It has been common to hear claims that the Irish Protocol gives businesses the best of both worlds. Under the Protocol, firms retain their former access to the EU single market without customs controls or regulatory barriers (for goods but not for services). It is also claimed that they enjoy unfettered access to the GB market, but this applies only to their exports. Imports of goods from GB have to pass through customs declarations and a range of customs checks both electronic and physical. Since imports of goods from GB are much larger (at 23% of GDP) than NI’s exports of goods to the EU (at 11% of GDP) the balance of advantage is against NI. Some firms will also have to pay tariffs on goods entering NI from GB where goods contain significant foreign content. In addition, firms in Northern Ireland are subject to EU commercial regulations. Since an aim of Brexit is to make regulations lighter and less intrusive it is possible that the continued application of EU regulations to NI will prove a constraint on company competitiveness in future.
The most fraught objections to the Protocol are political rather than economic but economic considerations will undoubtedly play a role. Nationalists in Northern Ireland support the Protocol primarily because it keeps NI close to the Republic of Ireland, and in setting up customs barriers between NI and GB it makes the Province a semi-detached part of the UK with future Irish unification more likely. Unionists oppose the Protocol for parallel reasons. Many businesses have chafed at the problems, delays and additional costs imposed by the new customs regime but on the whole prefer that the open land border remains in place. This is especially true of the dairy industry which – unlike most other sectors – operates on an all-island basis.
The UK Government is threatening to invoke article 16 of the Protocol to reform the existing agreement if it is unable to achieve a similar outcome through the ongoing negotiations with the EU. The criteria for invoking article 16 are to prevent “serious economic, societal or environmental difficulties that are liable to persist, or to diversion of trade”. Since there has already been clear diversion of trade (note that article 16 does not stipulate that this should be considered serious) and relatively serious societal difficulties involving rioting and the burning of buses, the UK Government says that grounds for invocation already exist. The importance of the wider economic impact is that ‘serious economic difficulties’ would form a further justification for invoking article 16. More generally, if the impact of the Protocol is shown to be negative it will affect public opinion and could lead to the Protocol being voted down in four years’ time under article 18 which deals with democratic consent.
Until recently it has not been possible to assess whether the Protocol is aiding or damaging the NI economy. The short time period since it was introduced in January 2021 and the fact that so-called ‘grace periods’ have limited the scope of application, mean that limited evidence is available. As quarterly GDP data have emerged over the current year, the Financial Times has claimed that output (gross value added) had declined less in NI than in other UK regions between 2019q4 and the 2021Q3. This is based on experimental constructed figures from ONS with wide margins of error, but even so the better performance of NI related to 2020 when NI had a lower incidence of covid than GB. During the period of the Protocol from January 2021 the NI economy is estimated to have expanded less rapidly than all English regions except the North East, although slightly faster than Scotland or Wales. Owen Polley says that on Twitter that the FT presented this data with the caption ‘Northern Ireland prospers from the Protocol’. Actual data (rather than the ONS’s constructed figures) for the first half of 2021, using NISRA’s composite economic index (NICEI), show Northern Ireland lagging the rest of the UK. In the latest data, for the second quarter of 2021, Northern Ireland is the slowest growing of all UK regions. Much of this is likely to reflect the impact of, and the recovery from, Covid and is unlikely to tell us much about the impact of the Protocol. What is absolutely clear however is that the data, whether actual or constructed, do not support a view that the Northern Ireland economy is benefiting from the Protocol. Not that this will stop nationalists in NI from claiming the opposite. SDLP MLA (and former Cabinet Office civil servant) Matthew O’Toole claimed (Nov 29th) that ‘one thing is clear and irrefutable. NI is outperforming every other region’. He stated that unionist and UK Government claims that the Protocol is damaging the NI economy are ‘a truly extraordinary distortion’.
What can be said about the likely long-term impact? A newly published study from academics at Strathclyde and Edinburgh universities gives the best insights yet. The study was originally completed in September 2020 prior to the Protocol coming into force but was subsequently revised and published in November 2021. It predicts that the Protocol would lead to a loss of 2.6%-3% of GDP in the long-run if fully implemented. It concludes that ‘there appears to be no evidence that the Protocol allows NI to enjoy the best of both worlds.’ This result emerges from a detailed model of trade and macro-economic outcomes. Although a standard type of approach used by the Treasury, Department of International Trade and widely across the world, it is nonetheless rather abstract and theoretical and necessarily uses key parameters which are not based on experience in Northern Ireland. Since we have misgivings about the Treasury or DIT models which, it should be pointed out, do not rely on statistical relationships estimated from historical data, we hold a similar view for this model. (We should note that both the current and previous Chancellors are preventing the Treasury from issuing further model-based assessments of Brexit). The precise numbers for NI should thus be taken with a pinch of salt, but the approach is useful since it considers a wide range of influences which the Protocol can have on the NI economy. These influences range much more widely than the relatively simple and short-term considerations cited by some businesses in media assessments of the Protocol.
The paper has some odd features with several references to the sea between GB and NI as the ‘North Sea’ (the confusion may be that the sea between Scotland and NI is the North Channel) and is somewhat naïve in viewing the main purposes of the Protocol as ‘safeguarding the Good Friday Agreement’ and ‘preserving economic stability on both islands’, but is otherwise thorough in the scope of its economic analysis. It focuses on trade and not on wider issues including labour migration, productivity and foreign direct investment. The omission of productivity effects is significant because the 2016 Treasury study of the long-term impact of Brexit estimated around half of the projected negative impact came from productivity losses generated as a knock-on impact of losses in trade with the EU. However, our view is that such knock-on effects are in practice minimal for the UK and hence the omission of productivity is sensible here (although the small size and peripheral position of NI means some effects might be possible). The model also omits feedback effects from GB – but these are likely to be relatively small.
The study estimates the likely impact of the non-tariff barriers imposed by the Protocol (including customs administration and product checks) and tariffs on goods crossing from GB to NI due to the EU’s rules of origin (i.e. tariffs on the proportion of goods from GB which are imported into GB from outside the EU). It also estimates the likely scale of substitution by NI firms of goods from GB for those from the rest of the world including the EU.
The model considers 18 separate sectors and uses the NI Input-Output table which records sales and purchases of each sector (and is itself partly estimated). Each sector is treated as if it were a single firm which attempts to maximise its profits by purchasing inputs from the cheapest source and by adjusting its labour and capital. In addition, a representative household in NI adjust its consumer purchases to reflect changes in prices. The so-called elasticities of substitution for firms and consumers between different sources of supply are unknown for NI and hence equivalents from UK and international studies are used instead. The real wage also falls if unemployment rises. None of this is fully realistic but it captures the main pressures for change including the direction of change and an approximate estimate of the size of changes in output, trade and consumption for each sector.
These pressures and likely directions of change are worth emphasizing. Higher customs admin and checking costs (non-tariff barriers) of 8-9% for goods depend on how much is bought in from GB and hence translate into effective costs of 5-9% for goods sectors and 0.5-1% for services.. Firms in NI pass on the higher costs in raised prices which reduces their competitiveness in international and domestic markets and causes exports and output to fall. Lower output leads to a decline in employment and to a reduction in wage rates as employees’ bargaining power is reduced. Lower incomes in turn reduce consumer purchases in NI. Reduced demand for exports and domestic consumption also causes investment to fall. Lower wages partially offset some of the price increases due to additional customs costs but prices remain elevated, reducing the purchasing power of NI consumers.
The authors estimate quite large falls in sales to GB (6-8%) and the rest of the world (9-10%) due to the loss of competitiveness. Because imports from GB become more expensive than those from elsewhere, firms in NI reduce their purchases from GB (by 8%) and increase their purchases from the rest of the world including the EU. The study thus predicts a substantial diversion of trade and provides a justification for invoking article 16 of the Protocol (to enable the UK to reduce customs costs on goods coming into NI from GB). The overall impact is a reduction of GDP, employment and wages in Northern Ireland which the study estimates as close to 3% for GDP, 1-2% for employment and around 4% for real wages. This level of precision should be treated with caution, but the direction of change is undoubtedly correct. The study does itself investigate the impact of using different parameters in the model, and concludes that even a wide range of alternatives still generate a negative impact on the NI economy.
The study recognizes that if UK commercial regulations diverge from those of the EU then the cost of customs arrangements will rise leading to even larger losses of GDP. It does not however calculate a direct cost of more restrictive EU regulations. Nor does it consider a loss of consumer welfare if some brands and commodities become unavailable in NI as firms in GB find it uneconomic to supply the small NI market while customs costs remain high.
The magnitude of the projected impact on the NI economy is smaller than for equivalent studies of the impact of the TCA on the UK economy. For instance, the University of Sussex’s UK Trade Observatory estimates that the TCA will reduce UK GDP by 4.4% or nearly twice what is projected here for Northern Ireland. We are unsure why the Sussex figure is so high but assume it is because they have included knock-on reductions in productivity associated with lower exports to the EU. As outlined above, we believe that such productivity effects are unlikely to occur especially at the scale of a large developed economy like the UK.
If the outcome is unfavourable for firms in NI, why do some of them prefer to maintain the Protocol unchanged? This study calculates that agriculture and food and drink sectors will be most damaged by the Protocol, along with other export-oriented sectors in manufacturing. Despite this, it is the agriculture and in particular the dairy sector which most resists reforming the Protocol. The highly integrated nature of dairy production in Ireland leads to a fear that customs administration would be applied to milk transport across the border. They also fear that the EU would resist attempts to find alternative means of protecting their single market which avoided new customs arrangements. More generally, while firms may feel that they can pass on higher costs into higher prices, they may be underestimating the impact of the consequent loss of competitiveness.
In conclusion, there are good reasons to believe that the Protocol will lead to losses for the NI economy and these are likely to grow over time. The costs of customs administration will cause prices to rise and to a loss of competitiveness for NI firms. It is also possible that the negative impact could be larger than estimated in this study. NI is currently integrated into efficient large scale UK distribution systems at present including for food and groceries. Taking NI out of these is sure to mean higher prices as local or Irish systems are much smaller and less efficient. Estimates are that grocery costs in the Republic are 15-20% higher than in GB or NI. Bringing goods from EU via Ireland must also be much more expensive than from GB (just the transport cost alone will be higher).
In our view there is strong case for an agreed change to the Protocol along the lines of the UK Command paper of July 2021. This would remove all customs administration and checks for goods from GB destined for the NI market. Goods destined for the Republic via NI would still undergo customs procedures. This reform would remove most of the additional costs predicted in this study. The move would also remove most of the political objections of NI Unionists to a Protocol which places an internal customs border between NI and the rest of the UK.
