 to mid-2016.

Following the PCA, a cluster analysis was conducted using the components in order to identify which component would allow the most suitable grouping. Component 2 and 3 did not allow for much variation and suggested creating several groups, but one main and dominant group with almost all local authorities assigned to it. Component 4 allowed for a lot of different and fairly evenly distributed groups. However, as the indicators are only based on demographics and exclude all other features that are crucial for rural Scotland, this component was also put aside. Component 1 however suggested a fairly even distribution of groups and takes a wide range of indicators into account.

Component 1 of the PCA suggests including both the share of the population in pensionable age and the share of the population in working age. These variables express two different features: A higher share of pensioners and thus an ageing population aims to capture the potential burden on services and the relative number of dependents. The share of the population in working age on the other hand captures not only the size of the workforce, but also indicates out-migration. Both variables are correlated, but not interchangeable. According to the PCA, a large working age population and a relatively small share of pensioners are features of more urban areas.

Population in settlements under 10,000 and the share of the population not assigned to either a settlement or locality are also correlated and a high share of each is a feature of more rural areas. While the first variable directly expresses the share of the population in rural settlements and thus expresses rurality, the second variable describes remoteness as it describes the population living in areas so small, that they are not picked up by the data.

According to the PCA, a large population and high population density is an indicator for more urban areas. This is also highly correlated with access to services, employment opportunities, remoteness and other rural features.

Broadband connectivity is a key variable to include when describing rural economies as the lack of high-speed broadband does not only affect homes, but also enterprises and their business. Access to high speed broadband is described as a key enabler. However, while in urban areas nearly all households can access at least 10 Mbit/s, this is not the case for more rural areas.

Lastly, following the PCA local government employment is taken into account as well. High absolute numbers of FTE government employees indicate greater human resources and thus more capacity for policy-making and the development of new strategies. This variable needs to be taken into account if we want to compare the size of rural economies.

Thus, component 1 was pursued further.

In a last step, the results from the PCA were used in order to create an index for the rural economy. All data was standardized and various weights were applied. However, because weighting did not change the results crucially, the decision was made not to apply any weights. Following the results for component 1, the additive index was calculated as follows:

Rurality=

(Share of population in pensionable age) + (Share of population that not assigned to either a settlement or locality) + (Share of premises unable to receive 10Mbit/s) + (Population in settlements under 10,000) + Access to Services 40% most deprived areas in LA) - (Population) - (Population per km ²) - (Share of 16-64 year olds) - (Local government employment)

Scores for the index range from -20.1 in Glasgow and thus least rural to 15.5 making N-ah Eileanan Siar most rural. In order to now group local authorities, an additional lens was applied in order to incorporate the existing and more detailed 8-fold Scottish Government classification. Firstly, the local authorities with the lowest scores are Glasgow, Edinburgh, Aberdeen and Dundee. These four LAs do not only have a distinctively higher score than the next highest LAs, but are also the only Scottish cities with more than 90% of the population living in large urban areas. Thus, these four LAs are classed as 'Larger Cities'.

The second group is classed as 'Urban with Substantial Rural areas' and includes all LAs with a population of more than 50% living in urban and other urban areas. Thus, while Stirling is considered 'Urban with Substantial Rural areas', East Ayrshire is not. The next cut-off point aims to distinguish between Mainly Rural and 'Islands & remote'. Here, the Islands and Argyll & Bute stand out both due to their high RESAS Classification score and due their share of population living in urban areas of less than 20%, and even 0% on the islands. Thus, while Scottish Borders is considered Mainly Rural, Argyll & Bute is classed as 'Islands & remote'. South Ayrshire and Angus are exceptions to this rule as they both have more than 50% of their population living in large urban and other urban areas and thus should be classed Urban with Substantial Rural. However, because according to their RESAS score they clearly are more rural than Stirling and other more urban areas, they are classed as Mainly Rural.

Accordingly, local authorities are classed as follows:

Local authority	RESAS Classification	Label
Na h-Eileanan Siar	15.5	Islands & Remote
Orkney Islands	12.7
Shetland Islands	10.8
Argyll & Bute	9.1
Scottish Borders	6.9	Mainly Rural
Dumfries & Galloway	6.6
Highland	4.9
Perth & Kinross	4.5
Angus	3.2
Moray	3.1
South Ayrshire	2.4
East Lothian	1.8
Clackmannanshire	1.4
Aberdeenshire	1.3
East Ayrshire	1
Stirling	0.7	Urban with Substantial Rural areas
East Dunbartonshire	0.4
North Ayrshire	0.2
Midlothian	-0.6
West Dunbartonshire	-1
Inverclyde	-1.3
East Renfrewshire	-1.6
Falkirk	-2.1
Renfrewshire	-3.5
West Lothian	-3.9
South Lanarkshire	-4.8
Fife	-4.9
North Lanarkshire	-6.8
Dundee City	-10	Larger Cities
Aberdeen City	-10.1
City of Edinburgh	-15.6
Glasgow City	-20.1