International exports of services from subnational areas of the UK: 2017

• London contributed over 40% of service exports in 2017.

• Most of the 12 NUTS1 regions in the UK accounted for between 40% and 45% of service exports to the EU.

• Financial and insurance services formed the biggest industry exporting services for 10 NUTS1 regions.

• The joint authorities outside London that had the largest service exports were Greater Manchester (£8.2 billion) and Edinburgh and the South East of Scotland (£7.0 billion).

• Of the NUTS3 areas, South Teesside (North East) had the highest percentage of exports to other EU countries (74%), while Central Valleys in Wales had the lowest percentage (11%).

• As a result of methodological changes, our outputs are now consistent with the UK Balance of Payments and all estimates are on an industry basis.

In this article, we provide estimates of the value of international exports of services from 2017 from the Nomenclature of Units for Territorial Statistics (NUTS)¹ NUTS1 countries and regions of the UK. We also present all results on a complete industry basis, moving away from the previous hybrid approach, including breakdowns by smaller geographies and by destination of export. The move away from the hybrid “functional” categories means that all categories now align to industry groups based upon the Standard Industrial Classification (SIC) 2007; we therefore no longer refer to product groups. This is part of our continued analysis of subnational trade through the Office for National Statistics’ devolution programme.

The UK Balance of Payments shows the value of goods exports and of service exports, separately, from the UK to other countries. HM Revenue and Customs’ Regional Trade Statistics provides estimates of the value of goods exports from various subnational geographies. In this article, we present estimates of the values of services exported by each NUTS1 area and joint authority² as well as some analysis of NUTS3 data. The dataset published with this release includes estimates for NUTS1, NUTS2, NUTS3 and joint authority areas.

Notes for: Introduction

1. The Nomenclature of Units for Territorial Statistics (NUTS) is a hierarchical classification of administrative areas, used across the EU for statistical purposes. There are 12 NUTS1 regions in the UK: Wales, Scotland, Northern Ireland and the nine former English Government Office Regions.
2. The joint authorities consist of the seven English Combined Authorities, the mayoral authorities Sheffield City Region and Greater London (split into Inner and Outer London), three Scottish City Deals, and two Welsh City Deals. Please see the Appendix for a list of the joint authorities.

This release uses many of the same processes as our previous articles, but we have made some notable improvements to our methodology based upon wider methodological developments and in order to meet user needs. As a result of these changes, this release is not entirely consistent with our previous subnational estimates and we therefore advise caution when comparing these results with previous results.

Much of the analysis has been conducted on the same methodological basis as our previous outputs. We use the International Trade in Services (ITIS) survey dataset alongside UK Balance of Payments information. For the ITIS survey, we use the data collected at the reporting unit level and apportion to the local unit level based on employment. While for previous releases we would group these results by enterprise and then allocate to the local unit, we have now chosen to allocate directly from the reporting unit. This is because the ITIS survey is collected on a reporting unit basis and adding up to the enterprise unit creates unnecessary extra inaccuracies.

For instance, say that an enterprise unit has two reporting units (1 and 2), of which the former exported £100 million and the latter £10 million, but they both had the same number of employees distributed across four local units: 1A, 1B, 2A and 2B. Each local unit has 25 employees and the local units belonging to reporting unit 1 (1A and 1B) are based in the North East while local units 2A and 2B are based in the North West. The previous method would have seen the enterprise unit total (£110 million) split equally four ways to allocate £55 to 1A and 1B in the North East and £55 to 2A and 2B in the North West. The current method removes this inaccuracy and allocates £50 million to 1A and 1B each, therefore allocating the full £100 million to the North East. The £10 million exported by reporting unit 2 gets split fifty-fifty between the local units 2A and 2B (both in the North West), therefore allocating the £10 million to the North West. Please note that this is a simplified example. Local units can be spread out across multiple regions and the number of employees is unlikely to be the same in every local unit.

Once broken down, we then aggregate the figures from each local unit to create totals for each geographical breakdown and each industry breakdown based upon information from each local unit. We use the industry codes of the local unit and constrain our totals to match recently published UK totals by industry. Note that the UK industry totals are based on the industry codes of the reporting unit but we have chosen to use the local unit to reflect local industrial compositions more accurately, while preserving consistency in our articles.

For exports not covered by the ITIS survey, we use employment figures from the Business Register and Employment Survey (BRES) as a proxy to derive regional and sub-regional estimates from the national-level figures presented in the Pink Book. For financial services, we also use Bank of England data as part of the process at the NUTS1 level. We have not used the International Passenger Survey (IPS) to reallocate as we had done in previous publications, as “Travel” as a category no longer exists (please see the next paragraph). However, we may review the use of the IPS future publications as it does shift some value away from London. For a detailed description of this methodology, please refer to our first publication.

Significant changes made in this release

The first significant change made in this release is creating outputs on a complete industry basis. In previous publications we used “functional categories”, which were a hybrid concept that included both industry and product categories to break estimates down based upon the data sources available. Information derived from the ITIS survey can be directly measured in terms of industry. Information derived from other sources, however, cannot. The recently published UK trade in services by industry, country and service type: 2016 to 2017 provides a new approach to converting non-ITIS survey product categories to an industry basis, which uses data-linking techniques and an additional data source, the Annual Survey of Goods and Services (ASGS), to approximate proportions of exports on a product basis to exports estimates on an industry basis. Applying this same methodology, we have converted the four product categories from our “functional” categories to industry categories comparable with the UK trade release. All outputs are now on an industry basis.

In the cases of the transport, finance and insurance product categories, the eponymous industries absorb much of this value, although some of these service products have been reallocated to other industries. However, the “travel” product category does not have such a natural home. A significant proportion of travel services that used to be presented on a product basis have now been allocated to the accommodation and food services industry, as well as to other industries such as education and administrative and support service activities. This means that previous estimates of services exported by such industries in prior publications would have excluded those services but will now include them. We cannot update the estimates for previous years, as the conversion has only been done for 2016 and 2017 thus far.

The second significant change is that we provide estimates broken down by two destinations: exports to the rest of the EU, and exports to the rest of the world (non-EU countries). Our previous publication from July 2017 also published service export figures by country of destination, but this was only at the NUTS1 level and only included the ITIS survey-based components of trade (covering just 45% of total service exports). This publication contains data on the service exports by industry and destination (EU and “rest of the world”) for all industries, although it provides less granularity in terms of destination. Future developments may include looking into providing more granularity of destinations for at least the NUTS1 regions.

For information sourced from the ITIS survey, the destination can be derived directly from the survey. It is subject to the same assumptions as our general methodology, namely that each local unit has contributed to the exports to this destination. For the non-ITIS components, we are currently limited to UK-wide information provided by the UK Balance of Payments. We therefore make the assumption that the exports of a particular industry to the EU and rest of the world are similar for each region as they are at the UK level. For example, if 40% of all UK service exports in the financial industry go to the EU, this percentage has been applied uniformly to the financial exports value in all regions, as we do not currently have a method to identify any differences. Users of the data should bear this assumption in mind when interpreting the non-ITIS destination data. Subject to data availability, future work may be done to review the use of this assumption. We have chosen not to provide more granularity in terms of destination largely because we do not currently have more accurate measures to estimate regional service exports by destination for the non-ITIS data, but also for reasons of confidentiality.

The third significant change is that these estimates have been constrained to match the UK Balance of Payments for the first time. This means that our estimates are consistent with the UK Balance of Payments and other National Accounts publications. To do so, we have used a method called “constraining”, to adjust breakdowns of values to match known totals, which means we have proportionally adjusted the subnational estimates to align with UK-level estimates. The process of constraining means the resultant outputs may no longer match the constituent component calculations, as values will have been added or subtracted between components during constraining. However, it means that subnational outputs match UK-level figures for each combination of industry and destination.

Please note that the first and the third methodological change may have a noticeable impact on some figures. For instance, the sum of service exports from London in 2017 is estimated to be £117 billion, which is the same as the 2016 estimate for London we published in October 2018. Considering the overall increase in the UK Balance of Payments estimates from £258 billion to £279 billion, one might expect the value for London’s exports to increase as well. This fairly small increase for London compared with our previous output may be a result of low growth in trade, or it may be, in part, explained by our methodological changes. Such differences are most notable when comparing region split by category rather than regional totals, in particular in the retail, administrative and support services industries and “other services”, which are subject to what appear to be large changes compared with previous years. These industries have changed largely as a result of reallocation from the functional categories, particularly the “travel” product, which does not have its own industry; all travel exports have therefore been added to other industries. The constraining process has also affected some industry totals. Caution must therefore be used when comparing the data presented here with previous estimates.

As shown in Figure 1, the largest share of service exports at the regional level came from London, which contributed 41.8% to the total (£116.6 billion), followed by the South East (£44.7 billion) and Scotland (£21.9 billion). London seems to be the NUTS1 area most dependent on service exports in comparison to their Gross Value Added (GVA), as service exports were about 27% as a percentage of their £431 billion balanced GVA in 2017. We also see fairly high percentages for the South East, for which service exports were 17% as a percentage of their GVA, and Scotland (16%).

In five regions, service exports were between 10% and 15% as a percentage of their GVA: North East, North West, East of England, South West and Wales. In four regions, service exports were estimated at less than 10% as a percentage of their GVA: the West Midlands (9%), Yorkshire and The Humber (8%), East Midlands (7%), and Northern Ireland (7%). While the figures presented in this output are experimental, this suggests that London, the South East and Scotland have a relatively high reliance on service exports compared with other regions. For the UK as a whole, service exports were about 15% as a percentage of GVA.

Most UK service exports were from the financial and insurance activities industry, generating more than a quarter of the total (£75.9 billion). Service exports in the professional, scientific and technical industry as well as the information and communication industry also formed substantial shares of the total (15.7% and 15.4%, respectively).

Figure 1: Subnational services by NUTS1 region and industry

UK trade in services by industry, country and service type

At the regional level, financial and insurance activities were the largest category in 10 out of the 12 NUTS1 regions and were in the top 3 in the remaining regions. The only regions in which financial and insurance activities did not form the largest category were the South East, for which professional, scientific and technical services formed the largest category, and Wales, where manufacturing was the largest industry at 32%. Manufacturing also formed a large part of the North East’s service exports, to which it contributed almost 20% or £1.1 billion. Please note that services exported by the manufacturing industry can be directly linked to manufacturing, such as the installation of machinery, as well as other services that may be performed by manufacturers such as 0% financing or offering extended warranties. It does not refer to the exports of manufactured goods.

Transportation and storage service exports also appeared among the industry groups with the highest value for many of the NUTS1 regions, being in the top three in four regions. Professional, scientific and technical activities formed a substantial portion of service exports in southern parts of England (including London, the South East and East of England), the East Midlands and Scotland, as this industry was one of the three highest value industries for these regions.

In fact, three industries formed a very large part of London’s service exports: information and communications; financial and insurance activities; and professional, scientific and technical activities. When added together, these three industries contributed nearly three-quarters (73%) of London’s overall service exports. Other areas saw a similar focus on a couple of industries, with Wales, for instance, selling 32% of its service exports in manufacturing and 28% in finance and insurance activities.

Looking at the regional composition by industry, London was the largest region for all but two industry groups: primary and utilities, and manufacturing. Scotland exported just over three-quarters of all service exports in primary and utilities, while the South East had the largest share in manufacturing service exports (just over a fifth). London and the South East generated more than half of the UK’s exports by several industries: wholesale and motor trades; information and communications; finance and insurance services; real estate; professional, scientific and technical services; and administrative and support.

This is the first time we are providing estimates by destination for the EU and non-EU. While for the International Trade in Services (ITIS) survey this can be estimated directly, we have had to make assumptions about the destination of the non-ITIS parts. Please see the Things you need to know about this release section.

For the UK overall, £163.9 billion went to outside the EU while £114.7 billion went to the EU-27¹. For most regions (except for Northern Ireland) exports to the rest of the world were estimated to be somewhat higher than exports to the EU. Most of the other regions typically exported between 55% and 60% of their service exports to non-EU countries. However, for two industry groups (retail and wholesale and motor trades), exports to the EU were higher than to the non-EU overall, which is reflected in most of the regions. The most equal distribution of total service exports was seen in the North East, which according to our estimates exported £2.7 billion to the EU and £2.8 billion to the rest of the world.

Notes for: Service exports from the NUTS1 regions by destination

1. The European Union consists of 28 members: Austria, Belgium, Bulgaria, Croatia, Cyprus, Czechia, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden and the United Kingdom. It also includes several other bodies, namely the European Central Bank, European Union institutions and European Stability Mechanism.

Together, the 15 joint authorities (the seven English Combined Authorities, the Sheffield City Region, Inner and Outer London, and three Scottish and two Welsh City Deals) contributed more than 60% (£171.8 billion) of the UK’s service exports. Inner London exported the highest value of services in 2017 at £97.0 billion, followed by Outer London (£19.5 billion) and Greater Manchester (£8.2 billion).

Figure 2: Heat map of service exports by joint authority

UK trade in services by industry, country and service type

Figure 2 shows the distribution of services per joint authority, with the areas in darker red shading highlighting high proportions of exports and the lighter shades of yellow representing small proportions of exports¹. Service exports were, as might be expected, particularly large in the service sectors. Financial and insurance activities were the largest industry in 10 joint authorities and in the top 3 industries for 12 of the 15 joint authorities, with the exceptions of Cambridgeshire and Peterborough, Aberdeen City Region, and Outer London. Professional, scientific and technical activities were also prominent and formed one of the top 3 industries in 5 joint authorities. It formed the largest industry group in Cambridgeshire and Peterborough and in Aberdeen and Aberdeenshire.

Looking at the production industries, manufacturing was the largest industry in terms of service exports value for Tees Valley and the Cardiff Capital Region. Aberdeen City Region formed the only joint authority for which non-manufacturing production services was one of the largest industries, as it had a share of £0.9 billion of its total of £4.7 billion.

For most industries, either Inner or Outer London were the largest joint authorities, except for manufacturing, for which Cardiff Capital Region had the largest value, and non-manufacturing production services, dominated by Aberdeen City Region.

Notes for: Service exports from the Joint Authorities

1. Please note that for the joint authorities, two changes have been made to the industry groups used at the NUTS1 level: primary and utilities and construction have been combined to form “non-manufacturing production” services, and real estate has been added to “other services”.

Similar to the NUTS1 regions, most joint authorities exported most of their services to non-EU countries, with the exceptions of Tees Valley (£0.5 billion exported to non-EU countries and £0.8 billion to the EU) and Outer London (£9.7 billion exported to non-EU countries and £9.8 billion to the EU). Inner London saw the biggest absolute difference between EU and non-EU exports, exporting £36.8 billion to the EU and £60.3 billion to non-EU countries.

Other big differences between EU and non-EU exports were seen in Cardiff Capital Region and Cambridgeshire and Peterborough. Cambridgeshire and Peterborough exported more than 50% more to non-EU countries than to the EU (£2.6 billion versus £1.7 billion). This was largely driven by professional, scientific and technical services, of which the joint authority exported approximately three times as much to the rest of the world as to the EU (£0.9 billion versus £0.3 billion). Cardiff Capital Region exported more than twice as much to non-EU countries as it did to the EU.

For the 168 NUTS3 areas in Great Britain, we only provide a split by EU versus non-EU and do not provide the industry dimension. This is largely because of concerns around reliability and confidentiality at this level of granularity. We are unable to process NUTS3 areas for Northern Ireland owing to limitations in our data and our estimates therefore only cover Great Britain. Please note that readers should be more cautious when relying on these numbers than on the NUTS1 regional figures. While all our figures are experimental, more granular figures are subject to more variation, particularly in the allocation of values from sources other than the International Trade in Services (ITIS) survey. This applies particularly to regions that are smaller in terms of value.

The average value of service exports from each NUTS3 area was approximately £1.6 billion, although this average drops to £1.1 billion if London is excluded. The average NUTS3 area in London exported approximately £5.6bn. Camden and The City, one of London’s NUTS3 areas, exported the largest value (£36.9 billion) followed by Westminster (£24.8 billion).

The following map provides information regarding the destination of the NUTS3 areas and whether this is to the EU or the rest of the world. The NUTS3 area that exported most to the EU as a percentage of its total was South Teesside in the North East, which exported 74% (of £0.5 billion) to EU countries. This was followed by Ealing in London (69% of a total of £1.7 billion) and Durham, also in the North East (64% of £0.8 billion).

Conversely, the NUTS3 areas that exported the highest percentage of their total to non-EU countries were Central Valleys in Wales (89% of £2.1 billion), South Ayrshire in Scotland (78% of £0.9 billion), and Lambeth in London (72% of £2.4 billion).

Figure 3: Percentage of service exports to the EU by NUTS3 region, 2017

Please note that these estimates provide some insight into the extent to which areas export to EU countries, but they do not provide any further granularity. While most regions export more to the rest of the world than to the EU, it should be kept in mind that the EU is one trading bloc consisting of only 27 countries (excluding the UK). The rest of the world, in comparison, consists of more than 160 countries.

These estimates remain experimental as development work continues. While we have met an important milestone by providing these estimates on an industry basis and including a destination breakdown, we intend to review our current processes and methodology in the coming months. We will try to ratify the accuracy and usefulness of the outputs, in particular through direct engagement with users.

Our aim is to create an ongoing annual publication containing the breakdowns provided here. We have also had demand to produce estimates of subnational service imports alongside this current publication. Separate analysis on this is ongoing, and we aim to eventually deliver outputs on subnational imports of services. This analysis may also inform further methodological changes in our exports work. In due time, we should therefore see a complete suite of outputs including both imports and exports on an industry basis and providing information on the origin or destination.

We welcome feedback from users through the statistical contact for this release, particularly suggestions toward improving the methodology or upon whether this approach meets user needs.