Large negative trade impact from The Windsor Framework

”There are three kinds of lies, lies, damned lies and statistics”, a phrase popularised by Mark Twain and often attributed (probably wrongly) to the 19^th Century Prime Minister Disraeli. What we do know is that government policy should be evidenced based. Carefully used statistics can be part of such an evidence base.

Few areas of UK policy are in greater need of a careful sifting of the data and evidence than consideration of the business impact of Northern Ireland’s post-Brexit trading arrangements: the Windsor Framework.

Since the start of the decade and beginning during the Pandemic the UK’s official statistical body (ONS) have produced a regular publication “Business Insights and Impact on the UK economy” reviewing how UK businesses of various types and sizes and regional locations have been coping with these uncertain times https://www.ons.gov.uk/businessindustryandtrade/business/businessservices/bulletins/businessinsightsandimpactontheukeconomy/previousreleases. Those ONS publications are in turn based on a twice monthly voluntary survey, the Business Insights and Conditions Survey (BICS), involving tens of thousands of businesses across the UK [Note 1, for technical details of that survey].

One of the many questions asked is to ask businesses operating in GB whether they are selling goods to customers in Northern Ireland (remember the Protocol/Windsor Framework’s “Irish Sea border” impacts much more on movement of goods from GB to Northern Ireland than the provision of services). The following Table summarises the results: 

Table 1: Change in the % of GB businesses, currently trading, selling goods to NI, 2020 compared to 2024-25

Note:*: For the 12 months before 25 January 2021-7 February 2021, i.e. “Wave 21” of BICS data. Because the question was asked to a survey of UK-wide businesses currently trading it is probable the figures are “inflated” by inclusion of some NI based businesses (though in how many cases would an N based business be selling a good from GB to NI?) but the figures as given are a close approximation to the “% businesses in GB”.

**: For the 12 months before 22 April 2025-4 May 2025, i.e. “Wave 131” of BICS data. Because the question was asked to a survey of UK-wide businesses currently trading it is probable the figures are “inflated” by inclusion of some NI based businesses (though in how many cases would an N based business be selling a good from GB to NI?) but the figures as given are a close approximation to the “% businesses in GB”.

Source: ONS 22 May 2025, Business Insights and Impact on the UK Economy 22 May 2025 and the Dataset 22 April 2025-4 May 2025 BICS Wave 131.

The proportion of GB based businesses selling goods to Northern Ireland dropped substantially between 2020 (the last year before the Protocol came into effect in January 2021) and the most recent data which relates to the 12 months up to end April 2025. For the manufacturing and retail sectors taken separately and for all businesses overall the percentage selling to Northern Ireland dropped by about one-third during the four-year period coinciding with the operation of the Protocol/Windsor Framework.

Similarly, across all size bands of firms in GB there were substantial declines in the percentages selling to Northern Ireland though perhaps most marked for the small and medium sized enterprises.

Whereas the previous data is indicative of the extent to which firms in GB stopped selling to Northern Ireland completely, the Survey also identified cases where businesses continued to sell something but changed the amount sold.

For the retail sector 14.2% of businesses in GB currently trading recorded declines in the sales to Northern Ireland in the 12 months to end April 2025 and 1.5 % recorded increases. A further 11.4% stopped selling to Northern Ireland completely. So, clearly weighted towards a reduced level of sales. For all businesses 15.1% recorded a decline, 6.2% an increase and 8.3% stopped completely. Amongst the size bands, the small firms with 10-49 employees recorded a marked bias towards reduced sales to Northern Ireland: 22.2% declined, 8.3% increased and 2.2% stopped completely.

Another interesting BICS question was one asking firms to identify the challenges they faced when it came to selling to other parts of the UK, the following Table shows the percentages of firms across the UK (i.e. both Northern Ireland plus GB) identified the Protocol (later Windsor Framework) as such a challenge:

Table 2: %’s of firms in GB plus NI identifying the Protocol (Windsor Framework) as a challenge in terms of intra-UK trading in goods

Note: *: The 12 months before 22 April 2025- 4 May 2025, i.e. BICS Wave 131.

**: The 12 months before 8 August 2022-21 August 2022, i.e. BICS Wave 63. This was the first time this question was asked.

Source: As in Table 1.

Substantial proportions of businesses identified the Protocol as a challenge: almost a quarter of manufacturing and over one-fifth in retail. Amongst all businesses in 2024-25 the proportion was about one-nine. To put that into perspective, that proportion was higher than the percentage of businesses which cited rules and regulations or supply chain disruption or transport capacity.

A limitation in the data is that it does not allow us to determine how many of these businesses were in Northern Ireland and how many in GB. However, given that the UK-wide proportions answering that the Protocol was a problem were so high this suggests that many of the respondents must have been in GB.

Summary

• The ONS survey results are both interesting and worrying.
• Since the start of 2021, that is since the Protocol came into force, there has been a substantial drop in the proportion of GB based businesses selling into Northern Ireland. Even when firms continue to sell, other BICS data indicated a clear tendency for sales to Northern Ireland to contract more often than they increase.
• These results produced by the UK Office for National Statistics are  consistent with previous data provided by the Northern Ireland Statistics and Research Agency/NISRA in December 2024. The latter showed that during 2020-23 the value of goods purchased by Northern Ireland from GB had grown by 24% whereas the value of goods purchased from the Republic of Ireland had grown by 51% (NISRA, 11 December 2024, Northern Ireland Economic Trade Statistics https://www.nisra.gov.uk/statistics/business/ni-economic-trade )
• Especially given the about to be completed Review of the Windsor Framework by Lord Murphy, policy makers should note the very problematic implications of the current arrangements”.”
• Note 1: The BICS is conducted twice a month, every sixth Survey, or Wave, included the questions about GB-NI trading. The responses (which were voluntary, i.e. not a mandatory business survey) were substantial in number: from about 39,000 business locations across the UK (of which about 1,200 in Northern Ireland). “Business locations” because there were some multi-unit businesses (so the number of firms responding was a smaller number). In the 22 April 2025-4 May 2025 BICS the number of respondents in Northern Ireland was 704 size band 0-99, 174 for size band 100-249 and 297 size band 250+.
• The BICS has a widespread coverage of the private sector but does not include agriculture, oil and gas extraction, energy production and public administration and defence. ONS classifies these data as “official statistics in development” implying not necessarily reaching the quality/reliability standing of “official statistics”. Nevertheless, the sample sizes at both the UK and regional levels are substantial.

Dr Esmond Birnie   Senior Economist Ulster University