1. Overview of new Foreign Direct Investment Survey questions

  • Analysis of the questions that were recently added to the Foreign Direct Investment (FDI) Survey as part of our development work shows a very similar pattern of responses in 2021 as previously reported for 2020.

  • Just over a half of respondents allocated all of their unquoted equity to one or more regions of the UK, whereas around 45% did not answer or could not allocate their values.

  • The survey questions are more likely than data-linking to allocate FDI to a single region and results from the survey questions are more weighted towards London than in our data-linked experimental statistics.

  • Companies were least able to answer the subnational equity question of the UK total FDI position in the mining and quarrying (54.7% inward, 66.4% outward), retail, wholesale and transportation (54.1% inward, 61.9% outward) and information and communication (70.7% outward) industries.

  • The analysis in this report suggests that there are limits to what companies can tell us about their FDI by UK subnational region or on the type of FDI; although the data from these questions can provide useful indicative information, they are not comprehensive, and we will investigate how the subnational survey responses complement our data-linking methods.


This progress report includes indicative FDI results that weight UK subnational totals using survey response components. The percentages give an indication of where in the UK respondents allocated their reported values. None of the values in this report should be used for analysis or decision making.

Back to table of contents

2. New survey questions

We have been developing our foreign direct investment (FDI) statistics to enhance the quality of these estimates to meet user needs. This includes providing timelier, more granular FDI statistics.

We have added new questions to the annual and quarterly FDI surveys. We ask companies to report the subnational location of a component for their stock of FDI (position) and profits (earnings).

We have also added questions asking for the type and funding of FDI activity. The type of activity was split into the following four categories, and further split by acquisitions and disposals:

  • new direct investments (greenfield)

  • extensions of capacity

  • acquisitions and disposals

  • other

This information was collected for 2020 and 2021 covering both UK-resident companies with foreign parent companies (inward), and UK-resident companies that control affiliates outside the UK (outward).

Back to table of contents

3. Responses to the new survey questions

Initial results from the new questions were summarised in our Foreign direct investment, overview of new survey questions: March 2022 article. This report now includes revised results from 2020 and current 2021 responses that were used in our Foreign direct investment involving UK companies: 2021 annual bulletin.

Foreign direct investment (FDI) by UK country and region

The new questions ask companies for their International Territorial Level 1 (ITL1) region for two components of FDI: unquoted equity and profits. Companies can also report that values cannot be allocated to any ITL1 country or region.

Inward investment

Nearly half of companies completing the subnational questions in the annual inward FDI Survey for 2021 allocated all their value (equity) to one region. Around 47% of respondents either did not answer the question or could not allocate their values to a UK country or region (Table 1). These proportions were similar to the 2020 results.

A higher proportion (almost two-thirds) of inward companies could not or did not answer the income question. The 2021 proportions of respondents in each category for income were similar to those in 2020.

Outward investment

The comparison of responses for outward FDI largely mirror patterns for inward FDI. Just over 56% of respondents for outward FDI allocated unquoted equity values to one region in both 2020 and 2021, respondents with one-quarter not answering the question (Table 2). In 2021, fewer respondents provided subnational information for outward income (45.9%), compared with for equity (57.4%). There were also a higher proportion of companies not answering the outward subnational income question (38.6%), compared with subnational unquoted equity (25.3%).

In the open-ended responses to these questions, companies still reported that their internal financial systems did not recognise UK locations. Instead, they recorded results for their operations in the UK as a whole.

FDI by type or source of funding

FDI by type provides information on the purpose of transactions, such as new investments or extensions of capacity. The source questions show how these transactions were funded, for example, from own funds or internal loans.

For 2020, most respondents did not provide any information in these questions. The situation remained largely unchanged for 2021. Outward acquisitions by type answers in 2021 were similar to 2020, whereas no respondents reported any inward acquisitions values by type or source of funding in 2021. As was found in our 2020 analysis, more information was collected for disposals than acquisitions in 2021, but the vast majority (at least 94%) of respondents did not answer these questions.

The low amount of information collected makes it difficult to interpret results. We will continue exploring why respondents are not answering these questions, along with considering other methods for compiling FDI by type statistics.

Back to table of contents

4. Indicative results by UK country and region

We used reported component information by UK country and region from foreign direct investment (FDI) companies to weight the total FDI positions or earnings. The percentages give an indication of where in the UK respondents allocated their reported values compared with our experimental subnational estimates that use data-linking.

FDI positions

Most companies providing information on the subnational composition of their FDI position allocated it to London. The proportion for inward FDI (69.6%) was higher than outward FDI (51.8%) in 2021 (Figure 1). The South East had the second highest proportions, 15.1% for inward FDI and 20.7% for outward FDI. This implies responses for all the other regions combined accounted for 15.3% and 27.5% of inward and outward FDI positions, respectively.

FDI positions by industry

While the results only cover the indicative weighted distribution of FDI positions and earnings for respondents, it does show some differences to our experimental statistics from data-linking. Part of this will reflect the companies that responded to the new FDI Survey questions, compared with those that could not or did not answer the question. For example, the proportion of 2020 FDI positions allocated to London (69.5% for inward and 72.8% for outward) were higher for both directions compared with our data-linked results from 2020 (of 43.1% and 42.8% respectively).

There were also differences in response types between industries, as well as inward compared with outward illustrative results (Figure 2). For inward FDI, companies that could not answer the question were most prominent in the mining and quarrying, retail, wholesale and transportation, and manufacturing industries, making up 54.7%, 54.1% and 38.5% of each industry's total FDI position respectively. In all three cases, companies that could not, or did not, answer accounted for over half of that industry group's total FDI position.

The biggest difference in the illustrative results between outward and inward FDI in 2021 was in the manufacturing, and information and communication industries. For manufacturing, companies answering the question with a single location accounted for the highest proportion (72.4%) of the total outward FDI position among those industries. However, for information and communication industries, companies accounting for 70.7% of those industries' total outward FDI position could not answer. For outward FDI, respondents not answering, or unable to answer, the question accounted for more than half of those industries' total position value in the mining and quarrying, information and communication, and retail, wholesale and accommodation industries.

These differences suggest subnational FDI results using survey responses are likely to underestimate the subnational values for any region where mining and quarrying, information and communication, or retail, wholesale and transportation industries accounted for a greater proportion of that area's economic activity. For example, based on our data-linked results for 2020, this suggests that North East Scotland would be particularly affected for both inward and outward FDI from the mining and quarrying links. Inward FDI in Lincolnshire and outward in Bedfordshire and Hertfordshire would likely be affected from retail, wholesale and transportation responses.

Back to table of contents

5. Comparing approaches for estimating inward subnational foreign direct investment (FDI)

We can compare 2020 survey responses with those from data-linking for the same companies. Our data-linked results combine FDI microdata with information from the Inter-Departmental Business Register (IDBR), Business Register and Employment Survey (BRES) and additional insights from a commercial data source. We will refer to this second approach as "data-linking".

Our data-linked results are based on company information from all components of FDI positions and earnings, whereas our survey questions ask for unquoted equity and profits. While it is difficult to compare these values directly, we can qualitatively compare results for individual companies. We compared inward subnational FDI survey responses for each business with the distribution of inward FDI positions, that we arrived at through data-linking (Table 3).

For 731 of the comparable businesses (62.6%), the inward FDI measures were apportioned to the same regions using both approaches. For 78.6% of the businesses, both approaches apportioned the largest proportion of the inward FDI position to the same International Territorial level 1 (ITL1) region. In 85.7% of cases, both approaches apportioned some of the business's inward FDI position to at least one of the same ITL1 regions.

Where the results for a business did not match exactly, this was often because the surveyed businesses apportioned their FDI across fewer regions than data-linking (Figure 3). This can be seen by the index values for results across two or more ITL1 regions. They were all above 1 for data-linked results, compared with survey responses (that equal 1). For example, data-linking identified nearly three times as many companies with a presence in two ITL1 regions, and more than three times as many with a presence in three regions.

Although the highest business counts for both data-linking and survey responses were for companies with a single ITL1 location, there were more instances where companies allocated all of their FDI to a single region on the survey when compared with our data-linked results.

Back to table of contents

6. Future developments

The analysis in this report suggests there are limits to what companies can tell us about their foreign direct investment (FDI) by UK subnational region, or on the type of FDI. The low response rates to the FDI by type questions indicate that alternative measures to estimate these values are likely to be more appropriate for these statistics. However, we will investigate how the subnational survey responses complement our data-linking methods and provide updates on our plans for these data in future reports.

Back to table of contents

8. Cite this article

Office for National Statistics (ONS), released 24 March 2023, ONS website, article, Foreign direct investment, overview of new survey questions, UK: March 2023

Back to table of contents

Contact details for this Article

Andrew Jowett
Telephone: +44 1633 455357