High streets in Great Britain

• This article has been produced as a joint project between the Office for National Statistics (ONS), the UK national statistics institute, and Ordnance Survey, the national mapping agency.

• There are nearly 7,000 high streets in Great Britain included in the analysis, defined by a cluster of 15 or more retail addresses within 150 metres.

• In 2017, around 10.3 million people, 16% of the British population, lived within 200 metres of a high street.

• High streets have seen higher recent population growth, compared with non-high street areas; between 2012 and 2017, high street areas saw a 6% population growth, compared with 3% in non-high street areas.

• 62% of residential property sales on the high street were for flats in 2017, up from 40% in 1997.

• Median residential property prices on the high street are generally lower than in other areas because of the different housing mix between high street areas and non-high street areas.

• The number of businesses on the high street increased by 15% between 2012 and 2017, compared with a 22% increase in non-high street areas.

• Between 2012 and 2017, the number of retail businesses on the high street fell (negative 2%) while those not based on the high street grew (positive 6%).

• In 2017, the retail sector provided between 25% and 31% of high street employment in all regions and countries of Great Britain except London (18%).

• Between 2012 and 2017, the number of high street retail jobs fell in every country and region except London (positive 6%).

• Between 2012 and 2017, employment on high streets in the accommodation and food sector increased by over 20% in every country and region.

• Employment growth in “other service sectors” between 2012 and 2017 was over 20% in the West Midlands, the East Midlands and London but declined in the North East, and Yorkshire and The Humber.

Introduction

The high street has long been central to the British retail experience. However, high streets are not just made up of shops. They also provide residential housing and offices for business as well as a range of local services, accommodation, and food and drink to residents and visitors alike.

The high street is also seen to be in a period of change. The closure of branches of retailers across many high streets has led to worries about the decline of retail on the high street, and in turn anxiety about how high streets will develop in the future.

In this context, it is important that good data on high streets are available to monitor the changes and inform policy responses. Therefore, the Office for National Statistics (ONS) and Ordnance Survey (OS) have joined together to identify high streets across Great Britain and to produce some initial analysis that investigates a range of data comparing the recent trends on the high street with those in non-high street areas.

In particular, this first article focuses on providing regional analysis to highlight some of the main high street trends. Based on the identification of almost 7,000 unique high streets across Great Britain, the article provides insights on the main different types of economic activity on the high street such as retail, food, and accommodation and offices. The article also examines data on population levels around high streets and information on the recent changes to residential housing on high streets.

It should be noted that this high streets project is very much a work in progress. The aim of this article, therefore, is to provide an introduction to the work, to provide some initial results, and to seek feedback to inform the next stages of the project. While the data come from a range of suppliers, including official statistics sources, the fact that the approach to this work is still being developed means that the data and results in this article are Experimental Statistics.

Looking ahead, our aims at the ONS and OS include further developing the high street geography to enable it to be widely used across government as a means for assessing high street topics, and also to expand on the initial results in this article with further articles to provide more detailed breakdowns of the data including at more local geographic levels.

The high street geography

The ONS has been working with the national mapping agency OS to locate, for the first time, where high streets are “on the map” for the whole of Great Britain. Given that no definition of high streets currently exists, OS has created an experimental dataset, using an automated methodology, which aims to depict the extent of Britain’s high streets and their land use.

The dataset is a high street-specific dataset that provides a street-level definition. It bounds retail clusters using street naming and aims to deliberately exclude other retail functions like retail parks, industrial estates, and isolated shopping centres. To be included in the dataset, a high street must be a named street predominately consisting of retailing, defined by a cluster of 15 or more retail addresses within 150 metres.

The high street extents and data are experimental and only a first step towards a more complete definition. We also plan to match more statistics from a wider range of sources in future, but we are keen to publish early findings to help inform the public and policy discussion about high streets.

It is important to note that we are only looking at today’s geographic extents and that the high street may have expanded or contracted in terms of its physical size over time. Some of the more striking figures in this article are also likely to indicate changing shopping patterns and we hope to explore this, as well as how high street extents have changed physically, in future research that will also bring in retail centres and destination shopping.

Technical information about the work can be found towards the end of this article in Section 8. Within Section 7 there is a link to an interactive map showing the location of high streets captured in the OS methodology. You can put a postcode or place name in the search box and see how your high street compares with others. If you click on an individual street you can see more information, like address types, or number of buildings.

What statistics are included

The data in this article go beyond just looking at retail function and include information on commercial and residential aspects of the high street as well as data on population change, house prices and transactions, and business and employment levels. The data originate from both OS and ONS sources, while the statistical data definitions differ by topic. The following paragraphs provide more detail.

The high street features data are provided by OS. As well as information on the number of high streets, and names of high streets, there are also data on the proportion of addresses that are for residential, retail, leisure and recreation, offices or community usage.

Please note that retail usage for the OS high streets data is classified by land use type as defined by the Ministry of Housing, Communities and Local Government. In other words, it differs from the retail definition used for the statistics in the business and employment sections of the output (see following), which are based on the definitions of the Standard Industrial Classification. So care should be taken when comparing the retail data between these two sections. More details of the definitions can be found in the Quality and Methodology section.

The figures in the remainder of the article are based on data for geographical areas that closely approximate the high street extents (and in some cases the areas around the high streets).

The figures for house prices and transactions are from HM Land Registry who provide data on the price paid and number of residential property transactions for properties sold in England and Wales. The data are based on a postcode lookup to very closely match the geographic high street extents.

The data for businesses and employment are taken from the ONS Inter-Departmental Business Register (IDBR). As with the housing data, figures have been compiled based on a postcode lookup to very closely match the geographic high street extents.

The data for population deliberately looked beyond just the high street itself and have been designed to include people living within approximately 200 metres from each high street. We calculated these figures for England and Wales using ONS Census 2011 Output Area population-weighted centroids and the latest 2017 mid-year population estimates. For Scotland we used the same population estimates for Scottish Government Census 2011 Data Zone population-weighted centroids.

Therefore, when comparing the data across different sections of the article, please be aware that while data on housing, business and employment are calculated to match as closely as possible to the geographic extents shown in the map included, the population data by contrast include a wider 200 metres area around the high street to capture local populations with a short walk to the high street area.

In summary, the geographic high street extents are produced on a consistent basis for Great Britain by OS, but the statistical data definitions used to attach data vary depending on the topic as discussed previously. We give breakdowns of the data in the article by region. We hope that users will find this work useful and we welcome comments to help us make future improvements.

How many high streets and what type of addresses?

We have identified high street extents based on clusters of 15 retail addresses within 150 metres, linked to roads. This gives us 6,969 individually-named shopping streets in Great Britain, some of which are neighbours, representing what we talk about in this article as “high streets”. The number of high streets by country and region is shown in Table 1.

Previously published data show that the longest continuous shopping street in England, calculated using the Ordnance Survey (OS) methodology, is London Road in Southend-on-Sea, at 2,983 metres. The longest high street in Scotland is Dumbarton Road in Glasgow at 1,706 metres and in Wales it is High Street in Bangor at 1,265 metres. The length of individual British shopping streets varies, but they are typically between 200 metres and 500 metres long.

Around a third of addresses on the high street in Great Britain belong to retail shops. More than half of the addresses on British high streets are residential. In London this figure rises to 64%.

In most regions, around 10% of addresses are offices and around 2% to 3% are leisure or community facilities. These address data are classified by land use type as defined by the Ministry of Housing, Communities and Local Government. Figure 1 shows the mix of addresses by regions.

You can see address data for individual high streets by linking to an interactive map on the OS website, see Section 7 of this article. You can put a postcode or place name in the search box and see how your high street compares with others. If you click on an individual street you can see more information, like the number of buildings and the proportion of retail to other address types.

Shopping street names

Clearly, not all British shopping streets are named “High Street”, although among the streets we identified it was the most popular name. There are three “market” references and other names like London Road, Station Road and Church Street suggest why shopping streets have built up where they are. Table 2 shows the top 30 shopping street names in Great Britain in 2019.

Population, 2017

This section of the article, examining population data, differs from other sections of the article in that it includes areas within approximately 200 metres of a high street as well as the high street itself. In doing so, this section aims to provide information on the population living within a short walk of a high street. Using this approach, we see 16% of Great Britain’s population living on or around the high street in 2017.

Unsurprisingly, London has a much higher proportion of its population living close to high streets, 38%. The North West, and Yorkshire and The Humber are close to the national figure, while other English regions, Wales and Scotland have between 11% and 13% of their population living around the high street. Figure 2 shows the proportion of each country or region’s population living on and around high streets in 2017.

Population change, 2012 to 2017

All countries and regions have seen growth in their population between 2012 and 2017, but it is particularly notable around the high street, compared with non-high street areas.

The largest difference is in the East Midlands where the population grew from 514,953 in 2012 to 554,264 in 2017, a rise of 8%. Population growth in areas beyond the high street in the East Midlands was 4%.

In England, six of the nine regions saw population growth of at least 6% around the high street. Wales and Scotland also saw higher population growth around their shopping streets, compared with other areas, although the gap was narrower than in England. Figure 3 shows the population change between 2012 and 2017 across countries and regions.

Residential property transactions

The data in this section come from HM Land Registry (LR) and provide statistics on the price paid and number of residential property transactions for properties that were sold in England and Wales. Rental properties and those sold at a discount to market level, such as properties sold under the Right to Buy scheme, are not included in these statistics.

Sales of high street dwellings represented 2% of all residential property transactions recorded by HM LR in England and Wales in 2017. The proportion of all transactions that occur on high streets has remained largely unchanged over the last 20 years.

As you might expect, a high proportion of the dwellings sold on the high street in England and Wales are flats. The mix of dwelling types has changed notably over the past 20 years.

In 1997, flats represented 40% of all sales in high street areas, terraced housing represented 41%.

By 2017, flats accounted for 62% of residential property transactions on the high street, compared with 18% outside of these areas. This could be explained by the conversion of existing terraced stock, but also by new purpose-built blocks of flats, aimed at students, young professionals and older residents in many city and town centres.

Terraced housing had fallen to 25% of property sales on the high street by 2017, much closer to the 28% seen in other areas. Figure 4 shows the change in the proportion of transactions by dwelling type since 1997.

Although the share of flats on the high street is higher than in the past, there remain notable proportions of terraced house transactions in some regions. The total transaction figures are fairly small, but the split is still clear.

In the North East, the North West, Yorkshire and The Humber, the East Midlands and Wales, at least 40% of the dwelling stock sold on the high street in 2017 was terraced housing. Figure 5 shows the proportion of flats and terraced houses, on the high street, across English regions and Wales in 2017.

Median property prices

The median price paid for residential property on high streets was lower than elsewhere, across all English regions and Wales, in 2017.

The difference was smallest in London where high street residential property prices were 10% lower than elsewhere. The largest difference was in the South West where prices on the high street were a third lower. Figure 6 shows the median price paid for all dwellings at regional level in 2017.

These differences seem to be influenced by the proportion of flats and terraced houses within the mix of property transactions on the high streets, compared with the wider area. Flats and terraced houses tend to be cheaper than detached and semi-detached housing, which feature strongly outside of town centres. So, the lower prices on the high street at regional level do not necessarily mean that dwellings are less desirable or cheaper because of their location, rather that high streets tend to be dominated by cheaper dwelling types.

Figure 7 shows how the prices of flats and terraced houses compare on the high street with the same dwelling type in other areas in the region in 2017. Using this comparison, we see that there are some regional variations in whether the same type of residential property is more or less expensive on the high street.

In four regions, flats achieved higher prices on the high street. The median price was 14% higher in the North East, in the East of England it was 6% higher and in the North West and the West Midlands the median price was 4% higher.

The median price paid for terraced housing was 7% higher on the high street in the South West. Figure 7 shows the percentage price difference for flat and terraced housing on the high street.

Number of businesses

In 2017, we estimate that there were 404,145 businesses¹, or local units, registered in postcodes in the immediate vicinity of a British high street. This represents 13% of all businesses in the country.

London had the largest number of British high street businesses, 127,715, and these accounted for 23% of the capital’s total businesses, a notably higher share than in other regions. High streets represented between 10% and 13% of total businesses in all other English regions, Wales and Scotland.

Table 3 shows the regional and country breakdown of the number of businesses on the high street and elsewhere in 2017. Note that totals may not sum, due to rounding.

We have looked at businesses over a five-year period, to give a feel for change on the high street, compared with wider areas. It is important to remember that, as with our other data, the high street extents we are working with are a snapshot from 2019 and streets may have changed in physical size.

The number of high street businesses has grown in all English regions, Wales and Scotland between 2012 and 2017, but at a lower rate than businesses beyond high street areas. This difference is more than 10 percentage points in six of the nine English regions and between 6 and 8 percentage points elsewhere.

The West Midlands was the only region to see higher growth in businesses on the high street, compared with elsewhere, but this seems to be influenced by non-retail, notably transportation and storage businesses. Figure 8 shows the percentage change in the number of businesses on the high street, compared with other areas in 2012 to 2017.

Types of businesses

Our research confirms the general view that the number of retail businesses on the high street has been falling.

At a national level, retail has fallen as a proportion of all high street businesses from 29% to 25% between 2012 and 2017. We see lower numbers of high street retail businesses over the last five years across all regions, except in London where they grew by 4% and in the North West where they grew by 7%. These areas also saw relatively high levels of growth in retail businesses outside of the high street.

The number of retail businesses has grown outside of traditional high streets in most parts of Great Britain between 2012 and 2017, although some areas saw little change or a slight fall in numbers. Figure 9 shows the percentage change in the number of retail businesses on the high street, compared with the wider region or country in 2012 to 2017.

While the number of retail businesses have declined in most regions, the number of businesses in many other sectors have been increasing over the same period. Table 4 lists the different industry sectors, showing both their importance to the high street (shown as the share of total high street businesses that each industry contributes) and the changes in the number of businesses by sector from 2012 to 2017.

The data show that the number of business units on the high street in Great Britain in the accommodation and food service activities sector increased by 20% over the same 2012 to 2017 period in which the number of retail businesses on the high street declined by 2%. At the same time, the number of businesses in other service sectors (including sectors that typically operate from office locations) also increased sharply, rising by 25%.

Employment on the high street

Figure 10 shows the share of employment on high streets in 2017 by industry (using the same industry breakdown as in Table 4). For most regions and countries of the UK, the retail sector provides between 25% and 31% of high street employment (the figure is lower in London at 18%). A further 13% to 18% of high street employment in all regions and countries is in the accommodation and food sector.

The other service sector category, which includes sectors that typically operate from offices in addition to banks and estate agents, makes up between 29% of high street employment in Yorkshire and The Humber and 38% in the East of England to 46% in London.

The largely public sector-based industries of health, education and public administration provide between 10% of high street employment (in the South East and the South West of England) and 19% in Yorkshire and The Humber.

Figure 11 shows that there was mixed performance in terms of employment growth in “other service industries” over the 2012 to 2017 period. In the West Midlands, the East Midlands and London, employment on high streets in these mostly office-based sectors increased by over 20% and in Wales and Scotland by over 10%. However, there was a slight decline in employment in the North East, and Yorkshire and The Humber regions.

Employment on the high street in the retail sector, meanwhile, declined in every region and country except for London (where it grew by 6%). In contrast, employment in the accommodation and food sector on high streets expanded between 2012 and 2017 across the whole of Great Britain. All regions and countries experienced employment growth of at least 20% in the sector. In the West Midlands, and Yorkshire and The Humber, this growth was over 40%.

Notes for: Businesses and employment on the high street

1. "Businesses" in this article refers to local units, which are individual business sites. An enterprise may have more than one local unit across the country. However, in the context of high streets, it is the number of local units that are of interest, rather than the number of enterprises, and therefore the article and data focus on the number of local business units.

You can see the location of high streets captured in the Ordnance Survey methodology by viewing the interactive map on the OS website.

You can put a postcode or place name in the search box and see how your high street compares with others. If you click on an individual street you can see more information, like address types, or number of buildings.

High street coverage

Not all streets that may be considered a high street in Great Britain are captured in the Ordnance Survey (OS) geography. For example, there is no high street visible in our map for Hugh Town on the Isles of Scilly because the density of the retailing does not meet the criteria OS has defined. We have decided to not make a manual intervention to include any shopping street here, as this would make the classification incomparable across other similar areas, such as for some islands in Scotland.

Likewise, a small selection of high streets are not visible in the map for the centre of cathedral cities, like Norwich. In this case, reworking the methodology to automatically capture the streets results in the inclusion of many smaller extents that do not meet the general expectation of a high street definition. There may be other exceptions, or indeed inclusions, that users find unexpected; we welcome your comments, please get in touch.

Ordnance Survey high streets geography technical specification summary

During the last year, OS has been working with the Office for National Statistics (ONS) and the Ministry of Housing, Communities and Local Government to improve our joint understanding of high streets.

Given that no definition of high streets currently exists, OS has created an experimental dataset, using an automated methodology, which aims to depict the extent of Britain’s high streets and their land use. The dataset is a high street-specific dataset that provides a street-level definition. It bounds retail clusters using street naming and aims to deliberately exclude other retail functions including retail parks, industrial estates and isolated shopping centres.

The dataset is derived from data products licensed under the Public Sector Mapping Agreement (PSMA) and One Scotland Mapping Agreement (OSMA) and is available for evaluation to PSMA or OSMA members by request only.

The high street dataset considers a high street as an extent along a named street predominately consisting of retailing. Spatial clustering of retail address point geometries from the dataset OS AddressBase Plus on a street-by-street basis was used to form candidate retail clusters from which a high street extent could be derived. Retail addresses were identified using the local authority-defined address classification.

Spatial clustering has been implemented using the data clustering algorithm DBSCAN (Density-based spatial clustering of applications with noise). Candidate retail clusters required a minimum of 15 retail address point geometries within 15 buildings and a maximum of 150 metres between the address geometries in the cluster.

In order to evaluate the land use profile of each high street extent the retail clusters were expanded or infilled to collect all address point geometries and their associated building polygon geometries, on the same named street within 50 metres of a retail cluster. All high street attributes are derived from the infilled clusters and not the original retail clusters. Building geometries are sourced from the OS dataset OS MasterMap Topography Layer.

The red geometries in the images in Figure 14 describe building polygon geometries in a candidate high street cluster containing retailing address point geometries. The blue geometries in the second image describe building polygon geometries added to the cluster following the infilling process.

Figure 12: A candidate high street retail cluster visualised using the buildings that the retail addresses are associated with

Source: Ordnance Survey

Download this image Figure 12: A candidate high street retail cluster visualised using the buildings that the retail addresses are associated with

.png (274.6 kB)

Figure 13: A final high street cluster collecting all property on a street within the bounds of a candidate retail cluster. The blue buildings are joined to the red buildings as part of an infilling process

Source: Ordnance Survey

Download this image Figure 13: A final high street cluster collecting all property on a street within the bounds of a candidate retail cluster. The blue buildings are joined to the red buildings as part of an infilling process

.png (280.2 kB)

Other retail functions such as retail parks, industrial estates and isolated shopping centres have deliberately been excluded from the dataset using a series of tests. Each test evaluates each cluster against a combination of form and function criteria.

Whilst every effort has been made to exclude other retail functions we are aware that there are a small number of non-high street geographies remaining in the dataset. For more information on the exclusion tests please see the full technical specification that is available from the Ordnance Survey link in Section 8.

The attributes associated with the address records that make up a high street extent have been used to derive a series of lookups, which relate a high street geography to a number of statistical and geographical identifiers, including the Unique Property Reference Number (UPRN). The ONS has also used a 200 metre intersect of 2011 Census Output Areas in England and Wales and Data Zones in Scotland and the high street building geometries to match some data where postcode-level data are unavailable.

Whilst this work explores high street geography for 2019 only, OS is also considering approaches to deriving additional retail geographies and insights around the changes to the retail landscape over time.

Population estimates

We calculated population figures for England and Wales using ONS Census 2011 Output Area population-weighted centroids and mid-year population estimates. For Scotland we used the same population estimates for Scottish Government Census 2011 Data Zone population weighted centroids.

We have looked at estimates of the population living within approximately 200 metres of a high street. It is a starting point for analysis while other methods are possible like walking distance or driving catchment areas. Further information on the quality of ONS small area population estimates is available.

House price statistics

This article uses postcode-level data and the same methodology as ONS house price statistics for small areas (HPSSAs). These data are taken from HM Land Registry and provide statistics on the price paid and number of residential property transactions for properties that were sold in England and Wales. Rental properties and those sold at a discount to market level, such as properties sold under the Right to Buy scheme, are not included in these statistics. Further information is available in the Quality and Methodology Information report.

Inter-Departmental Business Register

The Inter-Departmental Business Register (IDBR) is a comprehensive list of UK businesses used by government for statistical purposes. It is fully compliant with the European Union regulation on harmonisation of business registers for statistical purposes (EC No 177/2008) (PDF, 90KB).

The IDBR provides the main sampling frame for surveys of businesses carried out by the Office for National Statistics (ONS) and other government departments. It is also an important data source for analyses of business activities. The IDBR covers 2.6 million businesses in all sectors of the UK economy, other than very small businesses (those without employees and with turnover below the tax threshold) and some non-profit making organisations.

The two main sources of input are the Value Added Tax (VAT) system from HMRC (Customs) and Pay As You Earn (PAYE) from HMRC (Revenue). Additional input comes from Companies House, Dun and Bradstreet and our business surveys.

Standard Industrial Classification (SIC) 2007

The current Standard Industrial Classification (SIC) (2007) is used in classifying business establishments and other statistical units by the type of economic activity in which they are engaged. The classification provides a framework for the collection, tabulation, presentation and analysis of data, and its use promotes uniformity. In addition, it can be used for administrative purposes and by non-government bodies as a convenient way of classifying industrial activities into a common structure.

Ministry of Housing, Communities and Local Government Land Use Classification

The latest land use classification categories are the result of a collaborative development and data specification agreement between the Ministry of Housing, Communities and Local Government, and Ordnance Survey. More information is available within the Land Use Change Statistics 2013 to 2014 Methodology changes guidance (PDF, 1.77MB).

Ordnance Survey can provide more information on the new high street geography, including the full technical specification and a map with more explorable data.